Knee Osteoarthritis and Aquatic Cycling

Knee osteoarthritis and aquatic cycling

Citation for published version (APA):

Rewald, S. (2018). Knee osteoarthritis and aquatic cycling: Development and evaluation of an exercise programme. Datawyse / Universitaire Pers Maastricht. https://doi.org/10.26481/dis.20180711sr

Document status and date: Published: 01/01/2018

DOI: 10.26481/dis.20180711sr

Document Version: Publisher's PDF, also known as Version of record

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement: www.umlib.nl/taverne-license

Take down policy If you believe that this document breaches copyright please contact us at: [email protected] providing details and we will investigate your claim.

Download date: 23 Sep. 2021

Knee Osteoarthritis and Aquatic Cycling

Development and evaluation of an exercise programme

Stefanie Rewald The research presented in this dissertation was conducted at the Care and Public Health Research Institute (CAPHRI), Department of Epidemiology, Maastricht University. CAPHRI participatesintheNetherlandsSchoolofPrimaryCareResearch(CaRe). The research described in this dissertation was funded by the Netherlands Organisation for ScientificResearch(NWO),grantnumber022.003.036. Allrightsarereserved.Nopartofthisbookmaybereproducedortransmittedinanyformorby anymeans,withoutthewrittenpermissionfromtheauthoror,whereappropriate,thepublisher ofthearticle. ©StefanieRewald,Maastricht2017 dŚĞƉŚŽƚŽŐƌĂƉŚŽŶƚŚĞĐŚĂƉƚĞƌƉĂŐĞƐƐŚŽǁƐƚŚĞĂƌƚŝŶƐƚĂůůĂƚŝŽŶ͞ĞŽŶƚƐŶĂƉƉŝŶŐ͟ĨƌŽŵƚŚĞĞůŐŝĂŶ ĂƌƚŝƐƚƌŝŬEĂŐĞůƐ͘dŚĞĂƌƚŝŶƐƚĂůůĂƚŝŽŶǁĂƐĚŝƐƉůĂǇĞĚĂƚƚŚĞďĞĂĐŚŝŶKŽƐƚĞŶĚĞŝŶƚŚĞƐƵŵŵĞƌŽĨ ϮϬϭϯ͘/ƚƐŚŽǁƐĂůĂƌŐĞŐƌŽƵƉŽĨůŝĨĞͲƐŝǌĞƐĐƵůƉƚƵƌĞƐŽĨƉƌŽĨĞƐƐŝŽŶĂůĐǇĐůŝƐƚƐƚŚĂƚǁĞƌĞĐƌĞĂƚĞĚŝŶƚŚĞ ŚŽŶŽƵƌŽĨƚŚĞϭϬϬƚŚĂŶŶŝǀĞƌƐĂƌǇŽĨƚŚĞdŽƵƌŽĨ&ůĂŶĚĞƌƐŝŶϮϬϭϯ͘&ŽƌŚŝƐĂƌƚǁŽƌŬ͕ƚŚĞĂƌƚŝƐƚǁĂƐ ŝŶƐƉŝƌĞĚďǇƚŚĞƐŵĂůů͕ďƌŝŐŚƚůǇĐŽůŽƵƌĞĚƉůĂƐƚŝĐƚŽǇĐǇĐůŝƐƚƐƚŚĂƚǁĞƌĞĂĚĚĞĚĂƐĂŐŝŵŵŝĐŬƚŽǁĂƐŚŝŶŐ ƉŽǁĚĞƌŝŶƚŚĞ^ŝǆƚŝĞƐĂŶĚ^ĞǀĞŶƚŝĞƐ͘ Layout: TinyWouters Cover: AnnikaRewald Production: Datawyse:|UniversitairePersMaastricht

P UM UNIVERSITAIRE ISBN:978Ͳ94Ͳ6295Ͳ963Ͳ7 PERS MAASTRICHT KneeOsteoarthritisandAquaticCycling

ĞǀĞůŽƉŵĞŶƚĂŶĚĞǀĂůƵĂƚŝŽŶŽĨĂŶ ĞǆĞƌĐŝƐĞƉƌŽŐƌĂŵŵĞ

Dissertation toobtainthedegreeofDoctoratMaastrichtUniversity, ontheauthorityoftheRectorMagnificus, Prof.dr.RianneM.Letschertinaccordancewiththedecision oftheBoardofDeans,tobedefendedinpublic onWednesday11July2018,at14:00hours by StefanieRewald Supervisors Prof.dr.R.A.deBie Prof.dr.A.F.Lenssen CoͲSupervisors Dr.E.P.EMesters Dr.P.J.Emans AssessmentCommittee Prof.dr.R.J.E.M.Smeets(chair) Prof.dr.A.E.R.C.H.Boonen Prof.dr.J.Dekker(VUUniversityMedicalCenter(VUmc)Amsterdam) Dr.A.J.A.Köke Dr.P.J.vanderWees(RadboudUniversityMedicalCentre(Radboudumc) Nijmwegen) TABLEOFCONTENTS

Chapter1 Generalintroduction 7 Chapter2 Aquaticcircuittrainingincludingaquaticcyclinginpatients 21 withkneeosteoarthritis:afeasibilitystudy Chapter3 Effectofaquaticcyclingonpainandphysicalfunctioning 35 comparedwithusualcareinpatientswithkneeosteoarthritis: studyprotocolofarandomisedcontrolledtrial Chapter4 Aquaticcycling–whatdoweknow? 63 AscopingreviewonheadͲoutaquaticcycling Chapter5 Efficacyofaquaticcyclingonkneepainandphysical 107 functioninginpatientswithkneeosteoarthritis: arandomisedcontrolledtrial Chapter6 Generaldiscussion 133 Summary 155 Samenvatting 161 Zusammenfassung 167 Valorisationaddendum 173 Acknowledgements 183 Abouttheauthor 191 Listofpublications 195

CHAPTER1

Generalintroduction

8 Generalintroduction

This dissertation describes the development and evaluation of an aquatic cycling 1 programme for patients with knee osteoarthritis. The introduction summarises key facts about knee osteoarthritis, explains the development of aquatic cycling, and presentshypothesesofwhycyclingonastationarybikewhileimmersedinwatermight beavaluableformofaquaticexerciseforpatientswithkneeosteoarthritis.

KNEEOSTEOARTHRITIS

Osteoarthritis(OA)isachronicdiseasethatcanoccurinanyjointofthehumanbody. OA involves damaging and thinning of the cartilage, structural changes in the surroundingboneandosteophyteformation,ligamentouslaxity,lossofstrengthinthe periarticularmusclesandsynovialinflammation.1,2 The knees and hands are commonly affected by OA. However, OA is also known to affecthips,facetjointsandfeet.3Ͳ5Becauseoftheuniversalinvolvementofkneeand hipjointsduringtaskssuchaswalkingandstairclimbing,itisnotsurprisingthatknee andhipOAwasrankedatnumberelevenoutof291diseasescontributingtoglobal disabilityin2010.6Bothincreasingageandtheobesityepidemicareprimaryfactors contributing to the increasing prevalence of OA. By 2020, on a global scale OA is expectedtorankinthetopfiveofmostdisablingdiseases.7,8 ThelifetimeriskfordevelopingkneeOAisapproximately45%andseveralfactorsseem to cause knee OA.9 However, the development of knee OA appears to be largely influencedbytheinterplaybetweenmechanicalandsystemicriskfactors.Examplesof systematic risk factors are increasing age, obesity, female sex, physical (in)activity, ethnicity and genetics.1,3,8,10 Furthermore, individuals with a history of knee injury and/ormalalignmentofthekneeorlowerlegmuscleweakness,whichareallexamples ofmechanicalfactors,mighthaveanincreasedriskfordevelopingOA,1,3,8 The diagnosis and severity of knee OA is based on a clinical evaluation of the knee joint.11Atypicalclinicalevaluationconsistsofananamnesis(i.e.,exploringriskfactors and symptoms knee OA) and a physical examination assessing the existence of crepitus,impairedrangeofmotionandbonyenlargement.11Radiographicimagescan beusedtodetectchangesinthekneejoint.TypicalsignsofkneeOAincludenarrowing ofthejointspace,changesinsubchondralboneandformationofosteophytesatjoint margins.11TheKellgrenLawrencescore12isawidelyusedtooltoclassifyradiographic changesinkneeOAusingfivegradesrangingfromzero(nopresenceofOA)tofour (severeOA). Ingeneral,kneeOAdevelopsslowlyfromearlytolatestagescausingkneepainand disability.1Throughoutthetimecourseofthedisease,symptomsdevelopfromloadͲ dependentkneepain(e.g.afterexercise)tochronickneepainduringthedayoreven atnight.13Problemsperformingactivitiesofdailylivingmightresultinrestrictionsin

9 Chapter1 physical and social participation. These impairments can eventually affect mood and qualityofsleepandlife,whichmakeskneeOAawholepersondisease.1

(AQUATIC)EXERCISEFORPATIENTSWITHKNEE OSTEOARTHRITIS

Many of the pathogenic factors and symptoms associated with knee OA may be lessenedand/orimprovedwithlifestylechanges.Forexample,regularparticipationin exerciseisamainstayinvolvedinconservativetreatment.14Therenowissubstantial evidence suggesting routine exercise can alleviate symptoms of knee OA, and has a beneficial influence on cardiovascular risk factors and comorbidities.15Ͳ17 However, exercisetrainingisunderͲutilizedinindividualswithkneeOA.18Oneofthebarriersto exercise is that performing exercise with the osteoarthritic knee appears counterintuitivetopatientsbecausethismayprovokepainfulsymptoms.19 For individuals, as OA patients, who have limited painͲtolerance for weightͲbearing activities, waterͲbased exercise is frequently recommended. The buoyancy of water lessens the effects of body weight and results in joint decompression. Evidence suggeststhataquaticexercisepositivelyaffectskneepain,physicalfunctionandquality of life in patients with OA.20 In most instances, aquatic exercise for OA patients is performed in shallow water and does not include fullͲbody under water immersion. These headͲout aquatic exercise sessions include aerobic exercise, strengthening exercise or a combination of both. Equipment can be used to increase exercise intensityandvariation.21Ͳ23Inadditiontosmallsizeaquaticequipmentsuchasbench steppingͲplatforms, other landͲbased exercise equipment such as treadmills and stationary bikes are available in waterͲproof versions. More recently, aquatic cycling hasemergedasatrendyfitnessactivity.24,25

AQUATICCYCLING

AquaticcyclingisaheadͲoutaquaticexerciseactivityperformedonastationarybikein aswimmingpool.Participantsareimmersedatawaterlevelbetweennavelandchest height to ensure that the legs are immersed during the whole cycling movement withoutfloatingoffthesaddle.Inadditiontolegexercises,whichincludebothseated andoutͲofͲsaddlepositions,armandupperbodyexercisesagainstwaterresistanceare incorporated in a typical aquatic cycling session. These descriptions hold true for an uprightcyclingposition,whichismostcommon.However,therearealsoaquaticbikes thatareconstructedforsemiͲrecumbentcyclingandsometimeslegworkiscombined

10 Generalintroduction with arm ergometer exercises. This dissertation will focus on aquatic cycling in the 1 uprightpositionforpatientswithkneeOA.

Historicalperspective Scientific research on aquatic cycling started in the late sixties, peaked in the midͲ eightiesandfrom2007onwardsarenewedinteresthasemerged(Figure1.1). Number ofannual publications about aquatic cycling 6 5 4 3

publications 2 of 1 Nr. 0 196819731978198319881993 1998200320082013 Figure1.1 Numberofannualpublicationsbetween1968and2016 Inthelatesixties,astronautswerepreparingforthefirstspaceflightsandresearchers usedwaterasamediumtomimicweightlessnessandtostudyeffectsthatspacemight have on physiological functions of the human body.26Ͳ32 Another research interest duringthisperiodwastheexposureofhumanstodifferentwatertemperatures.33Ͳ42 For example, coldͲwater exposure was used to simulate a boating accident and to investigate the influence of physical movement on body temperature. In the aforementioned studiesaquatic cycling was used toisolate the effectsof immersion andtostudythermoregulatoryresponses,ratherthantostudytheeffectsofaquatic cyclingasaformofaquaticexercise. As interest in health and fitness increased in the 1970s and 1980s, aquatic aerobics increased in popularity. In the midͲ1980s several organisations such as the Aquatic ExerciseAssociation(AEA)andtheAquaticTherapyandRehabilitationInstitute(ATRI) cameintoexistencetoprovideprofessionalguidanceforthedevelopmentofaquatic exerciseandaquatictherapy.ThebenefitsoflowͲimpactaerobicexerciseprogrammes in water became more apparent and were promoted by athletes who used aquatic exerciseasaformofsupplementarytrainingorduringrecoveryfrominjuries.43Atthe sametime,aquaticbikeswerealsobeingrecognizedasatoolusefulforaquaticfitness. For example, cardiovascular and thermoregulatory responses to aquatic cycling in pregnantwomenandpatientswithcardiovasculardiseasesbecameresearchtopics.29, 44Ͳ47

11 Chapter1

In2007andonward,aquaticcyclinghasgrowninpopularityasahealthfitnessactivity. Inadditiontopublicinterest,therehasbecomearegainedresearchinterestinaquatic cyclingwiththefocusshiftingbeyonditsvalueasjustanexerciseactivity,buttoits applicationasatherapeutictool.50 Accompanying the increased interest in aquatic therapy, the design of aquatic bikes has improved. Up until the 1990s, most aquatic bikes were constructed from landͲ based cycle ergometers.51 Today, manufacturers design bikes specifically for the purposeofaquaticcycling.Themajorityofbikesarecreatedforuprightcyclingandthe bikeshavefins(variableinlength)orcups(variableinnumber)ontheflywheelthat increasepedalresistanceandhastobesetonlandpriortothesession. Aquaticbikes like the one shown in Photograph 1.1A are relatively low in cost (approximately 1.700€),easytostoreandneedlowtonomaintenance.Thesebikesenableswimming pools to provide aquatic cycling as a groupͲbased fitness activity. In the context of aquatic therapy and used in scientific research, aquatic bikes are equipped with magnetically or electromagnetically resistance mechanisms (Photograph 1.1BͲD), enablingadjustmentoftheresistanceduringpedallinginstead. Photograph1.1Aquaticbikeswith A B differentresistancemechanisms: A ProfixAquaBike™from Hydrorider®, B AquaCruiserII™from AquaKinetiqs, C Aquafit360™fromAquafit Technologie™, D RehaͲAquabikefrom SwissrehamedGmbH C D

12 Generalintroduction

AquaticcyclingforpatientswithkneeOA 1 Aquaticimmersionand,hence,aquaticcyclinginitiatesphysiologicalchangesthatare beneficialinthemanagementofkneeOA.Thesephysiologicaladjustmentstranslateto the following perceptions described by patients with knee OA exercising in warm water:feelingofweightlessness,relaxationandtheabilitytomovewithmorefreedom and ease coupled with less pain compared with landͲbased exercise.52Ͳ54 As such, immersion into water is suggested to essentially affect all homeostatic systems (Figure1.2). Figure1.2 Effectsofwaterimmersiononthehumanbody(usedwithpermissionofBruceE.Becker,MD) The biophysiological aspects of water immersion combined with potential effects of aquaticcyclingthataredirectly(i.e.themusculoskeletalsystem)orindirectlyrelated (i.e. cardiovascular system, central nervous system) to knee OA will be discussed in moredetail.

13 Chapter1

DƵƐĐƵůŽƐŬĞůĞƚĂůƐǇƐƚĞŵ ThepainͲdiminishingeffectofwaterimmersioniscausedbyavarietyoffactors.For example, during resting immersion approximately 50% of blood pumped out of the heartisredistributedtoskinandmuscles55supportinganincreaseinoxygendelivery. Thisissuggestedtohelpinthedissipationofalgogenicchemicalsandmetabolicwaste productscausingpain.56Furthermore,mildwaterpressureandthediureticeffectof water immersion can help to reduce oedema, with pain also possibly decreasing via reductioninjointswelling.57Inaddition,thepedalmovementassociatedwithaquatic cyclingalsoactivatesthecalfͲmusclepumpwhilesupportinglymphaticcirculation.58,59 Moreover, the increase in blood flow and jointdecompression may facilitatemuscle relaxation,wherebycontributingtopainͲrelief.56Withaquaticcyclingparticipantsare beingimmerseduptothexiphoidprocessandduetothebuoyancyofthewaterthe bodyweightisoffͲloadedbyapproximately50%.Asaresult,jointcompressionofthe lower back and the hips, knee and ankle joints is decreased. Another advantage of cyclingimmersedinwateristhereductionofstrainontheimmersedandmovingbody parts due to the viscous properties of water. Because the density of water is approximately800timeshigherthanair,itprovidesanequalamountofresistancein allplanes ofmotion. Thisresistancedampensthespeed of movement and prevents jerkymovementsanddeadspotsduringthepedallingmotion. In addition to the benefits of the aquatic environment, cycling is also a frequently recommendedexerciseactivityforpatientswithkneeOA.Cyclingisasimpleactivity and easy to learn with most adults in Europe learning to ride a bicycle in early childhood.RecentresearchshowedthatDutchpatientswithkneeOAcyclemoreoften thanpeoplewithoutkneeOA.60SeatedbicyclingisalowͲimpactexerciseandtheknee is kept stable while the knee is exercised through a large range of motion.61 Furthermore, the largest muscle groups of the lower extremities are used during cycling. LandͲbased stationary cycling interventions for OA patients have shown to improve typical symptoms of knee OA.62,63 These groupͲbased interventions led to significantreductionsinkneepain,improvedphysicalfunctioningandgait.62,63

ĂƌĚŝŽǀĂƐĐƵůĂƌƐǇƐƚĞŵ Aquaticimmersioninfluencesthedistributionofthebloodvolumeduetotheexternal pressurethatwaterexertsonthebody.Thismeansthatduringimmersionuptothe xiphoidprocessthecentralbloodvolume,whichisthebloodvolumeinthethorax,is increasedbyapproximately600ml.64Approximately,twoͲthirdsofthebloodisshifted tothelungcirculationandthehearttakesuponeͲthirdoftheredistributedblood.65 Consequently,agreatervolumeofbloodisenteringtheheartduringthediastoleand causesanincreasedstretchofthemyocardium.Asaresponsetotheincreasedvolume andstretch,thebloodvolumeejectedduringthefollowingsystole,whichiscalledthe stroke volume, is also increased. Interestingly, the stroke volume during resting

14 Generalintroduction immersioninwateriscomparabletothestrokevolumeduringaerobicexerciseonland 1 insedentaryindividuals.52,66Thus,justsittingontheaquaticbikeinthewatermight alreadybealightaerobictrainingassumingthatmostpatientswithkneeOAdonot meet the recommendations for physical activity.67,68 Another reason why aquatic exerciseisidealfordeconditionedpeopleisthatimmersiondecreasestheperipheral resistanceandsotheefforttocirculatethebloodisdecreasedandtheheartworks moreefficient.

ĞŶƚƌĂůŶĞƌǀŽƵƐƐǇƐƚĞŵ Waterimmersionandexerciseaffectthecentralnervoussystemandtherebymodulate pain sensation and mood. First, warm water temperatures and the hydrostatic pressureontheskinarethoughttostimulatethethermalandmechanoreceptorsand thisnonͲpainfulinputmightdiminishordecreasepainsensation.56,69Second,evidence suggeststhatexercisehasbeneficialeffectsonmentalhealthandimprovessymptoms of depression, anxiety and stress.70 Several physiological, immunological and psychologicalmechanismsseemtoberesponsibleforthesepositiveeffectsofexercise onmoodstates.70AlthoughmostresearchinvestigatedlandͲbasedaerobictraining,70 thereisalsosomeevidenceshowingthatbothpassiveimmersionandaquaticexercise mighthaveabeneficialinfluenceonmoodandrelaxation.53,54,71

AIMANDOUTLINEOFTHISDISSERTATION

Therapy involving cycling exercises in a heated pool may prove to be useful as a therapeutic tool for individuals with knee osteoarthritis. However, this therapeutic approach for patients with mildͲ toͲ moderate knee osteoarthritis has not been investigated. Therefore, the main project in this dissertation aimed to both develop and evaluate an aquatic cycling programme that is tailored to the needs of patients withkneeosteoarthritis.Afullaquaticcyclingtrainingprogrammewasdevelopedusing a twoͲstep model. First, the feasibility of aquatic cycling was evaluated (CChapter 2), Second, a 12Ͳweek aquatic cycling programme was developed (CChapter 3). We discovered during the conception and design phase of the exercise programme a dearthofsystematicreviewsofresearchfocusedonaquaticcycling.Therefore,aspart ofthisthesis,abroadscopingreviewwasconductedtoidentifytheavailableresearch on aquatic cycling (CChapter 4). In a randomised controlled trial design the 12Ͳweek aquatic cycling programme was compared with usual care for patients with mild to moderatekneeosteoarthritis(CChapter5).Weevaluatedthetherapeuticeffectsofour novel aquatic cycling programme on measurements of pain and physical function. Finally,thisdissertationconcludeswithageneraldiscussionofthemainfindingsand implicationsforfutureresearch(CChapter6).

15 Chapter1

REFERENCES

1. Roos EM, Arden NK. Strategies for the prevention of knee osteoarthritis. EĂƚƵƌĞ ZĞǀŝĞǁƐ͗ ZŚĞƵŵĂƚŽůŽŐǇ.2016;12:92Ͳ101. 2. GlynͲJones S, Palmer AJ, Agricola R, Price AJ, Vincent TL, Weinans H, et al. Osteoarthritis. >ĂŶĐĞƚ. 2015;386:376Ͳ87. 3. Johnson VL, Hunter DJ. The epidemiology of osteoarthritis. ĞƐƚ WƌĂĐƚŝĐĞ Θ ZĞƐĞĂƌĐŚ͗ ůŝŶŝĐĂů ZŚĞƵŵĂƚŽůŽŐǇ.2014;28:5Ͳ15. 4. Litwic A, Edwards MH, Dennison EM, Cooper C. Epidemiology and burden of osteoarthritis. ƌŝƚŝƐŚ DĞĚŝĐĂůƵůůĞƚŝŶ.2013;105:185Ͳ99. 5. GommerA,PoosM,HulshofT.Prevalentieenaantalnieuwegevallenvanartrose[Internet].Bilthoven: RIVM;2016[cited2017Dec28].Availablefrom:https://www.volksgezondheidenzorg.info/onderwerp/ artrose/cijfersͲcontext/huidigeͲsituatieͲnodeͲprevalentieͲenͲaantalͲnieuweͲgevallenͲvanͲartrose. 6. CrossM,SmithE,HoyD,NolteS,AckermanI,FransenM,etal.Theglobalburdenofhipandknee osteoarthritis: estimates from the global burden of disease 2010 study. ŶŶĂůƐ ŽĨ ƚŚĞ ZŚĞƵŵĂƚŝĐ ŝƐĞĂƐĞƐ.2014;73:1323Ͳ30. 7. Woolf AD, Pfleger B. Burden of major musculoskeletal conditions. ƵůůĞƚŝŶ ŽĨ ƚŚĞ tŽƌůĚ ,ĞĂůƚŚ KƌŐĂŶŝǌĂƚŝŽŶ.2003;81:646–56. 8. Murphy LB, Moss S, Do BT, Helmick CG, Schwartz TA, Barbour KE, et al. Annual Incidence of Knee Symptoms and Four Knee Osteoarthritis Outcomes in the Johnston County Osteoarthritis Project. ƌƚŚƌŝƚŝƐĂƌĞΘZĞƐĞĂƌĐŚ.2016;68:55Ͳ65. 9. MurphyL,SchwartzTA,HelmickCG,RennerJB,TudorG,KochG,etal.Lifetimeriskofsymptomatic kneeosteoarthritis.ƌƚŚƌŝƚŝƐĂŶĚZŚĞƵŵĂƚŝƐŵ.2008;59:1207Ͳ13. 10. AllenKD.Racialandethnicdisparitiesinosteoarthritisphenotypes.ƵƌƌĞŶƚKƉŝŶŝŽŶŝŶZŚĞƵŵĂƚŽůŽŐǇ. 2010;22:528Ͳ32. 11. ZhangW,DohertyM,PeatG,BiermaͲZeinstraMA,ArdenNK,BresnihanB,etal.EULARevidenceͲbased recommendations for the diagnosis of knee osteoarthritis. ŶŶĂůƐ ŽĨ ƚŚĞ ZŚĞƵŵĂƚŝĐ ŝƐĞĂƐĞƐ. 2010;69:483Ͳ9. 12. KellgrenJH,LawrenceJS.RadiologicalassessmentofosteoͲarthrosis.ŶŶĂůƐŽĨƚŚĞZŚĞƵŵĂƚŝĐŝƐĞĂƐĞƐ. 1957;16:494Ͳ502. 13. Neogi T. The epidemiology and impact of pain in osteoarthritis. KƐƚĞŽĂƌƚŚƌŝƚŝƐ ĂŶĚ ĂƌƚŝůĂŐĞ. 2013;21:1145Ͳ53. 14. NelsonAE,AllenKD,GolightlyYM,GoodeAP,JordanJM.Asystematicreviewofrecommendationsand guidelinesforthemanagementofosteoarthritis:Thechronicosteoarthritismanagementinitiativeof theU.S.boneandjointinitiative.^ĞŵŝŶĂƌƐŝŶƌƚŚƌŝƚŝƐĂŶĚZŚĞƵŵĂƚŝƐŵ.2014;43:701Ͳ12. 15. NueschE,DieppeP,ReichenbachS,WilliamsS,IffS,JuniP.Allcauseanddiseasespecificmortalityin patientswithkneeorhiposteoarthritis:populationbasedcohortstudy.ƌŝƚŝƐŚDĞĚŝĐĂů:ŽƵƌŶĂů;ůŝŶŝĐĂů ZĞƐĞĂƌĐŚĚͿ.2011;342:d1165. 16. HawkerGA,CroxfordR,BiermanAS,HarveyPJ,RaviB,StanaitisI,etal.AllͲcausemortalityandserious cardiovasculareventsinpeoplewithhipandkneeosteoarthritis:apopulationbasedcohortstudy.WůŽ^ KŶĞ.2014;9:e91286. 17. FransenM,McConnellS,HarmerAR,VanderEschM,SimicM,BennellKL.Exerciseforosteoarthritisof theknee:aCochranesystematicreview.ƌŝƚŝƐŚ:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐDĞĚŝĐŝŶĞ.2015;49:1554Ͳ7. 18. Hinman RS, Nicolson PJ, Dobson FL, Bennell KL. Use of nondrug, nonoperative interventions by communityͲdwellingpeoplewithhipandkneeosteoarthritis.ƌƚŚƌŝƚŝƐĂƌĞΘZĞƐĞĂƌĐŚ.2015;67:305Ͳ9. 19. DobsonF,BennellKL,FrenchSD,NicolsonPJ,KlaasmanRN,HoldenMA,etal.Barriersandfacilitators toexerciseparticipationinpeoplewithhipand/orkneeosteoarthritis:synthesisoftheliteratureusing behaviorchangetheory.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨWŚǇƐŝĐĂůDĞĚŝĐŝŶĞĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.2016;95:372Ͳ89. 20. Bartels EM, Juhl CB, Christensen R, Hagen KB, DanneskioldͲSamsoe B, Dagfinrud H, et al. Aquatic exercise for the treatment of knee and hip osteoarthritis. ŽĐŚƌĂŶĞ ĂƚĂďĂƐĞ ^ǇƐƚ ZĞǀ. 2016;3:CD005523.

16 Generalintroduction

21. LuM,SuY,ZhangY,ZhangZ,WangW,HeZ,etal.Effectivenessofaquaticexercisefortreatmentof 1 knee osteoarthritis: Systematic review and metaͲanalysis. ĞŝƚƐĐŚƌŝĨƚ Ĩƺƌ ZŚĞƵŵĂƚŽůŽŐŝĞ. 2015;10.1007/s00393Ͳ014Ͳ1559Ͳ9. 22. WallerB,OgonowskaͲSlodownikA,VitorM,LambeckJ,DalyD,KujalaUM,etal.Effectoftherapeutic aquatic exercise on symptoms and function associated with lower limb osteoarthritis: systematic reviewwithmetaͲanalysis.WŚǇƐŝĐĂůdŚĞƌĂƉǇ.2014;94:1383Ͳ95. 23. BarbosaTM,MarinhoDA,ReisVM,SilvaAJ,BragadaJ.PhysiologicalassessmentofheadͲoutaquatic exercisesinhealthysubjects:aqualitativereview.:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐ^ĐŝĞŶĐĞΘDĞĚŝĐŝŶĞ.2009;8:179Ͳ89. 24. Giacomini F, Ditroilo M, Lucertini F, De Vito G, Gatta G, Benelli P. The cardiovascular response to underwaterpedalingatdifferentintensities:acomparisonof4differentwaterstationarybikes.:ŽƵƌŶĂů ŽĨ^ƉŽƌƚƐDĞĚŝĐŝŶĞĂŶĚWŚǇƐŝĐĂů&ŝƚŶĞƐƐ.2009;49:432Ͳ9. 25. Yazigi F, Pinto S, Colado J, Escalante Y, ArmadaͲdaͲSilva PA, Brasil R, et al. The cadence and water temperatureeffectonphysiologicalresponsesduringwatercycling.ƵƌŽƉĞĂŶ:ŽƵƌŶĂůŽĨ^ƉŽƌƚ^ĐŝĞŶĐĞ. 2013;13:659Ͳ65. 26. ChristieJL,SheldahlLM,TristaniFE,WannLS,SagarKB,LevandoskiSG,etal.Cardiovascularregulation duringheadͲoutwaterimmersionexercise.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1990;69:657Ͳ64. 27. Connelly TP, Sheldahl LM, Tristani FE, Levandoski SG, Kalkhoff RK, Hoffman MD, et al. Effect of increased central blood volume with water immersion on plasma catecholamines during exercise. : ƉƉůWŚǇƐŝŽů;ϭϵϴϱͿ.1990;69:651Ͳ6. 28. Sheldahl LM, Tristani FE, Connelly TP, Levandoski SG, Skelton MM, Cowley AW, Jr. FluidͲregulating hormones during exercise when central blood volume is increased by water immersion. ŵĞƌŝĐĂŶ :ŽƵƌŶĂůŽĨWŚǇƐŝŽůŽŐǇ.1992;262:R779Ͳ85. 29. Hanna RD, Sheldahl LM, Tristani FE. Effect of enhanced preload with headͲout water immersion on exercise response in men with healed myocardial infarction. dŚĞ ŵĞƌŝĐĂŶ :ŽƵƌŶĂů ŽĨ ĂƌĚŝŽůŽŐǇ. 1993;71:1041Ͳ4. 30. SheldahlLM,TristaniFE,CliffordPS,HughesCV,SobocinskiKA,MorrisRD.EffectofheadͲoutwater immersion on cardiorespiratory response to dynamic exercise. :ŽƵƌŶĂů ŽĨ ƚŚĞ ŵĞƌŝĐĂŶ ŽůůĞŐĞ ŽĨ ĂƌĚŝŽůŽŐǇ.1987;10:1254Ͳ8. 31. Sheldahl LM, Tristani FE, Clifford PS, Kalbfleisch JH, Smits G, Hughes CV. Effect of headͲout water immersiononresponsetoexercisetraining.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1986;60:1878Ͳ81. 32. SheldahlLM,WannLS,CliffordPS,TristaniFE,WolfLG,KalbfleischJH.Effectofcentralhypervolemia oncardiacperformanceduringexercise.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1984;57:1662Ͳ7. 33. Toner MM, Sawka MN, Foley ME, Pandolf KB. Effects of body mass and morphology on thermal responsesinwater.:ƉƉůWŚǇƐŝŽů;ϭϵϴϱͿ.1986;60:521Ͳ5. 34. TonerMM,SawkaMN,HoldenWL,PandolfKB.Comparisonofthermalresponsesbetweenrestandleg exerciseinwater.:ƉƉůWŚǇƐŝŽů;ϭϵϴϱͿ.1985;59:248Ͳ53. 35. Toner MM, Sawka MN, Pandolf KB. Thermal responses during arm and leg and combined armͲleg exerciseinwater.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1984;56:1355Ͳ60. 36. YoungAJ,SawkaMN,LevineL,BurgoonPW,LatzkaWA,GonzalezRR,etal.Metabolicandthermal adaptationsfromendurancetraininginhotorcoldwater.:ƉƉůWŚǇƐŝŽů;ϭϵϴϱͿ.1995;78:793Ͳ801. 37. YoungAJ,SawkaMN,QuigleyMD,CadaretteBS,NeuferPD,DennisRC,etal.Roleofthermalfactorson aerobiccapacityimprovementswithendurancetraining.:ƉƉůWŚǇƐŝŽů;ϭϵϴϱͿ.1993;75:49Ͳ54. 38. Avellini BA, Shapiro Y, Fortney SM, Wenger CB, Pandolf KB. Effects on heat tolerance of physical traininginwaterandonland.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1982;53:1291Ͳ8. 39. McArdleWD,MagelJR,LesmesGR,PecharGS.Metabolicandcardiovascularadjustmenttoworkinair andwaterat18,25,and33degreesC.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1976;40:85Ͳ90. 40. McArdleWD,MagelJR,SpinaRJ,GergleyTJ,TonerMM.ThermaladjustmenttocoldͲwaterexposurein exercisingmenandwomen.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1984;56:1572Ͳ7. 41. McArdleWD,TonerMM,MagelJR,SpinaRJ,PandolfKB.Thermalresponsesofmenandwomenduring coldͲwater immersion: influence of exercise intensity. ƵƌŽƉĞĂŶ :ŽƵƌŶĂů ŽĨ ƉƉůŝĞĚ WŚǇƐŝŽůŽŐǇ ĂŶĚ KĐĐƵƉĂƚŝŽŶĂůWŚǇƐŝŽůŽŐǇ.1992;65:265Ͳ70. 42. Shapiro Y, Avellini BA, Toner MM, Pandolf KB. Modification of the Monark bicycle ergometer for underwaterexercise.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1981;50:679Ͳ83.

17 Chapter1

43. Becker BE, Cole AJ. Comprehensive Aquatic Therapy. 3rd ed. Pullmann, WA: Washington State UniversityPublishing;2010. 44. Katz VL, McMurray R, Berry MJ, Cefalo RC. Fetal and uterine responses to immersion and exercise. KďƐƚĞƚƌŝĐƐĂŶĚ'ǇŶĞĐŽůŽŐǇ.1988;72:225Ͳ30. 45. KatzVL,McMurrayR,BerryMJ,CefaloRC,BowmanC.Renalresponsestoimmersionandexercisein pregnancy.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨWĞƌŝŶĂƚŽůŽŐǇ.1990;7:118Ͳ21. 46. KatzVL,McMurrayR,GoodwinWE,CefaloRC.Nonweightbearingexerciseduringpregnancyonland andduringimmersion:acomparativestudy.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨWĞƌŝŶĂƚŽůŽŐǇ.1990;7:281Ͳ4. 47. McMurray RG, Fieselman CC, Avery KE, Sheps DS. Exercise hemodynamics in water and on land in patientswithcoronaryarterydisease.:ŽƵƌŶĂůŽĨĂƌĚŝŽƉƵůŵŽŶĂƌǇZĞŚĂďŝůŝƚĂƚŝŽŶ.1988;8:69Ͳ75. 48. BansiJ,BlochW,GamperU,KesselringJ.TraininginMS:influenceoftwodifferentendurancetraining protocols(aquaticversusoverland)oncytokineandneurotrophinconcentrationsduringthreeweek randomizedcontrolledtrial.DƵůƚŝƉůĞ^ĐůĞƌŽƐŝƐ.2012;19:613Ͳ21. 49. Bansi J, Bloch W, Gamper U, Riedel S, Kesselring J. Endurance training in MS: shortͲterm immune responses and their relation to cardiorespiratory fitness, healthͲrelated quality of life, and fatigue. :ŽƵƌŶĂůŽĨEĞƵƌŽůŽŐǇ.2013;260:2993Ͳ3001. 50. Garzon M, Gayda M, Garzon L, Juneau M, Nigam A, Leone M, et al. Biomechanical analysis to determinetheexternalpoweroutputonanimmersibleergocycle.ƵƌŽƉĞĂŶ:ŽƵƌŶĂůŽĨ^ƉŽƌƚ^ĐŝĞŶĐĞ. 2015;15:271Ͳ8. 51. Dressendorfer RH, Morlock JF, Baker DG, Hong SK. Effects of headͲout water immersion on cardiorespiratory responses to maximal cycling exercise. hŶĚĞƌƐĞĂ ŝŽŵĞĚŝĐĂů ZĞƐĞĂƌĐŚ. 1976;3: 177Ͳ87. 52. BeckerBE.Aquatictherapy:scientificfoundationsandclinicalrehabilitationapplications.WDΘZ͗ƚŚĞ ũŽƵƌŶĂůŽĨŝŶũƵƌǇ͕ĨƵŶĐƚŝŽŶ͕ĂŶĚƌĞŚĂďŝůŝƚĂƚŝŽŶ.2009;1:859Ͳ72. 53. RobinerWN.Psychologicalandphysicalreactionstowhirlpoolbaths.:ŽƵƌŶĂůŽĨĞŚĂǀŝŽƌĂůDĞĚŝĐŝŶĞ. 1990;13:157Ͳ73. 54. PeriniR,MilesiS,BiancardiL,PendergastDR,VeicsteinasA.Heartratevariabilityinexercisinghumans: effect of water immersion. ƵƌŽƉĞĂŶ :ŽƵƌŶĂů ŽĨ ƉƉůŝĞĚ WŚǇƐŝŽůŽŐǇ ĂŶĚ KĐĐƵƉĂƚŝŽŶĂů WŚǇƐŝŽůŽŐǇ. 1998;77:326Ͳ32. 55. Epstein M. Renal effects of headͲout water immersion in humans: a 15Ͳyear update. WŚǇƐŝŽůŽŐŝĐĂů ZĞǀŝĞǁƐ.1992;72:563Ͳ621. 56. HallJ,SwinkelsA,BriddonJ,McCabeCS.Doesaquaticexerciserelievepaininadultswithneurologicor musculoskeletal disease? A systematic review and metaͲanalysis of randomized controlled trials. ƌĐŚŝǀĞƐŽĨWŚǇƐŝĐĂůDĞĚŝĐŝŶĞĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.2008;89:873Ͳ83. 57. Gianesini S, Tessari M, Bacciglieri P, Malagoni AM, Menegatti E, Occhionorelli S, et al. A specifically designed aquatic exercise protocol to reduce chronic lower limb edema. WŚůĞďŽůŽŐǇ. 2016;10.1177/0268355516673539. 58. StickC,GrauH,WitzlebE.OntheedemaͲpreventingeffectofthecalfmusclepump.ƵƌŽƉĞĂŶ:ŽƵƌŶĂů ŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇĂŶĚKĐĐƵƉĂƚŝŽŶĂůWŚǇƐŝŽůŽŐǇ.1989;59:39Ͳ47. 59. Stick C, Heinemann W, Witzleb E. Slow volume changes in calf and thigh during cycle ergometer exercise.ƵƌŽƉĞĂŶ:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇĂŶĚKĐĐƵƉĂƚŝŽŶĂůWŚǇƐŝŽůŽŐǇ.1990;61:428Ͳ32. 60. HerbolsheimerF,SchaapLA,EdwardsMH,MaggiS,OteroA,TimmermansEJ,etal.PhysicalActivity PatternsAmongOlderAdultsWithandWithoutKneeOsteoarthritisinSixEuropeanCountries.ƌƚŚƌŝƚŝƐ ĂƌĞΘZĞƐĞĂƌĐŚ.2016;68:228Ͳ36. 61. LiebsTR,HerzbergW,RutherW,HaastersJ,RussliesM,HassenpflugJ.Ergometercyclingafterhipor knee replacement surgery: a randomized controlled trial. dŚĞ :ŽƵƌŶĂů ŽĨ ŽŶĞ ĂŶĚ :ŽŝŶƚ ^ƵƌŐĞƌǇ ŵĞƌŝĐĂŶǀŽůƵŵĞ.2010;92:814Ͳ22. 62. MangioneK,McCullyK,GloviakA,LefebvreI,HofmannM,CraikR.TheeffectsofhighͲintensityand lowͲintensitycycleergometryinolderadultswithkneeosteoarthritis.:ŽƵƌŶĂůƐŽĨ'ĞƌŽŶƚŽůŽŐǇ^ĞƌŝĞƐ͗ ŝŽůŽŐŝĐĂů^ĐŝĞŶĐĞƐĂŶĚDĞĚŝĐĂů^ĐŝĞŶĐĞƐ.1999;54:M:184Ͳ90. 63. SalacinskiA,KrohnK,LewisS,HollandM,IrelandK,MarchettiG.Theeffectsofgroupcyclingongait and painͲrelated disability in individuals with mildͲtoͲmoderate knee osteoarthritis: a randomized controlledtrial.:ŽƵƌŶĂůŽĨKƌƚŚŽƉĂĞĚŝĐĂŶĚ^ƉŽƌƚƐWŚǇƐŝĐĂůdŚĞƌĂƉǇ.2012;42:985Ͳ95.

18 Generalintroduction

64. Arborelius M, Jr., Ballidin UI, Lilja B, Lundgren CE. Hemodynamic changes in man during immersion 1 withtheheadabovewater.ĞƌŽƐƉĂĐĞDĞĚŝĐŝŶĞ.1972;43:592Ͳ8. 65. RischWD,KoubenecHJ,BeckmannU,LangeS,GauerOH.Theeffectofgradedimmersiononheart volume,centralvenouspressure,pulmonaryblooddistribution,andheartrateinman.WĨůƺŐĞƌƐƌĐŚŝǀ ƵƌŽƉĞĂŶ:ŽƵƌŶĂůŽĨWŚǇƐŝŽůŽŐǇ.1978;374:115Ͳ8. 66. VellaCA,RobergsRA.Areviewofthestrokevolumeresponsetouprightexerciseinhealthysubjects. ƌŝƚŝƐŚ:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐDĞĚŝĐŝŶĞ.2005;39:190Ͳ5. 67. deGrootIB,BussmannJB,StamHJ,VerhaarJA.ActualeverydayphysicalactivityinpatientswithendͲ stage hip or knee osteoarthritis compared with healthy controls. KƐƚĞŽĂƌƚŚƌŝƚŝƐ ĂŶĚ ĂƌƚŝůĂŐĞ. 2008;16:436Ͳ42. 68. FarrJN,GoingSB,LohmanTG,RankinL,KasleS,CornettM,etal.Physicalactivitylevelsinpatientswith earlykneeosteoarthritismeasuredbyaccelerometry.ƌƚŚƌŝƚŝƐĂŶĚZŚĞƵŵĂƚŝƐŵ.2008;59:1229Ͳ36. 69. BenderT,KaragulleZ,BalintGP,GutenbrunnerC,BalintPV,SukenikS.Hydrotherapy,balneotherapy, andspatreatmentinpainmanagement.ZŚĞƵŵĂƚŽůŽŐǇ/ŶƚĞƌŶĂƚŝŽŶĂů.2005;25:220Ͳ4. 70. MikkelsenK,StojanovskaL,PolenakovicM,BosevskiM,ApostolopoulosV.Exerciseandmentalhealth. DĂƚƵƌŝƚĂƐ.2017;106:48Ͳ56. 71. WatanabeE,TakeshimaN,OkadaA,InomataK.ComparisonofwaterͲandlandͲbasedexerciseinthe reductionofstateanxietyamongolderadults.WĞƌĐĞƉƚƵĂůĂŶĚDŽƚŽƌ^ŬŝůůƐ.2000;91:97Ͳ104. 72. BoidinM,LapierreG,PaquetteTanirL,NigamA,JuneauM,GuilbeaultV,etal.Effectofaquaticinterval trainingwithMediterraneandietcounselinginobesepatients:resultsofapreliminarystudy.ŶŶĂůƐŽĨ WŚǇƐŝĐĂůĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶDĞĚŝĐŝŶĞ.2015;58:269Ͳ75.

19 Chapter1

20 Chapter1

CHAPTER2

AquaticcircuittrainingincludingaquaͲcyclingin patientswithkneeosteoarthritis:afeasibilitystudy

StefanieRewald IlseMesters PieterJ.Emans J.J.ChrisArts A.F.TonLenssen RobA.deBie :ŽƵƌŶĂůŽĨZĞŚĂďŝůŝƚĂƚŝŽŶDĞĚŝĐŝŶĞ͘ϮϬϭϱ͖ϰϳ;ϰͿ͗ϯϳϲͲϯϴϭ

20 Chapter2

ABSTRACT

Background Aquaticcyclingiseasytolearn,acceptable,andsafeforpatientswithkneeosteoarthritis.Itcan thereforebeanidealcomponentofaquaticcircuittraining. Objective To investigate the feasibility of a small groupͲbased aquatic exercise programme including aquaticcycling. Design Afeasibilitystudyusingquantitative(preͲpost)andqualitative(crossͲsectional)assessments. Participants Avolunteercohortof10womenandmen,agerange46–77years,withkneeosteoarthritis. Methods Focus group interviews explored participants’ experience with the training. PreͲ and postͲ exercisekneepain,attendance,progressionintraining,andadverseeventswereregistered. Results Seventy per cent of patients attended all sessions. Focus groups revealed high levels of satisfaction with the selection of exercises, and participants valued the immediate pain relief experienced. Participants progressed well. However, aquatic cycling in an outͲofͲtheͲsaddle positionwastoodemandingformostparticipants. Conclusion An aquatic circuit training that includes aquatic cycling is feasible for patients with knee osteoarthritis.Participantsreportedpainreductionandwerepositiveaboutthediverseexercise programme.Aquaticcyclinginaseatedpositionisasafeandcontrolledtypeofmovement.

22 AquaticcircuittrainingincludingaquaͲcyclinginpatientswithkneeosteoarthritis

INTRODUCTION

WaterͲbasedexerciseisfrequentlyrecommendedforpatientswithkneeosteoarthritis 1 2 (OA),especiallywhenlandͲbasedtrainingisrestrictedbytheexperienceofpain. The hydrostatic pressure, temperature and buoyancy of water result in relief of body weight,musclerelaxation,decreasedjointcompression,andpainreduction.2Aquatic cyclingisanupcomingfitnessͲtrendinEuropeandcombinesstationarycyclingwiththe advantagesofexercisinginanaquaticenvironment.Previousresearchonrehabilitation after anterior cruciate ligament reconstruction and total knee surgery showed that aquaticcyclinginadditiontousualcareresultedingreaterimprovementinrangeof motion(ROM)andmorerapidreductioninkneejointswellingcomparedwithstandard care.3,4AsmallpreͲpostteststudywith19patientswithrheumaticdiseasesshoweda positiveinfluenceonwellͲbeing,strengthandmobilityaftertenweeksaquaticcycling.5 Basedontheresultsofpreviousstudiesitishypothesizedthatparticipantswithknee OAwouldalsoacceptthistypeofexercise.However,aquaticbikesareexpensive.Most therapypoolsdonothaveenoughspacetostoreseveralaquaticbikes.Wedeveloped anaquaticcircuittrainingforsmallgroupsofthreeparticipantsconsistingofaquatic exercisesandaquaticcycling.Thepresentstudyaimstoevaluatethefeasibilityofan eightͲweek aquatic circuit training in terms of adherence, possibility to progress in exercise level, occurrence of adverse events, operational aspects and patient acceptance.

METHODS

Design A convergent mixedͲmethods design was used to describe different aspects of feasibility.ThestudywasaproofofconceptofafullͲscaletrial,whichwasapprovedby thelocalethicscommittee(NL42617.068.12,NTR3766).

Participants Recruitment was carried out from February to April 2011 at Maastricht University MedicalCentre.Participantswereidentifiedbytheirorthopaedicsurgeonandsigneda writteninformedconsent.EligibleparticipantswerediagnosedwithkneeOAandhad an indication for conservative therapy including pharmacological and exercise treatment. Exclusion criteria were: a planned total knee arthroplasty, acute infection/inflammation, neuromuscular disease, and severe cardioͲrespiratory problems.

23 Chapter2

Measuresanddatacollectionprocedure Focusgroupinterviewswerechosentoexploreparticipants’experienceswithaquatic circuit training. The question guide (Table 2.1) was based on typical questions for formativeprogrammeevaluationandincludedquestionsaboutpositiveandnegative aspectsofthetrainingandaspectsthatshouldbechanged,droppedorfineͲtuned. Table2.1 Focusgroupquestions /ŶƚƌŽĚƵĐƚŽƌǇƋƵĞƐƚŝŽŶ 1.Whatdidyouexpectfromthetraining? dƌĂŶƐŝƚŝŽŶƋƵĞƐƚŝŽŶ 2.Whydidyouparticipate? <ĞǇƋƵĞƐƚŝŽŶƐ 3.IfyoucomparethistrainingwithlandͲbasedtraining,whatisthedifferencefromlandͲbasedtraining? 4.Whatdidyoulikebestabouttheprogramme? 5.Whatdidyouliketheleastabouttheprogramme? 6.Whatshouldbechanged? 7.Whatshouldbecontinuedjustasitisnow? 8.WhatshouldbecontinuedbutfineͲtuned? 9.Whatshouldbedropped? The focus groups were recorded and documented in a transcript. Only the physiotherapistandtheinterviewerhadaccesstotherecordsandfulltranscripts.The statementsofparticipantswereseparatedfrompersonaldatabythefollowingcode: gender(male/female),age,andfocusgroupnumber.Thefocusgroupstookplaceina meetingroomofMaastrichtUniversity.Bothfocusgroupslastedfor1.5hourswitha breakafter45minutes. QuantitativedatawascollectedonselfͲreportedpreͲandpostͲexercisekneepainona numeric pain rating scale (NRS).6,7 Furthermore, attendance, progression in training and adverse events was registered. In addition to patientͲrelevant information the physiotherapistregisteredexperienceswithexecutionofthetrainingandsupervision ofthegroupsandlogisticalaspectswiththeinstallationofthecircuittraining.

Intervention Participants trained in small groups of maximal three participants, once a week for 45minutesoveraneightͲweekperiod.Theaquaticcircuittrainingwasofferedfreeto thepatients;theyonlyhadtomeetthecostoftransport.Thetrainingwascarriedout in a heated therapy pool (32°C) of the physiotherapy department of Maastricht UniversityMedicalCentre+,supervisedbyaphysiotherapistwhowasalsointhepool herself.ThecircuittrainingcomprisesaquaticcyclingontheAquaCruiserII™,functional exercises, such as stepping and chair stands, ROM and strength exercises. Gait exerciseswereperformedasawarmͲupandcoolͲdown.Adetaileddescriptionofthe

24 AquaticcircuittrainingincludingaquaͲcyclinginpatientswithkneeosteoarthritis contentandgoalsoftheexercisesaccordingtotheframeworkfromvanderLeedenet al.isgiveninTable2.2.8 Table2.2 Descriptionoftheaquaticcircuitprogramme 2 ǆĞƌĐŝƐĞƐ͗WaterͲbasedcircuittrainingingroupsof3patients,onceaweek,45minpersession,supervised by1physiotherapistwhoisalsointhepool. WƌŝŵĂƌǇŝŶƚĞƌǀĞŶƚŝŽŶŐŽĂů͗ Ͳfeasibilityofaquaticcycling: x easeoflearningofcorrectaquaticcyclingtechniques x exploringpedallingfrequencyandresistanceduringaquaticcycling ^ĞĐŽŶĚĂƌǇŝŶƚĞƌǀĞŶƚŝŽŶŐŽĂůƐ;/&Ϳ͗ Ͳb770:gaitpatternfunction(exercise1) Ͳb710:mobilityofjointfunctions(exercise1,2,5,7) Ͳb715:stabilityofjointfunctions(exercise5,3,4,9) Ͳb740:muscleendurancefunctions(exercise2,3,5) Ͳb760:controlofvoluntarymovementfunctions(exercise1–9) Ͳb620:proprioceptivefunction(exercise3,4,5,6) >ĞǀĞůϭ 1. Gaittraining(2–4lapseachdirection): 1.1.Forwardര=രnormalgaitandsteplengthandcorrectfootmovement(heeltoforefoot) 1.2.Backwardര=രreversefootmovement(forefoottoheel) 1.3.Sideways=രsteptotheleftandbringrightfoottotheleftv.v. 2. Aquaticcycling:cyclingforward;moveextendedarmsalternatingforward/backward(10min) 3. Steppingonanunderwaterstep:stepdownandup 4. Squats 5. Flexionandextensionoftheunloadedknee(1Ͳlegstance,barwork:1Ͳhandhold) 6. Hipabduction(barwork:1Ͳhandhold) 7. Cyclelegs:Sittingonpoolnoodleorhangingonit(noodleintheback) 8. Stretchingm.quadriceps,m.iliopsoas,adductors,hamstrings,calf(10s,3–4reps) 9. Poolnoodlebalance:stepovertheimmersedpoolnoodle,movethepoolnoodlearoundthebody,sitting onthepoolnoodle >ĞǀĞůϮ 1. Gaittraining(2–4lapseachdirection): 1.1. Forwardര=രnormalgaitandsteplengthcorrectfootmovement(heeltoforefoot),1lapheelsup, 1lapheelstobuttock 1.2.Backwardര=രreversefootmovement(forefoottoheel) 1.3.Sidewaysര=രsteptotheleftandbringrightfoottotheleftv.v. 2. Aquaticcycling:cyclingforward/backward,changepedallingdirectionafter2min(totalduration10min) 3. Steppingonanunderwaterstep:stepupthenover,turnandrepeat 4. Squats:goupanddownin3steps(holdeachpositionfor5s) 5. Flexionandextensionoftheunloadedknee(1Ͳlegstance,freestanding) 6. Hipabduction/adduction(barwork:1Ͳhandhold) 7. Cyclelegs:sittingonpoolnoodleorhangingonit(noodleintheback) 8. Stretchingm.quadriceps,m.iliopsoas,adductors,hamstrings,calf(10s,3–4reps) 9. Poolnoodlebalance:stepovertheimmersedpoolnoodle,movethepoolnoodlearoundthebody,sitting onthepoolnoodle

25 Chapter2

Table2.2 (continued) >ĞǀĞůϯ 1. Gaittraining(2–4lapseachdirection): 1.1. Forwardര=രnormalgaitandsteplengthcorrectfootmovement(heeltoforefoot),1lapheels up,1lapheelstobuttock 1.2. Backwardര=രreversefootmovement(forefoottoheel) 1.3. Sideways=രരsteptotheleftandbringrightfoottotheleftv.v. 2. Aquaticcycling:cyclinginsitting*andhalfͲsitting**position,changepositionafter4min*,1min**, 2min*,1min**,2min* 3. Stepping:goupanddowninsideways 4. Squats,whilepushingandpullingapoolnoodlewitharmsonshoulderheight 5. Pushpoolnoodledown(underthefootofthefreeleg)whileextendingkneeandhip,letthepoolnoodle comeupgentlyuntilhipandkneereach90°flexion(barwork:1Ͳhandhold) 6. Lungesideways 7. Cyclelegs:sittingonpoolnoodleorhangingonit(noodleintheback) 8. Stretchingm.quadriceps,m.iliopsoas,adductors,hamstrings,calf(10s,3–4reps) 9. Poolnoodlebalance:stepovertheimmersedpoolnoodle,movethepoolnoodlearoundthebody,sitting onthepoolnoodle 'ĞŶĞƌĂůŝŶĨŽƌŵĂƚŝŽŶ ͲMainfocuson: • correctaquaticcyclingtechnique • kneeposition(neutralkneealignment,activecontrolofkneeposition)duringtheexercises Ͳ Duration:warmingͲup:5–10min,exercise1 exercise2–6:10minwithindividuallychosenpedallingfrequencyandresistance exercise3–6:1min(~15–20repetitions)eachwith1Ͳminrest coolingͲdown:5–10min,exercise1(inunevenweeks),7,8,9(evenweeks) Ͳ Restingtimebetweenexercises:1min(includingchangeofworkstation) Ͳ Intensity:exercise1,7–9:7–10onBorgscaleexercise2–6:11–13onBorgscale Ͳ Progressiontolevel2and3if: • kneepainduringandaftertheexercises<ര5onaNRSforpain • patientsperformedtheexerciseswithlighttomoderateexertion(Borgscale,noobservationforany visiblesignsofoverexertion) • patientsperformedtheexercisewithgoodqualityofperformance(basedonthevisualinspectionof thephysicaltherapist) Ͳ Trainingdevices: • Underwaterstep • Aquaticbike“AquaCruiserII™”(placedatthebottomofapool;participantsneededtobeimmersedto xiphoidprocesslevel) NRS:numericratingscale;v.v.:viceversa

Dataanalysis Focus group analysis was guided by the steps of framework analysis for descriptive accounts.9Therawdatawassummarisedperquestionfromtheinterviewguideand linkedwithillustrativequotesfromtherawdata.Toenhanceclarity,recurringthemes additional to the questions were developed in regular discussion between the physiotherapistandtheinterviewer. TheWilcoxonsignedͲranktestexamineddifferencesinpreͲandpostͲexercisescoresof knee pain. Results are presented as means and zͲscores (z). Standard deviation (SD)

26 AquaticcircuittrainingincludingaquaͲcyclinginpatientswithkneeosteoarthritis and 95% confidence interval (CI) were calculated. A significance level of a 2Ͳsided Ɖ൶<ര0.05wasset. 2 RESULTS

Between February and April 2011 the orthopaedic surgeon identified 24 eligible patients,ofwhom11werewillingtoparticipate.Oneparticipantfellonherkneeat home before signing the informed consent, which exacerbated her complaints and madeitimpossibleforhertoparticipate.Themainreasonsfornotparticipatingwere: not interested, no specific reason given (Ŷ൶=ര8); change of mind with respect to OA treatment,decidedtostopconservativetreatmentandtoundergototalkneesurgery (Ŷ൶=ര2);satisfiedwithphysiotherapytwiceaweek(Ŷ൶=ര2);andtoooccupied(Ŷ൶=ര1). Thefinalcohortcomprised10participants(7women)agedbetween46and77years (mean59.6(SD9.61)).Themajorityofthecohortwasemployed(Ŷ൶=ര6),3participants were retired, and 1 participant was seeking work. Radiological assessment of the tibiofemoraljointshowedKellgren/Lawrencescoresof2(Ŷ൶=ര2),3(Ŷ൶=ര4)and4(Ŷ൶=ര4).In addition,2patientshadgrade2patellarOA.ElapsedtimesincediagnosisofkneeOA rangedfrom1to180months(mean62months;SD69.06).Treatmenthistoryvaried from injections (Ŷ൶=ര5), physiotherapy (Ŷ൶=ര5), medical fitness training (Ŷ൶=ര4) and pain medication (Ŷ൶=ര3). Four people exercised regularly and continued their exercise routines(cyclingonahomeͲtrainer,exercisesforthelowerbackandmedicalfitness) duringthestudyperiod.Othershadexperienceswithmedicalfitness(Ŷ൶=ര2)andaquatic fitness(Ŷ൶=ര1),butwerenotparticipatingintheseactivitiesatthetimeofthestudy.

Feasibilityofthetrainingprotocol Theadherencerateforallsessionswas70%.Twopeoplemissedoneandfoursessions, respectively,becauseofholidays.ForworkͲrelatedreasonsoneparticipantcouldonly attendfivesessions. Alltheexercisesintheconditioningsectionconsistedofthreelevels.Exceptfortwo exercises, all patients could progress to level three. Pushing the pool noodle underwaterupanddownwasdifficultfortwoparticipantsbecausethebuoyancyof the pool noodle raised their supporting leg from the floor. Based on the physiotherapists’judgementofalowexercisecapacity(Ŷ൶=ര4)andtoofewsessionsor toolongbreaksbetweenthesessionstoprogress(Ŷ൶=ര3)allwomenremainedatlevel twowiththeaquaticcyclingexercise.Participantsperceivedresistancelevelstwoand threeascomfortableforpedalling,andscoredaquaticcyclingatanindividualchosen pedalling frequency as light (men) to moderate (women) on the Borg scale

27 Chapter2 independentfromtheresistance.CyclinginanoutͲofͲtheͲsaddleposition(level3)was evaluatedashardontheBorgscale. One adverse event occurred during the cooling down section of one session. One participant’sfootslippedduringgaittraining,scoringfiveonthepainscale.Kneepain decreasedquicklyaftertheendofthesession. Aswimmingpoolwithanadjustablefloorispreferable,becauseoftheweightofthe aquaticbike.Twopeopleareneededtoimmersethebikeifthefloorofthepoolisnot heightͲadjustableorhasanentrancefordisabledpeople.Transportationoftheaquatic bikeinsideandoutsidethepooliseasyasithastwowheelsatthefront.Thesaddle heightiseasilyadjustablebyprofessionalsorpatientsevenwhenthebikeisimmersed. A difference in body height of more than 20 cm can result in suboptimal levels of immersion among participants. In order to ensure that participant’s legs are underwaterduringcyclingparticipantsmustbeimmersedtoaminimumheightofthe navel and a maximum height of the xiphoid process. During shallowͲwater exercises immersiontochestheightispreferredsothatparticipantscanperformallexercises, suchassquattingwithoutimmersingtheirhead. ParticipantscouldsteponandofftheaquaͲbikewithoutassistanceandwereableto start exercising with minimal instruction. In addition, the combination of arm movementsandcyclingwaseasytolearnandparticipantsneedednoassistanceduring handͲfreecycling.Becausethecyclingpartlastedlongerthantheotherexercisesonly two participants changed workstations at the same time, which allowed time for tailoredfeedback.

Qualitativeevaluation Two focus group interviews were conducted, with four and three participants, respectively. Two people could not attend the interviews because they were on holiday;anotherparticipanthadotherappointments(Ŷ൶=ര1)onthedaythefocusgroups were held. Illustrative quotes from both interviews are shown in Table 2.3. Initially participantswereaskedabouttheirexpectations.Participantsansweredthattheywere motivatedtotryͲoutthetrainingbecauseexercisewasknownasameanstocontrol symptoms of OA. They knew that cycling and aquatic exercises are frequently recommendedtopeoplewithkneeOAbecauseoftheirlowimpactonthekneejoints. Participants described the training as a total body workout with the focus on joint mobility and a light to moderate exercise intensity. In comparison with landͲbased trainingthepainreliefeffectwashighlighted.Otherreportedeffectswerereleaseofa certain tension around the knee and higher selfͲefficacy about physical functioning. Participantsbelievedthatthewarmthandbuoyancyofthewaterhadagreatimpact onthereductionofpain,initiationandmaintenanceofmovements.Theyfeltthatthey could perform the cycling movements in the water more easily, although they experienced pedalling on the AquaCruiser II™ as more tiring. The participants were

28 AquaticcircuittrainingincludingaquaͲcyclinginpatientswithkneeosteoarthritis very positive about the selection of, and variety of, exercises. With regard to the aquaticcyclingsectionparticipantsstatedthattheywouldhavelikedtocycleforlonger than 10 min on the AquaCruiser II™. Nevertheless, cycling alone with no other exerciseswouldbetoomonotonous.Participantsevaluatedthetrainingassuitablefor 2 their needs and exercise capabilities. Even someone with fear of water could participate without problems. Participants only had few suggestions for further optimizationofthetraining.Participantswhowerenotemployedsuggestedscheduling the training in the morning. Furthermore, all participants would opt for a higher exercise frequency and a display on the aquatic bike with information about performancetomakethetrainingmoreefficient. Table2.3 Illustrativequotesfromthefocusgroupinterviews tŚĂƚĚŝĚǇŽƵĞǆƉĞĐƚĨƌŽŵƚŚĞƚƌĂŝŶŝŶŐ͍ • “IexpecteditcoulddonoharmandIthought:Let’ssee!”(Male57,group2) • “Ithought:itisformykneeandifitisnothelping,itcouldatleastdonoharm.”(Female58,group1) tŚǇĚŝĚǇŽƵƉĂƌƚŝĐŝƉĂƚĞ͍ • “…theysaidtomethatcyclingisakneeͲfriendlysport.”(Male54,group2) • “Weneedtostayactiveinordertocontrolpainandstiffness…”(Male57,group2) • “Itrytogotothegymeverymorning.Iammotivatedtodothisbecauseotherwisemy(health)condition willdeteriorateoverafewyears…”(Female58,group1) tŚĂƚŝƐƚŚĞĚŝĨĨĞƌĞŶĐĞǁŝƚŚůĂŶĚͲďĂƐĞĚƚƌĂŝŶŝŶŐ͍ • “Exercisinginwateriseasier.”(Female59,group2) • “WhiledoingmedicalfitnessIalwayskeepacertaintensioninmyknee.Thisisn’tchanging throughoutthewholeworkout.Thistrainingaimstostrengthenyourmuscles.Ididn’tfeelthis duringthewaterͲbasedtrainingthatwasarelief.”(Male54,group2) • “Traininginwateraimstoimproveandkeepyourflexibility.”(Male57,group2) ŽŵƉĂƌŝƐŽŶǁŝƚŚůĂŶĚͲďĂƐĞĚĞǆĞƌĐŝƐĞͬĐǇĐůŝŶŐ • “Pedallingisharderthanonausualbike,butatthesametimeitfeelssmoother.”(Female58,group1) • “Itiseasierthannormalcycling,itfeelseffortlessly.”(Female77,group1) • “Ioftentakethebike…butespeciallywhenIamcyclingwithatempoofduringuphillcyclingIfeela certainpainundermypatella.IneverhadthiswithaquaͲcyclingevenpedallingismoretiring.”(Male54, group2) tŚĂƚĚŝĚǇŽƵůŝŬĞďĞƐƚĂďŽƵƚƚŚĞƉƌŽŐƌĂŵŵĞ͍ WĂŝŶƌĞĚƵĐƚŝŽŶ • “Icouldmovewithalmostnopain.”(Female58,group1) • “AfterthetrainingIalwaysfeltbetter.TheimprovementdependedofcourseonhowIwasdoingduring theday,butitalwaysimproved.”(Male57,group2) • “Ihavenoticedthepositiveeffectsofthetraining.Sometimesyoumoveyourlegincertainwaysthat itcauseskneepain.Youcannotpreventthisduringeverydaylife.OnthosedaysIknewthatthepain wouldbereducedaftertheAquaCruiserII™training.”(Male54,group2) • “Itreducedthepain,butthedayafterthepainwasback.”(Female60,group1) • “…thepainreducingeffectlastedalittlewhileafterthetraining…”(Male57,group2) tĂƚĞƌƚĞŵƉĞƌĂƚƵƌĞ • “Thewarmwaterfeltverygood.”(Female58,group1) • “Itiswonderfulinthatwater.”(Female60,group1)

29 Chapter2

Table2.3 (continued) ^ĞůĞĐƚŝŽŶŽĨĞǆĞƌĐŝƐĞƐ • “Icamewithpleasuretoeverysession.Itwasfuntodotheexercises”(Male54,group2) • “Theexercisesincombinationwiththewarmthofthewatergaveagoodfeeling.”(Female58,group1) • “Ireallylikedthecombinationofexercises.”(Female59,group2) • “Iwouldliketodoaquaticcyclingatthebeginningandintheend,butalwaysincombinationwiththe otherexercises.” • “Itisatotalbodyworkout.”(Female77,group1) tŚĂƚĚŝĚǇŽƵůŝŬĞƚŚĞůĞĂƐƚĂďŽƵƚƚŚĞƉƌŽŐƌĂŵŵĞ͍ x “Everythingwasfine,itwasreallygreat!”(Female58,group1) tŚĂƚƐŚŽƵůĚďĞĐŚĂŶŐĞĚ͍ • “Nothing!Iamafraidofwaterandformeitreallywasanovercoming.InthebeginningIneededmuch assistance,butthisgotbetterwitheverysession.”(Female77,group1) DŽƚŝǀĂƚŝŽŶƚŽĐŽŶƚŝŶƵĞƚŚĞƚƌĂŝŶŝŶŐ • “IwouldliketoknowwhereIcancontinuethistypeoftraining.”(Female77,group1) • “Ilookedupinformationaboutaquaspinningclasses,butIamnotsureifthismatchesmyexerciselevel.” (Female51,group2) tŚĂƚƐŚŽƵůĚďĞĐŽŶƚŝŶƵĞĚďƵƚĨŝŶĞͲƚƵŶĞĚ͍ • “Thetrainingwasscheduledarounddinnertime.Iwouldprefertocomeinthemorningoraroundnoon.” (Female58,group1) • “Ithinkitwouldbeevenbettertoexercisetwotimesaweek.”(Female60,group1) • “Iexperiencedpainreductionthatlastedalittlewhileafterthetraining,butIdiditonlyonceaweek.I thinkyouneedtoexercisemoreoftentobeabletoevaluateitseffectiveness.”(Male57,group2) • “Adisplaywithmoreinformationabouttrainingparametersduringaquaticcyclingliketempo,pedalling frequencywouldbeusefulforfeedbackandtoguideprogression.”(Male54,group2) tŚĂƚƐŚŽƵůĚďĞĚƌŽƉƉĞĚ͍ Nosuggestionsweremade.

Quantitativeevaluation;selfͲreportedpreͲandpostͲexercisekneepain TheselfͲreportedpainscoresweresignificantlylowerafterthetrainingcomparedwith the scores before the training, ǌര=ര–2.524,Ɖ൶<ര0.05,ƌ൶=ര–0.21. Participants had a mean preͲexercisepainscoreof4.09(SD1.45;95%CI3.05–5.12).Afterthetrainingsessions participantshadameanpainvalueof3.18(SD1.33;95%CI2.23–4.13).

DISCUSSION

This study investigated the feasibility of groupͲbased aquatic circuit training for patients with knee OA. The training consisted of gait training, shallowͲwater toning exercises,flexibilityexercisesforthelowerlimbs,andaquaticcycling.DuetothesetͲup ofthetrainingonlyoneaquaticbikewasneeded. Seventypercentoftheparticipantsattendedalleightsessions.TheshorttimeͲperiod fortheresearchproject,summerholidaysandlimitedaccesshourstothetherapypool madeitdifficulttoreschedulesessionsforparticipantswhowerenotabletoattend sessionsduetoholidaysandworktime.Participantsperceivedthetrainingasatotal

30 AquaticcircuittrainingincludingaquaͲcyclinginpatientswithkneeosteoarthritis body workout, focusing on flexibility. They were positive about the immediate pain reductionandthefactthatmovementsfeltsmoother.Inaddition,selfͲreportedpain showedaoneͲpointreductionimmediatelyafterthetraining.Evidencesuggeststhat thedecompressingeffect,thewarmtemperatureandthemassageeffectduetothe 2 hydrostaticpressureincombinationwiththeexercisesmayexplainpainrelief.2,10 Theintensityofaquaticcyclingwasratedaslighttomoderate,althoughparticipants experienced underwater pedalling as more tiring than on a normal bike or homeͲ trainer.Onlythreepatients,basedonthephysiotherapists’judgement,progressedto anoutͲofͲtheͲsaddlecyclingpositionandperceivedthisasmoderatetoharddueto the fact that cycling in an outͲofͲsaddleͲposition is less controlled and less stable. It mightbethatmoretrainingisneededbeforeoutͲofͲsaddlemovementsarepossible.A goodtechniqueiscrucialtoholdastableposturewiththekneesinlinewithfeetand hips.Forinstance,Moserdevelopedanaquaticcyclingprogrammeforpatientswith rheumatic diseases and was able to introduce cycling in outͲofͲsaddle positions, but not before week seven of a tenͲweek programme.5 These findings suggest that ten minutesofaquaticcyclingisprobablytooshortatimetoachieveprogression,andthat cyclinginaseatedpositionispreferableduringearlyaquaticcircuittraining.Moreover, theperceptionofpatients,thataquaticcyclingwaseffortlessintheseatedposition, mightbeexplainedbythefactthatparticipantsinthepresentstudycycledataselfͲ selectedpace.Previousresearchonunderwatertreadmillwalkinginpatientswithknee OA showed that walking at a selfͲselected and comfortable pace resulted in lower energyexpenditurethanlandͲbasedwalking.11Thismightbeexplainedbythefactthat duringslowunderwaterwalkingbuoyancydominatesandthewaterresistanceisnot sufficienttoraisetheheartrate.Thus,participantsshouldfirstprogresstheirpedalling pacebeforecyclinginastandingposition.Cyclinginaseatedpositionisacontrolled movement and few instructions are needed, which creates more time for tailored feedbackforpatientsatotherworkstations.However,iftheperiodofaquaticcyclingis extended variation is needed to prevent monotony, which may require more supervision. Inconclusion,aquaticcircuittrainingincludingaquaticcyclingisfeasibleforpatients withOA.Aquaticcyclinginaseatedpositionisasafeandcontrolledmovementthat enables the physiotherapist to spend more time on supervision of other patients. Therefore, aquatic cycling is easy to incorporate in circuit training and enables institutions to provide small group trainings even with one aquatic bike. Further researchisneededtoinvestigatepatientacceptanceofahigherexercisefrequencyof twoorthreesessionsweeklyanditsimpactonsymptomsofkneeOA.Thefeasibility andeffectsonkneeOAofexerciseprogrammesincorporatingmoretimespentaquatic cyclinghavenotyetbeeninvestigated.

31 Chapter2

Acknowledgements TheAquaCruiserII™wasprovidedbyAquaKinetiqs©.TheauthorsthankWielWijnenat the Department of OrthopaedicSurgeryfor patientrecruitment. George Rooxkindly allowedtheuseofthepoolfacilitiesatMaastrichtUniversityMedicalCentre.

32 AquaticcircuittrainingincludingaquaͲcyclinginpatientswithkneeosteoarthritis

REFERENCES

1. TildenHM,ReicherterEA,ReicherterF.Useofanaquaticsprogramforolderadultswithosteoarthritis. Fromclinictothecommunity.dŽƉŝĐƐŝŶ'ĞƌŝĂƚƌŝĐZĞŚĂďŝůŝƚĂƚŝŽŶ͘2010;26:128Ͳ39. 2 2. BatterhamSI,HeywoodS,KeatingJL.SystematicreviewandmetaͲanalysiscomparinglandandaquatic exerciseforpeoplewithhiporkneearthritisonfunction,mobilityandotherhealthoutcomes.D DƵƐĐƵůŽƐŬĞůĞƚĂůŝƐŽƌĚĞƌƐ͘2011;12:123Ͳ36. 3. UlatkowskiM.UnterwasserfahrradversusherkömmlicheRehabilitationͲEineretrospektiveStudiemit und ohne Unterwasserfahrrad an implantierten Kniegelenksprothesen [Dissertation]. Heidelberg: HeidelbergUniversity;2009. 4. von Kathen M. ProspektivͲrandomisierte Vergleichsstudie zur Rehabilitation vorderer KreunzbandͲ plastiken zwischen koventioneller Therapie und Unterwasserfahrrad [Dissertation]. Bochum: RuhrͲ Universität;1999. 5. MoserS.EntwicklungundÜberprüfungeinesAquaͲCyclingͲProgrammsfürRheumapatienten[Master Thesis].Karlsruhe:UniversitätKarlsruhe;2009. 6. BellamyN,KirwanJ,BoersM,BrooksP,StrandV,TugwellP,etal.Recommendationsforacoresetof outcomemeasuresforfuturephaseIIIclinicaltrialsinknee,hip,andhandosteoarthritis.Consensus developmentatOMERACTIII.:ŽƵƌŶĂůŽĨZŚĞƵŵĂƚŽůŽŐǇ.1997;24:799Ͳ802. 7. BarbosaTM,MarinhoDA,ReisVM,SilvaAJ,BragadaJ.PhysiologicalassessmentofheadͲoutaquatic exercisesinhealthysubjects:aqualitativereview.:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐ^ĐŝĞŶĐĞΘDĞĚŝĐŝŶĞ.2009;8:179Ͳ89. 8. vanderLeedenM,StaalJB,BeekmanE,HendriksH,MestersI,deRooijM,etal.Developmentofa frameworktodescribegoalsandcontentofexerciseinterventionsinphysicaltherapy:amixedmethod approachincludingasystematicreview.WŚǇƐŝĐĂůdŚĞƌĂƉǇZĞǀŝĞǁƐ.2014;19:1Ͳ14. 9. RitchieJ,LewisJ.QualitativeResearchPractice:AGuideforSocialScienceStudentsandResearchers. London:SAGEPublicationsLtd;2003. 10. HallJ,SwinkelsA,BriddonJ,McCabeCS.Doesaquaticexerciserelievepaininadultswithneurologicor musculoskeletal disease? A systematic review and metaͲanalysis of randomized controlled trials. ƌĐŚŝǀĞƐŽĨWŚǇƐŝĐĂůDĞĚŝĐŝŶĞĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.2008;89:873Ͳ83. 11. Denning WM, Bressel E, DG D. Underwater Treadmill Exercises as a Potential Treatment for Adults WithOsteoarthritis./ŶƚĞƌŶĂƚŝŽŶĂů:ŽƵƌŶĂůŽĨƋƵĂƚŝĐZĞƐĞĂƌĐŚĂŶĚĚƵĐĂƚŝŽŶ.2010;4:70Ͳ80.

33 Chapter2

34 Chapter2

CHAPTER3

Effectofaquaticcyclingonpainandphysical functioningcomparedwithusualcareinpatientswith kneeosteoarthritis:studyprotocolofarandomised controlledtrial

StefanieRewald IlseMesters A.F.TonLenssen PieterJ.Emans WielWijnen RobA.deBie DDƵƐĐƵůŽƐŬĞůĞƚĂůŝƐŽƌĚĞƌƐ͘ϮϬϭϲ͖ϭϳ͗ϴϴ

34 Chapter3

ABSTRACT

Background Over the last decade aquatic exercise has become moreand more popular. One of the latest trendsisaquaticcycling,whereparticipantssitonawaterͲresistantstationarybikeand,while immersedchestdeepinthewater,combinecontinuouscyclingwithupperbodyexercisesthat utilise water resistance. Since stationary cycling and aquatic exercises are frequently recommendedtopatientswithkneeosteoarthritis,combiningbothwouldseemanobviousstep, andanaquaticcyclingexerciseprogrammeforpatientswithkneeosteoarthritishasindeedbeen developed.Thisstudyprotocolgivesadetaileddescriptionoftheexerciseprogrammeandthe methodologyofastudytocomparethisprogrammewithtreatmentinvolvingusualcareonly. Methods ThestudyisasingleͲblind,parallelͲgroup,randomisedcontrolledtrialofMaastrichtUniversity MedicalCentre+,theNetherlands.Inclusioncriteria:kneepainoffourtosevenona10Ͳpoint painratingscale;aKellgren/Lawrencescorebetweenonetothree;abilitytocycle;goodmental health;sufficientlanguageskills;indicationforphysicaltherapyinconjunctionwithimpairments due to OA. Exclusion criteria: any contraͲindication for aquatic exercise; planned total knee replacement; corticosteroid injection <3 months and/or hyaluronic acid injection <6 months; severejointcomplaints(otherthankneejoint);symptomaticandradiologicalapparenthipOA; inflammatoryjointdiseases;inabilitytosafelyenterandexitthepool;fearofwater.Participants will receive two 45Ͳmin moderate intense aquatic cycling sessions weekly over a period of 12weeks in addition to usual care or usual care only. Usual care consists of an individual intervention plan comprising lifestyle recommendations, medication routine and referral to a physical therapist. Participants will be assessed at baseline, and at 12 and 24 weeks after baseline. The primary outcome is selfͲreported knee pain and physical functioning. Secondary outcomes are lower limb muscle strength, functional capacity, selfͲreported disease severity, physicalactivitylevel,qualityoflife,selfͲefficacyandfearofmovement.Dailydiarieswillcollect informationonkneepain,physicalfunctioning,levelofphysicalactivity,painmedicationroutine and physical therapy (control group only) or exercise participation over two 30Ͳday periods (duringtheinterventionperiod). Discussion Toourknowledgethepresentstudyisthefirstrandomisedcontrolledtrialevaluatingtheeffects ofaquaticcyclinginthepreͲsurgicalstageofkneeosteoarthritis.Thistrialwilldemonstrateifthe newlydesignedaquaticcyclingintervention,insupplementtousualcare,canhelptoimprove impairmentsduetokneeosteoarthritis.

36 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA

INTRODUCTION

Aquatic cycling, which is cycling on a waterͲresistant stationary bike, might be a supplementtotheavailableexercisepossibilitiesforpatientswithkneeosteoarthritis (OA).KneeOA,acommonchronichealthcondition,affectsthedailylivesofmillionsof peopleworldwidebycausingkneepainanddifficultyperformingdayͲtoͲdayactivities.1 Alldimensionsofphysicalfunction,asdescribedbytheInternationalClassificationof 3 Functioning, Disability, and Health (IFC) framework, are affected by knee OA.2 For example,asareactiontoloaddependentjointpainthatcommonlyoccursduringdaily functional activities like walking or stair climbing, people tend to underuse the knee andbecomephysicallyinactive.3,4Avoidanceoftheseactivitiesgivesrisetoproblems with body functions and structures such as cardiovascular deconditioning, muscle weakness and reduced knee range of motion, but also to more general health problems such as a higher risk of comorbidity and premature mortality.5,6 Exercise therapy is crucial for maintaining good general health and alleviating the symptom progressionofkneeOA.4,6Inadditiontoexercise,patienteducationabouttreatment options, weight management and strategies to prevent capacity overload of the damagedknee,aswellaspharmacologicaltreatmentwithanalgesicsornonͲsteroidal antiͲinflammatory drugs (NSAIDs), are recommended for optimal conservative management of OA.6 However, only a small part of the population treats their complaints by participating in physical therapy or exercise therapy.7,8 The patients’ reasonsforexercising(ornot)dependontheir(perceptionoftheir)physicalabilityfor exercise;beliefsaboutexercise;motivationalfactorssuchasenjoyment,socialsupport, takingcontrolofthedisability;painandlimitationsofthelowerlimb.9Aquaticexercise enjoysagoodreputationamongpatientsbecauseexercisinginwaterfeelseasierand lesspainfulthanonland.10,11Thebuoyancyofthewaterresultsindecompressionof jointsandcausestheindividualtofeelweightlessandtomovemoresmoothlythanon land.11,12Inaddition,awarmwatertemperaturepromotesmusclerelaxation,possibly resulting in pain reduction and the perception of less joint stiffness.12,13 Recent systematicreviewsofaquaticexercisestudiesofindividualswithOAandotherchronic musculoskeletal disorders showed a small to moderate effect on joint pain, selfͲ reported functioning, and performance tests of physical functioning.14,15 These achievements are comparable to the results of landͲbased training.16 Growing recognition of the benefits of aquatic exercise and increasing public interest have resultedinmanyformsofaquaticexercise.OlderpatientswithOAvalueindividualised, expertͲsupervised shallowͲwater exercises, aqua jogging and hydrotherapy.17 The exercise possibilities in water range from simple vertical water exercise and water runningtomoreholisticprogrammessuchasWatsu®andtheadaptationoflandͲbased fitness trends like Zumba® to the aquatic environment.18 With the continual developmentandrefinementofwaterproofequipment,evenspinningisnowpossible inaswimmingpool.Aquaticcycling,whereparticipantsareimmersedchestdeepin

37 Chapter3 waterandpedalagainstwaterresistance,hasrecentlybecomeapopularwaterͲbased fitnessactivity.Itcombinestheadvantagesoftheaquaticenvironmentwiththoseof stationarylandͲbasedcycling,acombinationthatseemsidealforpatientswithknee OA.StationarycyclingisoftenusedinthetreatmentoflowerͲlimbinjuriesandchronic conditionslikeOAbecauseofthereducedjointload,therepetitivecircularpedalling movementthatcanbeusedtoimproverangeofmotion(ROM)inafunctionalmanner, andtheinvolvementofthelargestmusclegroupsofthelowerlimb.10Evidenceshows that stationary cycling can reduce knee pain and improve aerobic capacity, selfͲ reportedphysicalfunctioningandgait.19,20Sofar,onlyasmallnumberofstudieshave documented the therapeutic effects of aquatic cycling. Ulatkowski and von Kathen evaluated the additional effect of aquatic cycling during recovery from total knee surgeryandanteriorcruciateligamentreconstruction.21,22Inbothcases,patientswho did aquatic cycling showed greater improvements in kneeͲROM and a reduction in kneejointswellingcomparedwithpatientsreceivingusualcareonly.Furthermore,a small oneͲgroup preͲtest and postͲtest study on the effects of a 10Ͳweek aquatic cycling programme involving patients with rheumatic diseases showed a positive influence on strength, wellͲbeing and joint mobility.23 Another small study on the feasibility of aquatic cycling, as apart of an aquatic circuit training forpatientswith knee OA, evaluated aquatic cycling as a safe and controlled exercise regimen and reportedthatparticipantswereverysatisfiedwiththetraining.24 A12ͲweekgroupͲbasedaquaticcyclingtrainingformildtomoderatekneeOApatients wasdeveloped,becausecurrentlyonlyafewtherapeuticaquaticcyclinginterventions are available. Theresults ofthisstudy might provide guidance on theclinical use of aquaticcyclingandgreaterinsightintotheeffectivenessofaquaticcyclingmayhelpto broaden aquatic treatment possibilities. Furthermore, the study may support instructorsofcommunityaquaticcyclingclassesindealingwithparticipantswithknee OA.Forthesereasons,itisimportanttoexaminewhethera12Ͳweekaquaticcycling programme, in supplement to usual care, will result in better outcomes of selfͲ reported knee pain and physical functioning when compared with the relatively less intricateregimenusualcareonly. Thisarticleprovidesafulldescriptionofthestudy’srationale,designandmethodin accordancewiththeSPIRITguidelinesforreportingprotocolsofinterventiontrialsand theCONSORTguidelines.25,26

METHODS

Studydesign ThecurrentstudyisasingleͲblind,parallelͲgroup,randomisedcontrolledtrial(RCT)of Maastricht University Medical Centre+ (MUMC+). Due to the structure of the trial,

38 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA participantblindingisnotpossible.TodesignthetrialascostͲeffectivelyaspossible, theprogrammecoordinatorisinvolvedinmanyprojectactivitiessuchasrecruitment, data collection planningandexecution of the intervention, precluding blinding. Data collectionandentryisperformedbyblindedandindependentphysicaltherapistsand researchassistants.Thedatawillbeanalysedbyblindedanalysts. The randomisation procedure is performed by an independent research assistant of the Department of Epidemiology of Maastricht University using free, internetbased 3 softwaretogeneratetherandomallocationschedule(http://www.randomizer.org).A blockrandomisationwithaconstantblocksizeofeightpatientsandanallocationratio of1:1isusedtokeepsamplesizesequalacrosstheinterventionandcontrolgroup.

Settingandparticipants Participantswererecruitedinahospital(MUMC+)intheDutchprovinceofLimburg. PatientswererecruitedfromMarch2013untilOctober2015.Thesourcepopulation werepatientsdiagnosedwithmildtomoderatekneeOA.Theywerediagnosedbyan orthopaedic surgeon or nurse practitioner, and the diagnosis was based on clinical symptomsandxͲrays.Patientswithanindicationforconservativemanagementofknee OAwereofferedtheopportunitytoparticipateinthepresentstudy.Theorthopaedic specialist briefly explained the project and asked the patient for their agreement to share contact information (name and telephone number) with the programme coordinator.Researchonparticipationinselfmanagementprogrammesshowsthatthe recommendationofahealthprofessionalinfluencesthedecisiononwhethertotake partinaprogramme.27 NonͲparticipationhadnoconsequencesforfurthertreatment.

Inclusioncriteria Eligiblepatients1)ratedkneepainbetweenfourandsevenona10Ͳpointnumericpain ratingscale,2)hadaKellgren/Lawrencescorebetweenoneandthree,3)wereableto cycleonastationaryexercisebike,4)wereingoodmentalhealth(score<8foranxiety anddepressionontheHospitalAnxietyandDepressionScale,HADS),5)hadsufficient language skills and 6) had an indication for physical therapy in conjunction with impairmentsduetoOA.

Exclusioncriteria PotentialparticipantswithanycontraͲindicationforaquaticexercisetherapysuchas 1)severe,unstablecardiorespiratorycoͲmorbiditiesand2)openwounds,orpatients ona3)waitinglistfortotalkneesurgerywereexcludedfromparticipationinthisstudy. Furthermore, all potential participants who met one of the following criteria were excludedgiventhattheseconditionscouldlimitsafeandfullparticipationinthestudy

39 Chapter3 or impede the perception of symptoms of knee OA: 4) corticosteroid injection <3months and/or hyaluronic acid injection <6 months, 5) severe joint complaints (other than knee joint) that interfere the ability to participate in an exercise programme,6)symptomaticandradiologicalapparenthipOA,7)inflammatoryjoint diseases,8)inabilitytosafelyenterandexitthepooland9)fearofwater. Eligible patients first received verbal information by telephone. Interested patients were contacted by the programme coordinator after their consultation visit at the MUMC+. If their interest in participation continued after the telephone call, the programmecoordinatorsentadditionalinformationbymail.Eachpotentialparticipant couldconsiderparticipationforoneweekandwasinstructedtoholdoffanyphysical therapy until the randomisation results were known. If a candidate decided to participate,theyhad tosign an informedconsent form in whichtheydeclared their voluntary participation. The programme coordinator checked incoming applications, including two short questionnaires, to screen for any contraͲindications for physical activity using the physical activity readiness questionnaire (PARͲQ)and to screen for anxiety and depression using the HADS.28,29 In case of any doubts about patients’ mentaland/orphysicalhealth,thepatientconcernedwasadvisedtocontactamedical specialistforexaminationoradvice. Afterprovidingtheirinformedconsent,participantswererandomlyassignedtoeither the usual care control group or a 12Ͳweek aquatic cycling programme at MUMC+. Having completed the baseline assessment, participants in the intervention group startedtheaquaticcyclingprogramme(24sessions)andthecontrolgroupcouldstart withphysicaltherapyandcontinueotherusualcareroutines.ThepostͲprogrammeand followͲup measurements were scheduled after 12 and 24 weeks. After the last assessment, the control group was offered 12 sessions of aquatic cycling in a public swimming pool. The intervention group could also join this group after the 12Ͳweek interventionphase,buthadtopaytheregularrateforaquaticfitnesschargedbythe communitypool. AnoverviewoftheparticipanttimelineisgivenbelowinFigure3.1.

40 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA

MUMC+ Eligibility assessment Informed consent 3 Randomisation 0weeks (t )Baselineassessment 1 Aquatic cycling 1:1 Usual care

12weeks (t2)PostͲprogramme assessment

Usual care 24weeks (t3)FollowͲupassessment Maycontinueaquaticcycling atcommunitypoolifdesired, butwillhavetopaytheregular rate Reimbursed aquatic cycling Figure3.1 Participanttimeline

Intervention

ŽƚŚŐƌŽƵƉƐ Participantswerereferredbytheirgeneralpractitionerforaconsultationvisittothe orthopaedic specialist at MUMC+. Essentially, there were three different types of consultation visits. Participants from the area of Maastricht who have not yet been diagnosed with OA were referred to MUMC+ for further diagnostics of their knee complaints. Based on predictive values for severity of complaints, the MUMC+ scheduledpatientsforconsultationattheEarlyOAOutpatientClinicorthedepartment oforthopaedicsurgery.TheEarlyOAOutpatientClinicisresponsibleforthediagnosis andsecondarypreventioninpatientswithpreͲsurgicalkneeOA.Thediagnosisisbased

41 Chapter3 onrecentxͲraysandclinicalsymptoms.Subsequently,thenursepractitionerprovided patientswithpersonalisedinformationonOA,aninformationbookletonOAandan individual intervention plan consisting of lifestyle recommendations, medication routineandreferraltoaphysicaltherapist.Aftersixweeks,patientshadtheirsecond consultationvisittoevaluatethetreatment.PatientsalreadydiagnosedwithkneeOA and who came back for a followͲup visit at the department of orthopaedic surgery couldalsoparticipateinthestudyincaseofanindicationforphysicaltherapy.These wereusuallypatientswhohadlimitedsuccesswithpharmacologicaltreatmentssuch as injections and oral pain medication. In the case of injections, patients could still participate in the present trial after a washͲout period of three (in the case of a corticosteroid injection) to six months (in the case of a hyaluronic acid injection). If participants received an injection during the trial, the programme coordinator recordedthedateandtype.Occasionally,patientswerereferredtotheorthopaedic surgeonfordiagnosisandinthateventtheorthopaedicsurgeonprovidedadiagnosis, lifestylerecommendationsandatreatmentplan. Allparticipantswereinstructedtomaintaintheirusualcareroutine.Theprogramme coordinator kept track of changes in participants’ treatment plans by monitoring patientswithdiariesthatrecordedOAͲrelatedfunctionalproblems,kneepain,physical activity,physicaltherapyandmedicationuseinthefirstandthirdmonthafterbaseline assessment. Furthermore, a short interview by phone (control group) or in person (interventiongroup)wasscheduledaftersixweeks.Priortothelastassessment,the programme coordinator called participants and inquired about any changes in treatment.

ŽŶƚƌŽůŐƌŽƵƉ The control group was instructed to continue usual care, including working on prescribedlifestylerecommendations,medicationroutineandconsultationswiththeir orthopaedic surgeon during the 24Ͳweek trial. Furthermore, participants could start withphysicaltherapy,butthiswasnotnecessarytoparticipateinthepresentstudy. Use of and compliance with nonͲpharmacological interventions, such as physical therapyand exercise, is low in patientswith knee OA.7Motivation tostartwith and maintain such interventions is influenced by previous treatment experience and perceived effectiveness, attitudes towards exercise, perceived severity of knee symptomsandcomorbidity.30,31Inaddition,duetodifferencesinhealthcarecoverage, someparticipantswereunabletoaffordphysicaltherapy.Fundingconstraintsmadeit impossibletocoverthecostsfortheseparticipants.Tominimisedropout,participants were offered 12 weekly sessions of aquatic cycling after the end of a patient’s participationinthisstudy.Thesesessionswereheldinacommunitypoolbecauseof sizerestrictionsofthehospitalpool.

42 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA

/ŶƚĞƌǀĞŶƚŝŽŶŐƌŽƵƉ Participantsintheinterventiongroupalsocontinuedwithusualcare,thoughtheywere instructed not to start additional physical therapy during the 12Ͳweek intervention period. Supervised by a physical therapist, participants performed aquatic cycling exercises for45 minutes twice aweek over aperiod of12weeks.The training took place in a heated therapy pool (32° Celsius) at the MUMC+ department of physical 3 therapy. Depending on the body length of the participants, the water depth varied between1.20and1.30metresandparticipantswereimmersedbetweenthenaveland amaximumheightofthexiphoidprocess(Photograph3.1).Theaquaticbikeusedwas theAquaCruiserII™fromAquaKinetiqs(Photograph3.2). 3.13.2 Photograph3.1Basicpositionontheaquaticbike. Photograph3.2AquaCruiserIITM This bike differs from other aqua bikes used for recreational sporting activities by healthypeople(e.g.www.hydrorider.com).Differencesconsistsofcyclingbarefooted insteadofusingwatershoes,theAquaCruiserII™saddleismorecomfortable,andthe resistance can be adjusted during pedalling via a knob located below the handlebar insteadofbeingsetonlandpriortothesession.Theresistancecanbeadjustedbysix reproducible and equal increments by means of a magnetic braking system. Participants cycled on the aquatic bike throughout the whole session. Every session consistedofawarmͲup,aconditioningphaseandacoolͲdown.Adetailedoverviewof the programme, reported according to the framework of van der Leeden et al.,32 is providedinTable3.1.

43 Chapter3

Table3.1 Aquaticcyclingprogramme /ŶƚĞƌǀĞŶƚŝŽŶƐŐŽĂůƐ;/&Ϳ͗ Ͳb710:mobilityofjointfunctions Ͳb715:stabilityofjointfunctions Ͳb740:muscleendurancefunctions Ͳb760:controlofvoluntarymovementfunctions Ͳb620:proprioceptivefunction ǆĞƌĐŝƐĞƐ 1. CyclingatselfͲchosenrpm Ϯ͘ Mobilisationofupperbody 3. ϲϬͲƌƉŵĐǇĐůŝŶŐ=participantsfocusonpedallingataminimumof60rpm ϰ͘ >ŽǁĞƌůĞŐĞǆĞƌĐŝƐĞƐ;ϭͲϮĞǆĞƌĐŝƐĞƐƉĞƌƐĞƐƐŝŽŶͿ 4.1 oneͲlegpedalling 4.2 emphasisonupwardordownwardpedallingmovement 4.3 outͲofͲtheͲsaddleposition:standingclimb 4.4 outͲofͲtheͲsaddleposition:standingflat ϱ͘ ƌŵĞǆĞƌĐŝƐĞƐ;ϭĞǆĞƌĐŝƐĞƉĞƌƐĞƐƐŝŽŶͿ 5.1 shoulderabduction/adduction=armlifts 5.2 shouldertransverseabduction/adduction=flybacks 5.3 shoulderflexion/extension=walkingarms/armpendulum(oneͲsided) 5.4 elbowflexion/extension=curl 5.5 shoulderflexionandextension=armpendulum ¾ ĂƌŵĞǆĞƌĐŝƐĞƐǁŝůůďĞĐŽŵďŝŶĞĚǁŝƚŚĚŝĨĨĞƌĞŶƚŚĂŶĚƉŽƐŝƚŝŽŶƐ(fromlessintensetomoreintense): sweepingonwatersurface,handslicingsidewaysthroughthewater(‘cutting’),‘fisting’,cupped hands(‘scoop’),openhand(‘fan’) 6. Backwardpedalling 7. ROMknee=sittingontheaquabike,feetoutofthepedals,flexionandextensionofunloadedknees 8. Calfandhamstringstretching 'ĞŶĞƌĂůŝŶĨŽƌŵĂƚŝŽŶ͗ Ͳ Main focus is on correct aquatic cycling technique, i.e. cycling with a cadence of 60 rpm, a good alignmentofthelowerlegsandanuprightposture Ͳ SetͲup: x WarmͲup:exercise1,2 x Conditioning:exercise3,4,5 x CoolingͲdown:exercise1,6,7,8 Ͳ Totalprogrammeduration:12weeks(2sessionsperweek) Ͳ Frequency(exercisetime/repetitions)andrestingtime: x WarmͲup:5–10minutes x Conditioning:exercise3:5–8.20minutes exercise4:4setsof30–45seconds,1minuterestingtime exercise5:4setsof1minute(~20–40repetitions),1minuterestingtime x CoolingͲdown:5–10minutes Ͳ Intensity(conditioning):11–13BorgScale/70%ofmaximumheartrate((220Ͳage)x0.7)) Ͳ Progression: x exercise3:weeklyincreaseof15–20secondsincyclingtime x exercise 4: pedalling resistance (after session 6, depending on performance of the exercise plus no signsofoverloadinonͲgoingandprevioussessions) x exercise5:handposition>lengthofleverarm>speed/smalltobigamplitude>increasedsurfacearea usingaquaglovesordiscs(dependingonperformanceoftheexerciseplusnosignsofoverloadinonͲ goingandprevioussessions)

44 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA

Table3.1 (continued) Ͳ Trainingdevices: x Timer x BorgScale x Aquaticbike‘AquaCruiserII™’ x Aquadiscs x Aquagloves x Aquadumbbells 3 ICF=InternationalClassificationofFunctioning,Disability,andHealth;rpm=revolutionsperminute In the conditioning phase, participants cycled for 25 to 30 minutes at a moderate intensitylevelandcombinedcontinuouscyclingwithexercisesfortheupperbody.In addition, patients cycled in outͲofͲtheͲsaddle positions, did oneͲleg pedalling or emphasised one part of the pedal movement (e.g. by actively pulling the pedals upwards).Therefore,theconditioningphaseessentiallyconsistedofthreesegments: continuouscyclingataminimumcadenceof60revolutionsperminute(rpm),upper bodyexercisesandlowerbodyexercises.Thecontinuouscyclingsegmentconsistedof at least five minutes of cycling at a minimum pedalling cadence of 60 rpm.Exercise durationwasincreasedby15to20secondseachweek.Basedonaconditioningphase of 25 minutes, this is an increase of 1% per week, which is lower than the recommended weekly increase of 2.5% as advocated by the American Geriatric Society.33Thisisdeliberate,however,astheassumptionisthataquaticcyclingismore demanding than stationary cycling on land.24,33 Increased pedalling resistance was offeredwithcautionandonlyifaparticipantwasabletocyclecontinuouslyat60rpm without adverse reactions such as increased knee pain after the session, because increased workload results in increased knee load. This is in turn reported to be associatedwithworseningofkneepain.34 The upper body exercises were used as an active break for the lower limbs as the pedallingtempodecreaseswiththefocusshiftedfromthelegstotheupperbody.In addition, the upper body exercises enabled a varied exercise programme and prevented monotony, which might have occurred with 45 minutes of purely cycling. Theexercisesweretypicalexercisesusedinaquaticfitnesstostrengthenarms(biceps, triceps), shoulders (rotators, flexors, extensors) and upper back (e.g. rhomboids, latissimus).Asinglerepetitionmaximumasguidanceforexerciseintensitycannotbe transposed to the aquatic environment. Characteristically, aquatic strength exercises arerepeated20timesandmore.35Previousresearchhasshownthatthisisaneffective training method to increase muscular strength in chronic pain and OA patients.36Ͳ39 Additionally, the high number of repetitions allows time to rehearse the exercise to promote execution using strong, powerful movements with good technique and full ROM.40Floatingdevicesanddragequipmentwereusedtoincreaseresistanceandto provideavariedexerciseprogramme.Theequipmentusedhasnotbeensponsoredby themanufacturers.

45 Chapter3

ThemoreexhaustingupperͲbodyexerciseroutinewasfollowedbyexercisesfocusing onthelowerlimbs.PatientscycledinahalfͲseatedorstandingposition,emphasised theupwardsanddownwardspedallingmovementand/orcycledwithoneleg.Thereis currently no evidence regarding the influence of different body positions in aquatic cycling on knee joint load. Consequently, the results of biomechanical studies of stationarycyclingonlandhaveguidedthedevelopmentofthisexercisesegment.41,42 Research on the difference between seated and standing uphill cycling shows an increasedactivationofmonoarticularhipandkneeextensors.However,tokeepknee load as low as possible, standing positions should be limited during each session. DuringalandͲbasedspinningclass,outͲofͲsaddlepositionsaccountforapproximately 16% (~8 min) of the session’ s total time (50 min).43 In comparison, the time spent cyclinginstandingpositionsintheaquaticcyclingprogrammewas5to8%(2to4min) ofthetotalcyclingtime(~45min).ThecoolͲdownconsistedofslowlycyclingforward andbackward,kneeͲROMexercisesandstaticstretchingofthelowerlimbstodecrease theheartrate,prepareparticipantsforthechangeofbodyposition(e.g.fromsitting on the exercise bike to standing position) and environment (the pool floor slowly comes up during the stretching exercises), and to reduces experienced muscle soreness. The exercise intensity was moderate and was regulated by the patients themselves basedontheirperceivedexertionusingtheBORGscale.44,45Inaddition,heartratewas monitoredbyaPolarFt7,Wearlink®+Hybridcheststrapduringeachtrainingsession, andpeakandaverageheartratesareprotocolled.Anaverageheartrateof70Ͳ75%of the maximum heart rate is desirable and recommended by exercise guidelines for OA.46,47 Furthermore, the supervising physical therapist assessed the quality of the performancebyjudgingcompensationalmovements,posturalcontrol,safeexecution, and level of exertion (assessed by the talk test). In the event of any doubt about a participant’shealthstatus,thephysicaltherapistdiscontinuedthetrainingandreferred theparticipanttotheirgeneralpractitioner.

Outcomemeasures Thecurrentstudyinvestigatestheeffectofaquaticcyclingonimpairmentsduetoknee OA,suchaskneepain,reducedphysicalfunctioningoverthepreviousweekandonthe assessment day, and knee stiffness. It also seeks to make an overall assessment of disease severity and lower limb muscle strength compared with a control group receivingusualcare.Furthermore,thephenomenonofaquaticcyclingisexploredina more general health context through evaluating the effect of aquatic cycling on functionalcapacity,physicalactivitylevelandqualityoflife.Psychologicalmeasureson selfͲefficacy and fear of movement are also assessed. Outcomes are assessed in person, but to keep the number of missing values as small as possible, all

46 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA questionnaireswillbesentbymailtoanyparticipantsunabletocometotheMUMC+. AnoverviewofallmeasuresandtimingofassessmentisgiveninTable3.2. Table3.2 Overviewofmeasuresandtimingofassessment Timingofassessment(inweeks) Ͳ10 1Ͳ1212 24 RandomisationBaselineaquaticcycling/PostͲ FollowͲup usualcareonly intervention 3 Measures ^ĐƌĞĞŶŝŶŐ x Kellgren/Lawrencescore x x PhysicalActivityReadiness x Questionnaire x HospitalAnxietyandDepression x Scale x CumulativeIllnessRatingScale x WƌŝŵĂƌǇŽƵƚĐŽŵĞ x KneeInjuryandOsteoarthritis x x x OutcomeScore ^ĞĐŽŶĚĂƌǇŽƵƚĐŽŵĞƐ x LowerExtremityFunctionScalex x x x Numericpainratingscalesx x x x PatientGlobalAssessmentx x x x Isometricandisokineticmuscle x x x strength(hamstring,quadriceps) x TimedupandGox x x x 6ͲMinuteͲWalkingͲTestx x x x ShortQuestionnairetoAssess x x x HealthͲenhancingPhysicalActivity x RandͲ36HealthSurveyx x x x TampaScaleforKinesiophobiax x x x ArthritisSelfͲEfficacyScalex x x WƌŽĐĞƐƐŵĞĂƐƵƌĞƐ x DiariesonOAͲrelatedfunctional x problems,kneepain,physical activity,physicaltherapyand medicationuse x Attendanceataquaticcycling x x Adverseeventsduringaquatic x cycling x Participants’experiencewiththe x training x Averageandpeakheartrateduring x aquaticcycling

Primaryoutcome TheselfͲreportedscoreonkneepainandphysicalfunctioningassessedwiththeKnee Injury and Osteoarthritis Outcome Score (KOOS, http://www.koos.nu) is the primary outcome measure. The KOOS questionnaire is an extended version of the Western

47 Chapter3

Ontario and McMaster Universities Arthritis Index (WOMAC), which is a wellͲ recognised,validandresponsibleoutcomemeasureinkneeOAresearch.48Inaddition totheWOMACsubscalesforpain,stiffnessandphysicalfunction(initscompleteand originalformat),theKOOSalsotakesintoaccountdifficultieswithsportactivitiesand kneeͲrelatedqualityoflife.ThefivesubscalesarescoredonafiveͲpointLikertscale and final scores are modified to a 0 to 100 scale. A lower score is associated with higher impairments. The Dutch KOOS shows good, internal validity (Cronbach’ sɲ: 0.71),constructvalidity(SpearmancorrelationbetweenKOOSsubscalesandSFͲ36pain and physical function: 0.63, 0.75) and is a reliable (ICC: 0.45Ͳ0.89) measurement for patientswithmildtomoderatekneeOA.49TheKOOSisselfͲadministeredandpatients needapproximately10mintoanswerallquestions.50

Secondaryoutcomes The>ŽǁĞƌǆƚƌĞŵŝƚǇ&ƵŶĐƚŝŽŶ^ĐĂůĞ;>&^ͿisapatientͲreportedmeasureonphysical functioningonthetestday.51Thequestionnaireconsistsof20questionsandpatients cancompleteitwithinafewminutes.TheDutchversionoftheLEFShasfavourable psychometricproperties:goodinternalconsistency(0.96),reliability(ICC=0.86)anda good construct and discriminant validity.52 It is a diseaseͲspecific questionnaire and each item is scored on a fiveͲpoint Likert scale. The total score ranges from 0 to 80points.Ahigherscoreisassociatedwithbetterphysicalfunctioning. EƵŵĞƌŝĐƉĂŝŶƌĂƚŝŶŐƐĐĂůĞƐ;EWZ^ͿarefrequentlyusedtoassesspainintensityinOA and the NPRS has been recommended as a core outcome measure for chronic pain trials.53,54 Previous research showed that the NPRS is a valid and responsive tool for painmeasurementsinOApatientsandalsoareliabletool(ICC:0.64to0.86)inpatients with orthopaedic problems and musculoskeletal pain.55,56 The NPRS is a selfͲ administeredscale,completedinlessthan1minandalowerscoreindicateslesspain. Osteoarthritis research societies have defined a core set of outcome measures for clinicalOAtrials:pain,functionandWĂƚŝĞŶƚ'ůŽďĂůƐƐĞƐƐŵĞŶƚ;W'Ϳ͘57Participantswill beaskedtoconsiderallthewaysinwhichillnessandhealthconditionsareaffecting thematthetimeoftheassessmentandtomarkoneof21numberedcirclesonavisual analoguescale(VAS).58Ahigherscoresmeansthattheparticipantfeelsmoreaffected bytheirillnessandhealthconditions.ThePGAhasagoodtestͲretestreliability(ICC: 0.702)inpatientswithrheumaticarthritisandiscompletedbypatientswithinafew seconds.58 DataonŝƐŽŵĞƚƌŝĐĂŶĚŝƐŽŬŝŶĞƚŝĐŵƵƐĐůĞƐƚƌĞŶŐƚŚŽĨŚĂŵƐƚƌŝŶŐĂŶĚƋƵĂĚƌŝĐĞƉƐofthe affectedlegarecollectedwiththedynamometerBiodex®System3Pro.Theisometric quadricepsandhamstringmusclestrengtharetestedin30°and60°fixationwiththree repetitionseach.Isokineticquadricepsandhamstringmusclestrengtharemeasuredat 60°persecond(fiverepetitions)and180°degreespersecond(fiverepetitions).The

48 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA reliabilityofisometricandisokineticstrengthtestingismoderate(r=0.8–0.9)tohigh (r>0.9)inpatientswithmildkneeOA.59 ThedŝŵĞĚƵƉĂŶĚ'Ž;dh'Ϳperformancetestmeasuresthetimeneededbyapatient to get out of a chair, walk three metres, return and get back into the chair. The guideline for physical therapy in patients with hip and knee OA of the Royal Dutch SocietyforPhysical Therapy recommends the use of the TUG in combination with questionnaires (e.g. 3 KOOS) to evaluate treatment goals for physical functioning.60,61 The interͲrater reliability between three physical therapists assessing patients with rheumatoid arthritiswashigh(ICC:0.97).62IntraͲsessionreliabilitywasalsosatisfactory:ICCof0.75 withatimeintervalofmorethan25weeksandanICCof0.87withatimeintervalof less than 1 week.63,64 In frail elderly patients and elderly patients undergoing orthopaedicrehabilitation,theTUGcorrelateswellwithgaitspeed(r=Ͳ0.61,0.745)and performance of dayͲtoͲday activities (r=Ͳ0.78) and correlates highly with the Berg BalanceScale(r=Ͳ0.81).65,66 TheϲͲDŝŶͲtĂůŬŝŶŐͲdĞƐƚ;ϲDtdͿisasimpletest,recommendedbytheDutchphysical therapy guideline for OA, to assess functional capacity.61,67 Over a period of 6 min, participantswalkataselfͲchosenspeedwiththeaimofcoveringasmuchgroundas possible.Participantshavetowalkinasquarewithatotallengthof44metres.ThissetͲ up deviates from the standard as recommended by the American Thoracic Society whichincludesa30Ͳmetrecorridororwalkwaywithconesplacedatthebeginningand end of the 30Ͳmetre boundary to indicate turns.68 In patients with fibromyalgia and thoserecoveringfromtotalhipandkneesurgery,the6MWTisareliabletestwithan ICC for testͲretest reliability of 0.94 and 0.98.63,69 In terms of validity, the oxygen uptake during the 6MWT shows a high correlation with peak oxygen uptake values (r=0.86)obtainedduringmaximumexercisetestinginpatientswithheartfailure.70 The ^ŚŽƌƚ YhĞƐƚŝŽŶŶĂŝƌĞ ƚŽ ^ƐĞƐƐ ,ĞĂůƚŚͲĞŶŚĂŶĐŝŶŐ ƉŚǇƐŝĐĂů ĂĐƚŝǀŝƚǇ ;^Yh^,Ϳ is a surveytoassesshabitualphysicalactivityandconsistsofelevenquestionsonphysical activityinfourdifferentcontexts:commuting,leisuretime,duringworkandhousehold activities. It is a short and simple questionnaire with proper reliability and validity.71 The SQUASH is used to evaluate adherence to the Dutch physical activity guideline, recommending 30 min or more of at least moderate intense physical activity for a minimum of 5 days per week.72 With regard to OA, only one study evaluated the SQUASH.Wagenmakersetal.foundagoodcorrelationwithanaccelerometer(r=0.56) inpatientswithhipOAaftersurgery.73 TheZĂŶĚϯϲͲŝƚĞŵ,ĞĂůƚŚ^ƵƌǀĞǇ;ZĂŶĚͲϯϲͿisagenerictooltomeasurehealthͲrelated quality of life (HRQoL).74 It consists of 36 items that cover eight HRQoL domains: physicalfunctioning,rolelimitationsbecauseofphysicalhealthproblems,bodilypain, general health perception, vitality, social functioning, role limitations because of emotional problems, and mental health. The total score ranges from 0–100, with a higher score indicating better health status. The RandͲ36 is almost identical to the

49 Chapter3

MedicalOutcomeStudy(MOS)ShortͲFormͲ36(SFͲ36),andbothhaveaprovensound responsiveness in patients with knee OA (SRM=0.528), and internal consistency (Cronbach’ sɲ: >0.70) and testͲretest reliability (ICC: 0.40Ͳ0.82) in a Dutch general population.75Ͳ79 ThedĂŵƉĂ^ĐĂůĞĨŽƌ<ŝŶĞƐŝŽƉŚŽďŝĂ;d^<Ϳisusedtoassessfearofinjury/reͲinjurydueto movement.80 It is a 17Ͳitemscale that is scored on a fourͲpoint scale from ‘strongly disagree’to‘stronglyagree’.ThepresentstudyusestheDutchversion,whichshows good psychometric properties in patients with acute low back pain: good internal consistency (Cronbach’ s alpha =0.70) and satisfactory testͲretest reliability (ICC: 0.76).81 The ƌƚŚƌŝƚŝƐ ^ĞůĨͲĨĨŝĐĂĐǇ ^ĐĂůĞ ;^^Ϳ is a valid and responsible measure providing informationonpatients’selfͲefficacytoperformatask(e.g.‘Howcertainareyouthat youcanwalk100feetonflatgroundin20seconds?’)ortoachieveaspecificbehaviour (e.g.‘Howcertainareyouthatyoucandecreaseyourpainquiteabit?’).82,83Intotal, thescaleconsistsof20itemsthataredividedintothreesubscales:selfͲefficacypain scale,selfͲefficacyfunctionscaleandselfͲefficacyothersymptomsscale(e.g.fatigues, enjoyment).TheitemsarescoredonatenͲpointLikertscaleresultinginatotalscore rangingfrom0to100.HigherscoresindicateabetterselfͲefficacy.TheASEShasbeen translatedandisavailableinDutch.84ThepresentstudymeasurestheselfͲefficacyfor function. This subscale has a good testͲretest reliability (ICC: 0.85) and internal consistency(Cronbach’salpha=0.89).85

Processmeasures ĂŝůǇ ĚŝĂƌŝĞƐ collect information on knee pain, physical functioning, level of physical activity, pain medication routine and physical therapy participation over two 30Ͳday periods (during the intervention period). Participants can fill in the diaries on a computer or in a printed booklet version. Information on physical functioning and physical activity is gathered by questions derived from the LEFS and SQUASH questionnaires.52,72KneepainismeasuredbyNPRS.55Thesectiononmedicationuse asks if pain medication is used for knee pain or other pain, the name of the pain medicationandthedosageandtimeͲpoint(s)oftakingthemedication.Participationin, durationandintensityofexerciseroutinesorphysicaltherapywillbedocumentedas well.Furthermore,fourquestions,derivedfromtheRANDͲ36questionnaire,willask abouttherestrictionsinphysicalrolefunctioning.Previousresearchonlyshedlighton the level of hindrance and/or avoidance of activities, but not on the type of hindrance.86Thedailyrepeatedmeasureswouldprovidemoreinsightintothecourse ofpain,physicalfunctioning,physicalactivityandmedicationuse.Thediarydatafrom the intervention group will yield important information on the development of impairments,levelofphysicalactivityandmedicationuseinresponsetotheaquatic cyclingprogramme.Thediariesofthecontrolgroupwillprovideapictureofthelevel

50 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA of physical activity, participation in exercise therapy and the development of impairmentsovertime.Previousresearchusingbookletdiariesandcomparablediary periodsshowedgoodcomplianceandalowdropoutrate,indicatingthatthismethodis acceptableforchronicpainpatients.87 WĂƌƚŝĐŝƉĂŶƚƐ͛ ĞǆƉĞƌŝĞŶĐĞ ǁŝƚŚ ĂƋƵĂƚŝĐ ĐǇĐůŝŶŐ will be assessed after the final training session by means of planned focusͲgroup sessions. SmallͲgroup interviews will be planned, and participants will be asked broad, openͲended questions about their 3 expectations,fulfilmentofexpectations,positiveandnegativeaspectsofthetraining and suggestions for further development of the training. Thirty per cent of the participants(~20participants)intheinterventiongroupwillbeinvitedtoattendsmallͲ groupinterviewstoprovidefeedback. ƚƚĞŶĚĂŶĐĞ͕ ĂĚǀĞƌƐĞ ĞǀĞŶƚƐ ĂŶĚ ĞǆĞƌĐŝƐĞ ƉƌŽŐƌĞƐƐŝŽŶ of the intervention group were registeredbythephysicaltherapist.Foreverypatientatraininglogbookexistswhere thephysicaltherapistdocumenteddateandnumberofthetrainingsessionsattended. Intotaleverypatientcouldattend24sessions.Pedallingtempoandresistance,heart rate and BORG scores were noted for every exercise during the conditioning phase. Furthermore,thephysicaltherapistdocumentedtheoccurrenceandtypeofproblems withtheperformanceofcertainexercisesinanindicatedopentextboxinthetraining logbook. Thus, the physical therapist described the type of problem, whether the participantswereabletocontinuetheexerciseandincaseofperformancerestrictions the alternative exercise was described. Also, adverse effects during or following the sessionsweredocumentedinthetraininglogbook.NonͲseriousadverseeffectswere defined as increased joint pain, stiffness, muscle soreness and/or fatigue occurring duringorimmediatelyafterthelasttrainingsession.88,89Iftheseadverseeventswere experienced longer than 24Ͳh or interfered with physical activities and social participationtheywereclassifiedassevereadverseeventsseriousadverseeventwas defined as an occurrence that resulted in permanent or severe disability, hospitalization,ordeath.90

Datacollectionandmanagement Thedatafromallmeasurementswillberecordedonpaperbytheblindedoutcome assessors.Patientsareinstructednottoinformtheassessoraboutgroupallocation. The outcome assessors are physical therapists of the MUMC+. The performance measures and strength assessments are part of their routine tasks and no special training prior to the study was needed. The purpose and scoring method of all questionnairesusedwasexplainedbytheprogrammecoordinatorpriortothestartof thestudy.Furthermore,theoutcomeassessorspractiseddatacollectionseveraltimes in order to get an idea of the time needed. The data are recorded on paper, with numbersusedtorepresenttherankorderwithintherecruitmentprocessinorderto guaranteethatthedataisanalysedseparatelyfrompersonaldata.Thedatainthese

51 Chapter3 papercasereportsaredigitisedbyresearchassistantsandtheprogrammecoordinator will enforce data integrity through range checks and crossͲvalidation between the samevariablesassessedonrepeatedoccasions.Inaddition,visualrecordverification willbedonebycomparingthefirsttenrecordsofadatasetwiththecorresponding paper case reports.91 If no inconsistency is found, the programme coordinator will check every tenth record until an incorrect record is found. After correction of the incorrectrecord,allfollowingrecordswillbecheckeduntilsuccessiverecordsfreeof inconsistenciesarefound.92 Alldataonpaperwillbestoredinalockedarchiveforamaximumof15years.Onlythe programme coordinator has access to personal data. After the analysis, other researchersoftheteam(RAB,IM,AFL,andPJE)willalsohaveaccesstoanonymous data.

Samplesize Thepresentstudyisthefirsttoevaluatetheeffectsofaquaticcyclinginpatientswith mildtomoderatekneeOA.Therearenopreviousdataonwhichtobasethesample size calculation. The estimation of the sample size is based on two factors: 1) the minimum clinically important difference (MCID) of WOMAC, and 2) studies with a similar design (aquatic therapy versus usual care) or intervention (one group preͲ test/postͲtestfeasibilitystudyofaquaticcyclingforrheumaticpatients).Althoughthe presentstudyusestheKOOSquestionnaire,theWOMACquestionnairehasbeenused toestimatethesamplesize.TheWOMACiswellrecognizedinOAresearchandthe questionnaire and minimum clinically important differences (MCID) of the WOMAC subscales are known. The MCID changes from baseline to postͲinterventions on the WOMACpainandfunctionscalerangefrom15%to18%forpainand12%to17%for physical function.75,93 The results of previous studies are in line with or exceed the MCIDsreferredto.Hinmanetal.showeda21%and29%improvementinWOMACpain and function scores for the hydrotherapy group.36 The usual care group did not improve. A feasibility study of an aquatic cycling programme for rheumatic patients showed an improvement of 14% in the postͲintervention score of selfͲreported physicalfunctioning.23BasedontheaboveͲmentioneddata,theaquaticcyclingtraining inthepresentstudyisexpectedtoachieveatleastsimilarresultsastheinterventions of Moser and Hinman, or even exceed those results because of a higher exercise frequencyandintensityandlongerdurationoftheintervention.23,36Thus,adifference of25%betweentheaquaticcyclinggroupandusualcaregroupintermsofreduction of knee pain and improvement in physical functioning is hypothesised as clinically meaningful. The statistical level of significance was set to an alpha (ɲ) of 0.05 and statisticalpowerto0.80.Thestandarddeviationis20%ofthemaximumscoreofthe WOMACsubscaleforpainandphysicalfunction.94Withanexpecteddropoutrateof 20%,thefinalnumberofparticipantsneededis168.

52 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA

Dataanalysis Data analysis will be performed using IBM SPSS Statistics 23. The effect of group membership(aquatic cycling versus usualcare) onprimaryandsecondary outcomes will be estimated and tested for significance with a significance level set at 0.05. Furthermore, any significant changes that occur over time will be examined. Demographicvariablesandclinicalbackgroundvariables(i.e.BMI,coͲmorbidities)will 3 beusedasgroupingvariablesforsubgroupanalysisorascovariates. Multilevelanalysiswillbeappliedwithrepeatedmeasures(level1)thatareclustered withinpersons(level2),andwithpatients(level1)clusteredwithingroups(level2). Usingmultilevelanalysisallowstheuseofalldataavailable,includingdropout,lossto followͲup,missedappointmentsandparticipantincapacity.Diarydatawillbeexamined forthetimecourseoflevelofphysicalactivity,physicalfunctioning,kneepainandpain medication use in the intervention and control group, and for betweenͲgroup differencesinchange.Inaddition,therelationshipbetweentheaquaticcyclingtraining andthe factors just referred to will beevaluated. Multilevel analysiswill beused to estimateandtestbetweenͲpersondifferencesandthewithinͲpersonprocesses.

Datamonitoring The content of the aquatic cycling intervention is comparable to existing physical activityprogrammesonland.Researchhasshownthattheseprogrammesinvolveno additional harm or risk to the patient.47,95 Aquatic cycling in rheumatic patients was evaluatedinastudyassafeandfeasible.23Inaddition,thereisadequateevidencethat aquatictrainingandstationarycyclingarebeneficialandsafeactivitiesforpatientswith kneeOA.14,19,20 Becausetheriskofanyadverseeventsfromparticipationintheinterventiongroupis smallandcomparabletotheverylowriskofadverseeventsfromparticipationinlandͲ basedOAexerciseprogrammes,47nodatamonitoringcommittee(DMC)isneeded.In case of a serious adverse event, the programme coordinator will inform all professionals involved in the study and report the event via a web portal to the accreditedMedicalEthicsBoardwithin24hours.

Ethics Ethical approval has been obtained from the Medical Ethics Board of MUMC+ (referencenumber12Ͳ2Ͳ075)on06Ͳ03Ͳ2013.Thetrialwasregisteredon21Ͳ12Ͳ2012in the Netherlands Trial Register (NTR3766). Any modifications to the protocol that influencetheexecutionofthetrialorparticipantsafety,i.e.changesofstudydesignor procedureswillbedescribedinaformalamendment.Allsubstantialamendmentswill requireapprovalfromtheMedicalEthicsBoardofMUMC+.Participantsinthestudy arecoveredbyaninsurancepolicythatincludescoveragainstresearchsubjectinjury

53 Chapter3 ordeathasaresultofthestudy.Theresearchprojectiscoveredbyliabilityinsurance, whichisinaccordancewithSection7,subsection6oftheMedicalResearch(Human Subjects) Act (WMO). A copy of the insurance certificate of MUMC+ is in the possessionoftheboardoftheMedicalEthicsCommittee.

DISCUSSION

This trial will demonstrate if the newly designed aquatic cycling intervention, in supplementtousualcare,canhelptoimproveimpairmentsduetokneeOA.Asfaras we know, the present study is the first randomised controlled trial evaluating the effectsofaquaticcyclinginthepreͲsurgicalstageofOA.Ifthistrainingprovestobe effective,theresultscanprovideguidanceontheuseofaquaticcyclinginclinicaland community exercise settings. Aquatic cycling could be used to increase the range of motionoftheknee,lowerlimbmusclestrengthandaerobiccapacityinallpopulations, whereaslandͲbasedtrainingistoopainful.Itmightalsoanoptionforpatientswhofeel uncomfortable with traditional aquatic exercise because of poor swimming skills or hydrophobia. Previous studies have shown that aquatic cycling is well accepted by patients who have hydrophobia.22,24 As aquatic cycling has become a recent fitness trendinEuropeandtheUS,manypublicswimmingpoolsofferaquaticspinningtoa healthy population. Exercise instructors in community exercise settings who are qualified to supervise classes with musculoskeletal disorders could use the training programmedescribed(ifproveneffective)toadaptaquaͲspinningclassestotheneeds of people with knee OA. The opportunity to participate in a modern and popular exerciseclassmightespeciallybeappealingtokneeOApatientswhowanttobeactive and/orareyoung.20 Thestrengthofthisstudyistheclosemonitoringconductedduringtheintervention phase with diaries, since selfͲreported measures might be sensitive to dayͲtoͲday variations not capturing the development of OA impairments throughout the intervention.19AnotherstrongaspectofthestudyisthefollowͲupassessmentthree months after the end of the programme, something that is rarely done in aquatic exerciseresearch.14EspeciallyinterestinginthefollowͲupassessmentistheevaluation of whether participants in the intervention group continued to aquaͲcycle in the communityswimmingpoolorstayedactiveinanotherway.Thiswillindicateifpeople are willing to continue aquatic cycling at their own cost or if it has helped them to becomemoreactive.Thecontrolgroupwillbeinvitedtoattendtwelvefreeaquatic cyclingsessionsinthecommunitypool.Duetolimitedaccesstothehospitalpooland thelimitednumberofaquabikes(n=4),itisnotpossibletotrainbothgroupsinthe hospital.Fundingrestraintsmakeitimpossibletobearthecostsof24sessionstwicea weekforthecontrolgroupinthecommunitypool.Astheprogrammecoordinatorwill

54 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA give the training, there will be no difference in terms of the training content and structure. However, this waitingͲlist control design can influence the results of the studyintwoways.Ontheonehand,participantsinthecontrolgroupmightbemore motivated to follow usual care instructions with regard to physical activity because theydonotwanttobelessactivethantheinterventiongroup.Inaddition,thisgroup willbemonitoredbymeansofdiariestoo,whichmightalsomotivatethemtobemore active.Ontheotherhand,itispossiblethatthecontrolgroupparticipantswillfollow 3 usualcarerecommendationslessstrictlyastheywillbewaitingfortheirturntotryout aquatic cycling. Furthermore, the fact the control group participants did not receive any immediate and free intervention might influence motivation for further participation.Therefore,wewillinformparticipantsaboutgroupallocationbeforethe baselineassessment.Theassessmentsofthepresentstudyarenotpartoftheclinical routineandparticipantshavetocomebackfortheassessmentsaftergivingconsent andbeingrandomised. By informing participantsabout groupallocation prior tothe baseline,wewishtopreventfrustrationaboutgroupallocationandpossibledropout. Nevertheless,thisstrategyincreasestheriskofdropoutbeforebaselineassessment. Inconclusion,thistrialwillincreasetheknowledgeofaquaticcyclingandmightbea useful addition to aquatic exercise training. As with aquatic treadmill training, it is possibletoadequatelymonitorandmodifyexerciseintensitysincepedallingrateand resistancecanbeadjusted.96Inaddition,theexerciseprogrammeisbasedonexercise guidelines and the exercise intensity will be measured by pedalling frequency and resistance, average heart rate, peak heart rate and perceived exertion during the different parts of the conditioning phase. Recent reviews strongly recommend using and reporting exercise intensity, as will be done in this study, in order to obtain a betterunderstandingofthedoseͲresponserelationshipinaquaticexercise.14,15

Disseminationpolicy Thescientificintegrityofthisresearchprojectrequiresthatallresultsofthisstudybe disclosedunreservedly.TheresultswillbesubmittedforpublicationtopeerͲreviewed scientific journals. Furthermore, the results will be presented at national and international congresses. Through to November 2016, four articles have to be submittedwiththeprogrammecoordinatorasfirstauthor.Thesearticleswillprovide the basis for the programme coordinator’s PhD thesis. All authors must contribute significantly to the conception of an article and/or the analysis or interpretation of data.Eachauthorneedstorevisetheconceptsofanarticlecriticallyandhastogive final approval of the manuscript that will be published. It is not the intention to collaboratewithprofessionalwriters.Theoutcomesofthestudywillbereleasedtothe referring orthopaedic nurse practitioner and orthopaedic surgeons, the participating physical therapists, the local community swimming pool and the general medical

55 Chapter3 community.Inadditiontothestudyresults,everyparticipantwillreceiveanindividual summaryofher/hisstudyresultsassoonaspossibleafterparticipation.

Acknowledgement We are grateful to theNetherlands Organisation for ScientificResearch (NWO),P.O. box93138,2509ACTheHague,forfundinga4ͲyearPhDpositionfortheprogramme coordinator. The funding covered the salary of the programme coordinator and all organisationalcostsofthetrial.TheNWOgrantnumberis022.003.036.Theaquatic bikes used to train the intervention group were funded by the Transmural Care of MUMC+.Thesefundingsourcesplayednoroleinthedesignofthisstudyorduringany stageofthetrial,i.e.execution,analysesandreporting. WewouldgratefullyacknowledgethedepartmentofphysicaltherapyoftheMUMC+ for providing the therapy pool and testing facilities. We thank Maastricht Sport and Geusseltbadforconfidenceinourstudyandforprovidingthetrainingfacilities.

56 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA

REFERENCES

1. Litwic A, Edwards MH, Dennison EM, Cooper C. Epidemiology and burden of osteoarthritis. ƌŝƚŝƐŚ DĞĚŝĐĂůƵůůĞƚŝŶ.2013;105:185Ͳ99 2. Dreinhofer K, Stucki G, Ewert T, Huber E, Ebenbichler G, Gutenbrunner C, et al. ICF Core Sets for osteoarthritis.:ŽƵƌŶĂůŽĨZĞŚĂďŝůŝƚĂƚŝŽŶDĞĚŝĐŝŶĞ.2004;10.1080/16501960410015498:75Ͳ80. 3. Pisters MF, Veenhof C, van Dijk GM, Dekker J. Avoidance of activity and limitations in activities in patients with osteoarthritis of the hip or knee: a 5 year followͲup study on the mediating role of 3 reducedmusclestrength.KƐƚĞŽĂƌƚŚƌŝƚŝƐĂŶĚĂƌƚŝůĂŐĞ.2014;22:171Ͳ7. 4. Holla JF, van der Leeden M, Knol DL, Roorda LD, Hilberdink WK, Lems WF, et al. Predictors and outcomeofpainͲrelatedavoidanceofactivitiesinpersonswithearlysymptomatickneeosteoarthritis: afiveͲyearfollowupstudy.ƌƚŚƌŝƚŝƐĂƌĞΘZĞƐĞĂƌĐŚ.2015;67:48Ͳ57. 5. Blair SN, Sallis RE, Hutber A, Archer E. Exercise therapyͲthe public health message. ^ĐĂŶĚŝŶĂǀŝĂŶ :ŽƵƌŶĂůŽĨDĞĚŝĐŝŶĞĂŶĚ^ĐŝĞŶĐĞŝŶ^ƉŽƌƚƐ.2012;22:e24Ͳ8. 6. ZhangW,MoskowitzRW,NukiG,AbramsonS,AltmanRD,ArdenN,etal.OARSIrecommendationsfor the management of hip and knee osteoarthritis, Part II: OARSI evidenceͲbased, expert consensus guidelines.KƐƚĞŽĂƌƚŚƌŝƚŝƐĂŶĚĂƌƚŝůĂŐĞ.2008;16:137Ͳ62. 7. Hinman RS, Nicolson PJ, Dobson FL, Bennell KL. Use of nondrug, nonoperative interventions by communityͲdwellingpeoplewithhipandkneeosteoarthritis.ƌƚŚƌŝƚŝƐĂƌĞΘZĞƐĞĂƌĐŚ.2015;67:305Ͳ9. 8. Smink AJ, van den Ende CH, Vliet Vlieland TP, Swierstra BA, Kortland JH, Bijlsma JW, et al. Beating osteoARThritis: Development of a stepped care strategy to optimize utilization and timing of nonͲ surgical treatment modalities for patients with hip or knee osteoarthritis. ůŝŶŝĐĂů ZŚĞƵŵĂƚŽůŽŐǇ. 2011;30:1623Ͳ9. 9. ThorstenssonCA,RoosEM,PeterssonIF,ArvidssonB.HowdomiddleͲagedpatientsconceiveexercise asaformoftreatmentforkneeosteoarthritis?ŝƐĂďŝůŝƚǇĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.2006;28:51Ͳ9. 10. WestbyMD.Ahealthprofessional'sguidetoexerciseprescriptionforpeoplewitharthritis:areviewof aerobicfitnessactivities.ƌƚŚƌŝƚŝƐĂŶĚZŚĞƵŵĂƚŝƐŵ.2001;45:501Ͳ11. 11. TildenHM,ReicherterEA,ReicherterF.Useofanaquaticsprogramforolderadultswithosteoarthritis. Fromclinictothecommunity.dŽƉŝĐƐŝŶ'ĞƌŝĂƚƌŝĐZĞŚĂďŝůŝƚĂƚŝŽŶ.2010;26:128Ͳ39. 12. BeckerBE.Aquatictherapy:scientificfoundationsandclinicalrehabilitationapplications.WDΘZ͗ƚŚĞ ũŽƵƌŶĂůŽĨŝŶũƵƌǇ͕ĨƵŶĐƚŝŽŶ͕ĂŶĚƌĞŚĂďŝůŝƚĂƚŝŽŶ.2009;1:859Ͳ72. 13. Wilcock IM, Cronin JB, Hing WA. Physiological response to water immersion: a method for sport recovery?^ƉŽƌƚƐDĞĚŝĐŝŶĞ.2006;36:747Ͳ65. 14. WallerB,OgonowskaͲSlodownikA,VitorM,LambeckJ,DalyD,KujalaUM,etal.Effectoftherapeutic aquatic exercise on symptoms and function associated with lower limb osteoarthritis: systematic reviewwithmetaͲanalysis.WŚǇƐŝĐĂůdŚĞƌĂƉǇ.2014;94:1383Ͳ95. 15. LuM,SuY,ZhangY,ZhangZ,WangW,HeZ,etal.Effectivenessofaquaticexercisefortreatmentof knee osteoarthritis : Systematic review and metaͲanalysis. ĞŝƚƐĐŚƌŝĨƚ Ĩƺƌ ZŚĞƵŵĂƚŽůŽŐŝĞ. 2015;10.1007/s00393Ͳ014Ͳ1559Ͳ9. 16. Barker AL, Talevski J, Morello RT, Brand CA, Rahmann AE, Urquhart DM. Effectiveness of aquatic exercise for musculoskeletal conditions: a metaͲanalysis. ƌĐŚŝǀĞƐ ŽĨ WŚǇƐŝĐĂů DĞĚŝĐŝŶĞ ĂŶĚ ZĞŚĂďŝůŝƚĂƚŝŽŶ.2014;95:1776Ͳ86. 17. FiskenA,WatersDL,HingWA,SteeleM,KeoghJW.Perceptionandresponsestodifferentformsof aquaͲbasedexerciseamongolderadultswithosteoarthritis./ŶƚĞƌŶĂƚŝŽŶĂů:ŽƵƌŶĂůŽĨƋƵĂƚŝĐZĞƐĞĂƌĐŚ ĂŶĚĚƵĐĂƚŝŽŶ.2014;8:32Ͳ52. 18. Schoedinger P. Watsu In Aquatic Rehabilitation. In: Becker Bruce E, Cole Andrew J, editors. Comprehensiveaquatictherapy137Ͳ51.3rded:WashingtionStateUniversityPublishing;2010:137Ͳ51. 19. MangioneK,McCullyK,GloviakA,LefebvreI,HofmannM,CraikR.TheeffectsofhighͲintensityand lowͲintensitycycleergometryinolderadultswithkneeosteoarthritis.:ŽƵƌŶĂůƐŽĨ'ĞƌŽŶƚŽůŽŐǇ^ĞƌŝĞƐ͗ ŝŽůŽŐŝĐĂů^ĐŝĞŶĐĞƐĂŶĚDĞĚŝĐĂů^ĐŝĞŶĐĞƐ.1999;54:M:184Ͳ90.

57 Chapter3

20. SalacinskiA,KrohnK,LewisS,HollandM,IrelandK,MarchettiG.Theeffectsofgroupcyclingongait and painͲrelated disability in individuals with mildͲtoͲmoderate knee osteoarthritis: a randomized controlledtrial.:ŽƵƌŶĂůŽĨKƌƚŚŽƉĂĞĚŝĐĂŶĚ^ƉŽƌƚƐWŚǇƐŝĐĂůdŚĞƌĂƉǇ.2012;42:985Ͳ95. 21. UlatkowskiM.UnterwasserfahrradversusherkömmlicheRehabilitationͲEineretrospektiveStudiemit und ohne Unterwasserfahrrad an implantierten Kniegelenksprothesen [Doctoral Dissertation]. Heidelberg:HeidelbergUniversity;2009. 22. von Kathen M. ProspektivͲrandomisierte Vergleichsstudie zur Rehabilitation vorderer Kreunzbandplastiken zwischen koventioneller Therapie und Unterwasserfahrrad [Doctoral Dissertation].Bochum:RuhrͲUniversität;1999. 23. MoserS.EntwicklungundÜberprüfungeinesAquaͲCyclingͲProgrammsfürRheumapatienten[Master Thesis].Karlsruhe:UniversitätKarlsruhe;2009. 24. RewaldS,MestersI,EmansPJ,ArtsJJ,LenssenAF,deBieRA.AquaticcircuittrainingincludingaquaͲ cycling in patients with knee osteoarthritis: A feasibility study. :ŽƵƌŶĂů ŽĨ ZĞŚĂďŝůŝƚĂƚŝŽŶ DĞĚŝĐŝŶĞ. 2015;10.2340/16501977Ͳ1937. 25. ChanAW,TetzlaffJM,AltmanDG,LaupacisA,GotzschePC,KrlezaͲJericK,etal.SPIRIT2013statement: definingstandardprotocolitemsforclinicaltrials.ŶŶĂůƐŽĨ/ŶƚĞƌŶĂůDĞĚŝĐŝŶĞ.2013;158:200Ͳ7. 26. SchulzKF,AltmanDG,MoherD.CONSORT2010statement:updatedguidelinesforreportingparallel grouprandomisedtrials.D:.2010;340:c332. 27. Ackerman IN, Buchbinder R, Osborne RH. Factors limiting participation in arthritis selfͲmanagement programmes:anexplorationofbarriersandpatientpreferenceswithinarandomizedcontrolledtrial. ZŚĞƵŵĂƚŽůŽŐǇ.2013;52:472Ͳ9. 28. CardinalBJ,EstersJ,CardinalMK.Evaluationoftherevisedphysicalactivityreadinessquestionnairein olderadults.DĞĚŝĐŝŶĞĂŶĚ^ĐŝĞŶĐĞŝŶ^ƉŽƌƚƐĂŶĚǆĞƌĐŝƐĞ.1996;28:468Ͳ72. 29. AxfordJ,ButtA,HeronC,HammondJ,MorganJ,AlaviA,etal.Prevalenceofanxietyanddepressionin osteoarthritis: use of the Hospital Anxiety and Depression Scale as a screening tool. ůŝŶŝĐĂů ZŚĞƵŵĂƚŽůŽŐǇ.2010;29:1277Ͳ83. 30. CampbellR,EvansM,TuckerM,QuiltyB,DieppeP,DonovanJL.Whydon'tpatientsdotheirexercises? UnderstandingnonͲcompliancewithphysiotherapyinpatientswithosteoarthritisoftheknee.:ŽƵƌŶĂů ŽĨƉŝĚĞŵŝŽůŽŐǇĂŶĚŽŵŵƵŶŝƚǇ,ĞĂůƚŚ.2001;55:132Ͳ8. 31. MarksR.Kneeosteoarthritisandexerciseadherence:areview.ƵƌƌĞŶƚĂŐŝŶŐƐĐŝĞŶĐĞ.2012;5:72Ͳ83. 32. VanderLeedenM,StaalJB,BeekmanE,HendriksH,MestersI,deRooijM,etal.Developmentofa frameworktodescribegoalsandcontentofexerciseinterventionsinphysicaltherapy:amixedmethod approachincludingasystematicreview.WŚǇƐŝĐĂůdŚĞƌĂƉǇZĞǀŝĞǁƐ.2014;19:1Ͳ14. 33. Exerciseprescriptionforolderadultswithosteoarthritispain:consensuspracticerecommendations.A supplementtotheAGSClinicalPracticeGuidelinesonthemanagementofchronicpaininolderadults. :ŽƵƌŶĂůŽĨƚŚĞŵĞƌŝĐĂŶ'ĞƌŝĂƚƌŝĐƐ^ŽĐŝĞƚǇ.2001;49:808Ͳ23. 34. Ericson MO BA, Nisell R, Nemeth R, Ekholm J. Load moments about the hip and knee joints during ergometercycling.^ĐĂŶĚŝŶĂǀŝĂŶ:ŽƵƌŶĂůŽĨZĞŚĂďŝůŝƚĂƚŝŽŶDĞĚŝĐŝŶĞ.1986;18:165Ͳ72. 35. Silva LE, Valim V, Pessanha APC. Hydrotherapy versus conventional landͲbased exercise for the managementofpatientswithosteoarthritisoftheknee:arandomizedclinicaltrial.WŚǇƐŝĐĂůdŚĞƌĂƉǇ. 2008;88:12Ͳ21. 36. HinmanRS,HeywoodSE,DayAR.Aquaticphysicaltherapyforhipandkneeosteoarthritis:resultsofa singleͲblindrandomizedcontrolledtrial.WŚǇƐŝĐĂůdŚĞƌĂƉǇ.2007;87:32Ͳ43. 37. FoleyA,HalbertJ,HewittT,CrottyM.Doeshydrotherapyimprovestrengthandphysicalfunctionin patientswithosteoarthritisͲͲarandomisedcontrolledtrialcomparingagymbasedandahydrotherapy basedstrengtheningprogramme.ŶŶĂůƐŽĨƚŚĞZŚĞƵŵĂƚŝĐŝƐĞĂƐĞƐ.2003;62:1162Ͳ7. 38. Valtonen A, Poyhonen T, Sipila S, Heinonen A. Effects of aquatic resistance training on mobility limitation and lowerͲlimb impairments after knee replacement. ƌĐŚŝǀĞƐ ŽĨ WŚǇƐŝĐĂů DĞĚŝĐŝŶĞ ĂŶĚ ZĞŚĂďŝůŝƚĂƚŝŽŶ.2010;91:833Ͳ9. 39. Dalichau S, Scheele K. Status of aquatic functional training in the therapy of chronic back pain. WŚǇƐŝŬĂůŝƐĐŚĞDĞĚŝǌŝŶZĞŚĂďŝůŝƚĂƚŝŽŶƐŵĞĚŝǌŝŶ<ƵƌŽƌƚŵĞĚŝǌŝŶ.2003;13:35Ͳ41. 40. AssociationAE.Aquaticfitnessprofessionalmanual.6thed.Leeds:HumanKinetics;2010.

58 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA

41. JohnstonTE,BarrAE,LeeSC.Biomechanicsofsubmaximalrecumbentcyclinginadolescentswithand withoutcerebralpalsy.WŚǇƐŝĐĂůdŚĞƌĂƉǇ.2007;87:572Ͳ85. 42. Li L, Caldwell GE. Muscle coordination in cycling: effect of surface incline and posture. :ŽƵƌŶĂů ŽĨ ƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1998;85:927Ͳ34. 43. CariaMA,TangianuF,ConcuA,CrisafulliA,MameliO.QuantificationofSpinningbike performance duringastandard50Ͳminuteclass.:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐ^ĐŝĞŶĐĞƐ.2007;25:421Ͳ9. 44. BarbosaTM,MarinhoDA,ReisVM,SilvaAJ,BragadaJ.PhysiologicalassessmentofheadͲoutaquatic exercisesinhealthysubjects:aqualitativereview.:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐ^ĐŝĞŶĐĞΘDĞĚŝĐŝŶĞ.2009;8:179Ͳ89. 45. BarkerKL,DawesH,HansfordP,ShamleyD.Perceivedandmeasuredlevelsofexertionofpatientswith 3 chronicbackpainexercisinginahydrotherapypool.ƌĐŚŝǀĞƐŽĨWŚǇƐŝĐĂůDĞĚŝĐŝŶĞĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ. 2003;84:1319Ͳ23. 46. KökeAJA,vandenEndeCHM,JansenMJ,SteultjensMPM,CV.Clinicalpracticeguidelineforaphysical activity interventions for patients with osteoarthritis [KNGFͲstandaard Beweeginterventie artrose]. Amersfoort:KoninklijkNederlandsGenootschapvoorFysiotherapie(KNGF);2008. 47. BennellKL,HinmanRS.Areviewoftheclinicalevidenceforexerciseinosteoarthritisofthehipand knee.:ŽƵƌŶĂůŽĨ^ĐŝĞŶĐĞĂŶĚDĞĚŝĐŝŶĞŝŶ^ƉŽƌƚ.2011;14:4Ͳ9. 48. BellamyN,BuchananWW,GoldsmithCH,CampbellJ,StittLW.ValidationstudyofWOMAC:ahealth statusinstrumentformeasuringclinicallyimportantpatientrelevantoutcomestoantirheumaticdrug therapyinpatientswithosteoarthritisofthehiporknee.dŚĞ:ŽƵƌŶĂůŽĨZŚĞƵŵĂƚŽůŽŐǇ.1988;15:1833Ͳ 40. 49. de Groot I, Favejee M, Reijman M, Verhaar J, Terwee C. The Dutch version of the Knee Injury and OsteoarthritisOutcomeScore:avalidationstudy.,ĞĂůƚŚĂŶĚYƵĂůŝƚǇŽĨ>ŝĨĞKƵƚĐŽŵĞƐ.2008;6:16. 50. RoosEM,LohmanderLS.TheKneeinjuryandOsteoarthritisOutcomeScore(KOOS):fromjointinjury toosteoarthritis.,ĞĂůƚŚĂŶĚYƵĂůŝƚǇŽĨ>ŝĨĞKƵƚĐŽŵĞƐ.2003;1:64. 51. Binkley JM, Stratford PW, Lott SA, Riddle DL. The Lower Extremity Functional Scale (LEFS): scale development, measurement properties, and clinical application. North American Orthopaedic RehabilitationResearchNetwork.WŚǇƐŝĐĂůdŚĞƌĂƉǇ.1999;79:371Ͳ83. 52. HoogeboomTJ,deBieRA,denBroederAA,vandenEndeCH.TheDutchLowerExtremityFunctional Scalewashighlyreliable,validandresponsiveinindividualswithhip/kneeosteoarthritis:avalidation study.DDƵƐĐƵůŽƐŬĞůĞƚĂůŝƐŽƌĚĞƌƐ.2012;13:117. 53. BellamyN,KirwanJ,BoersM,BrooksP,StrandV,TugwellP,etal.Recommendationsforacoresetof outcomemeasuresforfuturephaseIIIclinicaltrialsinknee,hip,andhandosteoarthritis.Consensus developmentatOMERACTIII.:ŽƵƌŶĂůŽĨZŚĞƵŵĂƚŽůŽŐǇ.1997;24:799Ͳ802. 54. Perrot S, Rozenberg S, Moyse D, Legout V, Marty M. Comparison of daily, weekly or monthly pain assessmentsinhipandkneeosteoarthritis.A29Ͳdayprospectivestudy.:ŽŝŶƚ͕ŽŶĞ͕^ƉŝŶĞ͗ZĞǀƵĞĚƵ ZŚƵŵĂƚŝƐŵĞ.2011;78:510Ͳ5. 55. PriceD,BushF,LongS,HarkinsS.Acomparisonofpainmeasurementcharacteristicsofmechanical visualanalogueandsimplenumericalratingscales.WĂŝŶ.1994;56:217Ͳ26. 56. GallaschCH,AlexandreNM.Themeasurementofmusculoskeletalpainintensity:acomparisonoffour methods.ZĞǀŝƐƚĂ'ĂƷĐŚĂĚĞŶĨĞƌŵĂŐĞŵ.2007;28:260Ͳ5. 57. PhamT,VanDerHeijdeD,LassereM,AltmanRD,AndersonJJ,BellamyN,etal.Outcomevariablesfor osteoarthritisclinicaltrials:TheOMERACTͲOARSIsetofrespondercriteria.:ŽƵƌŶĂůŽĨZŚĞƵŵĂƚŽůŽŐǇ. 2003;30:1648Ͳ54. 58. PincusT,BergmanM,SokkaT,RothJ,SwearingenC,YaziciY.Visualanalogscalesinformatsotherthan a 10 centimeter horizontal line to assess pain and other clinical data. :ŽƵƌŶĂů ŽĨ ZŚĞƵŵĂƚŽůŽŐǇ. 2008;35:1550Ͳ8. 59. Carpenter MR, Carpenter RL, Peel J, Zukley LM, Angelopoulou KM, Fischer I, et al. The reliability of isokineticandisometriclegstrengthmeasuresamongindividualswithsymptomsofmildosteoarthritis. :ŽƵƌŶĂůŽĨ^ƉŽƌƚƐDĞĚŝĐŝŶĞĂŶĚWŚǇƐŝĐĂů&ŝƚŶĞƐƐ.2006;46:585Ͳ9. 60. StratfordPW,KennedyDM.Performancemeasureswerenecessarytoobtainacompletepictureof osteoarthriticpatients.:ŽƵƌŶĂůŽĨůŝŶŝĐĂůƉŝĚĞŵŝŽůŽŐǇ.2006;59:160Ͳ7.

59 Chapter3

61. PeterWF,JansenMJ,HurkmansEJ,BlooH,DekkerJ,DillingRG,etal.Physiotherapyinhipandknee osteoarthritis: development of a practice guideline concerning initial assessment, treatment and evaluation.ĐƚĂƌĞƵŵĂƚŽůſŐŝĐĂƉŽƌƚƵŐƵĞƐĂ^ŽĐŝĞĚĂĚĞWŽƌƚƵŐƵĞƐĂĚĞZĞƵŵĂƚŽůŽŐŝĂ.2011;36:268Ͳ81. 62. NorenAM,BogrenU,BolinJ,StenstromC.Balanceassessmentinpatientswithperipheralarthritis: applicability and reliability of some clinical assessments. WŚǇƐŝŽƚŚĞƌĂƉǇ ZĞƐĞĂƌĐŚ /ŶƚĞƌŶĂƚŝŽŶĂů. 2001;6:193Ͳ204. 63. Kennedy DM, Stratford PW, Wessel J, Gollish JD, Penney D. Assessing stability and change of four performance measures: a longitudinal study evaluating outcome following total hip and knee arthroplasty.DDƵƐĐƵůŽƐŬĞůĞƚĂůŝƐŽƌĚĞƌƐ.2005;6:3. 64. WrightAA,CookCE,BaxterGD,DockertyJD,AbbottJH.Acomparisonof3methodologicalapproaches todefining major clinicallyimportantimprovement of4 performance measures in patientswithhip osteoarthritis.:ŽƵƌŶĂůŽĨKƌƚŚŽƉĂĞĚŝĐĂŶĚ^ƉŽƌƚƐWŚǇƐŝĐĂůdŚĞƌĂƉǇ.2011;41:319Ͳ27. 65. Podsiadlo D, Richardson S. The timed "Up & Go": a test of basic functional mobility for frail elderly persons.:ŽƵƌŶĂůŽĨƚŚĞŵĞƌŝĐĂŶ'ĞƌŝĂƚƌŝĐƐ^ŽĐŝĞƚǇ.1991;39:142Ͳ8. 66. FreterSH,FruchterN.Relationshipbetweentimed'upandgo'andgaittimeinanelderlyorthopaedic rehabilitationpopulation.ůŝŶŝĐĂůZĞŚĂďŝůŝƚĂƚŝŽŶ.2000;14:96Ͳ101. 67. ButlandRJ,PangJ,GrossER,WoodcockAA,GeddesDM.TwoͲ,sixͲ,and12Ͳminutewalkingtestsin respiratorydisease.ƌŝƚŝƐŚDĞĚŝĐĂů:ŽƵƌŶĂů;ůŝŶŝĐĂůZĞƐĞĂƌĐŚĚͿ.1982;284:1607Ͳ8. 68. ATSStatement:GuidelinesfortheSixͲMinuteWalkTest.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨZĞƐƉŝƌĂƚŽƌǇĂŶĚƌŝƚŝĐĂů ĂƌĞDĞĚŝĐŝŶĞ.2002;166:111Ͳ7. 69. Pankoff BA, Overend TJ, Lucy SD, White KP. Reliability of the sixͲminute walk test in people with fibromyalgia.ƌƚŚƌŝƚŝƐĂƌĞĂŶĚZĞƐĞĂƌĐŚ.2000;13:291Ͳ5. 70. FaggianoP,D'AloiaA,GualeniA,LavatelliA,GiordanoA.Assessmentofoxygenuptakeduringthe6Ͳ minute walking test in patients with heart failure: preliminary experience with a portable device. ŵĞƌŝĐĂŶ,ĞĂƌƚ:ŽƵƌŶĂů.1997;134:203Ͳ6. 71. WendelͲVos GC, Schuit AJ, Saris WH, Kromhout D. Reproducibility and relative validity of the short questionnaire to assess healthͲenhancing physical activity. :ŽƵƌŶĂů ŽĨ ůŝŶŝĐĂů ƉŝĚĞŵŝŽůŽŐǇ. 2003;56:1163Ͳ9. 72. deHollanderEL,ZwartL,deVriesSI,WendelͲVosW.TheSQUASHwasamorevalidtoolthantheOBiN forcategorizingadultsaccordingtotheDutchphysicalactivityandthecombinedguideline.:ŽƵƌŶĂůŽĨ ůŝŶŝĐĂůƉŝĚĞŵŝŽůŽŐǇ.2012;65:73Ͳ81. 73. Wagenmakers R, van den AkkerͲScheek I, Groothoff JW, Zijlstra W, Bulstra SK, Kootstra JW, et al. ReliabilityandvalidityoftheshortquestionnairetoassesshealthͲenhancingphysicalactivity(SQUASH) inpatientsaftertotalhiparthroplasty.DDƵƐĐƵůŽƐŬĞůĞƚĂůŝƐŽƌĚĞƌƐ.2008;9:141. 74. Ware JE, Jr., Sherbourne CD. The MOS 36Ͳitem shortͲform health survey (SFͲ36). I. Conceptual frameworkanditemselection.DĞĚŝĐĂůĂƌĞ.1992;30:473Ͳ83. 75. AngstF,AeschlimannA,SteinerW,StuckiG.Responsiveness oftheWOMAC osteoarthritisindexas compared with the SFͲ36 in patients with osteoarthritis of the legs undergoing a comprehensive rehabilitationintervention.ŶŶĂůƐŽĨƚŚĞZŚĞƵŵĂƚŝĐŝƐĞĂƐĞƐ.2001;60:834Ͳ40. 76. KosinskiM,KellerSD,WareJE,Jr.,HatoumHT,KongSX.TheSFͲ36HealthSurveyasagenericoutcome measure in clinical trials of patients with osteoarthritis and rheumatoid arthritis: relative validity of scalesinrelationtoclinicalmeasuresofarthritisseverity.DĞĚŝĐĂůĂƌĞ.1999;37:MS23Ͳ39. 77. VeenhofC,BijlsmaJW,vandenEndeCH,vanDijkGM,PistersMF,DekkerJ.Psychometricevaluationof osteoarthritis questionnaires: a systematic review of the literature. ƌƚŚƌŝƚŝƐ ĂŶĚ ZŚĞƵŵĂƚŝƐŵ. 2006;55:480Ͳ92. 78. Aaronson NK, Muller M, Cohen PD, EssinkͲBot ML, Fekkes M, Sanderman R, et al. Translation, validation,andnormingoftheDutchlanguageversionoftheSFͲ36HealthSurveyincommunityand chronicdiseasepopulations.:ŽƵƌŶĂůŽĨůŝŶŝĐĂůƉŝĚĞŵŝŽůŽŐǇ.1998;51:1055Ͳ68. 79. ZeevanderKI,SandermanR.HetmetenvandealgemenegezondheidstoestandmetdeRANDͲ36,een handleiding. Groningen: Rijksuniversiteit Groningen, Noordelijk Centrum voor GezondheidsvraagͲ stukken;1992. 80. KoriS,MillerR,ToddC.Kinesiophobia:anewviewofchronicpainbehavior.WĂŝŶDĂŶĂŐ.1990.

60 AquaticcyclingonpainandphysicalfunctioningcomparedwithusualcareinpatientswithkneeOA

81. SwinkelsͲMeewisseEJ,SwinkelsRA,VerbeekAL,VlaeyenJW,OostendorpRA.Psychometricproperties oftheTampaScaleforkinesiophobiaandthefearͲavoidancebeliefsquestionnaireinacutelowback pain.DĂŶƵĂůdŚĞƌĂƉǇ.2003;8:29Ͳ36. 82. MielenzTJ,EdwardsMC,CallahanLF.ItemResponseTheoryAnalysisofTwoQuestionnaireMeasures of ArthritisͲRelated SelfͲEfficacy Beliefs from CommunityͲBased US Samples. ƌƚŚƌŝƚŝƐ. 2010;2010: 416796. 83. LorigK,ChastainRL,UngE,ShoorS,HolmanHR.Developmentandevaluationofascaletomeasure perceivedselfͲefficacyinpeoplewitharthritis.ƌƚŚƌŝƚŝƐĂŶĚZŚĞƵŵĂƚŝƐŵ.1989;32:37Ͳ44. 84. Taal E, Riemsma RP, Brus HL, Seydel ER, Rasker JJ, Wiegman O. Group education for patients with 3 rheumatoidarthritis.WĂƚŝĞŶƚĚƵĐĂƚŝŽŶĂŶĚŽƵŶƐĞůŝŶŐ.1993;20:177Ͳ87. 85. Brady TJ. Measures of selfͲefficacy: Arthritis SelfͲEfficacy Scale (ASES), Arthritis SelfͲEfficacy ScaleͲ8 Item(ASESͲ8),Children'sArthritisSelfͲEfficacyScale(CASE),ChronicDiseaseSelfͲEfficacyScale(CDSES), Parent's Arthritis SelfͲEfficacy Scale (PASE), and Rheumatoid Arthritis SelfͲEfficacy Scale (RASE). ƌƚŚƌŝƚŝƐĂƌĞΘZĞƐĞĂƌĐŚ.2011;63Suppl11:S473Ͳ85. 86. DekkerJ,vanDijkGM,VeenhofC.Riskfactorsforfunctionaldeclineinosteoarthritisofthehiporknee. ƵƌƌĞŶƚKƉŝŶŝŽŶŝŶZŚĞƵŵĂƚŽůŽŐǇ.2009;21:520Ͳ4. 87. AllenKD,GolightlyYM,OlsenMK.Pilotstudyofpainandcopingamongpatientswithosteoarthritis:a daily diary analysis. :ŽƵƌŶĂů ŽĨ ůŝŶŝĐĂů ZŚĞƵŵĂƚŽůŽŐǇ ͗ WƌĂĐƚŝĐĂů ZĞƉŽƌƚƐ ŽŶ ZŚĞƵŵĂƚŝĐ Θ DƵƐĐƵůŽƐŬĞůĞƚĂůŝƐĞĂƐĞƐ.2006;12:118Ͳ23. 88. FransenM,McConnellS,HarmerAR,VanderEschM,SimicM,BennellKL.Exerciseforosteoarthritisof theknee:aCochranesystematicreview.ƌŝƚŝƐŚ:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐDĞĚŝĐŝŶĞ.2015;49:1554Ͳ7. 89. MaiersM,EvansR,HartvigsenJ,SchulzC,BronfortG.Adverseeventsamongseniorsreceivingspinal manipulationandexerciseinarandomizedclinicaltrial.DĂŶƵĂůdŚĞƌĂƉǇ.2015;20:335Ͳ41. 90. ICH harmonized tripartite guideline: Guideline for Good Clinical Practice. :ŽƵƌŶĂů ŽĨ WŽƐƚŐƌĂĚƵĂƚĞ DĞĚŝĐŝŶĞ.2001;47:45Ͳ50. 91. King DW, Lashley R. A quantifiable alternative to double data entry. ŽŶƚƌŽůůĞĚ ůŝŶŝĐĂů dƌŝĂůƐ. 2000;21:94Ͳ102. 92. DodgeHF.Samplingplansforcontinuousproduction./ŶĚƵƐƚƌŝĂůYƵĂůŝƚǇŽŶƚƌŽů.1947;14:5Ͳ9. 93. TubachF,RavaudP,BaronG,FalissardB,LogeartI,BellamyN,etal.Evaluationofclinicallyrelevant changesinpatientreportedoutcomesinkneeandhiposteoarthritis:theminimalclinicallyimportant improvement.ŶŶĂůƐŽĨƚŚĞZŚĞƵŵĂƚŝĐŝƐĞĂƐĞƐ.2005;64:29Ͳ33. 94. AngstF,AeschlimannA,StuckiG.Smallestdetectableandminimalclinicallyimportantdifferencesof rehabilitationinterventionwiththeirimplicationsforrequiredsamplesizesusingWOMACandSFͲ36 quality of life measurement instruments in patients with osteoarthritis of the lower extremities. ƌƚŚƌŝƚŝƐĂŶĚZŚĞƵŵĂƚŝƐŵ.2001;45:384Ͳ91. 95. EttingerWH,Jr.,BurnsR,MessierSP,ApplegateW,RejeskiWJ,MorganT,etal.Arandomizedtrial comparingaerobicexerciseandresistanceexercisewithahealtheducationprograminolderadults with knee osteoarthritis. The Fitness Arthritis and Seniors Trial (FAST). :D͗ƚŚĞũŽƵƌŶĂůŽĨƚŚĞ ŵĞƌŝĐĂŶDĞĚŝĐĂůƐƐŽĐŝĂƚŝŽŶ.1997;277:25Ͳ31. 96. BresselE,WingJE,MillerAI,DolnyDG.HighͲintensityintervaltrainingonanaquatictreadmillinadults withosteoarthritis:effectonpain,balance,function,andmobility.:ŽƵƌŶĂůŽĨ^ƚƌĞŶŐƚŚĂŶĚŽŶĚŝƚŝŽŶŝŶŐ ZĞƐĞĂƌĐŚ.2014;28:2088Ͳ96.

61 Chapter3

62 Chapter3

CHAPTER4

Aquaticcycling–whatdoweknow? AscopingreviewonheadͲoutaquaticcycling

StefanieRewald IlseMesters A.F.TonLenssen JensBansi JohanLambeck RobA.deBie BenjaminWaller W>Ž^KŶĞ͘ϮϬϭϳ͖ϭϮ;ϱͿ͗ĞϬϭϳϳϳϬϰ

62 Chapter4

ABSTRACT

Over the past few years, aquatic cycling has become a trending fitness activity. However, the literaturehasnotbeenreviewedexhaustively.Therefore,usingscopingreviewmethodology,the aimofthisreviewwastoexplorethecurrentstateoftheliteratureconcerningaquaticcycling. This study specifically focused on study designs, populations and outcomes. A comprehensive search of seven databases (PubMed, MEDLINE, Cinahl, Embase, PEDro,Web of Science, WorldCat)wasconductedupto30thSeptember2016.GoogleScholar,WorldCat,ResearchGate, specific aquatic therapy websites and aquatic therapy journals were searched to identify additionalliterature.FullͲtextpublicationsinEnglish,GermanorDutchwereincluded.Studies wereincludedwhentheinterventioninvolvedheadͲoutcyclingcarriedoutin10°to35°Celsius water.Exclusioncriteriaweretheuseofwetsuitsorconfoundinginterventionsthatwouldaffect participants’homeostasis.63articleswereincludedandthestudyparametersofthesestudies weresummarised.Usingthreegroupingthemes,includedstudieswerecategorisedas1)single session tests comparing aquatic versus land cycling, or 2) aquatic cycling only sessions investigating different exercise conditions and 3) aquatic cycling intervention programmes. Although the experimental conditions differed noticeably across the studies, shared characteristics were identified. Cardiovascular parameters were investigated by many of the studieswiththeresultssuggestingthatthecardiacdemandofaquaticcyclingseemssimilarto landͲbased cycling. Only six studies evaluated the effect of aquatic cycling interventions. Therefore, future research should investigate the effects of aquatic cycling interventions, preferably in individuals that are expected to gain health benefits from aquatic cycling. Moreover,thiscomprehensiveoutlineofavailableliteraturecouldserveasastartingpointfor systematic reviews or clinical studies on the effects of aquatic cycling on the cardiovascular responses.

64 Aquaticcycling–whatdoweknow?

INTRODUCTION

WaterͲbasedfitnessequipmenthasgainedpopularitywithinaquaticfitnessleadingto a development of dryland training machines, such as stationary exercise bikes and treadmills, into waterͲproof exercise gear. Although aquatic cycling has become a trending fitness activity, the modification of standard ergometer bicycles for aquatic programs is nothing new and stems from the late sixties. Researchers used water immersion as an effective simulation of prolonged weightlessness, moreover, the utilizationoftheaquaticenvironmenthasbeenrecognizedasusefulinrehabilitation.1,2 SimilartolandͲbasedcycling,therepetitivecircularmovementofpedallingagainstthe 4 waterresistanceensuresauseofalargerangeofmotion(ROM)ofthelowerlimbsto improvecardiovascularfitnessandmusclestrength.Thefactthatindividualsaresitting ontheaquaticbikecanbebeneficialforthosewhohaveproblemswithbalanceand independent gait. However, in contrast, while the sitting position and hydrostatic pressureassistwithposturalcontrol,thelossoffreemovementi.e.reducedchallenges to balance, and the few variation of the exercises may limit its effect on functional capacity.Asharedcharacteristicwithothertypesofaquaticexerciseisthedecreaseof jointloadingduetothebuoyancyofthewater.Duringaquaticcyclingparticipantsare immersedinwateruptothechestandthebuoyancyofthewaterunloadsthejointsof the lower extremities and the lower spine, a condition appealing for patients experiencing pain or problems with physical functioning during exercising on land.3,4 Despitethepotentialbenefitsofaquaticcyclinganditslonghistory,theapplicationof aquaticcyclinginanexerciseandclinicalcontextstillappearstobelow.Limitations that might prevent clinicians using aquatic cycling for therapeutic purposes could includetheinvestmentcosts,storagespacerequirements,andtheelaboratesetͲupof the aquatic bikes. In particular, getting the bikes in and out of the pool, without an adjustablefloor,isdemanding. Thescientificevidenceonthepotentialbenefitsofaquaticcyclingseemstobescarce aswell.Obvioussearchtermslikeaqua(tic)cycling,aqua(tic)bikeorwatercyclingyield veryfewrelevantresultsfromscientificsearchengines.Moreover,thesmallnumberof references about aquatic cycling, used in previously published reviews on aquatic exercise,furtheremphasizestheimpressionofascarcityofliterature.5Ͳ7Thesereviews summarize the effects on headͲout aquatic exercise, including aquatic cycling, or comparedphysiologicalresponsesofdifferenttypesofaquaticexerciseandswimming with each other.5Ͳ7 Further, the aquatic cycling interventions were not described in detailinthesepriorreviewswiththesereviewsonlyincludingcrossͲoverstudies. Thus, the questions remain how has aquatic cycling been investigated in previous research, and whether a search effort solely on “aquatic cycling” would reveal additional publications and research investigating the effects of aquatic cycling interventionprogrammes.AsystematicreviewwithametaͲanalysiswouldnotsuitthis aimandthereforeascopingreviewstudydesignwaschosen.Systematicreviewsare

65 Chapter4 guidedbyspecificresearchquestionsleadingtostrictinͲandexclusioncriteria.The primary aim for performing a scoping review is to map the available literature that meet a comprehensive research question combined without restricting inclusion criteria.8 Where systematic reviews evolve out of an initial understanding of the research field, scoping reviews are employed to identify research and explore their featuressuchastargetpopulations,interventions,studydesignsandoutcomes.8,9Asa result scoping reviews help to develop an understanding of the extent and possible gapsanduncertaintiesintheexistingliterature.Furthermore,ascopingreviewmight identifyasufficientamountofstudiesthatwouldfacilitateasystematicreview.9 Therefore, the main objective of this study was to identify the scope of available researchwithregardtoaquaticcyclingasanexerciseactivity.Specifically,thisscoping review aimed to explore the aquatic cycling exercises, study designs, comparison of training effects (if applicable), populations and outcomes utilised in research investigating aquatic cycling. To enable a comprehensive coverage of available literature the following research question was formulated: What is the available researchonheadͲoutaquaticcyclingexercise?

METHODS

Frameworkofascopingreview Theprocedureofperformingascopingreviewfollowssimilarstepsasthoseusedin systematicreviewapproacheswithoutlimitingforstudydesignofincludedstudiesand without a quantitative synthesis. The framework of Arksey and O’Malley for scoping reviewswasimplementedinthisstudy.9Theframeworkconsistsoffiveessentialstages andoneadditionalstage;1)identifyingtheresearchquestion,2)identifyingrelevant studies,3)studyselection,4)chartingthedata,5)collating,summarizingandreporting the results, and additionally 6) consultation of experts (optional). All stages can be performedinaniterativemannerallowingrefiningofsearchparameters.

Identifyingrelevantstudies AcomprehensiveliteraturesearchwasconductedinAugust2015andupdatedto30th September2016insevenelectronicacademicdatabases(PubMed,MEDLINE,Cinahl, Embase,PEDro,WebofScience,WorldCat).Thesearchstrategywasdocumentedby titleofthedatabasesearched,dateofthesearch,thecompletesearchstringthatwas usedandthenumberofarticlesfound(Table4.1).Thedevelopmentofeachsearch string was an iterative process and familiarisation with the literature revealed additionalsearchtermsforaquaticcyclingsuchas“immersedcycling”or“underwater pedalling”.Thesetermswerecombinedwithmoregeneraltermsforaquatictherapy

66 Aquaticcycling–whatdoweknow?

(e.g. hydrotherapy) the search included the following key terms: ergometer, immersion, hydrotherapy, aqua(tic), cycling, underwater (bi)cycle ergometer, immersedergocycle. Table4.1 Searchstrategyandresults Database Date Searchstring Results PubMed 30Ͳ09Ͳ16 ((ergometer[AllFields]AND(("immersion"[MeSHTerms]OR 120 "immersion"[AllFields]OR"underwater"[AllFields]OR"aquatic"[All Fields])OR("hydrotherapy"[MeSHTerms]OR"hydrotherapy"[All Fields])))OR((aqua[AllFields]ANDcycling[AllFields])OR"underwater bicycleergometer"[AllFields]OR"underwatercycleergometer"[All 4 Fields]OR"immersedergocycle"[AllFields]OR"aquaticbike"[All Fields]OR"waterbike"[AllFields]))AND"humans"[MeSHTerms] MEDLINE 30Ͳ09Ͳ16 1.((cyclingand(hydrotherapyoraquaticexerciseoraquatictherapyor 157 waterexerciseorimmersion))or(aquacyclingorunderwaterbikeor aquaticbikeorimmersedergocycleorunderwaterbicycleergometer orunderwatercycleergometerorunderwaterpedallingor underwatercyclingorwaterbike)).af. Cinahl 30Ͳ09Ͳ16 (TXergometerAND((aquatictherapyorhydrotherapyoraquatic 30 exerciseorwaterexercise)ORimmersion))ORunderwatercycle ergometerORimmersedergocycleORaquacyclingORunderwater pedallingORunderwaterbikeORaquaticbikeORwaterbikeORaqua bike) Embase 30Ͳ09Ͳ16 1.((cyclingand(hydrotherapyoraquaticexerciseoraquatictherapyor 194 waterexerciseorimmersion))or(aquacyclingorunderwaterbikeor aquaticbikeorimmersedergocycleorunderwaterbicycleergometer orunderwatercycleergometerorunderwaterpedallingor underwatercyclingorwaterbike)).af. PEDro 30Ͳ09Ͳ16 (ergometerANDimmersion) 14 (ergometerANDwaterexercise) (ergometerANDhydrotherapy) (aquaticbike)OR(aquabike)OR(waterbike)OR(underwaterbike) Webof 30Ͳ09Ͳ16 TS=(((ergometerAND(immersionORhydrotherapy))OR((aquaAND 145 Science cycling)ORunderwaterbicycleergometerORunderwatercycle ergometerORimmersedergocycleoraquaticbikeorunderwater pedalingoraquabikeorwaterbike))) Refinedby:webofsciencecategories:(sportsciencesorclinical neurologyorrehabilitationorphysiologyormultidisciplinarysciences ormedicineresearchexperimentalorendocrinologymetabolismor neurosciencesorsurgeryorrespiratorysystemorpublic environmentaloccupationalhealthormedicinegeneralinternalor rheumatologyoroncologyororthopedics) WorldCat 30Ͳ09Ͳ16 ti:aquacyclingOR((kw:immersionANDsu:aquaͲcycling)OR 5 (ergometerANDhydrotherapy)OR(aquabike)OR(aquaticbike)OR (waterbike)OR(underwaterbike)) Totalnumberofrecords 674 af,allfields;TX,text;TS,topic;ti,title;kw,keyword;su,subject Additionally, ResearchGate, GoogleScholar and relevant aquatic therapy websites (http://www.wcpt.org/apti, http://www.atri.org, https://www.aeawave.com) were

67 Chapter4 examined.Moreover,thetableofcontentsoftheaccessiblekeyjournals‘International JournalofAquaticResearchandEducation’and‘JournalofAquaticPhysicalTherapy’of the American Physical Therapy Association were checked for additional literature. Finally,referencelistsofallincludedarticleswerehandͲsearchedfornewarticlesand theauthorsofthispaper,allexpertsinthefieldofaquatictherapyandaquaticfitness, checked their own libraries for additional literature. The table of contents and reference lists were screened for the key words related to cycling and (immersion) exercise(testing)onlandandinwater.Throughoutthesearchprocessitwasnoticed thatnoconsistentterminologyexistswithregardtoaquaticcycling.Toensurethatthe search terms used were correct and complete, the terminology used in included articles was reͲevaluated. This postͲhoc analysis (Appendix 4.1) addressing the terminologyusedtodescribeaquaticcyclingconfirmedourchoiceofsearchterms.

Studyselection Theinclusionandexclusioncriteriaweredevelopedintwostages.Inphaseone,the authorsagreedtoincludeallformatsoffullͲtextreportsthatfocusedontheeffectsof headͲout aquatic cycling exercise on the human body (Table 4.2, stage one). After familiarisation with the literature the selection criteria were further specified (Table4.2,stagetwo).Ineachstepoftheselectionproceduretwoormorereviewers wereinvolvedandinclusiondiscrepanciesweresolvedbydiscussion.Screeningoftitles and abstracts was performed by two reviewers (BW and SR) with the online programme “Covidence” (Covidence systematic review software, Veritas Health Innovation, Melbourne, Australia, available at:www.covidence.org). Next, allauthors wereinvolvedwiththefullͲtextscreeningandallresultswereindependentlyimported into a Microsoft Excel file and compared after completion of the review process. Information on the twoͲstage development of the inclusion criteria is available in a supportingfile(Appendix4.2).

Chartingthedata DescriptivedatawereextractedintoMicrosoftExceltablesincludingnameofthefirst author,yearofpublication,primaryresearchquestion,samplesize,age,gender,health statusofparticipants,exerciseparameters,mainresultsreportedintheabstract,water temperatures,aquaticbikeusedandlevelofbodyimmersion.Informationoneffectsof restingimmersionwasnotdiscussedforthisreview,butmighthavebeenpartofthe experimental setͲup of the included studies. The tables were organised by the body position on the aquatic bike (upright and semiͲrecumbent), because physiological responses might vary with immersion level related to the body position on the ergometer.2 All tables include information on interventions with healthy participants and patients. If patients were involved, information on the disease characteristics is

68 Aquaticcycling–whatdoweknow? reportedinthetables.Articlesthatoriginatedfromthesamedataset,butfocusingon differentoutcomes,weresummarisedandrepresentedasonestudyinthetables,but referencesfromallstudiesareincludedtoaididentificationoftheseparatearticles. Table4.2 Twostageexpertconsensusoninclusionandexclusioncriteria INCLUSION StageI x FullͲtextarticlesormasterordoctoraltheseswritteninEnglish,Dutch,German x Mostofthefollowingisdescribed:intensity,durationofthesession,bodypositiononthe bicycle,watertemperature,andtypeofaquaticbikeused x EffectofheadͲoutaquaticcyclingonthehumanbodyisdescribed 4 StageII x ParticipantshavetobeseateduprightorsemiͲrecumbentduringimmersedexercise Theexercisinglimbhastobefullyimmersedinwater EXCLUSION StageI x FullͲbody(abovehead)immersionofparticipants x UseofselfͲcontainedunderwaterbreathingapparatus(SCUBA) StageII x Longdurationrestingimmersion(>30min)priortoexercise x Confoundinginterventionsthatwouldaffectparticipantshomeostasise.g.manipulationof participants’glucoseleveloroxygensaturation x Watertemperaturesbelow10°Corabove41°Cforrestingimmersionandwatertemperatures above35.5°Cforexerciseconditions x UseofwetͲsuits

RESULTS

Thesearchrevealed465potentialstudies.Afterscreeningofthetitlesandabstracts, 350 studies were excluded and the fullͲtext versions of 115 publications were read (Figure 4.1). Finally, 63 articles met the inclusion criteria. The reasons for exclusion during the fullͲtext screening and the references of these excluded articles are presentedinasupportingfile(Appendix4.3).Nevertheless,someofthesepublications mightcontainusefulinformationandwerethereforeusedassupportiveliterature.All included articles were published in peerͲreviewed journals. Three of the included articleswerepublishedinGermanwithanEnglishabstract.10Ͳ12 Theincludedarticleswerecategorisedinthreegroupsaccordingtotheintervention characteristics. The first group consisted of comparisons using the aquatic bike as a toolforevaluatinglandversusaquaticcycling.Thesecondgroupconsistedofstudies on the physiological responses to single sessions of aquatic cycling under different exercise conditions (e.g. different water temperatures). Research on the effects of multiple aquatic cycling sessions was clustered in a third group. According to these

69 Chapter4 three grouping themes the extracted data was organised in three tables (Table 4.2, Table4.3andTable4.4). Figure4.1 Flowdiagramofidentifiedpublications *Onepublicationwasallocatedintwocategories

LandͲbasedcyclingcomparedtoaquaticcycling ThirtyͲonestudiescomparedaquaticcyclingwithlandcycling(Table4.3).Halfofthe studies(n=15)usedamaximalincrementalexercisetesttoinvestigatethephysiological responses during immersion versus on land exercise testing.11,13Ͳ26 Submaximal incrementalexercisetestswereconductedinsixstudies.27Ͳ32Incrementsweremostly achievedbyanincreaseinpedallingfrequency.Sevenstudiesoftheaforementioned studies controlled exercise intensity by electronically regulated pedalling resistance.10,21,24Ͳ27,31 An additional six studies compared submaximal continuous aquaticcyclingwithlandcycling.33Ͳ38Threeotherstudiesevaluatedaquaticcyclingasa mean for active recovery after an extensive exercise bout on land.39Ͳ41 Furthermore, onestudycomparedtheeffectofmoderateintensedrylandcyclingwithhighͲintensity intervaltraining(HIIT)onlandandinwater.42TwoͲthirdoftheaquaticcyclingsessions (n=22) were conducted in an upright body position. Nine studies11,24Ͳ26,30,31,34Ͳ36

70 Aquaticcycling–whatdoweknow? comparedsemiͲrecumbentcyclingonlandandinwater.FoursemiͲrecumbentbikes also had arm pedals.11,24,30,36 The level of body immersion of the participants varied from chest level to chin level. The water temperature during the exercise sessions rangedfrom18°Cto35°C. All but three studies used a crossͲover design to compare both environments. Additional study designs were a randomised controlled trial41,42 and a quasiͲ experimentalstudy.27In19outof31studiesparticipantswereyoung,healthymales. Five studies included healthy participants of both sexes13Ͳ16,26,36 and three studies includedpregnantwomen.27,37,38InfourotherstudiesparticipantsweremiddleͲaged men,29 males with cardiovascular diseases28,32 and men and women with 4 hypertension.42 Studies (n=21) investigating the difference in cardiovascular responses between aquatic versus land cycling compared oxygen consumption (VO2), heart rate (HR), strokevolume,cardiacoutputandbloodpressure.15Ͳ23,26Ͳ33,37,38,42Intotaleightstudies investigatedthemaximumVO2responseduringlandandaquaticcycling,withallbut 15 one study reporting equivalent VO2max values achieved by the participants on land and in water.17Ͳ23,26 Maximal HR was found to be lower during aquatic cycling at intensities higher than approximately 80% of the VO2max in seven from ten studies.20Ͳ23,26,29,30TheremainingthreestudiesreportedsimilarmaximalHRfortheland andwaterconditions.16Ͳ18Inmen,followingrecoveryfromamyocardialinfarction,no difference in submaximal HR on land and in water was found.28 McMurray et al. reported a trend towarda lower HR at submaximal intensities inwater in men with coronary heart disease.32 In pregnant women moderate aquatic cycling resulted in lowermaternalandfoetalHRcomparedtolandͲbasedcycling.38Fourstudiesreported higher stroke volume and cardiac output in the aquatic cycling group consiting of healthyparticipants.15,21,29,30Systolicbloodpressurewassimilarinhealthymalesduring an incremental exercise test when using aquatic versus landͲbased cycling.18,21 In pregnantwomenandinmenwithcoronaryarterydiseasethesystolicbloodpressure was reported to be lower during submaximal aquatic cycling.27,32,37,38 Sosner et al. reported a similar postͲexercise reduction in blood pressure in patients with hypertensionafterahighͲintensitycyclingsessiononlandandinwater.42 Other key outcomes were ventilation parameters,23,31,33 lipid mobilisation and oxidation,18,34 sympathoadrenal response,18,20,34 lactate accumulation and removal17,18,20,39,40 and thermoregulatory responses.35Ͳ37 Further outcomes were the developmentofpredictionequationstoestimateoxygenconsumptionfrompedalling rate during aquatic cycling13,14 and to calculate external power output of aquatic cycling.13Fenzletal.comparedthegasexchangemeasurementswiththeheartrate variabilitytoestimatetheventilatorthresholdonanarmͲlegaquaticbike.11

71 Chapter4

Immersion depths Chest Chest Chest bike Aquatic used Hydrorider® Hydrorider® Aquarider® water 30°C 30°C T 31°C in on LC late AC and rest in wave the Ͳ BP* and in AC HR performance, of 24hr HR pulse passive to recovery) hour soreness Ͳ between state of recovery regard 24 5 recovery Ͳ and of between decreased 2 in differ deceleration with AC physical not AC minute modified damage and (minute did rapid* and LC AC findings first differences difference max No HR More the No phase HIIT HIIT velocity land muscle Key perceived in peak of and (seated) lying 15s an 100% rpm until power land by POSITION output exercise): 70 exhaustion (HIIT) rpm 25W) after on 15s min passive passive peak of 40 BODY until until (vs. ext power min at bout land) sets P 50% every 4 10min: 25W AC protocol set on to rpm rpm at 60 (moderate of interspersed peak 5 25W 10 up was Ͳ UPRIGHT by exercise 30min sets water 50% steady between 75 parameters Ͳ 2 recovery, position output at protocol: workload: rpm: 65 minimal warm protocol: protocol and Rpm: Exercise: supine strenuous Duration: Land Initial Increments: exhaustion Rpm: Water Initial (corresponding Increments: thereafter Land 24min Land 6min output, power passive recovery Exercise # M (F/M: 24.4±2.2 30±7 65±10 (F/M: age: n=20 n=5 2/13) age: n=42 21/22) BP>130/85 mmHg age: Sample Cycling a HIIT a LC effect soreness early Aquatic LC, response vs. physical the muscle after the BP recovery, LC AC on muscle passive HR incremental on versus of HIIT of aim vs. moderate perceived and compare compare investigate AC and state damage, performance, recovery To decay marker parasympathetic reactivation, maximal exercise To after AC of Study To Cycling Ͳ based Ͳ Cross over RCT RCT Study design Land 2016 2016 2016 Year 16 42 4.3 41 Author Garzon Wahl Sosner Table

72 Aquaticcycling–whatdoweknow? level Immersion depths Xiphoid process Chest Xiphoid process bike Aquatic used Hydrorider® Hydrorider® Hydrorider® water 30°C 30°C 30°C T Ͳ 1.5 2 and = = the SV and rpm at v)O2 v)O2 AC SEE 0.0004 x Ͳ Ͳ 4 water land, = p<0.0001) in AC reserve C(a in C(a while SEE=0.319 on 0.99, W, ex rpm P AC = and and ,%HR 2 2 2 during on (r higher* =0.91, LC max ƌ AC VO VO ( equal SEE=7.6 during 0.329 obtained ext and %HR P were an based higher* (W) 0.739 of (L/min)=0.000542 AC during was ext 2 + P ext =0.99, ext 2 P recovery, for P water VO lower** (r were fraction in rpm DE Q generate means the 0.0001) the x × lower** 2.993 < 2reserve to comparable findings (W) p and a ext 13.91 Similar %VO Predicted 0.026 L/min) At were SV comparable During remained ejection When W, rpm (rpm) Key P and and and rpm rpm rpm until until until 70 70 70 exhaustion exhaustion exhaustion rpm rpm rpm 25W) 25W) 25W) min min min of of 40 40 40 of until until until until until until ext ext ex at at at 60 P P P every every every 25W 25W 25W of set set set to to to rpm rpm rpm rpm rpm rpm 60 60 5 5 5 25W 10 25W 10 25W 10 was was was by by by parameters protocol: protocol: protocol: workload: rpm: workload: rpm: workload: rpm: minimal minimal minimum protocol: protocol: protocol: exhaustion Rpm: Increments: Land Initial Increments: exhaustion Rpm: Water Initial (corresponding Increments: thereafter Land Initial Increments: exhaustion Rpm: Water Initial (corresponding Increments: thereafter Land Initial Increments: Water Initial (corresponding thereafter Exercise # (F/M: (F/M: (F/M: 33±10 33±10 32±7 n=20 6/24) age: n=33 5/28) age: n=20 2/18) age: Sample Ͳ a the in of to C(a AC drag the and central intensity , and 2 and the on recovery different model during relationship LC. establish VO incremental exercise a for for level ext Ͳ to rates P parameters prescription the exercise and for exercise exerted aim and AC during chest study compare develop AC legs. forces To between relative in method intensity AC To hemodynamics v)O2 maximal subsequent hemodynamic after with immersion pedalling accounting To mathematical Study calculate Ͳ Ͳ Ͳ Cross over Cross over Study design Cross over (continued) 2015 2015 Year 2015 14 13 15 4.3 Author Garzon Garzon Garzon Table

73 Chapter4

Xiphoid process Immersion depths Xiphoid process NR – Brazil used Sculptor RGS, Aquatic bike Hydrorider® Water Bike® Ͳ water 32.4°C 27°C, 31°C 28 30°C T in on in LC and AC with Ͳ Ͳ in test period, and scores F pre AC between non lower* recovery AC different recovery to trials pregnant the and sig. in AC were recovery cadences of comfort in comfortable passive AC not pregnant similar Ͳ passive between difference end 2 to values minute no non pregnant more Ͳ the VO were were BL during thermal 2 in lower** in at water F between and five submaximal 2max water VO the were in and T water to VO were women, first in BL compared values and and LC LC lower** and lower higher** 15min the BP HR, and pregnant pregnant findings Ͳ , difference differences and and in trials land was 2 values max AC pregnant exercise HR AC BL VO were compared Participants AC After AC land No on BP and No Key non After both of 70 in 3min first the or until to 30min until until + ergometer every the land a 3min of on rpm water): on 2min 3min water 5 + 85% in every by Test bike to every every (land corresponding 25W 75W position (water): rpm up threshold rpm, protocol: aqua HR at 25W 35W 10 50 threshold threshold the AC the supine thereafter Anaerobic parameters recovery Water at in of protocol: recovery on 1: 2: (floating) workload: workload: rpm: 50 70 + protocol: protocol: exhaustion and ventilator Rpm: ventilator Increments: Land Series Initial first Series 30min Land Initial Increments: exhaustion Rpm: Water Initial Increments: rpm until Land Wingate Passive 60min water Active 30min anaerobic sitting Exercise Ͳ at ±5.5 F (non # (10 wk n=0 M cyclists F 31.9±3.1 22±1 26.2 pregnant Ͳ 29 Ͳ and pregnant) pregnant gestation) (pregnant), 32.3±2.8 27 age: non F n=10 age: n=10 age: n=20 Sample 2 land VO and during Ͳ warm on during AC and non and and lactate BP LC between BL, and women, during comfort recovery LC water and aim neutral in and in compare compare compare AC pregnant pregnant responses To cardiorespiratory response, thermal AC water To removal passive and Study To Ͳ Ͳ Ͳ Cross over Cross over Study design Quasi experim ent (continued) 2013 2011 Year 2011 40 4.3 17 Author Yazigi Finkelstein 27 Ferreira Table

74 Aquaticcycling–whatdoweknow?

Immersion depths Xiphoid process Xiphoid process Xiphoid process in used authors Aquatic bike Hydrobike Evolution® Hydrorider ® EM designed the laboratory Ͳ water 28°C 30 31°C 33°C T for of water were were AC lower* in during in AC

time 4 reduced was were anaerobic intensity respiratory in be lower* exercise comparable the to lower* at LC 2 recovery were peak had volume, exercise was were AC inspiratory VO AC of in and attenuated** AC concentrations increased** in tidal tidal levels and concentrations AC min during achieve with was in AC quotient BP AC 15 variables workload and to acid and in higher** 2max higher** and glucose peak AC immersion 1: 2: 6 exercise groups fatty VO release ventilation, was findings systolic 2 and at was AC HR, threshold comparable Respiratory BL during Free increased** Water (nor)epinephrine peak ANP BL in Series Ventilatory both Ergometric during 60% Series VO Min frequency, higher** BL Key max NR at at HR LC 2min LC above increased Rpm: until NR and an and 10% protocol: by AC AC 6min estimated Rpm: 30min; treadmill: flywheel ventilatory speed for a 65% (n=9), the every (n=9), land 50W recovery protocol: at at 30min; to increased on LC 50W 122W for fins individual was 6min or of water water subjects subjects bout parameters of 2max AC protocol: up, the 1: 2: workload: NR Ͳ protocol: protocol: and and age) VO Ͳ of Land Initial Increments: exhaustion Water Workload number Rpm: Land Exercise warm the threshold Land 15min (220 Land Series 60% Series workload Exercise 1, # M M M (Series 31±3.6 22.7±1.9 30±8, n=17 age: n=11 age: n=15 age: 29±8 2) Sample AC LC on effect or ANP active the AC during mobilization lactate ventilator with oxidation immersion induced during lipid Ͳ aim LC lipid metabolic investigate compare compare water To of exercise release, and To removal recovery To and requirements and Study Ͳ Ͳ Ͳ Cross over Cross over Cross over Study design (continued) 2010 2007 1999 Year 18 33 39 4.3 Author Wiesner DiMasi Bréchat Table

75 Chapter4

notch Immersion depths Suprastern al Xiphoid process Shoulder EM EM EM bike Ͳ Ͳ Ͳ Aquatic used Modified Monark (Morlock& Dressen dorfer) Modified Monark (Morlock& Dressen dorfer) Modified Monark (Morlock& Dressen dorfer) water 31°C 30°C 32.5° C T at and AC were 38 higher sweat analysis AC and was was AC during in be AC when and the to depression and between medication during 38 from 2Peak during lower* 2Peak osmolality AC recovery BP storage VO response body differ VO was concentrations for in segment tendency concentration potassium Ͳ concentration blocker a heat Ͳ mean ST not during during T 40% 37 systolic and separately at beta did exercise LC activity and and and peptide in to SV AC greater showed difference difference AC AC induced vasopressin with diuresis Ͳ in HR rectal 2Peak 38 sodium aldosterone renin in in HR T and excluded LC VO LC findings Q 37 change group group caused HR, and No patients exercise were Plasma Lower* Higher* Foetal after Lower compared LC loss plasma concentrations recovery Plasma Natriuretic lower* Arginine lower* No 40% higher* Key No Ͳ 80 rpm 35 75% 60, 59 to 2peak increase until exhaustion, (range and VO subject’s subject’s an match of 52 2max 2max to until by the the 6min rpm 46, according of of VO VO in , 75% 5min 5min 39, every 2max 40% 40% 40% 40% protocol: 2max until adjusted VO VO every every 25W increase Dressendorfer was of 20min 6min 70% water 60 & parameters Ͳ averaging: protocol: protocol: 50 workload: workload: workload: workload: predetermined 55 2peak protocol: and protocol: 100% every 2peak 2peak VO rpm RPM: Rpm: Water Initial Land Initial VO Increments: of Water Initial VO Increments: 46) Land Duration: Intensity: RPM: Morlock Increments: Increments: resistance in and Exercise Land Initial wk MI 25 with of # M at M 49±3 30±1 NR F 10 = n=15 history age: n=7 gestation age: age: n Sample of Ͳ and AC a SV effects fluid the during effect influence with Q, on on the foetus the the and on hormones LC LC water men and submaximal MI in rest) aim and and out Ͳ LC (at examine evaluate compare AC AC regulating of To To head immersion HR graded and healed To of mother Study Ͳ Ͳ Ͳ Cross over Cross over Cross over Study design (continued) 1992 1993 1990, 1993 Year 19 4.3 28 , 38 Author Hanna Sheldahl Katz McMurray 37 Table

76 Aquaticcycling–whatdoweknow?

notch Immersion depths Shoulder Xiphoid process Suprastern al EM EM EM bike Ͳ Ͳ Ͳ ) Aquatic used Modified Monark (Morlock& Dressen dorfer) Modified Monark (Morlock& Dressen dorfer Modified Monark (Morlock& Dressen dorfer) water 32.5°C 30°C 32,5°C T Ͳ in AC but was LC LC left in mild less were AC arterial and systolic than and similar in Ͳ AC BP 100% exertion rpm

4 2 greater* stages, index, AC AC AC in levels during end 59 VO between 2 and were was peak of AC work VO 80 and and BL systolic at stroke concentration in pulmonary during , 2 at at higher* 52 between AC between and 100% VO less AC 46, index, in resistance were be for diastolic and comparable greater at Ͳ differ differ pressure, to submaximal 80 end not was reduced* not at particularly at HR cardiac lower* AC BP indexes slightly lower* norepinephrine epinephrine LC did did LC, LC l/min matched for in arterial peripheral were 1 findings were was and 2peak 2max 2max were Arterial groups HR VO pressure, ventricular volume Right Plasma reduced* Plasma AC both HR VO Trend exercise When lower Q during than Total during VO Key 80 80 rpm , rpm that by 60, 60, 59 2max 60 until Ͳ increase until exhaustion, VO 6min, 36 and an match of 6min of of 52 2max 2max 2max 2max to until by increments increments matched 6min controlled 3 3 46, VO VO VO VO 100% every rpm was 5min 5min 39, every least least 40% 40% 40% 40% 25W 30 and 2max 2max rpm adjusted at at increments increments 80 VO VO 3 3 every every 25W 10 increments of of was resistance of of increase 60, 60 60 parameters Ͳ Ͳ averaging: protocol: protocol: protocol: 6min, workload: workload: workload: workload: workload: workload: 55 55 NR protocol: protocol: protocol: 100% 100% rpm Land Initial workload electronic Rpm: Water Initial Land Initial Increments: resistance and Rpm: Water Initial Increments: in Land Initial Increments: completion Rpm: Water Initial Increments: completion Increments: matched Increments: every and Exercise with 35 36 # Ͳ Ͳ disease M M 21 22 52 M age: n=10 n=9 age: n=10 coronary artery age: Sample and in LC LC responses responses LC coronary the the graded and testing and to AC with dynamic AC disease aim exercise compare compare compare cardiovascular during AC To To sympathoadrenal response dynamic To cardiovascular during patients artery Study Ͳ Ͳ Ͳ Cross over Cross over Cross over Study design (continued) 1990 1990 1988 Year 20 21 32 4.3 Author Connelly Christie Mc Murray Table

77 Chapter4 chin +

Shoulder Immersion depths Shoulder Neck EM EM EM bike Ͳ Ͳ Ͳ Modified Monark (Morlock& Dressen dorfer) Aquatic used Modified Monark (Morlock& Dressen dorfer) Modified Monark (Morlock& Dressen dorfer) water 31°C 31°C 30°C T 2max LC than time differ VO were of exercise not and left 80% 2max was all unit groups AC did posture at VO AC HR systolic and per HR mean in Ͳ AC trials ventilation / land 40 80% in gas at l/min, at water between between water in AC 2.4 AC upright greater* diastolic in in conditions workloads Ͳ of voluntary differ differ expired and greater* 2 the of end were not not VO in posture AC land was lower* in did did greater* SV mean maximum findings volume was 2max 2max submaximal a submaximal upright was At between ventricular dimension At VO in At greater** Q HR Mean intensities HR, and lower* VO Key a . to until 5min 2max 75 to workloads VO to until 4 6min of 3min every within 40% maximal m to Ͳ every kp 50W 35 protocol: protocol: protocol: corresponded 25W 150 exhaustion that prescribed water water water parameters 2max workload: workload: and NR NR load and and VO achieve Land exhaustion Land Increments: Initial Initial Increments: work 80% Rpm: Land Individual to Rpm: Exercise # M M 26.3±3.9 48±8 27 M age: n=12 n=19 age: n=7 age: Sample in AC on of land effect and of effect aged Ͳ on blood posture the the volume effects dynamic maximal in water levels on supine to to the middle in out on work posture performance Ͳ blood shift in submaximal upright aim LC in investigate assess determine different head water upright and exercise central cardiac during of To To central volume cardiorespiratory responses and men To of immersion cardiorespiratory responses aerobic Study Ͳ Ͳ Ͳ Cross over Cross over Cross over Study design (continued) 1984 1987 1976 Year 29 22 Ͳ 4.3 23 Author Sheldahl Sheldahl Dressen dorfer Table

78 Aquaticcycling–whatdoweknow?

Xiphoid process Xiphoid process Xiphoid process Immersion depths Ͳ Ͳ Ͳ used Reha Aquabike® Reha Aquabike® Reha Aquabike® Aquatic bike Ͳ Ͳ water 27 28°C 27 28°C 28°C T for is LC Ͳ no strong LC a and is exercise post

exchange 4 and with similar and AC LC is exercises two, threshold gas LC AC parameters of showed in and arm and during workload, baseline AC increased* AC both HRV AC as to in in epinephrine threshold relationship adding in exercise determined were with AC by comparison different* leg insulin rate between AC between is acids in expressed higher** compared with in cost and increase 2 ventilatory work rate O fatty was leg – respiratory findings Ͳ 2 the difference Similar decrease exercise Free VO arm Extra lower** At capacity, lower** The heart Quantitative measurements correlation Key ANP and gas leg leg Ͳ Ͳ POSITION intensity state cycling Ͳ until until adjustment arm arm ratio threshold. BODY for 1:3 2min 2min steady during a moderate protocol: protocol: of ratio: every every 50W 75W anaerobic protocol: protocol protocol: arms reach of testing the 25W 25W to RECUMBENT testing Ͳ of at water water water 20% of parameters workload workload SEMI workload: workload: 70 70 NR and and and leg set Ͳ exercise: 60min workload Ͳ 10min Ͳ exchange of Exercise Land leg Initial Increments: exhaustion Contribution exercise: Rpm: 11 with Rpm: Land Initial Increments: exhaustion Arm Rpm: Land 0 45 # Ͳ M M 35.1±5.4 26 40.2±5.4 M overweight age: n=12 age: n=12 age: n=6 Sample in in to after vagally time release fatty rates gas changes the prolonged of water the during release free short rate work parameter in and ventilatory the LC and aim work during Ͳ and with investigate compare 2 LC ANP an To VO relationship increasing AC Comparison exchange modulated variability establish threshold acids AC Study To of Ͳ Ͳ Ͳ Cross over Cross over Study design Cross over (continued) 2015 2012 Year 2013 4.3 24 34 11 Author Fenzl Fenzl Fenzl Table

79 Chapter4

level Chin Neck Immersion depths Clavicles EM EM EM ) & & used Modified Collins (Craig Dvorak) Modified Collins (Craig Dvorak) Modified Monark (Chen Aquatic bike water 33°C 30°C 21°C, 25°C, 29°C T in Ͳ and in water showing 21°C medium semi , and LF 2 esophageal water positions at frequency for T during change the VO warm and exercise AC 25°C no resulted high 29°C in AC with than LC duration of in spectrum lower* 29°C and total was in maximal between in is occurring AC for higher 21°C and rectal output there linearly power , T differ HRV lower* similar of exercise be LC in max during of position in LC frequency to positions not to higher* HR total power were water and intensities exercise did LC and rose* recovery at increased AC upright changes central findings resulted 2max high tendency rectal rectal and The AC greater* maximal recumbent VO The distribution exercise The peak a to During T T during After water compared Key duration, in could land 6min on 6min the for 30min workload 29W 70W, for above and subjects of or for but as below rpm rpm until 2max water 60 60 25W VO (males) above, increments 0W 50W 0W 40, protocol: protocol: 2min of protocol 29°C set as increment 1 20, 44W 50W and 60% was exhaustion maintain every water water Same LC by same parameters EM in protocol: 2: 1: 2: workload: workload: workload: 25°C 60 60 50 protocol: and and until the longer Rpm: Exercise Land Series1: Initial Increments: each Series Initial Increments: followed each Rpm: Water Series Series on values no Land Workload: 21°C, Rpm: Land Increments: (females) Initial # (F/M: 22,0.9 26.8±4 30.6±6.5 M M n=7 age: (SEM) n=5 age: n=10 3/7) age: Sample of that on land core the AC HRV a on on EM the water of Ͳ effect with T exercise a with during used exercise the semi underwater a occurs EM and standard spectrum water attenuate immersion aim with LC in that rest) evaluate determine compare To water power (at and To would rise cardiovascular upright land and tests recumbent exercise To Study Ͳ Ͳ Ͳ Cross over Cross over Study design Cross over (continued) 1998 1989 Year 1996 4.3 25 35 26 Author Perini Chen Israel Table

80 Aquaticcycling–whatdoweknow? , 2 VO HIIT, thoracic thoracic thoracic reported; st st st vertebra 1 vertebra 1 vertebra Immersion depths 1 female; not F, EM EM EM & & & NR, used temperature; T, Modified Collins (Craig Dvorak) Modified Collins (Craig Dvorak) Aquatic bike Modified Collins (Craig Dvorak) ergometer; water minute(s); 20°C, 24°C, 28°C 25°C, 30°C, 35°C T 18°C, 25°C, 33°C volume; EM, min, LC and in in LC in l 33°C and stroke AC each for water water in decrease at 1.7 with Celsius; for 0.14l/min SV, 4 a larger* greater 18°C LC

at 25°C similar C, 33°C than in and was in infarction; and response was similar water exercise SV during mean; , averaged retarded water 2 exercise than 2 was higher* 18°C of somewhat exercise or 35°C and VO in VO during difference; of was workload and AC water 2 the error warmer women myocardial water VO seemed in lower* 30°C levels given 25°C women LC and prevented MI, relationship a during water in 33°C 1 thermoregulatory submaximal 2 Ͳ <0.01 was than and water in and water for AC with findings VO men rectal min 2 max similar standard 25°C ͼ T – 2 male; water Q For O in Similar men T VO and In more Ventilation cold Key water HR 18°C than LC At During and 25°C value arteriovenous Ͳ p M, deviation. SEM, at 84, v)O2, Ͳ 60, 60min 1.5min Error; C(a cycling; 84W 36, for workload 1.5, 18, and significant 36W 0, based Ͳ each 60 at 5,3,3, mean±standard **, Standard pressure; 36, as land protocol: protocol: protocol: SE, each 18, LC, between <0.05; exercise workload: 0, blood years 5min water water water workloads: parameters EM EM in BP, for minute; 30 30 30 value 10min and and and leg leg Ͳ Ͳ Ͳ p per variability; at Exercise Land Arm Rpm: Land Workloads: Duration Rpm: 120W Rest: Rpm: Land Arm lactate; presented rate 29 Ͳ is # 19 blood (F/M: 23.1, 26±5.5 NR revolution students M age heart significant BL, *, n=18 8:10) age: range: n=2 age: n=6 age: Sample and Rpm, HRV, and to in years, Ͳ on peptide; rate; females AC healthy yrs, output; and different exercise and metabolic thermo are in to exercise heart response during water T and males HR, cardiac natriuretic aim week(s); in cardiovascular Q, land compare compare compare water wk, T adjustment on and To To regulatory continuous different land To cardiorespiratory responses LC Study atrial participants training; Ͳ Ͳ Ͳ output; ANP, Watts; Cross over Cross over Cross over Study design W, interval otherwise power (continued) cycling; 1976 1984 1969 Year 36 30 uptake; stated 4.3 intensity 31 external Ͳ aquatic Ardle Ardle , not ext If Author Mc Mc Craig oxygen Table AC, # high P

81 Chapter4

Aquaticcyclingunderdifferentexerciseconditions TwentyͲfive studies investigated the effect of several different exercise conditions during aquatic cycling (Table 4.4). The comparisons are based on crossͲover studies withhealthyyoungmaleswiththeexceptionthathealthy(nonͲpregnant)femaleswere included in three studies43Ͳ45 and one study used a quasiͲexperimental design to compare ageͲmatched healthy controls with heart disease patients.46 Common coreoutcomes were cardiovascular,12,23,44,45,47Ͳ50 metabolic36,51Ͳ55 and thermal response43,49,50,52,56Ͳ61 to different exercise conditions. Furthermore, approaches to estimateandregulateexerciseintensityduringaquaticcyclingwereevaluated.62Ͳ64 Different exercise conditions were created mostly by changes in water temperature12,23,43,48Ͳ52,57,60,61 and different exercise intensities (high versus low)23,43,44,50,58Ͳ61,63Ͳ65 With regard to the exercise parameters intensity and duration, studies(n=11)utilisedcontinuous,submaximalexercise(40and60%ofVO2max)witha durationof30to60minutes.12,23,43,49,51,52,57,60,61Exerciseintensitieswereeitherbased ongradedexercisetestingonland12,23,44,45,47,50,51,53Ͳ56,58orinwater.23,43,46,48,52,57,62Ͳ64The watertemperaturesthatwerecomparedrangedfromcold(18Ͳ20°C)andcool(25°C)to thermoneutral(30Ͳ35°C).Otherstudiescompareddifferentlevelsofbodyimmersion [46],differenttypesofexercise(intervalversuscontinuouscycling,armversusarmͲleg versuslegexercise)45,49,61anddifferentaquaticbikeswitheachother.44Furthermore, thematernalandfoetalresponsetosubmaximal(60%ofVO2max)aquaticcyclingduring differentstagesofpregnancywasstudied.47,53Ͳ56 Fifteenstudiesuseduprightaquaticbikes.23,44Ͳ47,51Ͳ57,62Ͳ64Inallthesestudiespedalling frequencyregulatedexerciseintensitywhiletwostudiesfocusedontheinfluenceof pedallingresistanceprovidedbyadditionalfinstotheflywheel.44,64Sogabeetal.used theadditionalfinstoincreasepedallingresistanceinsemiͲrecumbentcycling.65Inall othersemiͲrecumbentbikesintensitywassetwithelectronicallycontrolledpedalling resistancemechanisms.43,48Ͳ50,58Ͳ61

82 Aquaticcycling–whatdoweknow?

hip, NR Calf, xiphoid process Xiphoid process Immersion depths used Hydroride r® Hydroride r® Hydroride r® Aquatic bike Water 30°C 29°C 30°C T 2max and HR, both phase the of 85 in in

4 trial at 92%VO double pedalling rpm cadence at at threshold between exercise BP, 60 and ventilation 79% higher** and 2max trial continuous by determination higher** arterial 50 concentration between was anaerobic difference the group %VO the BL 40, HR, , was 2 of reduced RPE no the in of for VO and RPE to , 2max continuous was disease immersion) max hyperpnoea findings differences the or of related Immersion 1 cadences heart No protocols product Central of Peripheral 92%VO %HR There methods Key rest at in a , AC 2max an until max cadence min passive VO or every HR per to 2min was POSITION positions interval different Ͳ 92% min scale at maintain and beats per and every BODY Borg saddle Ͳ calculated and 85 100 following of rpm beat of Ͳ the rpm immersion 80, comparison 10 15 the on out 40 reach exhaustion 85% in UPRIGHT 31min 75, of incremental, continuous 100 body parameters 16 to Ͳ and of rpm: workload: of 80 until one and Exercise Exercise: levels Initial Increments: least obtained: score inability Exercise: seated Duration: Intensity: Rpm: 2min Initial Increments: = n conditions # (heart F M 10/24) 64.7±7.8 32.8±4.8 34 21 10 27 (controls) = = = = n participants (F/M: n disease), 12 age: (heart disease), 61.0±7.8 (controls) n age: age:22.5±2.4 n Sample exercise in and to effect and heart various of point whether heart the the the a a responses exercise perception on physiological the cycling threshold method to levels of with participants (under the with the AC aim deflection type and determine investigate compare aquatic effort only To of different immersion respiratory healthy people disease To the affects response AC of during To rate method determine anaerobic ventilator Study Ͳ Ͳ Ͳ cycling Quasi Cross Cross Study experim ent over over design Aquatic 2016 2011 2015 Year 46 4.4 45 62 Author Dionne Pinto Brasil Table

83 Chapter4

are and Ͳ Hips thighs immersed Neck Immersion depths Mid sternum Ͳ Ͳ bikes used different 4 aqua Modified Monark EM (Morlock& Dressen dorfer) Aquatic bike Modified Monark EM (Morlock& Dressen dorfer) Water 25°C 20°C, 25°C, 30°C, 35°C T 20°C, 33°C 4 trial AC the the rpm resting water 25°C higher* T partial for were volitional trials in peak and water generated decreased to trial partial from at was across HR peak were AC ** and similar water bikes rpm HR were 35°C* and temperature , core 2 exercise T intake AC core was water and peak peak VO related** to and 35°C in T water bikes water water in VO VO aquatic different** increased and norepinephrine was 20°C 4 in in water 20°C 20°C energy in during 35°C HR core was in change and the T water different cold neutral change compared exhaustion in and cortisol expenditure and during rpm to the exercise and and 30°C Ͳ findings gender difference difference 2max 70 exercise 30°C No for No between Time exhaustion bikes At different** Change Plasma to in Dopamine higher exercise and VO correlated* BMI correlated* Post after Energy Key cold 4 the to the on until test with added resting maintain 2min bottom vs. to exercise the every resistance, resistance 2max 2max to cycling cycling no unable VO VO rpm rpm with 5 40 was with added 30min 45min 60% 60% steady steady incremental parameters axle, rpm: NR NR bike: pedals rpm Exercise Exercise: aqua resistance bracket the Initial Increments: participant set Exercise: immersion Duration: Intensity: Rpm: Rpm: Exercise: Duration: Intensity: 25 # 8/8) Ͳ M M = 31.7±5.8 17 25.6±5 n=16 participants (F/M age: n=11 age: (range) age: Sample n=11 Ͳ 2 EM on and VO and post intake acute and the water water on and men urine the different in during plasma of HR cold in energy exercising and of on of aim different assess investigate describe 2max water To responses women four To effects anthropometrics VO cortisol excretion catecholamine dopamine exercise T To effect exercise temperature Study Ͳ Ͳ Ͳ Cross Cross Cross Study over over over design (continued) 2005 2009 1994 Year 51 4.4 Author Giacomini 44 White McMurray 52 Table

84 Aquaticcycling–whatdoweknow?

Xiphoid process Immersion depths Ͳ used Aquatic bike Modified Monark EM (Morlock & Dressen dorfer) Water 30°C T in all 56 to 55 55 total at , at 47 wk during of 10min 4 55 change during 55 those 35 AC exercise postpartum, wk similar exercise workload with not th pregnancy lower* land and and increased 53 within from 35 after than but unaffected with did on were 25 pregnancy was exercise gestation and at evaporation 47 was wk than 56 trials, skin during water T reactive decreased advancing volume 25 and elevated during all slightly* output seen conditions immersion partum, 47 Ͳ unchanged pregnancy at cold water for all higher* with low were were in in post and for plasma lowered pregnancy storage during activity cardiac was values during declined partum tests decreased* Ͳ of 54 similar levels loss heat weeks declined alphafetoprotein lower* wk 54 2max , normal concentrations remained similar post 54 was levels 54 uterine storage, stress BL 10 skin resting immersion exercise VO greater heat T was calculated ages exercise AC and was and to resistance no to 55 were serum 54 heat cases and HR was 60% and cortisol exercise different HR was glucose triglyceride mean 23 exercise partum exercise , during during Ͳ Ͳ Ͳ findings of 2 maternal rectal evaporative reduced across compared gestational peripheral Compared VO Post exercise Blood pregnancy Blood Diuresis Post Plasma immersion There gestation T Exercise T Post Maternal Key during Foetal 21 rest natriuresis pregnancy achieve 2max cycling VO 20min 60% steady parameters NR Exercise Rpm: Intensity: Exercise: Duration: wk wk 15, and 35 # at 12 F partum 30±3(SE) to Ͳ and 8 n=12 25 gestation post age: at Sample and renal and and women changes the response foetal responses, immersion aim pregnant in investigate uterine thermoregulation, metabolic cardiovascular during AC changes, To Study Ͳ Cross Study over design (continued) 1990 1988 Year 4.4 53,54 Author Katz 47,55,56 McMurray Table

85 Chapter4

Neck Neck Immersion depths Ͳ used Aquatic bike Modified Monark EM (Shapiro) Modified Monark EM (Morlock& Dressen dorfer) Ͳ Water 26 29°C 20°C, 25°C, 30°C, 35°C T = = a Ͳ in 2 r the speeds during constant is greater* flywheel Ͳ a with for n was better , the 2 water rates 1 Ͳ between were and to VO 0.506n 0.000266n pedalling min when 35°C ͼ + l rate + of , sweat resistance 3 maintain h water and 1 to 1.64 runners water 0.25, attached coefficient range predicted to = + 20°C 30°C higher* b fins b rpm in than pedalling metabolic 0.00667n 0.00104n up 35°C thermoregulated in and were ; 3 of in + had in 40 Ͳ Ͳ for 2 (rpm) fins water 29 6 = correlation preferable findings runners 2 increase to 20°C exercise Swimmers 1 to VO 0.00164 0.00002n 0.104n number The measured 0.98 The was speed Changes for swimmers Runners Key = = fin rpm’s fins + fins 6 one and and to 40 rpm with 63, Ͳ 12min 1 Ͳ fins fins three 20 60 of AC 15 of of = = 6 50, Ͳ every 50, 2max to 20 2max different fins fins 1 AC 40, = VO rpm VO six intensity of with 10 30, 1:no fins to 1hr 1hr 30min <85% 60% high steady AC combination combinations parameters two four 1: 2: NR Series 60, 45, Ͳ Ͳ Exercise Series Exercise: combinations rpm’s Duration: Intensity: Increments: Rpm: 20 20 Series Exercise: different Duration: Increments: different Rpm: Exercise: Duration: Intensity: (SE) # M 25.8±2.1 20.8±1.1 (n=5 M n=6 age: (SE) n=11 athletes age: (SE) runners), 18.5±0.5 (n=6 swimmers) Sample be in exercise periods could trained AC swimmers the constant Monark applicable of a water rpm T and graded aim meter Ͳ AC prolonged modify compare moderate time which To ergo for in maintained for of different to runners Study thermoregulatory responses To Ͳ Ͳ Cross Cross Study over over design (continued) 1981 1979 Year 64 4.4 Author Shapiro McMurray 57 Table

86 Aquaticcycling–whatdoweknow?

+ Ͳ Neck Neck chin Immersion depths Shoulders Immersion depths Hand breadth above xiphoid process Ͳ Ͳ used used Combi Modified Monark EM (Morlock& Dressen dorfer) Modified Monark EM (Morlock& Dressen dorfer) Aquatic bike Aquatic bike Aerobike 330 ? Water water 30°C 25°C, 30°C, 35°C 18°C, 26°C, 34°C T T 26°C, 32°C, 35°C 26 a in °C 2max at of W), in 35°C 18 water rpm, VO higher* = to 0.996 water lower AC 4 than w grease T on = T 35° 2 to r was °C conditions , at in min 3 of (60–120 18 all monitor AC installation effect = per belt in l, to w compared between T product lower* at compared beats during similar was intensities friction switch differ installation water 15 34°C significant 35°C the regreasing was not or in no reed and 25°C frequency greater* of BP 8 highest* for 26 did AC 0.274+0.000008rpm had workload and water at = was findings findings was was 2 2peak 2 submaximal 34°C water T HR 30°C water VO At VO or Max. than HR Pressure during 26°C Systolic Modifications: nipples Key Key magnetic removal VO and 35°C the within every 60 POSITION intensity and for 60W, 50, 10W and 2max by 32°C 2min BODY 40, VO water exercise exercise exhaustion maximal T 20, then 26°C, workload every at 52% per in and 41, achieve 20W exhaustion initial 5min trials RECUMBENT to prescribed Ͳ by 10min 26, 3 incremental incremental levels rpm parameters parameters until 60 NR SEMI 5min four to Exercise Exercise Individual workloads 4 Exercise: Increments: increased first minute Rpm: Intensity: Rpm: Increments: maximal water Duration: Exercise: Exercise: 29 49 # # Ͳ Ͳ M M M 24 25 26 22±2 M 4 6 = = age: n=8 age: n age: n=10 age: n (range) (range) Sample Sample 2 BP, VO effect on max effects rpm on exercise HR the the of Water the standard underwater T water measure and product a and T temperature for to pressure of aim aim 2max EM formance function and and Ͳ VO determine investigate modify compare a water different To of on To effects cardiorespiratory responses per as HR frequency To land To of use Study Study Ͳ Ͳ Ͳ Ͳ Cross Cross Cross Cross Study Study over over over over design design (continued) 2010 1976 2016 1974 Year Year 48 63 Ͳ 23 4.4 12 Author Dressen dorfer Morlock Author Fujimoto Fenzl Table

87 Chapter4

st 1 thoracic vertebra NR Immersion depths Neck Neck Ͳ EM EM & used cycle Ͳ Modified electroni cally braked Siemens bicycle Modified Monark EM (Shapiro) Aquatic bike Modified Collins (Craig Dvorak) Modified row EM (Sogabe) water 20°C, 28°C 15°C 26°C T 31°C Ͳ 30 30 fat for of the rectal water 2000 and were T the Ͳ and of during body of the 10 body time: 400 20°C by of in higher greater* esophageal group small 2 fastening T greater** removal similar VO a water the immersion between 0.2°C of were and mass seat, immersion a prolonged achieved exercise fins: replacement a 20°C rectal rectal cranks, III T T body in resting was for the showed static plastic in exercise between women and of a aural trial to pedal II rate, that rpm large T with and during maintained with the the exercise and and level I of findings to different men modifications: decreases rectal attachment handle Preferred 40 Workload not mass AC For %, women level With women T decrease compared min Metabolic EM saddle fins ml/min Key = and , until 1 III Ͳ limb min various of ͼ resting 2 Level was speeds 10min 1 , Ͳ O 1 /0.05 Ͳ fins) (vs. rate with resting resting min l min every ͼ vs. vs. litres 2 700ml exercise tests same (large O 1.5 in = pedalling AC AC cycling 1 I rpm Ͳ leg 60 of of Ͳ the 2max 2 10 to 20 at min ͼ 1hr 40min 1hr Level VO VO 2 steady steady steady exercise 1250ml (arm parameters O = fins) fins rpm: 60 NR NR II of (no Rpm: Exercise Exercise: immersion Duration: Intensity: Level 1700ml Rpm: performed movement) Exercise: immersion Duration: Intensity: Rpm: Exercise: immersion) Duration: Intensity: combinations Exercise: size Initial Increments: 69 29 Ͳ (n=5 34 # Ͳ 19 n=5 body (F/M: M M body 23.3, 14 34.4±2.9 NR M n=16 8/8) age: range: n=15 age: (range) mass) age: small large mass, n=10 Sample age: n=7 (SE) cool role water in in Ͳ thermal exercise a and the thermal simple cold row static AC of the exercise leg the a applicable in on water horizontal to to of of women on water in EM aim mass and cold metabolic graded compare describe investigate describe morphology To influence intensity thermoregulation men and To response compared immersion of To responses and To modification body bicycle conventional for Study exercise Ͳ Ͳ Ͳ Ͳ Cross Cross Cross Cross Study over over over over design (continued) 1992 1987 1986 1987 Year 43 65 58 4.4 59 Toner Author McArdle Sogabe Golden Table

88 Aquaticcycling–whatdoweknow?

Neck Immersion depths Neck Neck used Modified Monark EM (Shapiro) Modified Monark EM (Shapiro) Modified Monark EM (Shapiro) Aquatic bike water 20°C, 26°C, 20°C, 26°C, 33°C 18/20 °C, 30°C T not HR of than final than to did in was high AC exercise with esophageal metabolic exercise 4 exercise types T at were 0.61) leg lower* leg = final and in during rectal (r exercise T exercise for difference compared in exercises were rectal intensity exercise between no correlated AC T lower , than values finale leg types low Ͳ skin trials water greater* was water intense T intensity leg Ͳ intensity T was differ water Ͳ water between ventilation during flow arm leg T Ͳ immersion low 20°C low high not were arm to during there rate rate, 20°C and exercise in moderately and arm heat intensities did for immersion trials water rectal water for between higher* flows T T T resting during during during and was findings 0.68) arm, leg difference 2peak high all all = skin were resting Heat with VO exercise RPE differ RPE lower RPE (r In between rate Final for for At lower* in No T Metabolic compared intensity Metabolic Key arm across low 2peak (40% leg Ͳ vs. for VO resting arm arm ) 1 low Ͳ arm vs. vs. (vs. 2peak, 2peak vs. matched min ) leg leg and l VO VO Ͳ Ͳ intensity water leg 1.6 cycling T arm arm output low (60% (60% of for 2 vs. vs. ) only 45min 45min 1hr high high VO across steady leg leg parameters 2peak power 40 40 NR VO ) 2peak,, water Rpm: Exercise Exercise: exercise Duration: Intensity: matched VO T Rpm: Exercise: exercise Duration: Intensity: (40% exercises, exercises Rpm: Exercise: immersion) Duration: Intensity: # 22.4±3.6 22.4±3.6 23.6±5.2 M M M n=8 age: n=8 age: n=9 age: Sample Ͳ of in arm arm, resting AC modes across water and metabolic metabolic the between the variables and and during during cold and and combined various temperature aim time and and exercise examine compare To relationship physiological perceptual over water during AC. Thermal response leg leg To thermal cool response immersion Study Ͳ Ͳ Ͳ Cross Cross Cross Study over over over design (continued) 1986 1984 1985 Year 4.4 49 60 61 Author Toner Toner Toner Table

89 Chapter4 Ͳ

p M, at cycling; Immersion depths Neck significant based *, Ͳ leg used Ͳ (Craig land Dvorak) EM & Modified arm Aquatic bike LC, week(s); Ͳ wk, water 24 35°C T variability; of an uptake; in rate min light water ear 30 T 32°C, that oxygen in , heart , workload ч with exercise 2 ear last 32°C T Ͳ VO temperature in with HRV, 28 the heavy heavy increase exercise of declined rectal rate; water and with with 24°C water light during of in T increase by in and heart with and temperature; with higher initial T, HR, 32°C exercising decrease an load continuously water exercise ч followed was findings 2 Error; rectal VO 24°C T with was persisted decreased work light After initial when Key female; F, = 2 Standard 0.70 (VO SE, = 2 pedalling was ergometer; in (VO deviation. minute; EM, workload per Workload AC increase workload heavy an Celsius; vs. by 60min light C, steady parameters mean±standard revolution 30 litres/min). as Rpm, resistance Rpm: increased 0.92 litres/min) Exercise Exercise: Duration: Intensity: pressure; years in blood # exertion; BP, M 27±5.8 presented is Sample n=10 age: index; perceived age of mass in and heavy rate water thermal body during RPE, healthy exercise warm BMI, <0.01 aim are and light investigate value reported Ͳ To regulation lactate; and Study cool p at not Ͳ participants blood NR, BL, Cross Study over design significant (continued) otherwise 1968 Year **, cycling; minute(s); 4.4 stated min, 50 <0.05; aquatic not Author Craig If value Table AC, # male;

90 Aquaticcycling–whatdoweknow?

Aquaticcyclinginterventionprogrammes In total eight intervention studies, investigating the effects of a multiple sessions aquatic cycling exercise programme, were found.66Ͳ73 The exercise programmes (Table4.5) lasted between three and 36 weeks with an exercise frequency between twoandfivetimesperweek.Thedurationofonesessionvariedbetween30and90 minutes.ExerciseintensitieswerebasedonlandͲbasedmaximalgradedexerciseͲtests andthetrainingintensitiesweresetbetween60and80%oftheVO2maxinallbutone study.66InaoneͲgrouptestͲreteststudy,Sheldahletal.assessedweightlossinobese 4 66 womenafteralowintense(30to40%ofVO2max)aquaticcyclingprogramme. Boidin et al.alsoevaluated the effects of aquatic cycling on cardiometabolic parameters in obese people.71 In this retrospective study the participants underwent an extensive lifestyle programme including highͲintensity aquatic cycling or land cycling. Furthermore, two randomised studies evaluated the cardiovascular effect of aquatic cyclingcomparedtolandcyclinginyounghealthymales68andpatientswithmultiple sclerosis.72,73 Two quasiͲexperimental studies investigated the influence of water temperatureonheattoleranceandaerobiccapacity.67,69,70 Four studies reported a significant improvement of cardiorespiratory parameters comparedtobaselineinhealthy(obese)peopleandmultiplesclerosispatients.68,71Ͳ73 Aquaticandlandcyclingevokedsimilarimprovementsincardiorespiratoryparameters. Further,moderatelandandaquaticcyclingachievedsimilarimprovementsinhealthͲ related quality of life and selfͲreported physical fatigue in patients with multiple sclerosis.72,73Boidinetal.reportedcomparableresultsinweightlossandreductionin fastingglycaemiaandtriglyceridelevelsinobesepeople.71Inobesewomen,aneight weekaquaticcyclingprogrammeincoldwaterdidnotleadtoweightloss.66 Inyoung,healthymales,therewasnosuperioreffectofcoldorwarmwateronthe improvements in cardiovascular parameters,67,69,70 lactate accumulation lactate accumulation,69drylandheattolerance67andmuscleglycogenutilization.69

91 Chapter4

Immersion depths NR 1,30m used Hydrorider ® Aquatic bike Aquarider Profession al® Water NR 28°C T 72 no over and 72 group thigh related Ͳ fatigue; total group groups tests baseline and no groups showed difference maximal and no were difference WC, health with 72 BP, glycaemia, difference QoL; improved** HR; 73 both physical group exercise between in group values mass, adaptions 73 no group no values and resting fasting abdominal fatigue resting related Ͳ improved body no in in in difference associated neurotrophins reported lactate Ͳ 2 in time levels; immune with health 73 and self VO unchanged difference intervention mass; group reduced Ͳ capacity, scores over endurance; term fat no and post findings group QoL time; difference associated Fatigue and remained Improved Cytokines Cardiorespiratory Short Reduction** trunk Improvement* aerobic difference Improvement* triglyceride Improvement* muscle Improved Key change increased no Ͳ 2peak VO usual + water + 60% AC 20min = POSITION AC 5x wk AC, wk MAP 3 36 steady HIIT BODY counselling 30min 34min 80% threshold training, diet 3/wk Ͳ RPE/ duration: duration: 5/wk 2 UPRIGHT exercise lactate session: session: 60 NR programme: parameters programme: Ͳ resistance 50 rehabilitation Rpm: Exercise Intensity: Frequency: Duration care Programme Rpm: Exercise Exercise based Mediterranean Programme Frequency: Duration resistance Intensity:15 range: range: 25 21 # 74 MS obese 18/10), 17/8), 19/2) 55/19) = = =28 = 55.1 56.3 50, 52, 58±9 55±7 Ͳ Ͳ n n 60 95 n n = = 44.6 LC: (F/M: 46.7 (F/M: age: age: patients AC: n n people AC: (F/M: age: LC: (F/M: age: Sample on in in LC programmes values or and effects and QoL, the AC intervention the exercise to of response, parameters neurothophin related patients Ͳ aim lifestyle intervention investigate compare a addition fatigue, concentrations cardiorespiratory health cytokine influence To To of in cardiometabolic exercise obese Study LC Ͳ cycling AC vs. LC C etro CT: esign tudy s. ohort: pective S d R s c A R v Aquatic 015 013 Year 2 2 4.5 71 72,73 Author Boidin Bansi Table

92 Aquaticcycling–whatdoweknow?

Neck Neck Immersion depths used Modified Monark EM Modified Monark EM (Shapiro) Aquatic bike Water 20°C vs. 32°C 35°C vs. 20°C T AC in AC no 69 rate for hot 3 min) of citrate ͼ mean water all Ͳ for rectal water (25%). glycogen group T groups min.

4 sweat in ͼ volume; group AC of beats cold no all in AC. cold 69,70 difference HR equal no reduction* lateralis in ; training beats sweat: (29 70 muscle Ͳ and only increase** water HR: was values 18 2max water group post volume; body vastus values VO training Ͳ increase* warm no erythrocyte and of difference decline 69 AC warm of in exercise in of warm Ͳ 70 an difference 14 total to training post Ͳ HR plasma only in group and training post 69 acclimation: acclimation intervention Ͳ activity; Ͳ water post greatest accumulation group LC no increase ; difference reduction* training post heat heat group. post increase** no Ͳ Ͳ Ͳ for cold findings increase** decreased 2max compared skin land groups, T (25%) Decrease* increased* Post demonstrated difference Higher* 0.9°C Post from Post LC Reduced* use; Lactate and 13% VO 4% group Unchanged difference 38% synthase HR Similar Key hot warm in in capacity, fat. divided AC AC body were % exercise 4wk 8wk steady steady and 60min 60min 2max 2max Groups maximal area, duration: duration: 5/wk 5/wk 75%VO 60%VO the session: session: water. programme: parameters programme: water on 40 NR surface cold cold Frequency: Duration Intensity: body Programme Rpm: Exercise or Programme Frequency: Duration Intensity: Rpm: and based Exercise Exercise n n n n 4.7 5, 20±1 20±1 = # M M n water: water: water: water: 23.2± age: age: age: age: 15 18 = = 5, 5, 9, 9, 20.8±1.8 20°C = 32°C = age: land: n Sample n 20°C = 35°C = 23.0±4.1 to land and in and on metabolic how warm water adaption of aerobic tolerance training training to on training cold heat aim water thermal and determine effect affects compared cold physical Study To Comparison and endurance hot its capacity Ͳ Ͳ vs. cold land water vs. Study design Quasi experi ment: warm vs. Quasi experi ment: hot cold water (continued) Year 1982 1995, 1993 67 69,70 4.5 Author Young Avellini Table

93 Chapter4

meter; oxygen , 2

m, Neck Shoulder Immersion depths VO Ͳ Ͳ male; used M, Modified Monark EM (Morlock & Dressen dorfer) Modified Monark EM (Morlock & Dressen dorfer) Aquatic bike temperature; cycling; T, Ͳ Water 17 22°C 31°C T based Ͳ Ͳ volume; and and land fat the AC AC LC, exercise fat, stroke exercise exercise rate; exercise SV, body between between group throughout both heart in submaximal AC minute; weight, submaximal HR, submaximal at 2max intake at and at per change HR difference difference VO body BP LC SV in no no in in in in not in weight caloric training; did body revolution findings change 2max interval intervention VO Constant free Increase* intensities; LC Increase** groups Decrease** intensities; LC Decrease* intensities Key No <0.01 Rpm, trial; value cold intensity Ͳ Ͳ p in 28°C, at in high AC AC 30°C, deviation. controlled HIIT, 8wk 12wk steady steady 2max 2max 31°C, significant VO 90min 30min in determined **, female; water. 40%VO 80% F, Ͳ Ͳ duration: duration: was tests 5/wk 3/wk randomised 30 60 20°C session: session: <0.05; programme: programme: parameters water mean±standard T RCT, NR NR as and value Ͳ life; ergometer; p Frequency: Duration Intensity: Programme water. Exercise 24°C preliminary Rpm: Exercise Programme Frequency: Duration Intensity: Rpm: Exercise years of at in EM, 4 F 9 = 9 = n = # quality n M n 49±8 obese Celsius; significant presented C, QoL, *, is n=7 age: 31.4±11.1 Sample n=22 age: water: land: control: age and to the in week(s); pressure; and reported; blood to AC land water wk, to in if not of to blood affects leads whether healthy NR, shift BP, due endurance adaptations loss are training aim water investigate lactate; weight cold To Comparison water determine cephalad volume immersion normal aerobic training Study sclerosis; circumference; AC participants blood Ͳ waist LC multiple BL, Study design RCT: vs. vs. control Single group test retest WC, MS, (continued) Year otherwise 1986 1982 cycling; Watts; 68 66 W, 4.5 stated minute(s); aquatic not Author Sheldahl Sheldahl If uptake; Table AC, # min,

94 Aquaticcycling–whatdoweknow?

DISCUSSION

This is the first review to scope the available literature on headͲout aquatic cycling exercise. The aim of this review was to describe the study parameters of available researchutilisingaquaticcyclingasanexercisemodality.SixtyͲthreepublicationswere identifiedandthereviewprovidesafullsummaryofthesetͲupofaquaticinterventions andpossiblecomparisons,coreoutcomes,involvedparticipantsandthestudydesigns utilisedincurrentliterature.Theexplorationoftheinterventionparametersrevealed greatvarietyontheuseandexecutionofaquaticcycling. 4 LandͲbasedcyclingversusaquaticcycling The main body of the current research on aquatic cycling focuses on cardiovascular outcomes and the core findings for the comparison between landͲbased and waterͲ based cycling showed similar trends. These latter studies17Ͳ23,26 reported comparable

VO2maxvaluesofaquaticandlandͲbasedcyclingandtherefore,thecardiacdemandof aquatic cycling seems similar to landͲbased cycling. The results for HR were less consistent with a tendency for a lower HR during aquatic cycling compared to landͲ basedcycling.20Ͳ23,26,29,30Further,cardiacoutputandstrokevolumewasreportedtobe higher during aquatic cycling.15,21,29,30 These results are in line with the general understanding concerning the effects of water immersion on the human body. Hydrostaticpressureexertsexternalpressureontheimmersedbody,whichincreases with increased depth.2,74. Due to the hydrostatic pressure exerted there is a shift of blood from the extremities to the chest cavity, increasing arterial filling, and thus cardiacoutputandstrokevolumeareincreased.2,74Becausecardiovascularparameters aremodifiedbyimmersion,thiscouldexplainwhytheliteratureisinconclusiveonthe optimal recommendations for exercise prescription during aquatic cycling. Another explanation may be as most aquatic bikes are not equipped with an electronically controlled pedalling resistance mechanism and approaches to estimate VO2 from aquatic cycling are often based on pedalling frequency, with or without additional resistance. However, these equations cannot be used for all aquatic bikes, as the design and drag resistance created by pedals and resistance fins vary considerably acrosstheaquaticbikes.

Aquaticcyclingunderdifferentconditions Duetotheheterogeneousnatureofaquaticcycling,manyvariablesareinvolvedwhen studying its impact on individuals, for example deviceͲspecific factors44,63Ͳ65 or environmental parameters as water temperature.12,23,43,48,49,51,52,56Ͳ61 Thus explaining why the cardiovascular response to different exercise conditions was frequently investigated.Forexample,itseemsthatVO2maxiscomparableacrossdifferentwater

95 Chapter4 temperatures and that participants perceived exercising in warm water as more exhaustive.23,48,49 Further, included studies concluded that exercise intensities up to maximallimitsareachievedbyanincreaseinpedallingfrequencyandthatVO2peakdoes not differ between the different types of aquatic bikes.44,64 However, highͲpedal frequenciesaredifficulttomaintainduringlongerexercisesessionswithacontinuous character.44,64 To avoid discomfort with maintaining high pedal frequencies, exercise intensity can be modified by an increase in pedalling resistance or by utilising an interval training.45 The latter was perceived less exhaustive than a continuous protocol.45

Aquaticcyclingasanintervention Onlysixstudiesinvestigatedtheeffectofmultipleaquaticcyclingsessions.66Ͳ73Infour studiesaquaticcyclingwasusedinaclinicalcontextforpatientswithmultiplesclerosis andasexercisetrainingforolderadultsandobeseindividuals.Researchshowedthat aquatic cycling was equally effective than landͲbased cycling for improving cardiovascular fitness.66,68,71Ͳ73 Furthermore, none of the included studies reported adverseeventsrelatedtothetraining,suggestingthataquaticcyclingisasafeexercise modality. Most of the exercise protocols of the aquatic cycling intervention programmes consistedofsteadycyclinginaseatedpositionwithmoderateintensity.OnlyBoidinet al.usedanintervalprotocolforthetrainingofobeseindividuals.71Itseemsthatthefull potential of aquatic cycling including outͲofͲsaddle positions and arm and trunk exercisesisnotpublishedyetinpeerͲreviewedjournals.7Additionoftheseelements mightpreventmonotonyespeciallyinmultiplesessionprogrammes75andresultsfrom supportive literature suggest that a full spectrum aquatic cycling programme is effectiveinpatientswithmusculoskeletaldisorders.76 Thisscopingreviewhasidentifiedanumberofareasforfurtherresearch.Mostofthe included studies have a crossͲover design with few cycling sessions and investigated the exercise response in young healthy males, because gender, body mass and morphologyareknowntoaffecttheresponsetoaquaticcycling.59,77,78Further,onlysix studies investigated the effect of an aquatic cycling intervention programme. To improve the use of aquatic cycling in healthcare, future studies, preferably RCTs, shouldinvestigatetheeffectsofaquaticcyclinginterventionsindifferentpopulations andonoutcomessuchas(joint)pain,musclestrengthorphysicalfunctioning,which areyettobeinvestigated.Ofspecificinterestmayalsobethebiomechanicsofaquatic cyclinganddifferencesofseatedandoutͲofsaddlecycling.Furthermore,theidentified literatureseemssuitableformoresystematicreviews.Forexampleitseemsworthyto synthesizetheavailableevidenceoncardiovascularresponsestoaquaticcycling.

96 Aquaticcycling–whatdoweknow?

TofurtherimprovetheunderstandingofacuteandlongͲtermphysiologicaladaptions to aquatic cycling training and facilitate between study comparisons, consistent reporting of the following parameters is recommended. Studies should describe the typeofaquaticbike,bodyposition,levelofimmersion,watertemperature,methods usedtocontrolandassessexerciseintensityi.e.trainingfrequency,duration,rpmand pedallingresistance.Furthermore,itshouldbestatedwhetherornotadverseevents occurred.Inadditiontoanaccuratedescriptionoftheaquaticcyclingintervention,an agreement of experts on uniform keywords to describe the exercise activity is also strongly advised since this would improve the search in scientific databases. In this reviewtheterms“aquaticcycling”and“aquaticbike”wereused,astheseexpressions 4 nowadaysarecommonlyassociatedwiththistypeofexercise. This review has strengths and weaknesses. The extensive search procedure in this review resulted in more than sixty publications on aquatic cycling only, which were summarisedanddisplayed.However,thepresentedstudiesshouldbeinterpretedwith caution,becausenoqualityassessmentoftheinternalvalidityoftheincludedstudies wasmadeinordertocoverabroadspectrumofliterature.Furthermore,thisreview providesaverygeneraloverviewoftheresearchonaquaticcyclingwithoutfocusingon certaindetailsoftheincludedstudies.Forexample,onlythemainoutcomesreported in the abstract of the included studies were reported in this review. Yet, this comprehensive outline of available literature in this scoping review could serve as a startingpointforsystematicreviewsorclinicalstudiesontheeffectsofaquaticcycling onthecardiovascularresponses.

CONCLUSION

ThisisthefirstscopingreviewtosummarisetheliteratureonheadͲoutaquaticcycling. Therearenumerousvariablesrelatedtoaquaticcyclinge.g.,thetypeofaquaticbikeor environmental factors e.g., water temperature or immersion level. As a result, the objectives of the identified studies in this review are heterogeneous. Most of the includedstudiescomparedaquaticcyclingwithlandͲbasedcyclingorexaminedhowto quantifyandmodifyexerciseintensity.Veryfewstudiesevaluatedtheeffectofaquatic cycling interventions. Cardiovascular parameters were investigated by many of the studies and the results suggest that the cardiac demand of aquatic cycling seems similartolandͲbasedcycling.Therefore,furtherresearchshouldsynthesizetheeffects ofaquaticcyclingoncardiovascularparametersinasystematicreview.Futurestudies should evaluate the effects of aquatic cycling interventions in a clinical and rehabilitativecontext.

97 Chapter4

REFERENCES

1. Duddy JH. The simulation of weightlessness using water immersion techniques: an annotated bibliography.,ƵŵĂŶ&ĂĐƚŽƌƐ.1969;11:507Ͳ40. 2. Becker BE, Cole AJ. Comprehensive Aquatic Therapy. 3rd ed. Pullmann, WA: Washington State UniversityPublishing;2010. 3. Barker AL, Talevski J, Morello RT, Brand CA, Rahmann AE, Urquhart DM. Effectiveness of aquatic exercise for musculoskeletal conditions: a metaͲanalysis. ƌĐŚŝǀĞƐ ŽĨ WŚǇƐŝĐĂů DĞĚŝĐŝŶĞ ĂŶĚ ZĞŚĂďŝůŝƚĂƚŝŽŶ.2014;95:1776Ͳ86. 4. Bartels EM, Juhl CB, Christensen R, Hagen KB, DanneskioldͲSamsoe B, Dagfinrud H, et al. Aquatic exercise for the treatment of knee and hip osteoarthritis. ŽĐŚƌĂŶĞ ĂƚĂďĂƐĞ ^ǇƐƚ ZĞǀ. 2016;3:CD005523. 5. Fenzl M, Schnizer W, Hartmann B, Villiger B., Knüsel O. Unterschiede in der maximalen SauerstoffaufnahmebeikörperlichenBelastungenimWasseroderanLand.^ĐŚǁĞŝǌĞƌŝƐĐŚĞĞŝƚƐĐŚƌŝĨƚ ĨƺƌDĞĚŝǌŝŶƵŶĚdƌĂƵŵĂƚŽůŽŐŝĞ.2005;53:172Ͳ8. 6. Meyer K, Leblanc MC. Aquatic therapies in patients with compromised left ventricular function and heartfailure.ůŝŶŝĐĂůĂŶĚ/ŶǀĞƐƚŝŐĂƚŝǀĞDĞĚŝĐŝŶĞDĞĚĞĐŝŶĞůŝŶŝƋƵĞĞƚǆƉĞƌŝŵĞŶƚĂůĞ.2008;31:E90Ͳ7. 7. BarbosaTM,MarinhoDA,ReisVM,SilvaAJ,BragadaJ.PhysiologicalassessmentofheadͲoutaquatic exercisesinhealthysubjects:aqualitativereview.:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐ^ĐŝĞŶĐĞΘDĞĚŝĐŝŶĞ.2009;8:179Ͳ89. 8. ArmstrongR,HallBJ,DoyleJ,WatersE.CochraneUpdate.'Scopingthescope'ofacochranereview. :ŽƵƌŶĂůŽĨWƵďůŝĐ,ĞĂůƚŚ.2011;33:147Ͳ50. 9. ArkseyH,O'MalleyL.Scopingstudies:towardsamethodologicalframework./ŶƚĞƌŶĂƚŝŽŶĂů:ŽƵƌŶĂůŽĨ ^ŽĐŝĂůZĞƐĞĂƌĐŚDĞƚŚŽĚŽůŽŐǇ.2005;8:19Ͳ32. 10. FenzlM,KrebsJ,VilligerB,ZauggT,GredigJ,WalterB,etal.LeistungsmessungenimWassermiteinem neuen KraftmessͲKurbelsystem. WŚǇƐŝŬĂůŝƐĐŚĞ DĞĚŝǌŝŶ ZĞŚĂďŝůŝƚĂƚŝŽŶƐŵĞĚŝǌŝŶ <ƵƌŽƌƚŵĞĚŝǌŝŶ. 2012;22:183Ͳ8. 11. FenzlM,SchlegelC.,VilligerB.,AebliN.,GredigJ.,J.K.EignungdeshochfrequentenSpektralbandes der Herzratenvariabilität für die Leistungsdiagnostik im Wasser. WŚǇƐŝŬĂůŝƐĐŚĞ DĞĚŝǌŝŶ ZĞŚĂďŝůŝƚĂƚŝŽŶƐŵĞĚŝǌŝŶ<ƵƌŽƌƚŵĞĚŝǌŝŶ.2013;23:225Ͳ30. 12. Fenzl M, Schnizer W, Knüsel O, Hartmann B. HerzͲKreislaufͲBelastung bei körperlicher Arbeit am Wasserfahrrad bei verschiedenen Wassertemperaturen gemessen am DruckͲFrequenzͲProdukt. WŚǇƐŝŬĂůŝƐĐŚĞDĞĚŝǌŝŶZĞŚĂďŝůŝƚĂƚŝŽŶƐŵĞĚŝǌŝŶ<ƵƌŽƌƚŵĞĚŝǌŝŶ.2010;20:27Ͳ31. 13. Garzon M, Gayda M, Garzon L, Juneau M, Nigam A, Leone M, et al. Biomechanical analysis to determinetheexternalpoweroutputonanimmersibleergocycle.ƵƌŽƉĞĂŶ:ŽƵƌŶĂůŽĨ^ƉŽƌƚ^ĐŝĞŶĐĞ. 2015;15:271Ͳ8. 14. GarzonM,GaydaM,NigamA,ComtoisAͲS,JuneauM.Immersibleergocycleprescriptionasafunction of relative exercise intensity. :ŽƵƌŶĂů ŽĨ ^ƉŽƌƚ ĂŶĚ ,ĞĂůƚŚ ^ĐŝĞŶĐĞ. 2015;http://dx.doi.org/10.1016/ j.jshs.2015.12.004. 15. GarzonM,JuneauM,DupuyO,NigamA,BosquetL,ComtoisA,etal.Cardiovascularandhemodynamic responsesondrylandvs.immersedcycling.:ŽƵƌŶĂůŽĨ^ĐŝĞŶĐĞĂŶĚDĞĚŝĐŝŶĞŝŶ^ƉŽƌƚ.2015;18:619Ͳ23. 16. Garzon M, Dupuy O, Bosquet L, Nigam A, Comtois AS, Juneau M, et al. Thermoneutral immersion exerciseacceleratesheartraterecovery:Apotentialnoveltrainingmodality.ƵƌŽƉĞĂŶ:ŽƵƌŶĂůŽĨ^ƉŽƌƚ ^ĐŝĞŶĐĞ.2017;17:310Ͳ6. 17. Yazigi F, Pinto S, Colado J, Escalante Y, ArmadaͲdaͲSilva PA, Brasil R, et al. The cadence and water temperatureeffectonphysiologicalresponsesduringwatercycling.ƵƌŽƉĞĂŶ:ŽƵƌŶĂůŽĨ^ƉŽƌƚ^ĐŝĞŶĐĞ. 2013;13:659Ͳ65. 18. Wiesner S, Birkenfeld AL, Engeli S, Haufe S, Brechtel L, Wein J, et al. Neurohumoral and metabolic responsetoexerciseinwater.,ŽƌŵŽŶĞĂŶĚDĞƚĂďŽůŝĐZĞƐĞĂƌĐŚ.2010;42:334Ͳ9. 19. Sheldahl LM, Tristani FE, Connelly TP, Levandoski SG, Skelton MM, Cowley AW, Jr. FluidͲregulating hormones during exercise when central blood volume is increased by water immersion. ŵĞƌŝĐĂŶ :ŽƵƌŶĂůŽĨWŚǇƐŝŽůŽŐǇ.1992;262:R779Ͳ85.

98 Aquaticcycling–whatdoweknow?

20. Connelly TP, Sheldahl LM, Tristani FE, Levandoski SG, Kalkhoff RK, Hoffman MD, et al. Effect of increased central blood volume with water immersion on plasma catecholamines during exercise. : ƉƉůWŚǇƐŝŽů;ϭϵϴϱͿ.1990;69:651Ͳ6. 21. ChristieJL,SheldahlLM,TristaniFE,WannLS,SagarKB,LevandoskiSG,etal.Cardiovascularregulation duringheadͲoutwaterimmersionexercise.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1990;69:657Ͳ64. 22. SheldahlLM,WannLS,CliffordPS,TristaniFE,WolfLG,KalbfleischJH.Effectofcentralhypervolemia oncardiacperformanceduringexercise.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1984;57:1662Ͳ7. 23. Dressendorfer RH, Morlock JF, Baker DG, Hong SK. Effects of headͲout water immersion on cardiorespiratoryresponsestomaximalcyclingexercise.hŶĚĞƌƐĞĂŝŽŵĞĚŝĐĂůZĞƐĞĂƌĐŚ.1976;3:177Ͳ 87. 24. FenzlM,KarnerͲRezekK,SchlegelC,GredigJ,VilligerB.VO2kineticsduringdifferentformsofcycling exerciseonlandandinwater.^ƉŽƌƚƐĂŶĚǆĞƌĐŝƐĞDĞĚŝĐŝŶĞͲKƉĞŶ:ŽƵƌŶĂů.2015;1:64Ͳ70. 25. PeriniR,MilesiS,BiancardiL,PendergastDR,VeicsteinasA.Heartratevariabilityinexercisinghumans: 4 effect of water immersion. ƵƌŽƉĞĂŶ :ŽƵƌŶĂů ŽĨ ƉƉůŝĞĚ WŚǇƐŝŽůŽŐǇ ĂŶĚ KĐĐƵƉĂƚŝŽŶĂů WŚǇƐŝŽůŽŐǇ. 1998;77:326Ͳ32. 26. Chen AA, Kenny GP, Johnston CE, Giesbrecht GG. Design and evaluation of a modified underwater cycleergometer.ĂŶĂĚŝĂŶ:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1996;21:134Ͳ48. 27. FinkelsteinI,deFigueiredoPA,AlbertonCL,BgeginskiR,SteinR,KruelLF.Cardiorespiratoryresponses during and after water exercise in pregnant and nonͲpregnant women. ZĞǀŝƐƚĂ ƌĂƐŝůĞŝƌĂ ĚĞ 'ŝŶĞĐŽůŽŐŝĂĞKďƐƚĞƚƌşĐŝĂ͗ZĞǀŝƐƚĂĚĂ&ĞĚĞƌĂĕĆŽƌĂƐŝůĞŝƌĂĚĂƐ^ŽĐŝĞĚĂĚĞƐĚĞ'ŝŶĞĐŽůŽŐŝĂĞKďƐƚĞƚƌşĐŝĂ. 2011;33:388Ͳ94. 28. Hanna RD, Sheldahl LM, Tristani FE. Effect of enhanced preload with headͲout water immersion on exercise response in men with healed myocardial infarction. dŚĞ ŵĞƌŝĐĂŶ ũŽƵƌŶĂů ŽĨ ĐĂƌĚŝŽůŽŐǇ. 1993;71:1041Ͳ4. 29. SheldahlLM,TristaniFE,CliffordPS,HughesCV,SobocinskiKA,MorrisRD.EffectofheadͲoutwater immersion on cardiorespiratory response to dynamic exercise. :ŽƵƌŶĂů ŽĨ ƚŚĞ ŵĞƌŝĐĂŶ ŽůůĞŐĞ ŽĨ ĂƌĚŝŽůŽŐǇ.1987;10:1254Ͳ8. 30. McArdleWD,MagelJR,LesmesGR,PecharGS.Metabolicandcardiovascularadjustmenttoworkinair andwaterat18,25,and33degreesC.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1976;40:85Ͳ90. 31. CraigABJ,DvorakM.Comparisonofexerciseinairandinwaterofdifferenttemperatures.DĞĚŝĐŝŶĞ ĂŶĚ^ĐŝĞŶĐĞŝŶ^ƉŽƌƚƐĂŶĚǆĞƌĐŝƐĞ.1969;1:124Ͳ30. 32. McMurray RG, Fieselman CC, Avery KE, Sheps DS. Exercise hemodynamics in water and on land in patientswithcoronaryarterydisease.:ŽƵƌŶĂůŽĨĂƌĚŝŽƉƵůŵŽŶĂƌǇZĞŚĂďŝůŝƚĂƚŝŽŶ.1988;8:69Ͳ75. 33. Brechat PH, Wolf JP, SimonͲRigaud ML, Brechat N, Kantelip JP, Berthelay S, et al. Influence of immersiononrespiratoryrequirementsduring30Ͳmincyclingexercise.ƵƌŽƉĞĂŶZĞƐƉŝƌĂƚŽƌǇ:ŽƵƌŶĂů. 1999;13:860Ͳ6. 34. FenzlM,SchnizerW,AebliN,SchlegelC,VilligerB,DischA,etal.ReleaseofANPandfatoxidationin overweight persons during aerobic exercise in water. /ŶƚĞƌŶĂƚŝŽŶĂů :ŽƵƌŶĂů ŽĨ ^ƉŽƌƚƐ DĞĚŝĐŝŶĞ. 2013;34:795Ͳ9. 35. IsraelDJ,HeydonKM,EdlichRF,PozosRS,WittmersLE,Jr.Coretemperatureresponsetoimmersed bicycle ergometer exercise at water temperatures of 21 degrees, 25 degrees, and 29 degrees C. :ŽƵƌŶĂůŽĨƵƌŶĂƌĞĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.1989;10:336Ͳ45. 36. McArdleWD,MagelJR,SpinaRJ,GergleyTJ,TonerMM.ThermaladjustmenttocoldͲwaterexposurein exercisingmenandwomen.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1984;56:1572Ͳ7. 37. McMurrayRG,KatzVL,MeyerͲGoodwinWE,CefaloRC.Thermoregulationofpregnantwomenduring aerobicexerciseonlandandinthewater.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨWĞƌŝŶĂƚŽůŽŐǇ.1993;10:178Ͳ82. 38. KatzVL,McMurrayR,GoodwinWE,CefaloRC.Nonweightbearingexerciseduringpregnancyonland andduringimmersion:acomparativestudy.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨWĞƌŝŶĂƚŽůŽŐǇ.1990;7:281Ͳ4. 39. Di Masi F, De Souza Vale RG, Dantas EHM, Lopes Barreto AC, da Silva Novaes J, Reis VM. Is blood lactateremovalduringwaterimmersedcyclingfasterthanduringcyclingonland?:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐ ^ĐŝĞŶĐĞΘDĞĚŝĐŝŶĞ.2007;6:188Ͳ92.

99 Chapter4

40. FerreiraJ,DaSilvaCarvalhoR,BarrosoT,SzmuchrowskiL,ŒledziewskiD.Effectofdifferenttypesof recovery on blood lactate removal after maximum exercise. WŽůŝƐŚ :ŽƵƌŶĂů ŽĨ ^ƉŽƌƚ ĂŶĚ dŽƵƌŝƐŵ. 2011;18:105Ͳ11. 41. WahlP,SannoM,EllenbergK,FrickH,BohmE,HaiduckB,etal.Aquacyclingdoesnotaffectrecovery ofperformance,damagemarkersandsensationofpain.:ŽƵƌŶĂůŽĨ^ƚƌĞŶŐƚŚĂŶĚŽŶĚŝƚŝŽŶŝŶŐZĞƐĞĂƌĐŚ. 2016;10.1519/JSC.0000000000001462. 42. SosnerP,GaydaM,DupuyO,GarzonM,LemassonC,GremeauxV,etal.Ambulatorybloodpressure reductionfollowinghighͲintensityintervalexerciseperformedinwaterordrylandcondition.:ŽƵƌŶĂůŽĨ ƚŚĞŵĞƌŝĐĂŶ^ŽĐŝĞƚǇŽĨ,ǇƉĞƌƚĞŶƐŝŽŶ.2016;10:420Ͳ8. 43. McArdleWD,TonerMM,MagelJR,SpinaRJ,PandolfKB.Thermalresponsesofmenandwomenduring coldͲwater immersion: influence of exercise intensity. ƵƌŽƉĞĂŶ :ŽƵƌŶĂů ŽĨ ƉƉůŝĞĚ WŚǇƐŝŽůŽŐǇ ĂŶĚ KĐĐƵƉĂƚŝŽŶĂůWŚǇƐŝŽůŽŐǇ.1992;65:265Ͳ70. 44. Giacomini F, Ditroilo M, Lucertini F, De Vito G, Gatta G, Benelli P. The cardiovascular response to underwaterpedalingatdifferentintensities:acomparisonof4differentwaterstationarybikes.:ŽƵƌŶĂů ŽĨ^ƉŽƌƚƐDĞĚŝĐŝŶĞĂŶĚWŚǇƐŝĐĂů&ŝƚŶĞƐƐ.2009;49:432Ͳ9. 45. Brasil RM, Barreto AC, Nogueira L, Santos E, Novaes JS, Reis VM. Comparison of physiological and perceptual responses between continuous and intermittent cycling. :ŽƵƌŶĂů ŽĨ ŚƵŵĂŶ ŬŝŶĞƚŝĐƐ. 2011;29A:59Ͳ68. 46. DionneA,LeoneM,AndrichDE,PerusseL,ComtoisAS.Acutebreathingpatternsinhealthyandheart disease participants during cycling at different levels of immersion. ZĞƐƉŝƌĂƚŽƌǇ WŚǇƐŝŽůŽŐǇ Θ EĞƵƌŽďŝŽůŽŐǇ.2017;235:1Ͳ7. 47. McMurray RG, Katz VL, Berry MJ, Cefalo RC. Cardiovascular responses of pregnant women during aerobicexerciseinwater:alongitudinalstudy./ŶƚĞƌŶĂƚŝŽŶĂů:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐDĞĚŝĐŝŶĞ.1988;9:443Ͳ7. 48. FujimotoT,SasakiY,WakabayashiH,SengokuY,TsubakimotoS,NishiyasuT.Maximalworkloadbut notpeakoxygenuptakeisdecreasedduringimmersedincrementalexerciseatcoolertemperatures. ƵƌŽƉĞĂŶ:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.2016;116:1819Ͳ27. 49. TonerMM,DroletLL,PandolfKB.Perceptualandphysiologicalresponsesduringexerciseincooland coldwater.WĞƌĐĞƉƚƵĂůĂŶĚDŽƚŽƌ^ŬŝůůƐ.1986;62:211Ͳ20. 50. Craig AB, Jr., Dvorak M. Thermal regulation of man exercising during water immersion. :ŽƵƌŶĂů ŽĨ ƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1968;25:28Ͳ35. 51. WhiteLJ,DressendorferRH,HollandE,McCoySC,FergusonMA.Increasedcaloricintakesoonafter exerciseincoldwater./ŶƚĞƌŶĂƚŝŽŶĂů:ŽƵƌŶĂůŽĨ^ƉŽƌƚEƵƚƌŝƚŝŽŶĂŶĚǆĞƌĐŝƐĞDĞƚĂďŽůŝƐŵ.2005;15:38Ͳ 47. 52. McMurrayRG,KocherPL,HorvathSM.AerobicpowerandbodysizeaffectstheexerciseͲinducedstress hormone responses to varying water temperatures. ǀŝĂƚŝŽŶ ^ƉĂĐĞ ĂŶĚ ŶǀŝƌŽŶŵĞŶƚĂů DĞĚŝĐŝŶĞ. 1994;65:809Ͳ14. 53. KatzVL,McMurrayR,BerryMJ,CefaloRC,BowmanC.Renalresponsestoimmersionandexercisein pregnancy.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨWĞƌŝŶĂƚŽůŽŐǇ.1990;7:118Ͳ21. 54. Katz VL, McMurray R, Berry MJ, Cefalo RC. Fetal and uterine responses to immersion and exercise. KďƐƚĞƚƌŝĐƐĂŶĚ'ǇŶĞĐŽůŽŐǇ.1988;72:225Ͳ30. 55. McMurrayRG,KatzVL,BerryMJ,CefaloRC.Theeffectofpregnancyonmetabolicresponsesduring rest, immersion, and aerobic exercise in the water. ŵĞƌŝĐĂŶ :ŽƵƌŶĂů ŽĨ KďƐƚĞƚƌŝĐƐ ĂŶĚ 'ǇŶĞĐŽůŽŐǇ. 1988;158:481Ͳ6. 56. McMurrayRG,BerryMJ,KatzVL,GraetzerDG,CefaloRC.Thethermoregulationofpregnantwomen duringaerobicexerciseinthewater:alongitudinalapproach.ƵƌŽƉĞĂŶ:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ ĂŶĚKĐĐƵƉĂƚŝŽŶĂůWŚǇƐŝŽůŽŐǇ.1990;61:119Ͳ23. 57. McMurrayRG,HorvathSM.Thermoregulationinswimmersandrunners.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ. 1979;46:1086Ͳ92. 58. Golden FS, Tipton MJ. Human thermal responses during legͲonly exercise in cold water. :ŽƵƌŶĂů ŽĨ WŚǇƐŝŽůŽŐǇ.1987;391:399Ͳ405. 59. Toner MM, Sawka MN, Foley ME, Pandolf KB. Effects of body mass and morphology on thermal responsesinwater.:ƉƉůWŚǇƐŝŽů;ϭϵϴϱͿ.1986;60:521Ͳ5.

100 Aquaticcycling–whatdoweknow?

60. TonerMM,SawkaMN,HoldenWL,PandolfKB.Comparisonofthermalresponsesbetweenrestandleg exerciseinwater.:ƉƉůWŚǇƐŝŽů;ϭϵϴϱͿ.1985;59:248Ͳ53. 61. Toner MM, Sawka MN, Pandolf KB. Thermal responses during arm and leg and combined armͲleg exerciseinwater.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1984;56:1355Ͳ60. 62. Pinto SS, Brasil RM, Alberton CL, Ferreira HK, Bagatini NC, Calatayud J, et al. NonͲInvasive DeterminationoftheAnaerobicThresholdBasedontheHeartRateDeflectionPointinWaterCycling. :ŽƵƌŶĂůŽĨ^ƚƌĞŶŐƚŚĂŶĚŽŶĚŝƚŝŽŶŝŶŐZĞƐĞĂƌĐŚ.2015;10.1519/JSC.0000000000001099. 63. Morlock JF, Dressendorfer RH. Modification of a standard bicycle ergometer for underwater use. hŶĚĞƌƐĞĂŝŽŵĞĚŝĐĂůZĞƐĞĂƌĐŚ.1974;1:335Ͳ42. 64. Shapiro Y, Avellini BA, Toner MM, Pandolf KB. Modification of the Monark bicycle ergometer for underwaterexercise.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1981;50:679Ͳ83. 65. SogabeY,MonjiK,NakashimaK,TajimaF,IwamotoJ.Modificationofaconventionalbicycleergometer forunderwateruse.:ŽƵƌŶĂůŽĨhK,.1987;9:279Ͳ85. 4 66. SheldahlLM,BuskirkER,LoomisJL,HodgsonJL,MendezJ.Effectsofexerciseincoolwateronbody weightloss./ŶƚĞƌŶĂƚŝŽŶĂů:ŽƵƌŶĂůŽĨKďĞƐŝƚǇ.1982;6:29Ͳ42. 67. Avellini BA, Shapiro Y, Fortney SM, Wenger CB, Pandolf KB. Effects on heat tolerance of physical traininginwaterandonland.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1982;53:1291Ͳ8. 68. Sheldahl LM, Tristani FE, Clifford PS, Kalbfleisch JH, Smits G, Hughes CV. Effect of headͲout water immersiononresponsetoexercisetraining.:ŽƵƌŶĂůŽĨƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.1986;60:1878Ͳ81. 69. YoungAJ,SawkaMN,QuigleyMD,CadaretteBS,NeuferPD,DennisRC,etal.Roleofthermalfactorson aerobiccapacityimprovementswithendurancetraining.:ƉƉůWŚǇƐŝŽů;ϭϵϴϱͿ.1993;75:49Ͳ54. 70. YoungAJ,SawkaMN,LevineL,BurgoonPW,LatzkaWA,GonzalezRR,etal.Metabolicandthermal adaptationsfromendurancetraininginhotorcoldwater.:ƉƉůWŚǇƐŝŽů;ϭϵϴϱͿ.1995;78:793Ͳ801. 71. BoidinM,LapierreG,PaquetteTanirL,NigamA,JuneauM,GuilbeaultV,etal.Effectofaquaticinterval trainingwithMediterraneandietcounselinginobesepatients:resultsofapreliminarystudy.ŶŶĂůƐŽĨ WŚǇƐŝĐĂůĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶDĞĚŝĐŝŶĞ.2015;58:269Ͳ75. 72. BansiJ,BlochW,GamperU,KesselringJ.TraininginMS:influenceoftwodifferentendurancetraining protocols(aquaticversusoverland)oncytokineandneurotrophinconcentrationsduringthreeweek randomizedcontrolledtrial.DƵůƚŝƉůĞ^ĐůĞƌŽƐŝƐ.2012;19:613Ͳ21. 73. Bansi J, Bloch W, Gamper U, Riedel S, Kesselring J. Endurance training in MS: shortͲterm immune responses and their relation to cardiorespiratory fitness, healthͲrelated quality of life, and fatigue. :ŽƵƌŶĂůŽĨEĞƵƌŽůŽŐǇ.2013;260:2993Ͳ3001. 74. BeckerBE.Aquatictherapy:scientificfoundationsandclinicalrehabilitationapplications.WDΘZ͗ƚŚĞ ũŽƵƌŶĂůŽĨŝŶũƵƌǇ͕ĨƵŶĐƚŝŽŶ͕ĂŶĚƌĞŚĂďŝůŝƚĂƚŝŽŶ.2009;1:859Ͳ72. 75. RewaldS,MestersI,EmansPJ,ArtsJJ,LenssenAF,deBieRA.AquaticcircuittrainingincludingaquaͲ cycling in patients with knee osteoarthritis: A feasibility study. :ŽƵƌŶĂů ŽĨ ZĞŚĂďŝůŝƚĂƚŝŽŶ DĞĚŝĐŝŶĞ. 2015;10.2340/16501977Ͳ1937. 76. MoserS.EntwicklungundÜberprüfungeinesAquaͲCyclingͲProgrammsfürRheumapatienten[Master]. Karlsruhe:UniversitätKarlsruhe;2009. 77. Stachenfeld NS, Taylor HS. Challenges and methodology for testing young healthy women in physiologicalstudies.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨWŚǇƐŝŽůŽŐǇ͗ŶĚŽĐƌŝŶŽůŽŐǇĂŶĚDĞƚĂďŽůŝƐŵ.2014;306:E849Ͳ 53. 78. Costello JT, Bieuzen F, Bleakley CM. Where are all the female participants in Sports and Exercise Medicineresearch?ƵƌŽƉĞĂŶ:ŽƵƌŶĂůŽĨ^ƉŽƌƚ^ĐŝĞŶĐĞ.2014;14:847Ͳ51.

101 Chapter4

APPENDIX4.1

PostͲhocanalysisofsearchterms Across all included articles it was explored how the exercise activity and the water exercise device (ergometer) were described. The majority (84%) of the studies described the type of exercise in a specific manner rather than using general expressions such as “water exercise” or “immersed leg exercise”. Also the word “exercise”(37%)or“cycling”(32%)wascombinedwithatermfortheexercisedevice and the exercise environment. Examples of these descriptions are “exercise on a bicycle ergometer (during immersion) in water”, “cycling in water immersion” or “watercyclingexercise”.Nineexpressions(15%)werefoundthatextendeddescription by specifying the exercising limb (e.g. performing leg cycle exercise in water) or the bodypositionontheexercisedeviceresultingindescriptionssuchas“uprightcycling exercisesinwater”. Likewisetotheexerciseactivitydescriptions,mostauthors(82%)clearlyindicatethat thedeviceisusedinwater.Inaddition,theterm“ergometer”isfrequentlyused(68%) andcombinedwiththeword“(bi)cycle”.Thisresultsindescriptionssuchas“(bi)cycle ergometer immersed/used in water”, “(under)water (bi)cycle ergometer” or “immersible/submersibleergometer”.InrecentpublicationsofGarzonetal.theshort description “immersible ergocycle” was introduced. Other descriptions like “whole body ergometer” or “arm / leg ergometer for use in water” focus on the exercising limb(s)oremphasizethattheergometercanbeusedonlandandinwaterbynaming thedevicean“airͲwaterergometer”.From2007onwardstheterms“aqua(tic)bike”, “water(stationary)bike”or“aquacycle”areoccasionallyusedandaccountfor13%of allidentifieddescriptions.

102 Aquaticcycling–whatdoweknow?

APPENDIX4.2

Developmentoftheinclusionandexclusioncriteria Theinclusionandexclusioncriteriaweredevelopedintwostages.Priortoscreening anyarticles,theauthorsagreedtoincludeallformatsoffullͲtextresearchreportsthat focusedontheeffectsofheadͲoutimmersedcyclingonthehumanbody(Table4.2, stage one). Studies that investigated the effect of fullͲbody immersed underwater cycling,possiblyincombinationwithaselfͲcontainedunderwaterbreathingapparatus (SCUBA),wereexcluded.Tworesearchers(BW,SR)screenedalltitlesandabstractsto 4 selectarticlesforfullͲtextreview.Theindependentscreeningandselectionofarticles was done with the online programme ‘Covidence’ (Covidence systematic review software, Veritas Health Innovation, Melbourne, Australia, available at www.covidence.org). Next, to become familiar with the literature and to check whethertheinͲandexclusioncriterianeededmorespecificationapilotfullͲtextreview ofarticlesidentifiedfromscientificdatabases(n=68)wascarriedout.Tosavetimeand to stimulate discussion the pilot fullͲtext screening was split between different reviewers.Thesixreviewershavedifferentbackgroundsinphysicaltherapy(AFL,RB, BW,SR),epidemiology(RB,IM),exercisephysiology(JB)andaquatictherapy(BW,SR). Thirteen to 14 fullͲtext versions were screened per reviewer and one reviewer (SR) screenedallarticles.Disagreementbetweenthereviewersconcernedwhetherornot toincludestudiesthatdonotfocusonaquacyclingasanexerciseactivity,buttostudy physiological responses as body temperature to immersion. Consequently, the reviewers agreed that these studies might contain useful information as long as the experiments were carried out under usual exercise conditions. Furthermore, the reviewersdecidednottoexcludeacertaintypeofergometer.However,thelimbshad tobeimmersedduringexerciseandthesubjectsshouldbeseatedinuprightorsemi recumbent position during the exercise. All inclusion and exclusion criteria are summarised in Table 4.2. Finally, two reviewers (SR, AFL) used the extended list of inclusiononallidentifiedfullͲtextarticles.

103 Chapter4 1997 1984 1996 1995 1996 2001 1990 1982 2007 1976 Date 1992 1997 1996 1998 1987 2007 2004 1996 1996 1987 1977 1991 2004 1996 H A O PW CG CE CE CE J IB L C TC JM F JS SG MC DS YS author AL H DT AC Park Raglin Veicsteinas Hesselberg Stocks Iwamoto First Rim Chu Johnston Hochachka Mekjavic O’Brien Newstead Wakabayashi Mourot Johnston Lee Pirnay Cross Yeon Ertl Passias Rhind Johnston Ͳ cold intensity mild Ͳ cooling low peptides. moderate moderate work core during subsets both exercise atmospheric during rate during cooling levels oxide muscular clamping: divers. natriuretic state Ͳ during and to leukocyte by accelerates various suit cooling Ͳ apparent at swimsuit humans nitrous and shoulder and same steady wet in core not heating the and 30% is responses temperature hindered rate is of thermal exercise female hip, a active are intensity core anxiety Ͳ leukocytes affecting man thresholds immersion by water. cooling in low on intensities knee, trait immersion during Korean respiratory at wearing inhalation in cold in core water without underwater by in response and during effects dehydration water Ͳ exercise out abolished Ͳ humans men is children differing during in cold exercise consumption two water humans immersion increases clamp: induced head in suits responses to and circulatory ventilation cool and divers immersion resting water in oxygen Ͳ on exercising wet rest thermal narcosis exercise post at a immersion thermoregulatory and scuba underwater cold hyperthermia temperature and and prepubescent threshold to thermal, response in to threshold of exercise hypoxia cold repeated on during during core water during tissue renal without resting of and with mechanisms man immersion in thermoregulatory cold exertional on exercise ventilation human and body shivering in responses on adjustments following of with regulation the temperature with insulation during during vasoconstriction temperature lowers insulation exercise between regulatory insulation shifts exercise the body hyperthermia measurement underwater metabolic lowers shivering. water body of vasomotor thermal insulation exercise exercise to shell pill habituation changes and accumulation of in hypoxia induction water in in physical hypoglycemia and and lowers temperature volume of of title term Ͳ water effects relationship Ͳ Effect Superficial Hypercapnia Human Insulation The cold Study Metabolic Telemetry The Influence Short pressures Plasma Endurance Eucapnic Succinate Thermal exercise intensity Cytokine Effect Changes Alcohol Decrease Changes neuroendocrine Metabolic Responses exposure s hot e l 3 or c exercise participants’ i t X r immersion of I a immersion, D d criteria (<10°C) aquatic N e water head d E duration cold u P l c P x Exclusion (>41°C) manipulation homeostasis Unusual above conditions: long very A E

104 Aquaticcycling–whatdoweknow? 2005 2004 2005 2003 1986 1999 2004 1971 2004 1986 1999 2009 1987 Date 2006 2009 1991 2015 2008 2006 2012 1983 2014 2014 2004 MA T D P IB T TM N PH KB J S BA K K K M DL F M KS author SH Meyer Katz Sheldahl Pandolf Meyer Yun Costill Barbosa First Park Christie Miyamoto Kinugasa Mekjavic Rewald Zamparo Graef Bréchat Avellini Crampton Leone Bermingham Wadell Fenzl Hayashi and Land and an volume Water cycling 4 exercise: COPD study. oder Cold blood sprint in with environment of review dimensional Ͳ divers. Wasser central land feasibility immersed two im A by to hold of Ͳ patients Immersion humans for restoration qualitative in minutes a breath relation during 30 ergocycle in assists accelerated Belastungen comparison modality cranking is osteoarthritis: Ͳ after water women subjects: players allowed? in better arm and knee Responses be training and immersible differences exercise: exercise Korean healthy soccer with körperlichen sitting an in during in land recovery to on bei Thermal during land on youth effective swimming intensity on on in watercraft patients an balance thermodilution in ͲͲ exercises and environment: immersion output and exercise work response water Ͳ immersion exercise reduction. and fluid driven Gender review Ͳ water the immersion. a high cycling water strategies water aquatic Ͳ in therapy – in and aquatic cardiac uptake acute water of programmes in pedal weight at out out Ͳ thoracic Ͳ water aqua and oximetry, aqua Sauerstoffaufnahme cold incremental after and with onset a training and recovery oxygen exercise head head Fick training the during to exertion of the prescription during should volume pregnancy. Morphology zone" at including during group patients match rehabilitation during Ͳ maximalen Herzinsuffizienz techniques physical "null locomotion delays Mass, stroke during failure: post bei der exercise of responses regulation of perceived requirements training increase cardiac validation of physical exercise in assessment for Body of heart water and immersion rate of arm in in echocardiography, in circuit ergometric requirements balance respiratory output immersion Ͳ lipids title rate temperature intensity heart comparison Exercise Wassertherapie Energy Special Influence Exercise Doppler Aquatic Cardiovascular Cardiovascular A Study Heart Core Hemodynamic Energy Including Blood Determination loading applications High Caution Cardio Unterschiede Physiological performance Power The Water of about immersed summary a not provided cycling information limb criteria intervention provides aquatic Article Exclusion information Wrong Insufficient the intervention elsewhere Exercising Reviews

105 Chapter4

106 Chapter4

CHAPTER5

Efficacyofaquaticcyclingonkneepainandphysical functioninginpatientswithkneeosteoarthritis: arandomisedcontrolledtrial

StefanieRewald IlseMesters A.F.TonLenssen PieterJ.Emans GerardvanBreukelen RobA.deBie ^ƵďŵŝƚƚĞĚ

106 Chapter5

ABSTRACT

Objective Toevaluatetheeffectsofaquaticcyclingonkneepainandphysicalfunctioninginpatientswith kneeosteoarthritis(OA). Methods A randomised controlled trial compared a twelveͲweek aquatic cycling exercise programme (24sessions)withusualcareforpatientswithmildtomoderatekneeOA.Aftertheendofthe trial period, the usual care group was invited to twelve aquatic cycling sessions. Primary outcomes were the Knee Injury and Osteoarthritis Outcome Score (KOOS)on knee pain and physical function. Secondary outcomes were KOOS scores on symptoms, sport, and diseaseͲ specificqualityoflife(QoL),diseaseseverity,physicalactivity,selfͲefficacy,kinesiophobia,RandͲ 36 QoL, timedͲupͲandͲgo test (TUG), sixͲminute walking test (6ͲMWT), and leg strength. Outcomes were assessed at baseline, postͲintervention and 24Ͳweeks followͲup. Multilevel (mixedregression)analysisexaminedtheeffects. Results Onehundredelevenpatientswererandomised,9patientswithdrewbeforebaseline,19patients were lost to followͲup. Average attendance rate for the aquatic cycling sessions was 80%. StatisticallysignificantdifferencesatpostͲinterventionandfollowͲupwerefoundforkneepain (KOOS:B=8.16,p=0.014),physicalfunctioning(KOOS:B=7.16,p=0.027;TUG:B=Ͳ0.91,p=0.001; 6ͲMWT:B=46.75,p=0.002)andkinesiophobia(B=Ͳ3.84,p=0.002)infavouroftheaquaticgroup. LikewiseashortͲtermimprovementindiseaseͲspecificQoL(B=13.03,p=0.001)wasdetectedin theaquaticgroup. Conclusion ThisisthefirsttrialinvestigatingagroupͲbasedaquaticcyclingprogrammeforpatientswithmild to moderate knee OA. Compared to usual care aquatic cycling seems effective in improving physicalfunctioningandkneepain.

108 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis

INTRODUCTION

Knee osteoarthritis (OA) is a degenerative disease of synovial joints and the most prominentsymptomsarepainandlossofphysicalfunction.InthearrayofnonͲsurgical treatment options, exercise therapy has been shown to reduce pain and improve physicalfunctioninginkneeOA.1AgroupͲbasedaquaticcyclingprogrammemightbea useful extension of exercise options for this population. Many OA patients prefer to exercise in water, especially when exercising on land is too painful.2 For example, stationarylandͲbasedcyclingisoftenrecommended3,4forpatientswithkneeOA,butis restricted to seated cycling positions to prevent weightͲbearing on the knee joints. Aquatic cycling provides a variety of exercises, because outͲofͲsaddle positions are possibleduetothebuoyancyofthewaterandupperbodyexercisecanbecombined withseatedcycling.OurfeasibilitystudyshowedthatkneeOApatientsfeelsecureon 5 thebikeandevaluateaquaticcyclingasapleasurableactivity.5However,inthisstudy aquaticcyclingwaspartofanaquaticcircuittraining;thereisnohighͲqualityevidence availableonthesoleeffectofaquaticcyclingtrainingontheimpairmentsofkneeOA.5,6 Therefore, we developed a full aquatic cycling training programme.7 The aim of the present randomised controlled trial (RCT) is to evaluate the efficacy of a 12Ͳweek groupͲbased aquatic cycling programme compared to usual care of an early OA outpatientcliniconkneepainandphysicalfunctioninginpatientswithkneemildto moderatekneeOA.

PATIENTSANDMETHODS

Design ThestudywasasingleͲblind,parallelͲgroupRCTinpatientswithmildtomoderateknee OA who consulted Maastricht University Medical Centre+ (MUMC+). The Medical EthicsCommitteeoftheMUMC+approvedthestudy.Thestudywasregisteredatthe NetherlandsTrialRegister:NTR3766.

Participants BetweenMarch2013andOctober2015participantswererecruitedfromtheEarlyOA Outpatient Clinic of the MUMC+. Patients with knee OA who rated their knee pain between4and7onatenͲpointnumericratingscale8andaKellgren/Lawrencescore9 between1and3wereeligibleforthestudy.Furthermore,eligiblepatientshadaclear indicationforconservativetreatmentofOA,includingaprimarycarephysicaltherapy referral. In addition, patients were able to cycle on a stationary bike, were able to understandandwriteDutchandscored8pointsorlowerontheHospitalAnxietyand

109 Chapter5

Depression Scale (HADS).10 Exclusion criteria were any contraͲindication for aquatic exercise therapy like open wounds, unstable angina or an acute flareͲup of joint inflammation. Further exclusion criteria were a planned total knee surgery, a corticosteroidinjectionlessthanthreemonthsagoorahyaluronicacidinjectionless thansixmonthsago,severejointcomplaintselsewhere,symptomaticandradiological provenhipOA,inflammatoryjointdiseases,andtheinabilitytosafelyenterandexit thepoolorfearofwater.

Randomisationandblinding After obtaining written informed consent participants were randomised with a constantblocksizeofeightandanallocationratioof1:1byanindependentresearch assistant of the Department of Epidemiology of Maastricht University to guarantee concealment(availableat:www.randomizer.org). Blindedphysicaltherapistscollectedalldataonpaperandblindedresearchassistants entered the paperͲbased data. Aquatic cycling trainers and patients could not be blinded.

Intervention Beforebaselineassessment,allparticipantsreceivedcareasusualfromtheEarlyOA OutpatientsClinicoftheMUMC+.UsualcareincludeddiagnosisofkneeOA,followed by information about knee OA (verbal information by the nurse practitioner and a booklet: https://centraal.mumc.nl/sites/central/files/23999Ͳ0711_Arthrose.pdf), indication of personal risk factors and a tailored treatment plan including pain medicationandareferralforprimarycarephysicaltherapy. hƐƵĂůĐĂƌĞŐƌŽƵƉ;hͿ TheUCcontinuedwiththetailoredtreatmentplanandcoulddecideiftheywantedto startwithphysicaltherapy.Participationinphysicaltherapywasnotobliged.Priorto postͲtestand24ͲweeksfollowͲupassessment,informationonusualcareroutinewas obtainedduringashorttelephoneinterview.Afterthelastmeasurementat24Ͳweeks, people in UC were invited to twelve weekly sessions of aqua cycling in a local communitypoolinMaastricht.

ƋƵĂƚŝĐĐǇĐůŝŶŐŐƌŽƵƉ;Ϳ Participantsintheinterventiongroupwereinstructednottostartadditionalphysical therapy during the intervention period of twelve weeks. Twice a week participants exercisedfor45minutesinsmallgroupsofmaximallyfourparticipantssupervisedbya physical therapist. Participants cycled in an upright position on the aqua bike „AquaCruiserIITM“throughoutthewholesession.Thedepthofthetherapypoolwas

110 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis adjustedtoensurethatthelegswereimmersedduringthewholemovement.Typically participants were immersed between the xiphoid process and the first rib in warm water(32°C).Themainpartofthetrainingconsistedofcyclinginasittingpositionwith goodposturalcontrol.Inaddition,outͲofͲtheͲsaddlepositions,legexercises,andupper bodyexerciseswereincorporated.ExerciseintensitywascontrolledbytheBorgScale, heartrate(220ͲminusͲageformula),andpedallingtempo.Adetaileddescriptionofthe aquaticcyclingprogrammecanbefoundinourpublishedstudyprotocol.7

Measurements Alloutcomeswereassessedatbaseline,12ͲweekspostͲinterventionandat24Ͳweeks followͲup. Physical therapists of the MUMC+ collected the data during a oneͲhour appointment. To minimise the number of missing data, especially on the primary 5 outcomes,thequestionnairesweresentbymailtoanyparticipantunabletocometo theMUMC+.

Primaryoutcome PrimaryoutcomesweretheselfͲreportedscoreonkneepainandphysicalfunctioning ofthepreviousweekassessedwiththeKneeInjuryandOsteoarthritisOutcomeScore questionnaire(KOOS,http://www.koos.nu).Thesumscorespersubscalewerelinearly transformedtoascalefromzero(severesymptoms)to100(nosymptoms).

Secondaryoutcomes SecondaryoutcomeswerescoresfortheKOOSsubscalessymptoms,sportsactivityand diseaseͲrelatedqualityoflife.Furthermore,physicalfunctioningonthetestdaywas assessedwiththeLEFS;LowerExtremityFunctionScale.11Painintensitywasmeasured withthetenͲpointNumericPainRatingScale(NRS)afterthesixͲminutewalkingtest.8 The Patient Global Assessment (PGA) evaluated patientͲrated general health.12 The ShortQUestionnairetoASsessHealthͲenhancingphysicalactivity(SQUASH)collected dataonhabitualphysicalactivityduringthepreviousmonth13,14ofwhichtotalminutes ofphysicalactivitywerecalculated.Moreover,qualityoflifewasmeasuredwiththe Rand36ͲitemHealthSurveyandthemental(RandͲ36MCS)andphysicalcomponent score (RandͲ36 PCS) were calculated.15 The Tampa Scale for Kinesiophobia (TSK) measured fear of (re)Ͳinjury to movement16 and four questions of the Arthritis SelfͲ EfficacyScale(ASES)assessedselfͲefficacyforphysicalfunctioning.17Thetotalscoresof theTSKandASESwerecomputed.Performancemeasuresofphysicalfunctioningwere the sixͲminute walking test (6ͲMWT)18,19 and the timedͲupͲandͲgo test (TUG).20,21 Isometric and isokinetic muscle strength of the affected leg was measured with the Biodex®System3Pro.22

111 Chapter5

Thepsychometricpropertiesoftheabovementionedoutcomeshavebeendescribed previously.7

Assessmentofcovariates,sessionattendance,andadverseevents Kellgren/Lawrence scores, body mass index (BMI), and comorbidities were obtained from medical records of the MUMC+. In addition, history of comorbidity was completed by a short telephone interview after participants had given informed consent.TheresearchersusedtheChronicIllnessResourcesSurvey(CIRS)tocategorize comorbidities.23 In the AC the physical therapists monitored the attendance in an exercise logbook. Theynotedthedate,thenumberofsessionsattended,pedallingtempoandresistance, BORGscoreandheartrateduringthemainpartofthetraining,andtheoccurrenceof adverseeventsorproblemswiththeperformanceofcertainexercises.

Samplesize Sample size calculation was based on the primary endpoint selfͲreported knee pain.1,24,25Asamplesizeof168seemedsufficienttodetectagroupdifferenceinknee painwithaneffectsizeofd=0.5.ThispreͲcalculatedsamplesizewasreportedinour study protocol and was based on the results of a study with a comparable design (aquatic therapy vs. control) and target group.26 Thus, with a twoͲsided significance levelof5%,astatisticalpowerof80%andanexpecteddropoutof25%,84patientsper arm(168intotal)wouldberequired.

Statisticalanalysis Data were analysed according to the intentionͲtoͲtreat principle by including all randomisedpatientswhoweremeasuredatleastonce,usingmultilevel(mixed)linear regressionanalysisperoutcome.Mixedlinearregressionmodelsconsistofarandom part(variancesof,andcorrelationsbetween,therepeatedoutcomemeasures)anda fixedpart(regressionweights,effectsofpredictorsonoutcomemeanpertimepoint). Fortherandompartofthemodel,anunstructuredcovariancematrixwasused.The fixed(predictor)modelpartconsistedoftime(usingdummycodingforpostͲtestand followͲup,withbaselineasreferencetimepoint),treatmentgroup(AC=1,UC=0),and patient characteristics as predictors. The latter were age (in years), sex (female=1, male=0),BMI,quadricepsstrength(inNewtonmeter(NM),correctedforbodyweight), tibiofemoral and patellofemoral Kellgren/Lawrence score and comorbidity count.27 Furthermore, the interaction of group with time was added to the model since this interaction represents the group difference in change from baseline to postͲ interventionandfollowͲup,andisthustheeffectofinterestinthisRCT.Moreover,to exploretheeffectsofphysicaltherapy,resp.theeffectofbeinginformedaboutthe

112 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis freeaquaticcyclingprogrammeafterfollowͲupintheUCarm(14UCpatientsdidnot receivetheinformationaboutthefreepostͲinterventionexerciseopportunitybefore randomisation), the variables ‘received PT’ resp. ‘received info’ and their interaction withtimewereaddedtothemodelaswell.Foreachoutcome,themixedmodelwas simplifiedstepwiseand,ifthegroup*timeinteractionsremainedstablefrompostͲtest to followͲup, the final model was further reduced by including a single term “group*postfollow”togetamorepreciseestimateoftheeffectofaquaticcyclingfor bothtimepoints.28Furthertechnicaldetailsonthemixedregressionanalysis,including thestepstakeninmodelreduction,aregiveninAppendix5.1. Alloutcomeanalyseswereadjustedforallbaselinevariablesbyincludingthelatteras covariates.TheprimaryoutcomesKOOSkneepainandKOOSphysicalfunctioningwere analysed separately and the treatment effect (i.e. group by time interaction) was testedusingɲ=0.05twoͲtailed.Inviewofmultipletesting,theɲforall21secondary 5 outcomes was 0.05/23=0.0022 (i.e. Bonferroni correction based on a total of 23outcomes).The effectsize (ES) of aquatic cyclingwascalculated forall outcomes (seeAppendix5.1).AllstatisticalanalyseswereconductedwiththeStatisticalPackage fortheSocialSciences(SPSS,version22.0).

RESULTS

OnehundredelevenparticipantswererandomisedtotheAC(n=55)andtheUC(n=56) arm. Nine participants (all UC) withdrew from the study before baseline assessment and were excluded from further analysis. To check for possible bias in the effect analysesarisingfromthisdropout,weconductedalogisticregressionanalysiswithin the UC group, with dropout before baseline (1=yes, 0=no) as outcome, and as predictors all measured baseline variables. The potential impact appeared to be ignorable(seeAppendix5.2). One hundred two participants completed the baseline assessment. Information on baselinecharacteristicscanbefoundinTable5.1. Table5.1 BaselinecharacteristicsofparticipantsinACandUC Characteristics UC AC (n=47) (n=55) Age,years,mean(SD) 61(7.4) 59(9.5) Females/males(n) 24/23 39/16 BMI,mean(SD) 29(5.4) 29(5.6) K/LTF,mean(SD) 2(0.5) 2(0.6) K/LPF,mean(SD) 0.7(1.0) 0.7(1.1) Countcomorbidity,mean(SD) 1(1.3) 2(1.7) AC, aquatic cycling group; BMI, body mass index; BW, body weight; K/L, KellgrenͲLawrence; KOOS, Knee injuryandOsteoarthritisOutcomeScore;NM,Newtonmeter;PF,patellofemoral;SD,standarddeviation;TF, tibiofemoral;UC,usualcaregroup

113 Chapter5

Ninety participants completed the postͲintervention assessment and 83 patients participated in the followͲup assessment. Figure 5.1 shows the participants’ flow, includingreasonsforwithdrawal. Figure5.1 Flowchart

114 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis

Primaryoutcome:KOOSkneepain The analysis for knee pain included 101 participants (one missing due to a missing baselineKOOSquestionnaireanddropoutafterbaseline).Sincethegroupdifference appeared to remain stable from postͲtest to followͲup, the final model was further reducedbyreplacingthegroup*posttestandgroup*followͲuptermswithasingleterm group*postfollow (Appendix 5.2), which comes down to constraining the group differencetobethesameatfollowͲupasatpostͲtest.Thisconstraintwasconfirmedby likelihoodratiotestingandgivesasingleestimateoftheoutcomedifferencebetween AC and UC, with more precision than estimates per time point (Appendix 5.3). Figure5.2Ashowstheobservedmeanscoreforkneepaininbothgroupsovertimeand Figure 5.2B shows the predicted mean scores based on the final mixed models presented in Table 5.2. In general, dropout can bias observed means and predicted 5 meanscorrectforthatasmuchaspossible.Here,bothplotsagreequitewellwitheach other.Appendix5.4entailsanoverviewoftheobservedmeans. Thegroupbytimeeffect(Table5.2,ModelA)showsthatthechangeinpainscorefrom baselineto12and24ͲweeksfollowͲupwassignificantlyhigherintheACarmthanin theUCarm(p=0.014,95%CI1.67to14.64;ES=0.50).Furthermore,ahigherBMIwas relatedtomorekneepain(p=0.046,95%CIͲ0.93toͲ0.01).Therewasnosignificant effectofage(p=0.088),butassaidinAppendix5.1,allcovariateswithp<0.10were keptinthemodel.Further,thetimeeffectspresentedinTable5.2arethosefortheUC control arm (group=0), showing absence of significant change in that arm. Last, the groupeffectshowninTable5.2(ModelA)isthegroupdifferenceatbaseline(time=0), showingabsenceofabaselinegroupdifferenceinKOOSkneepain.

Primaryoutcome:KOOSphysicalfunctioning TheanalysisofselfͲreportedphysicalfunctioningincludedfewerpatients(n=98)than theanalysisforkneepainduetomissingvaluesonthecovariate“strength”(meterout oforder,patientrefusal),whichwasinthefinalmodelforphysicalfunctioning(Table 5.2, Model B) but not in that for knee pain. As for the analysis of knee pain, the group*timeinteractiontermscouldbereducedtoonesingleestimate(SeeAppendix 5.2+5.3).Thepredictedmeans(Figure5.2D)ofthefinalmodel(seeTable5.2,Model B)agreequitewellwiththeobservedmeansofphysicalfunctioningpergroupandper time point (Figure 5.2C and Appendix 5.4). At baseline the AC and UC arm were comparableandtimehadnosignificantinfluenceontheUCarm.Thegroupbytime effectinTable5.2(ModelB)showsastatisticallysignificantgroupdifference(p=0.027, ES=0.43)infavouroftheAC.Justlikeforkneepain,ahigherBMIwasrelatedtomore limitations in physical functioning (p=0.006, 95%CIͲ1.26 toͲ0.21). Furthermore, quadriceps strength had a significant effect on physical functioning (p=0.001, 95%CI 0.01 to 0.20); participants with more quadriceps strength reported fewer problems with physical functioning. An intentionͲtoͲtreat analysis without covariates (so that

115 Chapter5 n=101insteadof98)resultedinagroup*timeeffectsimilartothemodelinTable5.2 (regressionweight=7.855,SD=3.18,p=0.015). Figure5.2 KOOSscoresonpain(upperhalf)andphysicalfunctioning(lowerhalf)overtime(atbaseline, post=12Ͳweek postͲintervention, followͲup: 24 weeks after baseline). The left half shows observedoutcomemeans,righthalfshowspredictedoutcomemeansbasedonTable5.2. Table5.2 FinalmixedmodelsforKOOSpainandphysicalfunctioning KOOSPAIN Fixedeffects(n=101) Estimate SE pͲvalue 95%CI Intercept 56.02 12.91 0.000 30.39,81.64 GroupͲ0.46 2.89 0.874Ͳ6.19,5.27 Age 0.25 0.15 0.088 Ͳ0.04,0.55 BMIͲ0.47 0.23 0.046Ͳ0.93,Ͳ0.01 PostͲtest Ͳ1.69 2.51 0.504 Ͳ6.67,3.30 FollowͲupͲ0.94 2.69 0.727Ͳ6.28,4.39 Group*Postfollow 8.16 3.27 0.014 1.67,14.64 KOOSPHYSICALFUNCTIONING Fixedeffects(n=98) Estimate SE pͲvalue 95%CI Intercept 80.04 9.57 0.000 61.05,99.02 GroupͲ3.43 3.38 0.313Ͳ10.13,3.28 BMI Ͳ0.74 0.26 0.006 Ͳ1.26,Ͳ0.21 Strength 0.11 0.05 0.029 0.01,0.20 PostͲtest Ͳ0.05 2.45 0.984 Ͳ4.81,4.91 FollowͲupͲ1.37 2.52 0.588Ͳ6.38,3.64 Group*Postfollow 7.16 3.19 0.027 0.83,13.49 BMI, body mass index; CI, Confidence Interval; KOOS, Knee injury and Osteoarthritis Outcome Score; SE, standarderror

116 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis

Secondaryoutcomes Table5.3presentskneespecificOAoutcomes.12,29Otherhealthoutcomes,likelevelof physical activity or general quality of life, are presented in Appendix 5.5. The group differencesformostsecondaryoutcomesremainedstablefrompostͲtesttofollowͲup assessmentandtherefore,thefinalmodelwasreducedtoonegroup*timeinteraction (seeAppendix5.2).TheobservedmeansforalltimeͲpointsofallsecondaryoutcomes arepresentedinAppendix5.4. At12ͲweekstheACgroupachievedsignificantlyhigherscoresattheKOOSqualityof lifesubscaleincomparisontobaselineandcomparedtotheUCgroup(13points,CI: 5.852, 20.215, p=0.001, ES=0.71). Further, the AC group performed the TUG test significantlyfasterthantheUCgroupatthepostͲandfollowͲupassessment(Ͳ0.9s,95% CI:Ͳ1.453,Ͳ0.369,p=0.001,ES=Ͳ0.62).And,participantsfromtheACinterventiongroup 5 achievedsignificantlybetterresultsatthe6MWTona44mͲsquaredtrackthantheUC groupatbothpostͲinterventionandfollowͲupassessment(46m:95%CI:17.60,75.90, p=0.002, ES=0.49). The analysis of the other selfͲreported outcomes, presented in Appendix5.5,showedasignificantimprovementinfearofmovementfortheACgroup compared to the UC group (Ͳ3.8 points: 95% CI:Ͳ6.23,Ͳ1.45, p=0.002, ES=0.14). All otheroutcomesshowednosignificantgroupdifference. Table5.3 GroupdifferenceforthesecondarykneeͲspecificoutcomes Outcome N Estimate SE pͲvalue 95%CI KOOSSymptoms 101 5.520 2.66 0.052 Ͳ0.039,10.540 KOOSSport97 3.884 4.24 0.362Ͳ4.551,12.318 KOOSQoLpostR 100 13.034 3.61 0.001* 5.852,20.215 KOOSQoLfollowͲupR 100 6.738 3.61 0.070Ͳ0.571,14.046 LEFS 99 5.960 2.05 0.005 1.890,10.030 NPRS(post6MWT)99 Ͳ1.169 0.57 0.042Ͳ2.295,1.514 PGA 99 Ͳ0.618 0.535 0.252 Ͳ1.678,0.446 6MWT9846.750 14.6400.002* 17.600,75.901 TUG 98 Ͳ0.911 0.272 0.001* Ͳ1.453,Ͳ0.369 4Ͳcepsstrength:isokinetic180°9912.4564.047 0.003 4.394,20.517 CI,ConfidenceInterval;KOOS,KneeinjuryandOsteoarthritisOutcomeScore;LEFS,LowerExtremityFunction Scale; NRS, Numeric Pain Rating Scale; PGA, Patient Global Assessment; QoL, Quality of life; SE, standard error; TSK, Tampa Scale for Kinesiophobia; TUG, TimedͲupͲandͲgo test; 4Ͳceps, quadriceps; 6MWT, SixͲ MinuteͲWalkingͲTest. *sig. pч0.0022 (the alpha used for all secondary outcomes to adjust for multiple testing), R=model reduction to one group*time interaction was not valid and therefore, both group*time interactionsarepresented

UtilisationofusualcareintheUCgroup FifteenparticipantsoftheUCgroup(32%)consultedaphysicaltherapistabouttheir kneeproblems.Mostpatients(n=9)visitedthephysicaltherapistbetweenoneandsix times during the active intervention period; three patients had three to

117 Chapter5

12appointmentsandonepersonhadmorethan12appointments.Oftwopatientsthe followͲup after the first consult is unknown due to dropout of these participants. Physicaltherapyconsistedofeducationandhomeexercise(n=9),medicalfitness(n=3) andmanualtherapy(n=2).Threeparticipantsreceivedacorticosteroidinjectionbefore thepostͲinterventionassessment(n=2)resp.thefollowͲupassessment(n=1).

ComplianceandcoͲinterventionsintheACgroup 62% of the AC participants completed all 24 sessions and 76% participated in twoͲ thirds(>16sessions)ofthesessions.Theaverageattendancerateforall24sessions was 80% (SD 31%). Common reasons for nonͲcompleting the AC intervention were vacations of three weeks or longer (n=5) and unplanned treatment of comorbidities (n=8). Other reasons were mensicusectonomy (n=2), problems with transportation (n=1),busyworkschedule(n=1),andoneparticipantdislikedexercisinginwarmwater inretrospect.ThefollowingcoͲinterventionswerecontinuedduringthetrial:wearinga brace(n=3)ororthopaedicshoes(n=2),receivingphysicaltherapyforothercomplaints (n=4)andconsultingadieticianforweightloss(n=1).Furthermore,threeparticipants received a corticosteroid injection before postͲintervention assessment and three participantsstartedphysicaltherapyfortheirkneeOAaftertheinterventionperiod.

Exerciseintensityoftheaquaticcyclingprogramme ThetrainingwasperceivedaslighttomoderatewithameanBorgscoreof12and50% ofHRmaxaccordingtothe220Ͳageformula.Participantsmanagedthetrainingwithout theneedofmajoradaptionsoftheexercises.Theycycledwithapproximately60rpm’s and with only water resistance or some extra pedalling resistance, which refers to resistancelevelonerespectivelytwointheaquaticbikeutilisedinthisstudy.

Adverseevents One serious adverse event occurred during the training in the AC group when a participantdidnotfeelwellattheendofthetrainingandhyperventilated.Duetothe cardiovascularhistoryofthepatientthepatientwashospitalizedforonenightandwas dischargedthefollowingdaywiththediagnosisofhyperventilationandcontinuedwith thetrainingaftertwoweeksofrest. Furthermore, four patients reported an exacerbation of symptoms after the Biodex strength assessment and refused to do this assessment in the followͲup measurements. Withregardtothetraining,10patientsreportedincreasedkneepainthedayafterthe trainingonce(n=9)ortwice(n=1).

118 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis

DISCUSSION

The results of this RCT indicate that a 12Ͳweek aquatic cycling training programme improves selfͲreported knee pain and physical functioning in patients with mild to moderatekneeOAcomparedtousualcare.Furthermore,ourstudyshowsamediumͲ term improvement in performanceͲbased physical functioning, a decrease in fear of movement as well as a shortͲterm increase in quality of life in the aquatic cycling group.Exceptforqualityoflife,effectsweremaintaineduntil12weeksaftertheendof the aquatic cycling training. To the authors’ best knowledge this is the first study evaluatingtheshortandmediumͲtermeffectsofagroupͲbasedaquaticcyclingtraining programmeforpatientswithkneemildtomoderatekneeOA.6 The magnitude of our treatment effects on selfͲreported knee pain (ES = 0.50) and performanceͲbased physical functioning (ES = 6MWT: 0.49, TUG:Ͳ0.62) were higher 5 than the reported standardised mean differences (pain: 0.26, 0.31; physical functioning: 0.22) of recent systematic reviews comparing aquatic therapy to nonͲ treatment control conditions.2,30 Possibly, the comparison of the intervention with a usualcarewaitingͲlistcontrolgrouphasinflatedoureffectestimates.However,several factorssuggestthataquaticcyclingiseffectiveforpatientswithmildtomoderateknee OA. Thedecreaseinperceivedkneepainintheaquaticcyclinggroupislikelytheresultof different aspects of the intervention. First, warm water temperature and buoyancy seem to contribute to pain relieve and feelings of wellͲbeing.31 Second, evidence suggests that aerobic exercise is effective in improving knee pain and physical functioninginpatientswithkneeOA.Incomparisonwithusualcare,theaquaticcycling groupsignificantlyimprovedtheirsixͲminutewalkingdistance;asurrogatemeasureof cardiovascularfunction.Thisfindingsuggeststhatthetrainingintensitywassufficient toachieveimprovementsinfunctionalaerobiccapacityandisinlinewithfindingsfrom aquaticcyclinginterventionsforobesepeople32andpatientswithmultiplesclerosis.33 TheimprovementofperformanceͲbasedphysicalfunctioningmightalsobeexplained byanimprovementinmuscularcoordinationofthelowerlimbs,whichforexampleis required for an efficient gait.34,35 Cycling requires coͲcoordination of the lower limbs36Ͳ38andpreviousstudiesonlandͲbasedcyclingprogrammesprovidedevidence that cycling is an effective treatment to improve gait pattern in patients with knee OA.3,4 The mediumͲterm followͲup assessment showed that improvements in the aquatic group could be maintained. Interestingly, the maintained treatment effect on the performanceͲbased assessments on physical functioning was small to moderate, whereas the lasting effect for patientͲreported physical functioning was small. This discrepancybetweenwhatpeopleƚŚŝŶŬtheycandoandwhatthey,infact,ĐĂŶĚŽis noticedinotherstudies.39Ͳ41Theexerciseexperienceduetoparticipationinouraquatic cycling programme might have reassured participants’ perception of their exercise

119 Chapter5 capabilities,whichissupportedbythesignificantreductioninfearofmovementinthe aquaticgroup.42Otherfeaturesthatmighthavecontributedtothemaintainedeffects werethesmallgroupsizeandthesetͲupoftheaquaticbikesclosetotheedgeofthe swimming pool. These aspects enabled the supervising physical therapist to provide personalised feedback and to educate participants on the benefits of exercise and physicalactivityduringthetraining.Studiesthatprovidedpersonalisededucationon OA and exerciseͲrelated topics43 or assured participants exercise abilities with an intensiveaquaticresistancetraining42alsofoundimprovementsevenaftercessationof theintervention. InpatientswithkneeOAqualityoflifeisassociatedwithkneepainanddisabilityand ourfindingsareinlinewithpreviousstudiessuggestingashortͲterm,smalleffectof aquatic exercise on quality of life.2,30 As stated earlier, the aquatic environment positivelyinfluencespainperceptionandcanincreasefeelingsofwellͲbeing.31Possibly, the cessation of the aquatic programme and the patients’ perception of small improvements in physical functioning explain the vanished effect at the followͲup assessment. Several aspects should be taken into account when interpreting the results of our study. First, the final number of included participants (n = 102) was lower than anticipated(n=126+25%dropout=168),andofthese102only90providedpostͲtest data.Withthissamplesize,wehad80%powertodetectaneffectsizeof0.60forthe primaryoutcomes(whereɲ=0.05).Second,thestudydesignmighthaveinfluenced thebehaviourintheUCarm.44Participantswereallocatedtothegroupsbeforethe baselinemeasurementandbydoingso;welostnineparticipantsbeforethebaseline assessment(likelyduetoanabsenceofalternativetreatmentatthetime).Asaresult, these participants could not be included in the effect analysis. Moreover, the final waitingͲlistcontroldesigncouldhavecreatedawaitͲandͲseeattitudeintheUCarm, whichinturnmighthaveinflatedoureffects.Inourtrial,68%ofthepatientsintheUC group decided not to visit a physical therapist, probably because most of these participants knew they could try out aquatic cycling after the end of their participation.44 However, the outcomes of our control group remained stable from baselineto24ͲweeksfollowͲup,whichisinlinewithacohortstudyshowingthatthe impairments of elderly people with hip or knee OA did not change over a fiveͲyear period.45 Another weakness of our study is that the possibility for the UC group to participateinafreeaquaticcyclingtraininginanewlybuiltpublicswimmingpoolafter studyparticipationbecameavailableafterthestartofthestudyandso14patientsdid nothavethisinformationduringtheassessments.Theoutcomeanalysesdidnotshow anysignificantoutcomedifferencebetweentheparticipantsintheUCarmwhoknew (n=33)andthe14participantswhodidnotknow,theycouldtryoutACafterwards.

120 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis

Since participants tolerated the training well, future studies should investigate the potentialofaquaticcyclingforpatientswithseverekneeOAandduringpostͲsurgical rehabilitation.Furthermore,anevaluationofthefeasibilityandcostͲeffectivenessofa communityͲbased aquatic cycling programme for individuals with knee OA seems indicated,becauseaquatictherapyisassociatedwithhighcostsandtheadherencetoa regularexerciseroutineiscrucialintheconservativemanagementofOA. Inconclusion,aquaticcyclingmightbeausefulextensionoftheexercisepossibilities for participants with mild to moderate knee OA. Compared to usual care, a light to moderate intense aquatic cycling training of 12Ͳweeks resulted in mediumͲterm improvementsofkneepain,selfͲreportedandperformanceͲassessedphysicalfunction, andfearofmovement.Furthermore,ashortͲtermeffectondiseaseͲspecificqualityof lifewasdetected. 5

Acknowledgement A special thanks goes out to the participants of this study and the therapists who performedtheassessments,thestudentassistantswhosupportedtheassessorsand trainerorwhopatientlyenteredthedata,ConnydeZwartwhowasresponsibleforthe randomisation,andWielWijnenforhisgreatsupportinpatientrecruitment.Wewould gratefullyacknowledgethedepartmentofphysicaltherapyoftheMUMC+,Maastricht SportandtheGeusseltbad for providing the training and testing facilities. The study was financially supported by the Netherlands Organisation for Scientific Research (NWO);grantnumber022.003.036.Noneoftheseorganisationsplayedaroleduring anystageofthetrial,i.e.execution,analysesandreporting.

121 Chapter5

REFERENCES

1. FransenM,McConnellS,HarmerAR,VanderEschM,SimicM,BennellKL.Exerciseforosteoarthritisof theknee.ŽĐŚƌĂŶĞĂƚĂďĂƐĞŽĨ^ǇƐƚĞŵĂƚŝĐZĞǀŝĞǁƐ.2015;10.1002/14651858.CD004376.pub3. 2. WallerB,OgonowskaͲSlodownikA,VitorM,LambeckJ,DalyD,KujalaUM,etal.Effectoftherapeutic aquaticexerciseonsymptomsandfunctionassociatedwithlowerlimbosteoarthritis:systematicreview withmetaͲanalysis.WŚǇƐŝĐĂůdŚĞƌĂƉǇ.2014;94:1383Ͳ95. 3. MangioneK,McCullyK,GloviakA,LefebvreI,HofmannM,CraikR.TheeffectsofhighͲintensityand lowͲintensitycycleergometryinolderadultswithkneeosteoarthritis.:ŽƵƌŶĂůƐŽĨ'ĞƌŽŶƚŽůŽŐǇ͕^ĞƌŝĞƐ͗ ŝŽůŽŐŝĐĂů^ĐŝĞŶĐĞƐĂŶĚDĞĚŝĐĂů^ĐŝĞŶĐĞƐ.1999;54:M:184Ͳ90. 4. SalacinskiA,KrohnK,LewisS,HollandM,IrelandK,MarchettiG.Theeffectsofgroupcyclingongait and painͲrelated disability in individuals with mildͲtoͲmoderate knee osteoarthritis: a randomized controlledtrial.:ŽƵƌŶĂůŽĨKƌƚŚŽƉĂĞĚŝĐĂŶĚ^ƉŽƌƚƐWŚǇƐŝĐĂůdŚĞƌĂƉǇ.2012;42:985Ͳ95. 5. RewaldS,MestersI,EmansPJ,ArtsJJ,LenssenAF,deBieRA.AquaticcircuittrainingincludingaquaͲ cycling in patients with knee osteoarthritis: A feasibility study. :ŽƵƌŶĂů ŽĨ ZĞŚĂďŝůŝƚĂƚŝŽŶ DĞĚŝĐŝŶĞ. 2015;10.2340/16501977Ͳ1937. 6. RewaldS,MestersI,LenssenAF,BansiJ,LambeckJ,deBieRA,etal.AquaticcyclingͲWhatdoweknow? AscopingreviewonheadͲoutaquaticcycling.WůŽ^KŶĞ.2017;12:e0177704. 7. RewaldS,MestersI,LenssenAF,EmansPJ,WijnenW,deBieRA.EffectofaquaͲcyclingonpainand physicalfunctioningcomparedwithusualcareinpatientswithkneeosteoarthritis:studyprotocolofa randomisedcontrolledtrial.DDƵƐĐƵůŽƐŬĞůĞƚĂůŝƐŽƌĚĞƌƐ.2016;17:88. 8. Perrot S, Rozenberg S, Moyse D, Legout V, Marty M. Comparison of daily, weekly or monthly pain assessmentsinhipandkneeosteoarthritis.A29Ͳdayprospectivestudy.:ŽŝŶƚ͕ŽŶĞ͕^ƉŝŶĞ͗ZĞǀƵĞĚƵ ZŚƵŵĂƚŝƐŵĞ.2011;78:510Ͳ5. 9. KellgrenJH,LawrenceJS.RadiologicalassessmentofosteoͲarthrosis.ŶŶĂůƐŽĨƚŚĞZŚĞƵŵĂƚŝĐŝƐĞĂƐĞƐ. 1957;16:494Ͳ502. 10. AxfordJ,ButtA,HeronC,HammondJ,MorganJ,AlaviA,etal.Prevalenceofanxietyanddepressionin osteoarthritis: use of the Hospital Anxiety and Depression Scale as a screening tool. ůŝŶŝĐĂů ZŚĞƵŵĂƚŽůŽŐǇ.2010;29:1277Ͳ83. 11. HoogeboomTJ,deBieRA,denBroederAA,vandenEndeCH.TheDutchLowerExtremityFunctional Scalewashighlyreliable,validandresponsiveinindividualswithhip/kneeosteoarthritis:avalidation study.DDƵƐĐƵůŽƐŬĞůĞƚĂůŝƐŽƌĚĞƌƐ.2012;13:117. 12. PhamT,VanDerHeijdeD,LassereM,AltmanRD,AndersonJJ,BellamyN,etal.Outcomevariablesfor osteoarthritis clinical trials: The OMERACTͲOARSI set of responder criteria. :ŽƵƌŶĂů ŽĨ ZŚĞƵŵĂƚŽůŽŐǇ. 2003;30:1648Ͳ54. 13. Wagenmakers R, van den AkkerͲScheek I, Groothoff JW, Zijlstra W, Bulstra SK, Kootstra JW, et al. ReliabilityandvalidityoftheshortquestionnairetoassesshealthͲenhancingphysicalactivity(SQUASH) inpatientsaftertotalhiparthroplasty.DDƵƐĐƵůŽƐŬĞůĞƚĂůŝƐŽƌĚĞƌƐ.2008;9:141. 14. deHollanderEL,ZwartL,deVriesSI,WendelͲVosW.TheSQUASHwasamorevalidtoolthantheOBiN forcategorizingadultsaccordingtotheDutchphysicalactivityandthecombinedguideline.:ŽƵƌŶĂůŽĨ ůŝŶŝĐĂůƉŝĚĞŵŝŽůŽŐǇ.2012;65:73Ͳ81. 15. WareJE,NewEnglandMedicalCenterH,HealthI.SFͲ36physicalandmentalhealthsummaryscales:a user'smanual.Boston:HealthInstitute,NewEnglandMedicalCenter;1994. 16. SwinkelsͲMeewisseEJ,SwinkelsRA,VerbeekAL,VlaeyenJW,OostendorpRA.Psychometricproperties oftheTampaScaleforkinesiophobiaandthefearͲavoidancebeliefsquestionnaireinacutelowback pain.DĂŶƵĂůdŚĞƌĂƉǇ.2003;8:29Ͳ36. 17. Taal E, Riemsma RP, Brus HL, Seydel ER, Rasker JJ, Wiegman O. Group education for patients with rheumatoidarthritis.WĂƚŝĞŶƚĚƵĐĂƚŝŽŶĂŶĚŽƵŶƐĞůŝŶŐ.1993;20:177Ͳ87. 18. Butland RJ, Pang J, Gross ER, Woodcock AA, Geddes DM. TwoͲ, sixͲ, and 12Ͳminute walking tests in respiratorydisease.ƌŝƚŝƐŚDĞĚŝĐĂů:ŽƵƌŶĂů;ůŝŶŝĐĂůZĞƐĞĂƌĐŚĚͿ.1982;284:1607Ͳ8. 19. Pankoff BA, Overend TJ, Lucy SD, White KP. Reliability of the sixͲminute walk test in people with fibromyalgia.ƌƚŚƌŝƚŝƐĂƌĞĂŶĚZĞƐĞĂƌĐŚ.2000;13:291Ͳ5.

122 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis

20. StratfordPW,KennedyDM,WoodhouseLJ.Performancemeasuresprovideassessmentsofpainand functioninpeoplewithadvancedosteoarthritisofthehiporknee.WŚǇƐŝĐĂůdŚĞƌĂƉǇ.2006;86:1489Ͳ96. 21. FreterSH,FruchterN.Relationshipbetweentimed'upandgo'andgaittimeinanelderlyorthopaedic rehabilitationpopulation.ůŝŶŝĐĂůZĞŚĂďŝůŝƚĂƚŝŽŶ.2000;14:96Ͳ101. 22. Carpenter MR, Carpenter RL, Peel J, Zukley LM, Angelopoulou KM, Fischer I, et al. The reliability of isokineticandisometriclegstrengthmeasuresamongindividualswithsymptomsofmildosteoarthritis. :ŽƵƌŶĂůŽĨ^ƉŽƌƚƐDĞĚŝĐŝŶĞĂŶĚWŚǇƐŝĐĂů&ŝƚŶĞƐƐ.2006;46:585Ͳ9. 23. vanDijkGM,VeenhofC,SchellevisF,HulsmansH,BakkerJP,ArwertH,etal.Comorbidity,limitationsin activitiesandpaininpatientswith osteoarthritisofthehiporknee.DDƵƐĐƵůŽƐŬĞůĞƚĂůŝƐŽƌĚĞƌƐ. 2008;9:95. 24. RoosEM,LohmanderLS.TheKneeinjuryandOsteoarthritisOutcomeScore(KOOS):fromjointinjuryto osteoarthritis.,ĞĂůƚŚĂŶĚYƵĂůŝƚǇŽĨ>ŝĨĞKƵƚĐŽŵĞƐ.2003;1:64. 25. AngstF,AeschlimannA,StuckiG.Smallestdetectableandminimalclinicallyimportantdifferencesof rehabilitationinterventionwiththeirimplicationsforrequiredsamplesizesusingWOMACandSFͲ36 quality of life measurement instruments in patients with osteoarthritis of the lower extremities. ƌƚŚƌŝƚŝƐĂŶĚZŚĞƵŵĂƚŝƐŵ.2001;45:384Ͳ91. 26. HinmanRS,HeywoodSE,DayAR.Aquaticphysicaltherapyforhipandkneeosteoarthritis:resultsofa 5 singleͲblindrandomizedcontrolledtrial.WŚǇƐŝĐĂůdŚĞƌĂƉǇ.2007;87:32Ͳ43. 27. Bastick AN, Runhaar J, Belo JN, BiermaͲZeinstra SM. Prognostic factors for progression of clinical osteoarthritisoftheknee:asystematicreviewofobservationalstudies.ƌƚŚƌŝƚŝƐZĞƐĞĂƌĐŚΘdŚĞƌĂƉǇ. 2015;doi:10.1186/s13075Ͳ015Ͳ0670Ͳx. 28. Verbeke G, Molenberghs G. Linear Mixed Models for Longitudinal Data: SpringerͲVerlag New York; 2000. 29. PeterWF,JansenMJ,HurkmansEJ,BlooH,DekkerJ,DillingRG,etal.Physiotherapyinhipandknee osteoarthritis: development of a practice guideline concerning initial assessment, treatment and evaluation.ĐƚĂƌĞƵŵĂƚŽůſŐŝĐĂƉŽƌƚƵŐƵĞƐĂ^ŽĐŝĞĚĂĚĞWŽƌƚƵŐƵĞƐĂĚĞZĞƵŵĂƚŽůŽŐŝĂ.2011;36:268Ͳ81. 30. Bartels EM, Juhl CB, Christensen R, Hagen KB, DanneskioldͲSamsoe B, Dagfinrud H, et al. Aquatic exercise for the treatment of knee and hip osteoarthritis. ŽĐŚƌĂŶĞ ĂƚĂďĂƐĞ ^ǇƐƚ ZĞǀ. 2016;3:CD005523. 31. BeckerBE.Aquatictherapy:scientificfoundationsandclinicalrehabilitationapplications.WDΘZ͗ƚŚĞ ũŽƵƌŶĂůŽĨŝŶũƵƌǇ͕ĨƵŶĐƚŝŽŶ͕ĂŶĚƌĞŚĂďŝůŝƚĂƚŝŽŶ.2009;1:859Ͳ72. 32. BoidinM,LapierreG,PaquetteTanirL,NigamA,JuneauM,GuilbeaultV,etal.Effectofaquaticinterval trainingwithMediterraneandietcounselinginobesepatients:resultsofapreliminarystudy.ŶŶĂůƐŽĨ WŚǇƐŝĐĂůĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶDĞĚŝĐŝŶĞ.2015;58:269Ͳ75. 33. Bansi J, Bloch W, Gamper U, Riedel S, Kesselring J. Endurance training in MS: shortͲterm immune responses and their relation to cardiorespiratory fitness, healthͲrelated quality of life, and fatigue. :ŽƵƌŶĂůŽĨEĞƵƌŽůŽŐǇ.2013;260:2993Ͳ3001. 34. BartholdyC,JuhlC,ChristensenR,LundH,ZhangW,HenriksenM.Theroleofmusclestrengtheningin exercise therapy for knee osteoarthritis: A systematic review and metaͲregression analysis of randomizedtrials.^ĞŵŝŶĂƌƐŝŶƌƚŚƌŝƚŝƐĂŶĚZŚĞƵŵĂƚŝƐŵ.2017;10.1016/j.semarthrit.2017.03.007. 35. Mills K, Hunt MA, Leigh R, Ferber R. A systematic review and metaͲanalysis of lower limb neuromuscular alterations associated with knee osteoarthritis during level walking. ůŝŶŝĐĂů ŝŽŵĞĐŚĂŶŝĐƐ;ƌŝƐƚŽů͕ǀŽŶͿ.2013;28:713Ͳ24. 36. BreugemSJ,HaverkampD,SiereveltIN,StibbeAB,BlankevoortL,vanDijkCN.Theimportantpredictors ofcyclinguseinthreegroupsofkneepatients.ŝƐĂďŝůŝƚǇĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.2011;33:1925Ͳ9. 37. JohnstonTE.Biomechanicalconsiderationsforcyclinginterventionsinrehabilitation.WŚǇƐŝĐĂůdŚĞƌĂƉǇ. 2007;87:1243Ͳ52. 38. LiebsTR,HerzbergW,RutherW,HaastersJ,RussliesM,HassenpflugJ.Ergometercyclingafterhipor knee replacement surgery: a randomized controlled trial. dŚĞ :ŽƵƌŶĂů ŽĨ ŽŶĞ ĂŶĚ :ŽŝŶƚ ^ƵƌŐĞƌǇ ŵĞƌŝĐĂŶǀŽůƵŵĞ.2010;92:814Ͳ22. 39. Zambon S, Siviero P, Denkinger M, Limongi F, Victoria Castell M, van der Pas S, et al. Role of Osteoarthritis, Comorbidity, and Pain in Determining Functional Limitations in Older Populations: EuropeanProjectonOsteoarthritis.ƌƚŚƌŝƚŝƐĂƌĞΘZĞƐĞĂƌĐŚ.2016;68:801Ͳ10.

123 Chapter5

40. Terwee CB, van der Slikke RM, van Lummel RC, Benink RJ, Meijers WG, de Vet HC. SelfͲreported physicalfunctioningwasmoreinfluencedbypainthanperformanceͲbasedphysicalfunctioninginkneeͲ osteoarthritispatients.:ŽƵƌŶĂůŽĨůŝŶŝĐĂůƉŝĚĞŵŝŽůŽŐǇ.2006;59:724Ͳ31. 41. Stratford PW, Kennedy DM. Performance measures were necessary to obtain a complete picture of osteoarthriticpatients.:ŽƵƌŶĂůŽĨůŝŶŝĐĂůƉŝĚĞŵŝŽůŽŐǇ.2006;59:160Ͳ7. 42. WallerB,MunukkaM,RantalainenT,LammentaustaE,NieminenMT,KivirantaI,etal.Effectsofhigh intensityresistanceaquatictrainingonbodycompositionandwalkingspeedinwomenwithmildknee osteoarthritis:a4ͲmonthRCTwith12ͲmonthfollowͲup.KƐƚĞŽĂƌƚŚƌŝƚŝƐĂŶĚĂƌƚŝůĂŐĞ.2017;25:1238Ͳ46. 43. FransenM,McConnellS,HarmerAR,VanderEschM,SimicM,BennellKL.Exerciseforosteoarthritisof theknee:aCochranesystematicreview.ƌŝƚŝƐŚ:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐDĞĚŝĐŝŶĞ.2015;49:1554Ͳ7. 44. CunninghamJA,KypriK,McCambridgeJ.Exploratoryrandomizedcontrolledtrialevaluatingtheimpact ofawaitinglistcontroldesign.DDĞĚŝĐĂůZĞƐĞĂƌĐŚDĞƚŚŽĚŽůŽŐǇ.2013;13:150. 45. PistersMFVC,vanDijkGM,HeymansMW,TwiskJW,DekkerJ.Thecourseoflimitationsinactivities over 5 years in patients with knee and hip osteoarthritis with moderate functional limitations: risk factorsforfuturefunctionaldecline.KƐƚĞŽĂƌƚŚƌŝƚŝƐĂŶĚĂƌƚŝůĂŐĞ.2012;20:503Ͳ10.

124 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis

APPENDIX5.1

Statisticalanalysis OutcomeswereanalysedwithatwoͲlevelmixedlinearregressionmodeltocapturethe nestingofrepeatedmeasureswithinpatients.Nestingofpatientswithinaquaticcycling groupswasnotfeasible,becauseweusedadynamicgroupformationapproachforthe aquatic cycling training that changed the composition of these groups frequently. Mixedlinearregressionmodelsconsistofarandompart(variancesof,andcorrelations between, the repeated outcome measures) and a fixed part (regression weights, effectsofpredictorsonoutcomemeanpertimepoint).

^ƉĞĐŝĨŝĐĂƚŝŽŶŽĨƚŚĞƌĂŶĚŽŵƉĂƌƚ 5 Fortherandompart,weassumedanunstructuredcovariancematrix,whichisquite flexible and still parsimonious for the present small number of repeated measures (baseline,postͲtest,followͲup).

^ƉĞĐŝĨŝĐĂƚŝŽŶŽĨƚŚĞĨŝǆĞĚƉĂƌƚ Thefixed(predictor)modelpartconsistedoftime(usingdummycodingforpostͲtest andfollowͲup,withbaselineasreferencetimepoint),treatmentgroup(AC=1,UC=0), andpatientcharacteristicsaspredictors.Thelatterwereage(inyears),sex(female=1, male=0),BMI,quadricepsstrength(inNewtonmeter(NM),correctedforbodyweight), tibiofemoral and patellofemoral KellgrenͲLawrence score and comorbidity count.27 Furthermore, to explore the effects of physical therapy, resp. the effect of being informedaboutthefreeaquaticcyclingprogrammeaftertrialparticipationintheUC arm(14UCpatientsdidnotreceivetheinformationaboutthefreepostͲintervention exerciseopportunitybeforerandomisation),thevariables‘receivedPT’resp.‘received info’wereincludedascovariatesaswell.Bothcovariateswerecodedsuchthatthey wereuncorrelatedwithtreatmentgroup.Thus,thegroupeffectinthemixedmodel stillrepresentedthecontrastbetweenthetotalACarmandthetotalUCarm,andnot between AC and some subgroup of UC.a Finally, the interactions of time (dummy indicators, with baseline as reference category) with treatment group and with the covariates“receivedPT”and“receivedinfo”completedthefixedpartofthemodel. The interaction of group with time represents the group difference in change from baseline to postͲintervention and followͲup, and is thus theeffect of interest in this RCT. The aboveͲmentioned covariates and interactions were used in all mixed

aMorespecifically,bothcovariateswerecodedaszeroforallpatientsintheACarm,andcodedas–pfor patientsintheUCarmnotgettingPTresp.info,andcodedas(1Ͳp)forpatientsintheUCarmgettingPT resp.info.Here,pwastheproportionpatientsintheUCarmgettingPTresp.info.Inthisway,theaverage covariatevaluewaszerointheUCarm,justlikeintheACarm.

125 Chapter5 regression analyses, except that in the analysis for strength, BMI was not included, becausetheoutcomestrengthwasalreadycorrectedforbodyweight.

DŽĚĞůƌĞĚƵĐƚŝŽŶĂŶĚůŝŬĞůŝŚŽŽĚƌĂƚŝŽŵŽĚĞůƚĞƐƚŝŶŐ Foreachoutcome,themixedmodelwassimplifiedstepwisebyfirstdroppingcovariate (“receivedPT”resp.“receivedinfo”)bytimeinteractionsifnotsignificant(usingɲ=0.10 twoͲtailedfordeletingterms,butnotfordrawingfirmconclusions)andthendropping nonͲsignificantcovariateslikewise(iftheywerenotinvolvedinanyinteraction).Ifthe treatmentgroupdifference(group*timeinteractions)remainedstablefrompostͲtest tofollowͲup,thefinalmodelwasfurtherreducedbyreplacingthegroup*posttestand group*followuptermswithasingletermgroup*postfollow(coded0forbaselineand1 foreachothertimepoint)togetasingleandmorepreciseestimateoftheeffectof aquatic cycling for both time points. Successive models were estimated with the maximumlikelihood(ML)methodtoallowlikelihoodratiotestingofmodelreductions, but the final model for each outcome was reͲrun with the restricted maximum likelihood (REML) method to obtain unbiased standard errors for the regression weightsandinparticularforthegroupbytimeeffectofinterest.28Eachoutcomewas checkedfornormalityandoutliersbyresidualplotspertimepoint.

ĂůĐƵůĂƚŝŽŶŽĨƚŚĞĞĨĨĞĐƚƐŝǌĞ Theeffectsize(ES)ofaquaticcyclingwascomputedfromthefinalmodelbydividing the regression weight of the group*posttest (resp. group*followup, resp. group*postfollow)interactionbythesquarerootoftheestimatedoutcomevarianceat postͲtest(resp.atfollowͲup,resp.averagedacrossbothtimepoints).

126 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis

APPENDIX5.2

Results

ƌŽƉŽƵƚďĞĨŽƌĞďĂƐĞůŝŶĞĂƐƐĞƐƐŵĞŶƚ Asstatedearlier,ninedropoutsintheUCarmoccurredbeforebaselineassessment.To check possible bias in theeffectanalysesarising from this dropout,we conducteda logisticregressionanalysiswithintheUCgroup,withdropoutbeforebaseline(1=yes, 0=no) as outcome, and as predictors all measured baseline variables (sex, age, BMI, Kellgren/Lawrence score the tibiofemoral and patellofemoral joint, count of comorbidities, HADS anxiety and depression score, and having received the informationaboutthefreetwelveͲweekaquacyclingprogrammeinalocalswimming 5 pool after participation in the study). Stepwise model reduction left only this last variableasapredictorofdropout(p=0.003).Morespecifically,theopportunityforthe controlgrouptoparticipateinafreetwelveͲweekaquacyclingprogrammearoseafter the project had started and so the first 14 participants in the control group did not receivethisinformation.Ofthese14,sevenparticipantsdeclinedparticipationbefore thebaselineassessment,againstonlytwodropoutsamongthe42participantswhodid receivetheinformation.Intheabsenceofarelationofdropoutbeforebaselinewith anyotherparticipantcharacteristics,andgiventhatoureffectanalysesadjustedforall measuredparticipantcharacteristicsanyway,thepotentialimpactofthisdropouton theeffectanalysisappearsignorable(notethatparticipantsdroppingoutĂĨƚĞƌbaseline werealwaysincludedintotheeffectanalyses).

DŽĚĞůƌĞĚƵĐƚŝŽŶŽĨƚŚĞƉƌŝŵĂƌǇŽƵƚĐŽŵĞƐ AlikelihoodratiotestconfirmedthestabilityofthegroupdifferencefrompostͲtestto followͲup (Knee pain: chiͲsquare=0.031, df=1, p=0.86, Physical Functioning: chiͲ square=0.026,df=1,p=0.87).Appendix5.3showsthefinalmodels(Model1A+1B)and itsreduction(Model2A+2B)assumingstabilityofthegroupdifference.

DŽĚĞůƌĞĚƵĐƚŝŽŶŽĨƚŚĞƐĞĐŽŶĚĂƌǇŽƵƚĐŽŵĞƐ The reduction of the group by time interaction was not supported for the KOOS subscalequalityoflife(chiͲsquare=3.99,df=1,p=0.0455).ThatmightbeatypeIerror duetomultipletesting,buttheeffectatpostͲtestappearedtobetwiceaslargeasat followͲup.Therefore,theeffectofaquaticcyclingonthisoutcomeisreportedpertime point(Table5.3).

127 Chapter5

KƵƚůŝĞƌƐ Alloutcomesweretestedforoutliers.Althoughsomeoutcomesshowedoneorafew outliers, repeating the analysis without the outlier(s) hardly affect the results (significant effects remained significant and nonͲsignificant results remained nonͲ significant,effectestimateschangedverylittle).

128 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis

APPENDIX5.3

Modelreductionoftheprimaryoutcomes

KOOSPAIN Fixedeffects(n=01) Estimate SE pͲvalue 95%CI MODEL1A Intercept 56.01 12.91 0.000 30.39,81.64 GroupͲ0.43 2.89 0.882Ͳ6.17,5.31 Age 0.25 0.15 0.089Ͳ0.04,0.55 BMIͲ0.47 0.23 0.046Ͳ0.93,Ͳ0.01 PostͲtestͲ1.82 2.62 0.489Ͳ7.02,3.38 FollowͲupͲ0.71 2.99 0.812Ͳ6.65,5.22 Group*PostͲtest 8.40 3.54 0.020 1.37,15.43 Group*FollowͲup 7.74 4.05 0.059Ͳ0.30,15.77 MODEL2A 5 Intercept 56.02 12.91 0.000 30.39,81.64 GroupͲ0.46 2.89 0.874Ͳ6.19,5.27 Age 0.25 0.15 0.088Ͳ0.04,0.55 BMIͲ0.47 0.23 0.046Ͳ0.93,Ͳ0.01 PostͲtestͲ1.69 2.51 0.504Ͳ6.67,3.30 FollowͲupͲ0.94 2.69 0.727Ͳ6.28,4.39 Group*Postfollow 8.16 3.27 0.014 1.67,14.64 KOOSPHYSICALFUNCTIONING Fixedeffects(n=98) Estimate SE pͲvalue 95%CI MODEL1B Intercept 80.08 9.57 0.000 61.09,99.08 GroupͲ3.51 3.41 0.307Ͳ10.28,3.27 BMIͲ0.74 0.26 0.006Ͳ1.26,Ͳ0.21 Strength 0.11 0.05 0.029 0.01,0.20 PostͲtest 0.15 2.53 0.953Ͳ4.88,5.18 FollowͲupͲ1.52 2.69 0.575Ͳ6.87,3.84 Group*PostͲtest 6.80 3.40 0.043 0.24,13.71 Group*FollowͲup 7.44 3.63 0.043 0.23,14.64 MODEL2B Intercept 80.04 9.57 0.000 61.05,99.02 GroupͲ3.43 3.38 0.313Ͳ10.13,3.28 BMIͲ0.74 0.26 0.006Ͳ1.26,Ͳ0.21 Strength 0.11 0.05 0.029 0.01,0.20 PostͲtestͲ0.05 2.45 0.984Ͳ4.81,4.91 FollowͲupͲ1.37 2.52 0.588Ͳ6.38,3.64 Group*Postfollow 7.16 3.19 0.027 0.83,13.49 BMI, body mass index; CI, Confidence Interval; KOOS, Knee injury and Osteoarthritis Outcome Score; SE, standarderror

129 Chapter5

46 46 46 46 35 35 33 44 46 33 49 46 46 ======AC n n n n n n n n n n n n n up 3.37±2.34 4.04±2.49 7.10±1.62 Ͳ 41.90±7.61 64.35±17.26 64.35±15.83 45.09±19.73 57.98±27.17 51.35±10.88 25.22±25.09 69.00±16.84 519.46±94.34 3350.43±2169.66 follow week Ͳ 24 39 38 38 38 31 32 31 38 39 33 39 38 38 ======UC n n n n n n n n n n n n n 4.48±2.25 4.15±2.58 6.88±1.28 57.24±19.16 62.50±16.04 40.74±19.62 65.55±30.05 42.05±11.69 49.41±10.90 21.45±16.96 65.42±17.98 515.91±111.39 3081.05±2049.52 49 49 49 46 49 49 49 43 44 44 46 49 49 ======AC n n n n n n n n n n n n n 2.94±2.00 7.19±1.65 3.84±2.31 42.43±8.11 51.35±10.88 63.55±15.33 66.61±11.23 47.47±18.26 67.64±29.04 24.80±23.05 70.14±17.52 525.25±86.09 3312.73±2602.69 intervention Ͳ post week 40 40 40 37 41 41 41 37 37 34 35 41 41 Ͳ ±19.15 ======UC 12 n n n n n n n n n n n n n 3.60±2.54 7.11±1.81 4.43±2.46 40.90±9.49 48.30±11.81 55.90±18.04 37.61±20.71 56.98±28.95 22.44±20.38 66.80±19.04 64.90 ) 505.78±120.09 3275.90±2670.58 D S ± n a e m ( 55 55 55 55 52 54 54 52 55 55 54 52 54 ======AC s t n n n n n n n n n n n n n n 4.07±2.20 8.08±2.28 5.27±2.02 i 40.33±6.49 45.15±11.69 56.96±12.96 58.87±11.97 38.67±15.72 50.29±29.77 20.19±17.96 61.89±17.15 492.91±91.68 o 2801.93±1687.15 p Ͳ e m Baseline i t 98 t n 47 47 45 46 43 44 44 44 43 43 44 43 44 e ======UC r n n n n n n n n n n n n n e f 3.74±2.33 6.98±1.73 4.81±2.05 f 39.38±7.75 49.16±11.76 57.89±15.26 62.20±16.84 39.77±19.15 57.98±27.17 66.32±16.28 i 21.51±17. 521.28±64.32 d 3041.38±2259.20 e h t t a s PA n 4 a . e min. 5 Functioning m x i 180° d d e weekly n v isok. r e Sport Pain Physical Symptoms QoL (post6MWT) e PCS p s 36 ceps: b p Ͳ Ͳ KOOS LEFS PGA TUG SQUASH: Outcomes KOOS KOOS KOOS KOOS NPRS 6MWT 4 R A O

130 Efficacyofaquaticcyclingonkneepainandphysicalfunctioninginpatientswithkneeosteoarthritis go Ͳ QoL, MCS, and Ͳ up 44 44 34 33 34 34 33 34 33 44 Scale; ======AC n n n n n n n n n n up 4.07±1.02 Ͳ timed 23.02±6.26 51.59±11.42 56.11±24.58 86.00±36.41 94.11±40.16 60.08±32.45 79.62±31.43 Assessment; 117.48±58.74 107.86±41.91 Function TUG, follow Global week Ͳ Extremity 24 Patient 29 38 37 31 31 31 29 31 31 38 ======UC n n n n n n n n n n Kinesiophobia; PGA, Lower 4.23±0.73 55.43±8.61 25.13±6.85 49.95±20.92 79.17±27.35 63.28±26.81 90.57±38.50 71.59±27.30 117.96±44.95 101.96±34.84 for LEFS, Score; Scale Score; 5 Tampa 42 43 43 46 46 44 44 44 43 46 Component ======AC n n n n n n n n n n TSK, Outcome 4.22±0.96 53.27±8.80 21.50±5.80 96.64±39.05 54.87±21.10 89.31±30.66 52.01±18.06 75.22±25.52 65.14±20.95 110.37±39.96 Physical intervention Ͳ activity; PCS, post Osteoarthritis physical week 32 32 34 35 35 36 34 36 34 35 Ͳ Activity; and ======UC 12 n n n n n n n n n n 4.17±0.82 25.86±6.71 55.60±27.10 97.04±44.56 97.44±40.01 53.49±11.47 50.46±21.30 77.70±31.98 66.76±24.76 110.90±44.79 injury enhancing Physical Ͳ Knee PA, Health Scale; KOOS, 51 51 53 53 52 52 54 54 54 51 ======AC ASsess n n n n n n n n n n to Rating 3.83±1.09 24.04±5.75 44.20±20.89 80.94±30.66 78.92±43.67 54.76±27.90 50.54±10.60 93.58±37.96 39.90±21.44 67.57±29.86 isometric; Pain Test. Ͳ isom., Baseline Numeric Questionnaire Walking Ͳ 45 45 44 44 43 42 45 44 45 43 ======NRS, UC Short n n n n n n n n n n isokinetic; 4.12±0.92 53.96±9.35 24.28±6.17 88.50±46.45 54.36±27.40 92.40±36.31 65.03±29.26 48.78±19.42 77.86±28.30 111.60±48.41 –Minute isok., Six minutes; SQUASH, Scale; min, 36; 6MWT, Rand Score; Efficacy 30° 60° Ͳ 60° 180° 36, Ͳ R Self 30° 60° 60° isok. isom. isok. isom. quadriceps; life; isok. isom. isom. of Component ceps, Ͳ Arthritis MCS 4 36 ceps: ceps: ceps: Ͳ Ͳ Ͳ Ͳ 4 4 Hamstrings: Hamstrings: Outcomes R TSK ASES 4 Hamstrings: Hamstrings: test; ASES, Mental Quality

131 Chapter5

APPENDIX5.5

Resultsofthefinalmixedmodelanalysisoftheothersecondaryoutcomes

Outcome N Estimate SE pͲvalue 95%CI SQUASH:minweeklyPA 99 431.422 355.00 0.228 Ͳ267.188,1139.032 RͲ36PCS 97 0.721 1.40 0.609Ͳ2.070,3.512 RͲ36MCS 99 1.161 1.70 0.497 Ͳ2.229,4.552 TSK 100Ͳ3.835 1.200.002*Ͳ6.225,Ͳ1.446 ASES 97 0.267 0.18 0.145 Ͳ0.093,0.627 4Ͳceps:isokinetic60° 100 10.231 6.19 0.103Ͳ2.119,22.582 4Ͳceps:isometric60° 99 14.938 7.04 0.037 0.913,28.963 4Ͳceps:isometric30° 97 10.072 4.33 0.023 1.435,18.710 Hamstrings:isokinetic180° 99 8.573 3.30 0.011 2.000,15.150 Hamstrings:isokinetic60° 100 6.717 4.40 0.131Ͳ2.050,15.483 Hamstrings:isometric60° 100 7.714 4.12 0.070 Ͳ0.637,16.065 Hamstrings:isometric30° 98 5.681 5.68 0.270Ͳ4.497,15.858 ASES, Arthritis SelfͲEfficacy Scale; CI, Confidence Interval; MCS, Mental Component Score; PCS, Physical ComponentScore;QoL,Qualityoflife;RͲ36,Rand36;SE,standarderror;SQUASH,ShortQuestionnaireto ASsess HealthͲenhancing physical activity; TSK, Tampa Scale for Kinesiophobia; 4Ͳceps, quadriceps. *sig. p=<0.0022(thealphausedforallsecondaryoutcomestoadjustformultipletesting)

132 Chapter5 APPENDIX5.5 Resultsofthefinalmixedmodelanalysisoftheothersecondaryoutcomes Outcome N Estimate SE pͲvalue 95%CI SQUASH:minweeklyPA 99 431.422 355.00 0.228 Ͳ267.188,1139.032 RͲ36PCS 97 0.721 1.40 0.609Ͳ2.070,3.512 RͲ36MCS 99 1.161 1.70 0.497 Ͳ2.229,4.552 TSK 100Ͳ3.835 1.20 0.002*Ͳ6.225,Ͳ1.446 ASES 97 0.267 0.18 0.145 Ͳ0.093,0.627 CHAPTER6 4Ͳceps:isokinetic60° 100 10.231 6.19 0.103Ͳ2.119,22.582 4Ͳceps:isometric60° 99 14.938 7.04 0.037 0.913,28.963 4Ͳceps:isometric30° 97 10.072 4.33 0.023 1.435,18.710 Hamstrings:isokinetic180° 99 8.573 3.30 0.011 2.000,15.150 Hamstrings:isokinetic60° 100 6.717 4.40 0.131Ͳ2.050,15.483 Hamstrings:isometric60° 100 7.714 4.12 0.070 Ͳ0.637,16.065 Hamstrings:isometric30° 98 5.681 5.68 0.270Ͳ4.497,15.858 ASES, Arthritis SelfͲEfficacy Scale; CI, Confidence Interval; MCS, Mental Component Score; PCS, Physical Generaldiscussion ComponentScore;QoL,Qualityoflife;RͲ36,Rand36;SE,standarderror;SQUASH,ShortQuestionnaireto ASsess HealthͲenhancing physical activity; TSK, Tampa Scale for Kinesiophobia; 4Ͳceps, quadriceps. *sig. p=<0.0022(thealphausedforallsecondaryoutcomestoadjustformultipletesting)

132 Chapter6

134 Generaldiscussion

The aim of this dissertation was to develop and evaluate an aquatic cycling training programmeforpatientswithmildtomoderatekneeosteoarthritisሺOAሻǤThisgeneral discussion presents an overview of the research preceding the development of an aquatic cycling training programme in addition to an evaluation of the intervention itself.Furthermore,methodologicalandpracticalconsiderationsarediscussedfollowed byabriefoutlineofimplicationsforthefuture.

MAINFINDINGS

In 2010, at the beginning of this project, research on understanding the effects of aquatic cycling exercises in both the clinical and nonͲclinical settings was scarce. Furthermore, neither primary nor secondary care institutions in the Maastricht area were familiar with aquatic cycling or possessed aquatic bikes. Nevertheless, a smallͲ scale aquatic bike manufacturer expressed interest in collaborating with Maastricht University. At the start of the collaboration, the manufacturer already had a single 6 aquaticbikeprototypedesignedwiththeintenttobeusedinpatientpopulations. Accordingly, the initial step was to design and conduct a feasibility study using the aquatic bike prototype during aquatic circuit training in patients with knee OA (CChapter2).Thetrainingprogrammecombinedaquaticcyclingwithtraditionalaquatic exercises(e.g.,gaittraining,shallowͲwatertoningexercises,andflexibilityexercisesfor the lower limb). Evaluation of the programme included use of traditional feasibility criteriasuchas:patientadherenceandacceptance,possibilitytoprogressinexercise level,occurrenceofadverseeventsandoperationalaspects.1 Themajorfindingsfromourstudysuggested:First,patientsdemonstratedanoverall highattendancerateindicatingacceptanceoftheprogramme,withattendancerates being consistent with other feasibility studies of aquatic training programmes for patientswithkneeOA.2,3Second,focusgroupinterviewssuggestedpatientsatisfaction withthevarietyandintensityofexercises.Participantsalsoreportedthepainrelieving effects of the programme, which occurred immediately during the training sessions. Third, participants performed well with shallow water exercises and seated aquatic cycling.However,performingaquaticcyclinginanoutͲofͲtheͲsaddlepositionwastoo demanding for most participants. This suggested that ten minutes of aquatic cycling once a week was probably too short to achieve progression to outͲofͲsaddle movements.Lastly,participantsexpresseddesiretocyclelongerthantenminutes,but emphasized, that exercise variation was needed to prevent monotony, particularly duringseatedcycling. Weusedthefeedbackandresultsfromourfeasibilitystudytodevelopafullaquatic cyclingprogrammeforpatientswithmildtomoderatekneeOA(CChapter3).Toprevent monotony,wecombinedarmexercises,legexercisesandoutͲofͲsaddlepositionswith

135 Chapter6 continuouscyclinginseatedpositions.However,weintroducedoutͲofͲsaddleexercises onlyafteraminimumoftwoweeksoftraining,andwekeptthetotaltimeofcyclingin standing positions low to prevent overload of knee joints. Due to a lack of prior research on the effects of full aquatic cycling programmes, we based exercise parameterssuchasduration,frequencyandexerciseintensityonexerciseguidelines for OA.4Ͳ6 Furthermore, we relied on evidence taken from knee loading during landͲ based cycling and spinning to design the composition and setͲup of different exercises.7Ͳ10 Chapter 3 outlines our training programme as well as the design of a singleͲblinded, twoͲarm, randomised controlled trial to compare the effects of the 12Ͳweekaquaticcyclingprogrammewithusualcare.Theprimaryoutcomesincluded selfͲreported knee pain and physical functioning because these are the most prominent symptoms of knee OA. In accordance with recommendations from the Dutch Physical Therapy Guideline for OA and international research societies, secondary outcomes included performance tests of physical functioning, muscle strength,qualityoflifeandpatientglobalassessmentofdiseasestatus.5,11Ͳ14Wealso included measurements of selfͲreport physical activity, fear of movement and selfͲ efficacyforphysicalfunctioning.DatameasurementtimeͲpointsincludedbaseline,at theendofthe12ͲweekinterventionperiodandafollowͲupat24ͲweekspostͲbaseline. Participantsinbothgroupsalsofilledoutdiariesduringtheinterventionphaseinorder toprovideinformationoncompliancetousualcare(e.g.medicationuse,utilisationand content of physical therapy, level of physical activity) while also capturing the developmentofOAimpairmentsthroughouttheintervention.However,theanalysisof thediarydataisnotpartofthisdissertation. During the developmental phase of the interventions included in this dissertation (Chapter2and3)weperformedsystematicsearchesofonlinescientificdatabasesto find evidence on the effects aquatic cycling exercise. Chapter 4 includes a scoping reviewofscientificdataonthetopicofaquaticcycling.Weidentified63publications with most research focusing on the comparison between aquatic cycling and land cycling,orstudiesinvestigatedtheeffectsofdifferentexerciseconditions(e.g.water temperature)duringaquaticcycling.Onlysixstudiesevaluatedtheeffectsofaquatic cyclingexerciseinterventions.Resultsfromourreviewconfirmedourinitialhypothesis that there is a dearth of scientific clinicalͲbased peerͲreviewed research that has investigatedtheeffectsofaquaticcyclingexercises,includingoutͲofͲsaddlepositions andupperbodyexercises. Chapter5describestheresultsofourrandomisedcontrolledtrial,whichtothebestof ourknowledgeisthefirststudytoevaluatetheeffectsofagroupͲbasedaquaticcycling trainingprogrammeforpatientswithmildtomoderatekneeOA.Ourresearchstudyis also the first to evaluate the effects of groupͲbased aquatic cycling exercise accompanied by a mediumͲterm followͲup period. According to recent systematic reviews,15Ͳ17 only three aquatic exercise studies to date have included a followͲup period.18Ͳ20Accordingly,theresultsofourrandomisedcontrolledtrialsuggestedthata

136 Generaldiscussion

12Ͳweek aquatic cycling exercise programme is effective for improving selfͲreported kneepainandphysicalfunctioningcomparedwithusualcare.Observationsfromour studyalsosuggestedthatourprogrammeresultedinanimprovementinperformanceͲ basedphysicalfunctioning,adecreaseinfearofmovementandanincreaseindiseaseͲ relatedqualityoflifeexclusivetotheaquaticcyclinggroup.Interestingly,participants maintainedfavourableeffectsoftheaquaticcyclingprogramme(exceptforqualityof life)atthreemonthsfollowingcessationoftheintervention.

METHODOLOGICALCONSIDERATIONS

The following section highlights four aspects important to the development and evaluation of the aquatic cycling training programme, including: recruitment and dropout,comparisonofaquaticcyclingwithusualcare,usualcarecomplianceandthe efficacyoftheaquaticcyclingexerciseprogramme. 6 Recruitmentanddropout When interpreting results from our RCT it is noteworthy that the final number of patients(n=102)includedinthisstudywaslowerthanthecalculatedapriorisample size(n=168).ItispossiblethatourrecruitmentsettingandthewaitͲlistcontroldesign impededrecruitmentandfacilitateddropout.

Recruitment GeneralpractitionersintheNetherlandstypicallymanagepatientsusingaconservative treatmentapproachforOA,largelybecausethereareonlyfewdocumentedreasons requiringareferraltoanorthopaedicsurgeon.Forexample,oneinstanceisifapatient responds unsatisfactory to conservative treatment or if the diagnosis is unclear.21 However,amajorityofpatientsarestillreferredtoanorthopaedicspecialistresulting inunnecessaryhospitalvisits.21 Therefore, to address this problem, the MUMC+ set up a twoͲvisit specialist care intervention programme (i.e. initial and evaluation consults), coordinated by an orthopaedic nurse practitioner, in 2011. The Early OA Outpatient Clinic supports patient selfͲmanagement for knee OA, but also provides guidance on available treatment options following a stepped care model.22,23 In the initial step, patients receiveeducationandlifestyleadvice.Ifadditionalcareisneeded,patientsarethen referred to primary care such as physical therapy or weight loss therapy (step 2). Patientsmayalsohavetheoptionofreceivinganinjectionwithhyaluronicacidorpain medication, receive other minor interventions or they may be referred to an orthopaedicsurgeon(step3).

137 Chapter6

Participants were recruited from the Early OA Outpatient Clinic of the MUMC+. Consistent with other aquatic exercise studies,18,24Ͳ26 our study population involved participantswithanindicationforconservativetreatment(e.g.notlistedfortotalknee surgery)includingareferralforphysicaltherapy(step2),andnocontraͲindicationsfor aquaticexercise.Advantagesofusingthisrecruitmentsettingandstrategywerethe availability of comprehensive diagnoses documentation, which included radiographs andpatienteducationdeliveredbyanexperiencednursepractitioner. The evaluation of the number of referred patients to the Early OA Outpatient Clinic showedthatonlyasmallfractionofthereferredpatientswaseligibleforparticipation inourtrial.Approximately250patientscontacttheclinicperyear,27andfromthispool ofpotentialparticipantsonly35%(n=88)demonstrateanindicationforprimarycare physical therapy.27 From our feasibility study we knew that 50% of eligible patients wouldbewillingtoparticipateandthiswouldresultinarecruitmentrateof44patients peryearandastudydurationofalmostfouryears.Theseestimateswereoptimisticas nearly 41% of all patients receive an injection of hyaluronic acid of corticosteroid, whichwasanexclusioncriterionforparticipationinourtrial(ifpatientsreceivedan injection in the last three to six months). Unfortunately, we did not have access to thesenumberswhenthedesignphaseofthistrialbegan.Therefore,weextendedour recruitment duration from 1.5 to 2.5 years, which enabled us to recruit 102participants. As shown above, this was the maximum number of patients that could have been recruited in this period. Due to funding restrictions, any further extensionoftherecruitmentphasewasnotpossible. A possible alternative recruitment strategy would have been the involvement of generalpractitioners,sincemostpatientswithmildtomoderatekneeOAaretreated inprimarycare.28However,thepatientsthatarereferredtotheEarlyOutpatientClinic are likely to be different compared with those not referred. In many cases, patients demonstratingasuddenflareͲuporthoseneedingamedicalevaluationforexclusionof otherdiagnosesarereferredtotheclinicbytheirgeneralpractitioner.

Dropouts Althoughtheactualdropoutrateof24%(n=27)wasclosetotheexpectedpercentage (25%) some dropouts may have been preventable. For example, nine participants in the usual care dropped out before the baseline assessment and were lost for the analysis.Becauseanextravisittothehospitalneededtobescheduledforthebaseline assessment,participantsrequestedtobeinformedabouttheirgroupallocationbefore the baseline assessment. Providing this information on the treatment allocation is a recommended strategy to improve recruitment for randomised controlled trials.29 Despite this, four participants declined because they were unsatisfied with being allocatedtothecontrolgroup.

138 Generaldiscussion

Mostofthedropouts(n=12)occurredfollowingthebaselineassessment.Similarto before,themainreasonfordropoutinthecontrolgroup(n=6)wasdisappointment about group allocation (n = 4). Therefore, in an attempt to minimise dropout we offeredparticipantsintheusualcaregroup12weeklysessionsofaquaticcyclingafter thelastfollowͲupmeasurementat24Ͳweeks.However,duetoorganisationalproblems (e.g.timingoftheopeningofthenewregionalswimmingpool)thisactionplancould notbedelivereduntilfourmonthsfollowingtheinitiationofrecruitment.Priortothe updated plan for controls, individuals were offered a oneͲday walking clinic as compensationforparticipation.AfterwecouldofferthetryͲoutsessionstotheusual grouponlyoneparticipantintheusualcaregroupwithdrewfromthetrialbecauseof disappointmentwiththegroupallocation. Consistentwithourstudy,othersreportingresearchstudyparticipationsatisfactionin cancer patients suggest that offering a shorter postͲtrial exercise programme is not able to adequately compensate for disappointment about usual care group allocation.30Incontrast,deRooijetal.offeredthefullprogrammeof20weekstailored exercise therapy (i.e. for patients with knee OA) to usual care group participants 6 following32weekstrialparticipationandthisresultedonlyinasingledropoutbecause ofdissatisfactionwithgroupallocation.31

AquaticCyclingcomparedtoUsualCare ThedesignofourRCTisconsistentwithotherstudiesasreportedinsystematicreviews comparing aquatic therapy with a control group associated with no specific interventionsuchasusualcare,educationorsocialattention.15,16 Inourstudy,theaquaticcyclingprogrammewascomparedwithusualcareprovidedby the Early OA Outpatient Clinic.All eligible patients received education and individual lifestyleadviceregardingOAandareferralforphysicaltherapyfromanorthopaedic nurse practitioner. Research shows that these specialist care visits are effective for improvingselfͲreportedpain,physicalfunctioningandhealthrelatedqualityoflifein patients with lower limb OA.22,32 These improvementsare also accompanied byhigh patientsatisfaction.22,27,32 Although we found no improvement in the usual care group, patient symptoms remained stable with no changes over time, despite the fact that only 32% of our participantsintheusualcaregroupconsultedwithaphysicaltherapist.Theseresults are in line with a longitudinal Dutch cohort study of patients with hip and knee OA receivingusualcare(e.g.medication,physicaltherapy)fromahospitalorrehabilitation centre.33 Unfortunately, underutilisation of physical therapy and other conservative treatmentmodalitiesforkneeOAiscommon.32,34Thereasonsforthelowutilisationof physicaltherapyinourtrialwillbediscussedinlatersections. Thephysicaltherapyprovidedtoparticipantsvariedfrommanualtherapy,education andhomeexercises,tosupervisedexercisetherapyincludingneuromuscular,aerobic

139 Chapter6 and/or strength exercises. The different types of exercise offered to our usual care groupalignwiththegeneralrecommendationthatexerciseiseffectiveforOA,andthat thereisnoevidencesupportingthebenefitsofacertaintypeofexercisemorethan others.12,35TheindividualisededucationandadviceofferedbytheEarlyOAOutpatient Clinicincombinationwiththedifferentphysicaltherapytreatmentsarealsoconsistent with the understanding that treatments should be tailored to the needs and preferencesoftheOApatient.35Therefore,theselectionofusualcareasacomparison groupseemsadequate.36 The results from our study cannot be used to answer the question whether aquatic cyclingissuperiortolandͲbasedexercise,whichiscommonlyappliedinprimarycare physicaltherapyforkneeOA.However,anycomparisonwithlandͲbasedexerciseisnot likelytoappropriateduetohighlydifferenttrainingmediums.16 Nevertheless, our findings add to the body of evidence suggesting that exercise is effectiveforimprovingpainandphysicalfunctioninginpatientswithkneeOA.11,12,35,37 However, our study design may have increased the risk of a positive bias towards aquatic cycling because participants were aware they received a free treatment compared with the usual care who did not receive free exercise therapy during the active study period.16 But our study design was appropriate with respect to our outlinedgoalsoftestingthepossibleeffectsofaquaticcyclinginpatientswithmildto moderatekneeOA.36

Usualcarecompliance Although stepped care models should theoretically reduce the underutilisation numbers of conservative treatment options,38 68% of our participants in the control groupdidnotconsultaphysicaltherapist.Eveniftheabsolutereasonsareunknown, wehaveseveralhypothesesthatmayaccountforourfindings. First,physicaltherapyforOAiscurrentlynotincludedinthebasichealthcarepackage intheNetherlands,andeveninupgradedpackagesphysicaltherapysessionsareoften limited.34,39 Financial costs are considered a barrier for exercise participation and adherenceinpatientswithkneeOA.34 Second,theeffectivenessofaninterventionplaysanimportantroleinthetreatment choiceofpatients.Lackofknowledgeonthebenefitsofexercisemightdetourpatients fromconsultingwithaphysicaltherapistaboutthebenefitsofexercise.40Despitethis, our patients received information on the benefits of physical therapy and exercise during their consultation at the early outpatient clinic. This brief exposure to new informationmightnotbestrongenoughtooutweighpersonalbeliefsandexperiences withexercise,whicharestrongpredictorsofexerciseparticipation.41Ͳ43 Third,theopportunitytotryaquaticcyclingfollowingstudyparticipationmighthave influenced the decision ofparticipants in the control group onwhether to consulta physicaltherapistornot.Inordertoparticipateinourtrial,participantsinthecontrol

140 Generaldiscussion group were not obliged to start with physical therapy. This aligns with recommendations to minimalize the influence of the study on usual care.44 Still, the waitͲlistaspectofourRCTdesignmayhavereducedparticipantmotivationtoconsult withaphysicaltherapist.45 Remarkably,outofall32patientswhoconsultedwithaphysicaltherapist,onlyone participated in a groupͲbased exercise programme according to the Dutch Physical Therapy Guideline for exercise programmes in OA.37 The other participants who consulted a physical therapist followed shortͲterm (three to six sessions) individual trajectories.5,28,46 Previous research on attitudes and beliefs of physical therapists regardingexerciseforkneeOAintheUnitedKingdomdemonstratedthatfourtofive sessionswereevaluatedadequatelyenoughtosetͲupanexerciseprogramme.47The physicaltherapistsfeltthatpatientsneededtocontinuewiththeexerciseprogramme ontheirownandtakeresponsibilityfortheirownhealth.47Inaddition,accesstomore sessionsofphysicaltherapymightbelimitedfromapatientperspective,includinglack offinancialresources39orconcernsthatphysicaltherapymightbeneededagaininthe future(e.g.forOAorcomorbidities).However,recentresearchsuggeststhattwelve 6 sessions of physical therapy are effective for creating an exercise routine, improve symptoms and reduce costs associated with more complex healthcare linked to worsening of symptoms.39 Therefore, the Dutch Ministry of Health and Welfare and Sportmadethedecisiontoaddtwelvephysicaltherapysessionsforpatientswithhip andkneeOAperyeartothebasichealthcareinsuranceasofJanuary1st2018.39

Exerciseprogramme Exercise treatments for knee OA should reduce pain and physical disability. Aquatic cycling may be effective for reducing these impairments. Participants in our studies demonstratedpositiveattitudesaboutthetypeofexerciseandvaluedtheimmediate pain relief. However, while this instant relief phenomenon is frequently reported acrossdifferentaquaticexercisestudies,25,26,48Ͳ50itismorelikelythatacutepainreliefis attributabletotheaquaticenvironment(e.g.jointͲspecificlocalphysiologicalresponses to buoyancy, hydrostatic pressure and warm water temperature)51,52 as opposed to typeofaquaticexercise. Nevertheless,accordingtoarecentCochranereview,16aquaticexercisecontributesto small shortͲterm improvements in pain and physical function, suggesting aquatic exerciseaddstothebeneficialeffectsoftheaquaticenvironment.

ĨĨĞĐƚŽĨĂƋƵĂƚŝĐĐǇĐůŝŶŐŽŶƉĂŝŶĂŶĚƉŚǇƐŝĐĂůĨƵŶĐƚŝŽŶŝŶŐ WeobservedamediumeffectonpainandsmallͲtoͲmoderateimprovementsinboth selfͲreported and performanceͲbased physical functioning. The reasons for the reportedimprovementsarelikelymultifaceted.

141 Chapter6

First, the aquatic cycling programme can be categorised as singleͲexercise aerobic training.Regardlessofwhetherthebodypositionchangedorthefocusoftheexercises shifted (e.g. to the upper body) during training, participants kept cycling for 45minutes.AsystematicreviewofJuhletal.53showsthatexercisesessionsfocusing ononeaim(e.g.aerobiccapacityormusclestrength)aremoreefficaciousthanmixed sessions. Our results are in line with studies on the effects of landͲbased aerobic walking or cycling exercises for OA, which showed a reduction of pain and physical limitationsinOApatients.7,8,54Ͳ56Anadvantageofaquaticcyclingisthatcyclingcanbe combinedwithexercisesoftheupperbodyanddifferentbodypositions,whichensures theinvolvementofthewholebodyduringonesession.AccordingtoFransenetal.,25 whole body movements might have a beneficial influence on selfͲreported physical functioning. Second, our findings suggest that exercise intensity was appropriate in leading to improvements in functional aerobic capacity and muscle endurance. Following the trainingperiod,participantsintheaquaticcyclinggroupwalkedmoremetersduring the sixͲminute walk test, a surrogate measure for cardiovascular function.57 These observations are consistent with research showing that aquatic cycling interventions improved cardiovascular fitness in patients with multiple sclerosis as well as healthy obese participants [58Ͳ60]. Furthermore, isokinetic quadriceps strength testing with highvelocities,anindicatorformuscularendurance,reachedaborderlinesignificance in favour of the aquatic cycling group and probably the real effect could not be detectedduetoinsufficientpower. Third,wecontrolledexerciseintensitytoensurealighttomoderateintensetraining. WeusedtheageͲpredictedequation(220Ͳage=max.heartrate)toestimatemaximum heartrate.Wemadenocorrectionsforpossibleeffectsofimmersionbecauseprevious researchshowedthatheartrateonlandandduringaquaticcyclingarecomparable.61A disadvantageofthisequationisthatitdoesnottakeintoaccountindividualvariation andmayunderestimatemaximumheartrateinolderadults.62Therefore,wealsoused the Borg scale and pedal speed. The Borg scale is commonly used to determine exerciseintensityofaquaticexercises63,64withevidencesuggestingtheBORGscaleis alsousefulfortheregulationofexerciseintensityduringaquatictreadmillexercise65 and aquatic cycling training.66,67 Since buoyancy might assist upward movements in water(e.g.duringcyclingwhenthepedalispulledupwardsduringacycle)controlof exercise intensity is crucial to ensure that water resistance offsets buoyancy.48 Therefore,weaimedtokeepparticipantsatapedalspeedof60rpmduringseated cycling,whichwasguidedandcontrolledbycountingthefrequenciesandsimulating pedal pace with rhythmic commands. This frequency was commensurate with a “somewhat hard” intensity (13, 6 to 20 Borg scale) and a heart rate of 50% of maximum,whichisconsistentwiththerecommendationsforexerciseintensityinknee OA.4,37

142 Generaldiscussion

Fourth,improvementsinphysicalfunctioninginourstudymightbeexplainedbythe factthat(outͲofsaddle)cyclingrequiresamuscularactivationandcoordinationsimilar to landͲbased walking and sitͲtoͲstand activities.7,8,68Ͳ70 Moreover, the aquatic environmentmighthaveledtoimprovementsinneuromuscularfunctionoftheknee joint,whichhasbeenobservedtoleadtolesssymptomsofkneeOA.71,72Throughout training sessions participants were repeatedly reminded to keep the knees aligned. However,waterturbulenceandpressurepresentedanextrachallengeintheabilityto keepthealignement,andthereforethismighthaveenhancedthetrainingeffect.

ĨĨĞĐƚŽĨĂƋƵĂƚŝĐĐǇĐůŝŶŐŽŶƋƵĂůŝƚǇŽĨůŝĨĞĂŶĚĨĞĂƌŽĨŵŽǀĞŵĞŶƚ In addition to improvements in pain and physical functioning, participants in the aquatic cycling group also reported an increase in quality of life and less fear of movement.WefoundashortͲtermmediumeffectsize(ES=0.71)forqualityoflife.A CochraneReviewonaquaticexerciseinhipandkneeOA16foundaneffectsizeof0.25 for quality of life. However, the Cochrane Review took general and diseaseͲspecific 6 qualityoflifequestionnairesintoaccount,whileweonlyfoundaneffectfordiseaseͲ specificqualityoflife.DiseaseͲspecifictoolsaremoreresponsiveandcandetectsubtle effectsonqualityoflife.73 Thepainrelievingeffectoftheaquaticenvironmentmightexplainthereductioninfear ofmovement.Patientsaresuggestedtoexperiencelesspainwithexerciseinwater, and this may have influenced participant perception of physical capabilities.74,75 Research on the effects of aquatic treadmill exercise also showed that the aquatic environmentmightserveasasupportingfactorintheabilitytoexerciseatacertain levelofintensitywithoutanincreaseinpain.49Thisisinlinewiththeexperienceofour participants,whoperceivedaquaticcyclingmoredemandingthanonland,butwhileat thesametimeexerciseinwaterappearedlesspainfulthanonland.

DĞĚŝƵŵͲƚĞƌŵĞĨĨĞĐƚƐŽĨĂƋƵĂƚŝĐĐǇĐůŝŶŐ Theeffectsonpain,physicalfunctioningandfearofmovementweremaintaineduntil the followͲup measurement period. While interpreting our results, the study setting and supervision should be taken into account since both may have influenced adherenceandthelastingeffectsobserved.76 AccordingtoDobson etal.,76 theenvironmental context influences participation and adherence.Theinterventionprogrammewasfreeofcost,conductedinasmallhospital therapypoolwithindividualshowerandchangingfacilitiesandthelocationwaseasily reachableeitherbybusorcar.Theimpactofthesefactorsbecameclearsinceonlyone participant in the intervention group continued regularly with aquatic cycling in the publicswimmingpool.Otherparticipantsfromtheinterventiongroupdidnotcontinue with the sessions in the public swimming pool or stopped after a few sessions. Reportedreasonsfordiscontinuingwithaquaticcyclingwerecosts(7€persessions),

143 Chapter6 location(e.g.difficulttoreachbybus),colderwatertemperatureandthelargergroup sizeofapproximately10Ͳ15participants.Anotherexplanationfornotcontinuingwith aquatic cycling may have been that experienced benefits did not outweigh the increased time and monetary investment related with the sessions in the public swimming pool. On the other hand the maintained improved functioning may have been sufficient to do daily activities so the additional training effort might not be consideredasneededatthattimepoint. Withregardtostudysupervision,physicaltherapistseducatedpatientsonthebenefits of exercise and encouraged patients during exercise. Guidelines emphasize that educationisanimportantelementinthepersistentmanagementofOAwithbothlandͲ based and aquatic exercise studies demonstrating favourable effects of including patient education.19,31,56 Moreover, the training setͲup (e.g. participants were sitting stableonbikesclosetotheedgeofthepool,whichfacilitatedeasycommunication) enabledthesupervisingphysicaltherapisttoeducateparticipantsduringtraining.This education might have altered participants perception of their physical ability and, in combinationwithimprovementsinfunctionalcapacity(e.g.progressioninsixminute walk test), participants may have perceived less restrictions during activities of daily living.Finally,thismighthaveresultedinanincreasedphysicalactivityduringactivities ofdailylivingevenaftercessationofourprogramme,whichwecanonlyhypothesizeat thistime,couldhavecontributedtothemaintainedeffectsobservedatthefollowͲup period.

PRACTICALCONSIDERATIONS

Therearepracticalconsiderationsthataccompanythequestionofwhetherinvestment inaquaticbikesinhealthcaresettingsorpublicswimmingpoolsisworthwhile.

AquaticcyclingͲatrendysportsactivity? Severalindicatorssuggestthataquaticcyclingisalastingtrendynichewithinthefield ofaquaticfitness.First,closelyfollowingthebeginningofthe21thcenturytherehas beenthedevelopmentofmanynewsportͲandfitnessconceptsandactivities.Sport hasdevelopedintoalifestyleandispresentinpopculturesas(social)media,music, fashionandmarketing.77Oftentimesthese‘new’sportsareamixofalreadyexisting sports.Forexample,“taebo”combinesdancing,boxingandaerobics.77Aquaticcycling isalsoa‘sampling’sportandexclusiveaquaticcyclingcentresinNewYork,Franceand the Netherlands offer aquatic cycling training in a trendy environment to meet the aesthetic dimension that is nowadaysoften related tosports.77To meet the current pluralityoflifestylestheaquaticcyclingsessionsofferedinthesecentresarefurther specialisedbycombiningaquaticcyclingwithyogaorprovidespecialclassesformen.

144 Generaldiscussion

Second,duetotheincreasinguseofwearablesandtechnologyinsports,peoplewant toexercisesmarterandquantifytheirachievements.AccordingtoEtkin,78“OveroneͲ inͲfiveU.S.adultsusesomeformofpersonalhealthtrackingdevice…andanestimated 485 million wearable computing devices will be in the market by 2018.”. The new generationofaquaticbikescomeswithdisplaysthatenablesparticipantstochoosean exerciseprogramme,receiveinformationonduration,speedandcalorieexpenditure andprovidestheopportunitytowatchtelevisionorlistentomusicduringtraining. Third, aquatic cycling has now been available to consumers on the open market for severalyears.AccordingtoLambrechtandStamm,79itislikelythatafterfiveyearsa respectivetrendinsportlosesitscharacterofbeinganewactivityandthenbecomes anestablished'normalsport.Thus,itislikelythataquaticcyclinghaslastingpowerin thefitnessmarket.

Aquaticcycling–advantagesanddisadvantages Aquaticcyclingprovidesseveraladvantagesovertraditional(aquatic)exerciseforknee 6 OA.First,themovementisverycontrolledandeasytolearn.Cheatingisnotpossible because participants must overcome water resistance with every move and this requiresselfͲpoweredcycling.Withtraditionalwaterexercisemovingthetargetlimb with less tension or speed can easily decrease the intensity of an exercise. Second, individualswithkneeOAoftenexperiencediscomfortatotherbodysites.Therefore, landͲbasedcyclingmaybeuncomfortable(e.g.duetolowerbackpainorwristpain).8,80 Assuch,withaquaticcycling,thelowerextremitiesandlowerbackareimmersedin warm water resulting in a relief of joint loading. Therefore, aquatic cycling is a promisingexercisealternativeforpeoplewithkneeOA.Third,aquaticcyclingcanbe extendedtoawholebodyworkoutbyinvolvingthearmsandupperbody.Fourth,due tothestationarycharacteroftraining,thesupervisingtrainercaneasilycommunicate with participants and readily assess the performance of each individual. Moreover, participants sit stable on the bike and are immersed in water to chest level, which could be a psychological advantage for individuals with poor swimming skills and/or embarrassment related to high body weight. Obese individuals are often times too buoyant in traditional aquatic exercise classes, whereas sitting on the aquatic bike preventsthisproblem.UseofapparelsuchasbikeshortsandtͲshirtsinordertolessen exposureofbodyregionsisacceptableduringaquaticcycling. In contrast, the stable and stationary character of aquatic cycling that prevents participantsfromfloatingalsolimitsvariationinexercise.Furthermore,aquaticbikes arealsocostly,takeupstoragespaceandthetransferfromthedecktothewatertakes extra time if patients are not able to place bikes in the pool themselves. Therefore, withrespecttooperationalaspects,ourexperiencesuggestedthataswimmingpool withanadjustablefloormightbepreferablenotonlyforsettingupbikes,butalsofor adjustingwaterdepthaccordingtoparticipantbodyheight.2WithtraditionalshallowͲ

145 Chapter6 wateraquaticexerciseequipmentisnotnecessarilyneededtoachievetheintended traininggoalsandexerciseintensities.81,82

AquaticcyclingforpatientswithkneeOA–betweentherapyandlifestyle intervention The results of our trial suggest that aquatic cycling can be used for primary care physical therapy. Since health guidelines recommend an individualised treatment strategies that meet the needs of a heterogeneous population of OA patients,12,35,37 aquaticcyclingisafeasibleoptionforthoseneedingtoimprovecardiovascularfitness, muscularenduranceand/orneuromuscularcoordination. Nevertheless, the availability of a pool and aquatic bikes in a primary care physical therapysettingisnotcurrentlyanoptionthatisreadilyoffered.Thus,aquaticcyclingin the current format is more likely to be effective as a communityͲbased intervention thanatreatmentoptioninprimarycaresettings.Consequently,thequestionremains astowhetherourprogrammecanbetranslatedtoacommunityͲbasedaquaticcycling programme. Research suggests that communityͲbased aquatic exercise programmes increase qualityoflife83Ͳ87andphysicalfunction.84,87Theseprogrammesarecommonlyopento abroadertargetpopulation(e.g.hipandkneeOA)andsupervisedbysportandlifestyle professionals. The latter provides a pragmatic alternative to the delivery of interventionsbyhealthprofessionals,suchasphysicaltherapists.However,toensure safeandbeneficialexerciseparticipationitisimportantthatfitnessprofessionalsare educatedwithregardtomusculoskeletalpathology.76Thisdissertationprovidesinsight intodifferenttrainingsetͲupsforpatientpopulations(chapter2Ͳ4),andbasedonthese findingsawellͲeducatedaquaticfitnessprofessionalshouldbeabletoadjusttraining sessionstomeettheneedsofaparticipantwithkneeOA.Bycontrast,patientsneedto beabletoparticipateincommunityͲbasedexerciseclasses.Forexample,patientswith restrictionsinjointfunctioninganddailyactivitiesshouldconsultaphysicaltherapist first.37Forpatientsthatdonotexerciseregularly,butwhoarewillingtoandableto start an exercise programme, physical therapists can provide guidance on how to design an exercise routine while also providing advice to participants in making the transitiontocommunityͲbasedexerciseprogrammes.37

IMPLICATIONSFORFUTURERESEARCH

This final section outlines ideas for future research with regard to aquatic cycling. Furtherresearchshouldsupportthreeimportantgoalsof(aquatic)exerciseresearchin OA: 1) optimising the use of aquatic exercise therapy, 2) assessing the (longͲterm) effectiveness of (aquatic) exercise on symptoms and structural changes of OA, and

146 Generaldiscussion

3)improvingpostͲinterventionmaintenanceofanexerciseroutineorlifestylephysical activity.15,16,88 Theuseandpotentialbenefitsofaquaticcyclinginatherapeuticcontextneedfurther research.Ourfindingssuggestthataquaticcyclingmaybeusedastherapeuticexercise forpatientswithmildtomoderatekneeOA.Basedonhighpatientacceptance,itis likelythataquaticcyclingmightalsobeanexercisepossibilityforpatientswithmore severe knee OA. Furthermore, aquatic cycling might be a valuable option for postͲ surgical treatment. Previous research indicates that aquatic cycling might be useful during rehabilitation following total knee surgery89 or anterior cruciate ligament reconstruction.90 Another patient group that might benefit from aquatic cycling are patients with hip OA since landͲbased cycling has been deemed a suitable type of exercise.91Tothebestofourknowledge,thereisnoevidenceavailableontheeffects ofaquaticcyclinginpatientswithhipOA. With regard to the longͲterm effectiveness of aquatic cycling, future studies should incorporate longer followͲup periods of six months and 12Ͳmonths postͲbaseline. In addition,theknowledgeregardingtheeffectsofaquaticcyclingshouldbeextended. 6 Forexample,itwouldbeinterestingtofindoutwhetheraquaticcyclingcouldbeused toimproveneuromuscularcontrolinpatientswithkneeOA.Theturbulenceofwater providesastimulatingenvironmentforneuromusculartraining,andthereforeproper alignment of the lower limbs can be trained in a functional manner. Additionally, researchontheeffectofexerciseoncartilagehealthandserumbiomarkersisstillinits infancy.92Researchinhumansandratssuggeststhattibiofemoralcartilagemightbe responsive to aquatic training,81,93 but further research is required to better understandtheexerciseͲinducedresponsesofcartilageanditsclinicalrelevance. Likewise,musclestrengthisanimportantcontributortotheimprovementofkneepain and physical function,94,95 but whether aerobic cycling interventions can achieve increasesinmusclestrengthremainsunclear.7,8,55,96Therefore,itwouldbeinteresting toinvestigatetheeffectsonmusclestrengthofanaquaticcyclingprogrammethatis complimented with resistance training. Another valuable extension of the aquatic cycling programme might be the addition of dietary therapy21 since obesity might influenceprogressionofOA.97 To improve our understanding of patient responsiveness to aquatic exercise, future studiesneedtobetterdescribethestudypopulation(e.g.age,sex,BMI,diseasestatus and duration, diagnostic criteria) and intervention (e.g. goal, intensity, duration and frequency).16Foraquaticcyclingstudies,thelattershouldincludeinformationontype of aquatic bike, immersion depth, water temperature, rpm, pedal resistance and external power output. External power output is a possibility to control exercise intensitybytheworkthatisgenerated,whichisusuallyexpressedinwattwithlandͲ based cycling. Due to the many factors influencing the power output in the aquatic environmentwewerenotabletoquantifyandcontrolthepoweroutput.Recently,a

147 Chapter6 simple equation was developed to estimate the external power output in aquatic cycling.98 In order to increase accessibility and support longͲterm adherence to exercise, researchontheimplementationofaquaticcyclingasacommunityͲbasedintervention for patients with OA seems indicated. Future research should establish criteria that ensure a safe transition to communityͲbased aquatic cycling interventions. For example, patients and aquatic cycling instructors should be properly educated to preventoverloadofthesymptomaticjointsduringcommunityͲbasedgroupclasses. AnotherareaofconcernshouldbethecostͲeffectivenessofcommunityͲbasedaquatic cyclingclasses.Aquatictherapy,particularlywhendeliveredoneͲtoͲoneviaaphysical therapist,isassociatedwithhighercoststhanlandͲbasedexercisetherapy.100However, communityͲbased programmes might be costͲeffective in terms of health care and societalcosts.20,101 Finally, lack of reliable care data and/or subsequent overly optimistic recruitment projectionsdelayedrecruitmentofparticipantsinmanytrialsincludingourown.99To ensure that future studies have sufficient power we recommend evaluating the recruitmentstrategyduringthedesignphaseofafuturestudy.

148 Generaldiscussion

REFERENCES

1. BowenDJ,KreuterM,SpringB,CoftaͲWoerpelL,LinnanL,WeinerD,etal.Howwedesignfeasibility studies.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨWƌĞǀĞŶƚŝǀĞDĞĚŝĐŝŶĞ.2009;36:452Ͳ7. 2. SegalNA,WallaceR.Toleranceofanaquaticpowertrainingprogrambyolderadultswithsymptomatic kneeosteoarthritis.ƌƚŚƌŝƚŝƐ.2012;2012:895495. 3. LauMC,LamJK,SiuE,FungCS,LiKT,LamMW.PhysiotherapistͲdesignedaquaticexerciseprogramme for communityͲdwelling elders with osteoarthritis of the knee: a Hong Kong pilot study. ,ŽŶŐ <ŽŶŐ DĞĚŝĐĂů:ŽƵƌŶĂůyŝĂŶŐŐĂŶŐzŝyƵĞĂŚŝ.2014;20:16Ͳ23. 4. Exerciseprescriptionforolderadultswithosteoarthritispain:consensuspracticerecommendations.A supplementtotheAGSClinicalPracticeGuidelinesonthemanagementofchronicpaininolderadults. :ŽƵƌŶĂůŽĨƚŚĞŵĞƌŝĐĂŶ'ĞƌŝĂƚƌŝĐƐ^ŽĐŝĞƚǇ.2001;49:808Ͳ23. 5. PeterWF,JansenMJ,HurkmansEJ,BlooH,DekkerJ,DillingRG,etal.Physiotherapyinhipandknee osteoarthritis: development of a practice guideline concerning initial assessment, treatment and evaluation.ĐƚĂƌĞƵŵĂƚŽůſŐŝĐĂƉŽƌƚƵŐƵĞƐĂ^ŽĐŝĞĚĂĚĞWŽƌƚƵŐƵĞƐĂĚĞZĞƵŵĂƚŽůŽŐŝĂ.2011;36:268Ͳ81. 6. BarkerKL,DawesH,HansfordP,ShamleyD.Perceivedandmeasuredlevelsofexertionofpatientswith chronicbackpainexercisinginahydrotherapypool.ƌĐŚŝǀĞƐŽĨWŚǇƐŝĐĂůDĞĚŝĐŝŶĞĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ. 2003;84:1319Ͳ23. 7. MangioneK,McCullyK,GloviakA,LefebvreI,HofmannM,CraikR.TheeffectsofhighͲintensityand lowͲintensitycycleergometryinolderadultswithkneeosteoarthritis.:ŽƵƌŶĂůƐŽĨ'ĞƌŽŶƚŽůŽŐǇ^ĞƌŝĞƐ͗ 6 ŝŽůŽŐŝĐĂů^ĐŝĞŶĐĞƐĂŶĚDĞĚŝĐĂů^ĐŝĞŶĐĞƐ.1999;54:M:184Ͳ90. 8. SalacinskiA,KrohnK,LewisS,HollandM,IrelandK,MarchettiG.Theeffectsofgroupcyclingongait and painͲrelated disability in individuals with mildͲtoͲmoderate knee osteoarthritis: a randomized controlledtrial.:ŽƵƌŶĂůŽĨKƌƚŚŽƉĂĞĚŝĐĂŶĚ^ƉŽƌƚƐWŚǇƐŝĐĂůdŚĞƌĂƉǇ.2012;42:985Ͳ95. 9. Ericson MO BA, Nisell R, Nemeth R, Ekholm J. Load moments about the hip and knee joints during ergometercycling.^ĐĂŶĚŝŶĂǀŝĂŶ:ŽƵƌŶĂůŽĨZĞŚĂďŝůŝƚĂƚŝŽŶDĞĚŝĐŝŶĞ.1986;18:165Ͳ72. 10. Caria MA, Tangianu F, Concu A, Crisafulli A, Mameli O. Quantification of Spinning bike performance duringastandard50Ͳminuteclass.:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐ^ĐŝĞŶĐĞƐ.2007;25:421Ͳ9. 11. FransenM,McConnellS,HarmerAR,VanderEschM,SimicM,BennellKL.Exerciseforosteoarthritisof theknee:aCochranesystematicreview.ƌŝƚŝƐŚ:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐDĞĚŝĐŝŶĞ.2015;49:1554Ͳ7. 12. ZhangWNG,MoskowitzRW,AbramsonS,AltmanRD,ArdenNK,BiermaͲZeinstraS,BrandtKD,CroftP, Doherty M, Dougados M, Hochberg M, Hunter DJ, Kwoh K, Lohmander LS, Tugwell P. OARSI recommendations for the management of hip and knee osteoarthritis: part III: Changes in evidence followingsystematiccumulativeupdateofresearchpublishedthroughJanuary2009.KƐƚĞŽĂƌƚŚƌŝƚŝƐĂŶĚ ĂƌƚŝůĂŐĞ.2010;18:476Ͳ99. 13. BellamyN,KirwanJ,BoersM,BrooksP,StrandV,TugwellP,etal.Recommendationsforacoresetof outcomemeasuresforfuturephaseIIIclinicaltrialsinknee,hip,andhandosteoarthritis.Consensus developmentatOMERACTIII.:ŽƵƌŶĂůŽĨZŚĞƵŵĂƚŽůŽŐǇ.1997;24:799Ͳ802. 14. Dobson F, Hinman RS, Roos EM, Abbott JH, Stratford P, Davis AM, et al. OARSI recommended performanceͲbased tests to assess physical function in people diagnosed with hip or knee osteoarthritis.KƐƚĞŽĂƌƚŚƌŝƚŝƐĂŶĚĂƌƚŝůĂŐĞ.2013;21:1042Ͳ52. 15. WallerB,OgonowskaͲSlodownikA,VitorM,LambeckJ,DalyD,KujalaUM,etal.Effectoftherapeutic aquaticexerciseonsymptomsandfunctionassociatedwithlowerlimbosteoarthritis:systematicreview withmetaͲanalysis.WŚǇƐŝĐĂůdŚĞƌĂƉǇ.2014;94:1383Ͳ95. 16. Bartels EM, Juhl CB, Christensen R, Hagen KB, DanneskioldͲSamsoe B, Dagfinrud H, et al. Aquatic exercise for the treatment of knee and hip osteoarthritis. ŽĐŚƌĂŶĞ ĂƚĂďĂƐĞ ^ǇƐƚ ZĞǀ. 2016;3:CD005523. 17. LuM,SuY,ZhangY,ZhangZ,WangW,HeZ,etal.Effectivenessofaquaticexercisefortreatmentof knee osteoarthritis : Systematic review and metaͲanalysis. ĞŝƚƐĐŚƌŝĨƚ Ĩƺƌ ZŚĞƵŵĂƚŽůŽŐŝĞ. 2015;10.1007/s00393Ͳ014Ͳ1559Ͳ9.

149 Chapter6

18. LundH,WeileU,ChristensenR,RostockB,DowneyA,BartelsEM,etal.Arandomizedcontrolledtrialof aquaticandlandͲbasedexerciseinpatientswithkneeosteoarthritis.:ŽƵƌŶĂůŽĨZĞŚĂďŝůŝƚĂƚŝŽŶDĞĚŝĐŝŶĞ. 2008;40:137Ͳ44. 19. StenerͲVictorinE,KruseͲSmidjeC,JungK.ComparisonbetweenelectroͲacupunctureandhydrotherapy, both in combination with patient education and patient education alone, on the symptomatic treatmentofosteoarthritisofthehip.ůŝŶŝĐĂů:ŽƵƌŶĂůŽĨWĂŝŶ.2004;20:179Ͳ85. 20. CochraneT,DaveyRC,MatthesEdwardsSM.RandomisedcontrolledtrialofthecostͲeffectivenessof waterͲbasedtherapyforlowerlimbosteoarthritis.,ĞĂůƚŚdĞĐŚŶŽůŽŐǇƐƐĞƐƐŵĞŶƚ.2005;9:iiiͲiv,ixͲxi,1Ͳ 114. 21. HofstedeSN,VlietVlielandTP,vandenEndeCH,NelissenRG,MarangͲvandeMheenPJ,vanBodegomͲ VosL.VariationinuseofnonͲsurgicaltreatmentsamongosteoarthritispatientsinorthopaedicpractice intheNetherlands.D:KƉĞŶ.2015;5:e009117. 22. Voorn VM, Vermeulen HM, Nelissen RG, Kloppenburg M, Huizinga TW, Leijerzapf NA, et al. An innovativecaremodelcoordinatedbyaphysicaltherapistandnursepractitionerforosteoarthritisof thehipandkneeinspecialistcare:aprospectivestudy.ZŚĞƵŵĂƚŽůŽŐǇ/ŶƚĞƌŶĂƚŝŽŶĂů.2013;33:1821Ͳ8. 23. Smink AJ, van den Ende CH, Vliet Vlieland TP, Swierstra BA, Kortland JH, Bijlsma JW, et al. Beating osteoARThritis: Development of a stepped care strategy to optimize utilization and timing of nonͲ surgical treatment modalities for patients with hip or knee osteoarthritis. ůŝŶŝĐĂů ZŚĞƵŵĂƚŽůŽŐǇ. 2011;30:1623Ͳ9. 24. FoleyA,HalbertJ,HewittT,CrottyM.Doeshydrotherapyimprovestrengthandphysicalfunctionin patientswithosteoarthritisͲͲarandomisedcontrolledtrialcomparingagymbasedandahydrotherapy basedstrengtheningprogramme.ŶŶĂůƐŽĨƚŚĞZŚĞƵŵĂƚŝĐŝƐĞĂƐĞƐ.2003;62:1162Ͳ7. 25. FransenM,NairnL,WinstanleyJ,LamP,EdmondsJ.Physicalactivityforosteoarthritismanagement:a randomized controlled clinical trial evaluating hydrotherapy or Tai Chi classes. ƌƚŚƌŝƚŝƐ ĂŶĚ ZŚĞƵŵĂƚŝƐŵ.2007;57:407Ͳ14. 26. HinmanRS,HeywoodSE,DayAR.Aquaticphysicaltherapyforhipandkneeosteoarthritis:resultsofa singleͲblindrandomizedcontrolledtrial.WŚǇƐŝĐĂůdŚĞƌĂƉǇ.2007;87:32Ͳ43. 27. Fiolet H, Huijnen B, Metsemakers J, van Hoef L, Schulpen G, Wesseling G et al. Substitutie in de ketenzorg. WƌĂŬƚŝũŬ. 2014:9 [cited 2018 Feb 18]. Available from: http://www.heritage.azm.nl/ afbeeldingen/praktijk/praktijk2014_2.pdf 28. Barten DͲJJA, Swinkels lCS, Dorsman SA, Dekker J, Veenhof C, de Bakker DH. Treatment of hip/knee osteoarthritis in Dutch general practice and physical therapy practice: an observational study. D &ĂŵŝůǇWƌĂĐƚŝĐĞ.2015;16:75. 29. Treweek S, Pitkethly M, Cook J, Kjeldstrom M, Taskila T, Johansen M, et al. Strategies to improve recruitment to randomised controlled trials. ŽĐŚƌĂŶĞ ĂƚĂďĂƐĞ ^ǇƐƚ ZĞǀ. 2010;10.1002/14651858.MR000013.pub5:Mr000013. 30. Courneya KS, Forbes CC, Trinh L, Sellar CM, Friedenreich CM, Reiman T. Patient satisfaction with participationinarandomizedexercisetrial:effectsofrandomizationandausualcareposttrialexercise program.ůŝŶŝĐĂůdƌŝĂůƐ;>ŽŶĚŽŶ͕ŶŐůĂŶĚͿ.2013;10:959Ͳ66. 31. deRooijM,vanderLeedenM,CheungJ,vanderEschM,HakkinenA,HaverkampD,etal.Efficacyof Tailored Exercise Therapy on Physical Functioning in Patients With Knee Osteoarthritis and Comorbidity:ARandomizedControlledTrial.ƌƚŚƌŝƚŝƐĂƌĞΘZĞƐĞĂƌĐŚ.2017;69:807Ͳ16. 32. BrandCA,AmatyaB,GordonB,TostiT,GorelikA.Redesigningcareforchronicconditions:improving hospitalͲbasedambulatorycareforpeoplewithosteoarthritisofthehipandknee./ŶƚĞƌŶĂůDĞĚŝĐŝŶĞ :ŽƵƌŶĂů.2010;40:427Ͳ36. 33. PistersMFVC,vanDijkGM,HeymansMW,TwiskJW,DekkerJ.Thecourseoflimitationsinactivities over 5 years in patients with knee and hip osteoarthritis with moderate functional limitations: risk factorsforfuturefunctionaldecline.KƐƚĞŽĂƌƚŚƌŝƚŝƐĂŶĚĂƌƚŝůĂŐĞ.2012;20:503Ͳ10. 34. SeltenEM,VriezekolkJE,GeenenR,vanderLaanWH,vanderMeulenͲDillingRG,NijhofMW,etal. Reasons for Treatment Choices in Knee and Hip Osteoarthritis: A Qualitative Study. ƌƚŚƌŝƚŝƐ ĂƌĞ Θ ZĞƐĞĂƌĐŚ.2016;68:1260Ͳ7.

150 Generaldiscussion

35. Fernandes L, Hagen KB, Bijlsma JW, Andreassen O, Christensen P, Conaghan PG, et al. EULAR recommendations for the nonͲpharmacological core management of hip and knee osteoarthritis. ŶŶĂůƐŽĨƚŚĞZŚĞƵŵĂƚŝĐŝƐĞĂƐĞƐ.2013;72:1125Ͳ35. 36. FreedlandKE,MohrDC,DavidsonKW,SchwartzJE.Usualandunusualcare:existingpracticecontrol groups in randomized controlled trials of behavioral interventions. WƐǇĐŚŽƐŽŵĂƚŝĐ DĞĚŝĐŝŶĞ. 2011;73:323Ͳ35. 37. KökeAJA,vandenEndeCHM,JansenMJ,SteultjensMPM,CV.Clinicalpracticeguidelineforaphysical activity interventions for patients with osteoarthritis [KNGFͲstandaard Beweeginterventie artrose]. Amersfoort:KoninklijkNederlandsGenootschapvoorFysiotherapie(KNGF);2008. 38. SminkAJ,DekkerJ,VlietVlielandTP,SwierstraBA,KortlandJH,BijlsmaJW,etal.Healthcareuseof patients with osteoarthritis of the hip or knee after implementation of a steppedͲcare strategy: an observationalstudy.ƌƚŚƌŝƚŝƐĂƌĞΘZĞƐĞĂƌĐŚ.2014;66:817Ͳ27. 39. vanRijn MJ.Kamerbriefover BasispakketZvwper2018. MinisterievanVolksgezondheid,Welzijnen Sport.DenHaag.2017. 40. ThorstenssonCA,RoosEM,PeterssonIF,ArvidssonB.HowdomiddleͲagedpatientsconceiveexercise asaformoftreatmentforkneeosteoarthritis?ŝƐĂďŝůŝƚǇĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.2006;28:51Ͳ9. 41. SmithTO,PurdyR,ListerS,SalterC,FleetcroftR,ConaghanPG.Attitudesofpeoplewithosteoarthritis towards their conservative management: a systematic review and metaͲethnography. ZŚĞƵŵĂƚŽůŽŐǇ /ŶƚĞƌŶĂƚŝŽŶĂů.2014;34:299Ͳ313. 42. CampbellR,EvansM,TuckerM,QuiltyB,DieppeP,DonovanJL.Whydon'tpatientsdotheirexercises? UnderstandingnonͲcompliancewithphysiotherapyinpatientswithosteoarthritisoftheknee.:ŽƵƌŶĂů 6 ŽĨƉŝĚĞŵŝŽůŽŐǇĂŶĚŽŵŵƵŶŝƚǇ,ĞĂůƚŚ.2001;55:132Ͳ8. 43. HendryM,WilliamsNH,MarklandD,WilkinsonC,MaddisonP.Whyshouldweexercisewhenourknees hurt? A qualitative study of primary care patients with osteoarthritis of the knee. &ĂŵŝůǇ WƌĂĐƚŝĐĞ. 2006;23:558Ͳ67. 44. Smelt AF, van der Weele GM, Blom JW, Gussekloo J, Assendelft WJ. How usual is usual care in pragmaticinterventionstudiesinprimarycare?Anoverviewofrecenttrials.ƌŝƚŝƐŚ:ŽƵƌŶĂůŽĨ'ĞŶĞƌĂů WƌĂĐƚŝĐĞ.2010;60:e305Ͳ18. 45. CunninghamJA,KypriK,McCambridgeJ.Exploratoryrandomizedcontrolledtrialevaluatingtheimpact ofawaitinglistcontroldesign.DDĞĚŝĐĂůZĞƐĞĂƌĐŚDĞƚŚŽĚŽůŽŐǇ.2013;13:150. 46. Holden MA, Nicholls EE, Hay EM, Foster NE. Physical Therapists’ Use of Therapeutic Exercise for Patients With Clinical Knee Osteoarthritis in the United Kingdom: In Line With Current Recommendations?WŚǇƐŝĐĂůdŚĞƌĂƉǇ.2008;88:1109Ͳ21. 47. HoldenMA,NichollsEE,YoungJ,HayEM,FosterNE.UKͲbasedphysicaltherapists'attitudesandbeliefs regarding exercise and knee osteoarthritis: findings from a mixedͲmethods study. ƌƚŚƌŝƚŝƐ ĂŶĚ ZŚĞƵŵĂƚŝƐŵ.2009;61:1511Ͳ21. 48. DenningWM,BresselE,DGD.UnderwaterTreadmillExercisesasaPotentialTreatmentforAdultsWith Osteoarthritis./ŶƚĞƌŶĂƚŝŽŶĂů:ŽƵƌŶĂůŽĨƋƵĂƚŝĐZĞƐĞĂƌĐŚĂŶĚĚƵĐĂƚŝŽŶ.2010;4:70Ͳ80. 49. BresselE,WingJE,MillerAI,DolnyDG.HighͲintensityintervaltrainingonanaquatictreadmillinadults withosteoarthritis:effectonpain,balance,function,andmobility.:ŽƵƌŶĂůŽĨ^ƚƌĞŶŐƚŚĂŶĚŽŶĚŝƚŝŽŶŝŶŐ ZĞƐĞĂƌĐŚ.2014;28:2088Ͳ96. 50. FiskenA,Waters DL,HingWA, SteeleM,KeoghJW. Perceptionandresponsestodifferentforms of aquaͲbasedexerciseamongolderadultswithosteoarthritis./ŶƚĞƌŶĂƚŝŽŶĂů:ŽƵƌŶĂůŽĨƋƵĂƚŝĐZĞƐĞĂƌĐŚ ĂŶĚĚƵĐĂƚŝŽŶ.2014;8:32Ͳ52. 51. HallJ,SwinkelsA,BriddonJ,McCabeCS.Doesaquaticexerciserelievepaininadultswithneurologicor musculoskeletal disease? A systematic review and metaͲanalysis of randomized controlled trials. ƌĐŚŝǀĞƐŽĨWŚǇƐŝĐĂůDĞĚŝĐŝŶĞĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.2008;89:873Ͳ83. 52. WestbyMD.Ahealthprofessional'sguidetoexerciseprescriptionforpeoplewitharthritis:areviewof aerobicfitnessactivities.ƌƚŚƌŝƚŝƐĂŶĚZŚĞƵŵĂƚŝƐŵ.2001;45:501Ͳ11. 53. Juhl C, Christensen R, Roos EM, Zhang W, Lund H. Impact of exercise type and dose on pain and disability in knee osteoarthritis: a systematic review and metaͲregression analysis of randomized controlledtrials.ƌƚŚƌŝƚŝƐΘƌŚĞƵŵĂƚŽůŽŐǇ.2014;66:622Ͳ36.

151 Chapter6

54. RoddyE,ZhangW,DohertyM.Aerobicwalkingorstrengtheningexerciseforosteoarthritisoftheknee? Asystematicreview.ŶŶĂůƐŽĨƚŚĞZŚĞƵŵĂƚŝĐŝƐĞĂƐĞƐ.2005;64:544Ͳ8. 55. AlkatanM,BakerJR,MachinDR,ParkW,AkkariAS,PashaEP,etal.ImprovedFunctionandReduced Pain after Swimming and Cycling Training in Patients with Osteoarthritis. :ŽƵƌŶĂů ŽĨ ZŚĞƵŵĂƚŽůŽŐǇ. 2016;43:666Ͳ72. 56. Brosseau L, Wells GA, Kenny GP, Reid R, Maetzel A, Tugwell P, et al. The implementation of a communityͲbased aerobic walking program for mild to moderate knee osteoarthritis: a knowledge translationrandomizedcontrolledtrial:partII:clinicaloutcomes.DWƵďůŝĐ,ĞĂůƚŚ.2012;12:1073. 57. RossRM,MurthyJN,WollakID,JacksonAS.Thesixminutewalktestaccuratelyestimatesmeanpeak oxygenuptake.DWƵůŵŽŶĂƌǇDĞĚŝĐŝŶĞ.2010;10:31Ͳ. 58. BansiJ,BlochW,GamperU,KesselringJ.TraininginMS:influenceoftwodifferentendurancetraining protocols (aquatic versus overland) on cytokine and neurotrophin concentrations during three week randomizedcontrolledtrial.DƵůƚŝƉůĞ^ĐůĞƌŽƐŝƐ.2012;19:613Ͳ21. 59. Bansi J, Bloch W, Gamper U, Riedel S, Kesselring J. Endurance training in MS: shortͲterm immune responses and their relation to cardiorespiratory fitness, healthͲrelated quality of life, and fatigue. :ŽƵƌŶĂůŽĨEĞƵƌŽůŽŐǇ.2013;260:2993Ͳ3001. 60. BoidinM,LapierreG,PaquetteTanirL,NigamA,JuneauM,GuilbeaultV,etal.Effectofaquaticinterval trainingwithMediterraneandietcounselinginobesepatients:resultsofapreliminarystudy.ŶŶĂůƐŽĨ WŚǇƐŝĐĂůĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶDĞĚŝĐŝŶĞ.2015;58:269Ͳ75. 61. Karnahl B. Vergleichende Untersuchungen von LeistungsͲ und Stoffwechselparametern im ergometrischenTestanLandundimWasser[Dissertation].Potsdam:UniversitätPotsdam;2010. 62. TanakaH,MonahanKD,SealsDR.AgeͲpredictedmaximalheartraterevisited.:ŽƵƌŶĂůŽĨƚŚĞŵĞƌŝĐĂŶ ŽůůĞŐĞŽĨĂƌĚŝŽůŽŐǇ.2001;37:153Ͳ6. 63. BarbosaTM,MarinhoDA,ReisVM,SilvaAJ,BragadaJ.PhysiologicalassessmentofheadͲoutaquatic exercisesinhealthysubjects:aqualitativereview.:ŽƵƌŶĂůŽĨ^ƉŽƌƚƐ^ĐŝĞŶĐĞΘDĞĚŝĐŝŶĞ.2009;8:179Ͳ89. 64. CarvalhoVO,BocchiEA,GuimaraesGV.TheBorgscaleasanimportanttoolofselfͲmonitoringandselfͲ regulationofexerciseprescriptioninheartfailurepatientsduringhydrotherapy.Arandomizedblinded controlledtrial.ŝƌĐƵůĂƚŝŽŶ:ŽƵƌŶĂů.2009;73:1871Ͳ6. 65. ShonoT,FujishimaK,HottaN,OgakiT,UedaT,OtokiK,etal.PhysiologicalresponsesandRPEduring underwater treadmill walking in women of middle and advanced age. :ŽƵƌŶĂů ŽĨ WŚǇƐŝŽůŽŐŝĐĂů ŶƚŚƌŽƉŽůŽŐǇĂŶĚƉƉůŝĞĚ,ƵŵĂŶ^ĐŝĞŶĐĞ.2000;19:195Ͳ200. 66. BrasilRM,PintoS,BorreaniS,BagatiniNC,BarretoAC,SenraAC,etal.CorrelationbetweenBorgscale (6Ͳ20) with a new water cycling scale (Brasil scale). DĞĚŝĐŝŶĞ ĂŶĚ ^ĐŝĞŶĐĞ ŝŶ ^ƉŽƌƚƐ ĂŶĚ ǆĞƌĐŝƐĞ. 2014;46:942. 67. FinkelsteinI,deFigueiredoPA,AlbertonCL,BgeginskiR,SteinR,KruelLF.Cardiorespiratoryresponses duringandafterwaterexerciseinpregnantandnonͲpregnantwomen.ZĞǀŝƐƚĂƌĂƐŝůĞŝƌĂĚĞ'ŝŶĞĐŽůŽŐŝĂ ĞKďƐƚĞƚƌşĐŝĂ͗ZĞǀŝƐƚĂĚĂ&ĞĚĞƌĂĕĆŽƌĂƐŝůĞŝƌĂĚĂƐ^ŽĐŝĞĚĂĚĞƐĚĞ'ŝŶĞĐŽůŽŐŝĂĞKďƐƚĞƚƌşĐŝĂ.2011;33:388Ͳ 94. 68. BreugemSJ,HaverkampD,SiereveltIN,StibbeAB,BlankevoortL,vanDijkCN.Theimportantpredictors ofcyclinguseinthreegroupsofkneepatients.ŝƐĂďŝůŝƚǇĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.2011;33:1925Ͳ9. 69. JohnstonTE.Biomechanicalconsiderationsforcyclinginterventionsinrehabilitation.WŚǇƐŝĐĂůdŚĞƌĂƉǇ. 2007;87:1243Ͳ52. 70. LiebsTR,HerzbergW,RutherW,HaastersJ,RussliesM,HassenpflugJ.Ergometercyclingafterhipor knee replacement surgery: a randomized controlled trial. dŚĞ :ŽƵƌŶĂů ŽĨ ďŽŶĞ ĂŶĚ ũŽŝŶƚ ƐƵƌŐĞƌǇ ŵĞƌŝĐĂŶǀŽůƵŵĞ.2010;92:814Ͳ22. 71. SkouST,RoosEM.GoodLifewith osteoArthritisinDenmark(GLA:D):evidenceͲbasededucationand supervised neuromuscular exercise delivered by certified physiotherapists nationwide. D DƵƐĐƵůŽƐŬĞůĞƚĂůŝƐŽƌĚĞƌƐ.2017;18:72. 72. Villadsen A, Overgaard S, HolsgaardͲLarsen A, Christensen R, Roos EM. Immediate efficacy of neuromuscularexerciseinpatientswithsevereosteoarthritisofthehiporknee:asecondaryanalysis fromarandomizedcontrolledtrial.:ŽƵƌŶĂůŽĨZŚĞƵŵĂƚŽůŽŐǇ.2014;41:1385Ͳ94. 73. BeaudartC,BiverE,BruyereO,CooperC,AlͲDaghriN,ReginsterJY,etal.Qualityoflifeassessmentin musculoͲskeletalhealth.ŐŝŶŐůŝŶŝĐĂůĂŶĚǆƉĞƌŝŵĞŶƚĂůZĞƐĞĂƌĐŚ.2017;10.1007/s40520Ͳ017Ͳ0794Ͳ8.

152 Generaldiscussion

74. TildenHM,ReicherterEA,ReicherterF.Useofanaquaticsprogramforolderadultswithosteoarthritis. Fromclinictothecommunity.dŽƉŝĐƐŝŶ'ĞƌŝĂƚƌŝĐZĞŚĂďŝůŝƚĂƚŝŽŶ.2010;26:128Ͳ39. 75. WallerB,MunukkaM,RantalainenT,LammentaustaE,NieminenMT,KivirantaI,etal.Effectsofhigh intensityresistanceaquatictrainingonbodycompositionandwalkingspeedinwomenwithmildknee osteoarthritis:a4ͲmonthRCTwith12ͲmonthfollowͲup.KƐƚĞŽĂƌƚŚƌŝƚŝƐĂŶĚĂƌƚŝůĂŐĞ.2017;25:1238Ͳ46. 76. DobsonF,BennellKL,FrenchSD,NicolsonPJ,KlaasmanRN,HoldenMA,etal.Barriersandfacilitatorsto exercise participation in people with hip and/or knee osteoarthritis: synthesis of the literature using behaviorchangetheory.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨWŚǇƐŝĐĂůDĞĚŝĐŝŶĞĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.2016;95:372Ͳ89. 77. SchwierJ.DotherightthingsͲTrendsimFelddesSports.ĚƐǀͲ/ŶĨŽƌŵĂƚŝŽŶĞŶ.1998;13:7Ͳ13. 78. EtkinJ.Thehiddencostofpersonalquantification.dŚĞ:ŽƵƌŶĂůŽĨĐŽŶƐƵŵĞƌƌĞƐĞĂƌĐŚ.2016;42:967Ͳ84. 79. Lamprecht M, Stamm H. Vom Avantgardistischen Lebensstil Zur Massenfreizeit: Eine Analyse Des EntwicklungsmustersVonTrendsportarten.^ƉŽƌƚǁŝƐƐĞŶƐĐŚĂĨƚ.1998;28:370–87. 80. WesselingJ,WelsingPM,BiermaͲZeinstraSM,DekkerJ,GorterKJ,KloppenburgM,etal.ImpactofselfͲ reported comorbidity on physical and mental health status in early symptomatic osteoarthritis: the CHECK(CohortHipandCohortKnee)study.ZŚĞƵŵĂƚŽůŽŐǇ;KǆĨŽƌĚ͕ŶŐůĂŶĚͿ.2013;52:180Ͳ8. 81. Munukka M, Waller B, Rantalainen T, Hakkinen A, Nieminen MT, Lammentausta E, et al. Efficacy of progressiveaquaticresistancetrainingfortibiofemoralcartilageinpostmenopausalwomenwithmild kneeosteoarthritis:arandomisedcontrolledtrial.KƐƚĞŽĂƌƚŚƌŝƚŝƐĂŶĚĂƌƚŝůĂŐĞ.2016;24:1708Ͳ17. 82. TorresͲRondaL,DelAlcazarXS.ThePropertiesofWaterandtheirApplicationsforTraining.:ŽƵƌŶĂůŽĨ ŚƵŵĂŶŬŝŶĞƚŝĐƐ.2014;44:237Ͳ48. 83. Cadmus L, Patrick MB, Maciejewski ML, Topolski T, Belza B, Patrick DL. CommunityͲbased aquatic 6 exerciseandqualityoflifeinpersonswithosteoarthritis.DĞĚŝĐŝŶĞĂŶĚ^ĐŝĞŶĐĞŝŶ^ƉŽƌƚƐĂŶĚǆĞƌĐŝƐĞ. 2010;42:8Ͳ15. 84. SuomiR,CollierD.Effectsofarthritisexerciseprogramsonfunctionalfitnessandperceivedactivitiesof daily living measures in older adults with arthritis. ƌĐŚŝǀĞƐ ŽĨ WŚǇƐŝĐĂů DĞĚŝĐŝŶĞ ĂŶĚ ZĞŚĂďŝůŝƚĂƚŝŽŶ. 2003;84:1589Ͳ94. 85. BelzaB,TopolskiT,KinneS,PatrickDL,RamseySD.Doesadherencemakeadifference?Resultsfroma communityͲbasedaquaticexerciseprogram.EƵƌƐŝŶŐZĞƐĞĂƌĐŚ.2002;51:285Ͳ91. 86. Patrick DL, Ramsey SD, Spencer AC, Kinne S, Belza B, Topolski TD. Economic evaluation of aquatic exerciseforpersonswithosteoarthritis.DĞĚŝĐĂůĂƌĞ.2001;39:413Ͳ24. 87. BarkerAL,TalevskiJ,MorelloRT,NolanGA,DeSilvaRD,BriggsAM.JumpingintothedeepͲend:results fromapilotimpactevaluationofacommunityͲbasedaquaticexerciseprogram.ůŝŶŝĐĂůZŚĞƵŵĂƚŽůŽŐǇ. 2016;35:1593Ͳ601. 88. FransenM,McConnellS,HarmerAR,VanderEschM,SimicM,BennellKL.Exerciseforosteoarthritisof theknee.ŽĐŚƌĂŶĞĂƚĂďĂƐĞŽĨ^ǇƐƚĞŵĂƚŝĐZĞǀŝĞǁƐ.2015;10.1002/14651858.CD004376.pub3. 89. UlatkowskiM.UnterwasserfahrradversusherkömmlicheRehabilitationͲEineretrospektiveStudiemit und ohne Unterwasserfahrrad an implantierten Kniegelenksprothesen [Dissertation]. Heidelberg: HeidelbergUniversity;2009. 90. von Kathen M. ProspektivͲrandomisierte Vergleichsstudie zur Rehabilitation vorderer Kreunzbandplastiken zwischen koventioneller Therapie und Unterwasserfahrrad [Dissertation]. Bochum:RuhrͲUniversität;1999. 91. Damm P, Dymke J, Bender A, Duda G, Bergmann G. In vivo hip joint loads and pedal forces during ergometercycling.:ŽƵƌŶĂůŽĨŝŽŵĞĐŚĂŶŝĐƐ.2017;60:197Ͳ202. 92. RobertsHM,MooreJP,GriffithͲMcGeeverCL,FortesMB,ThomJM.Theeffectofvigorousrunningand cycling on serum COMP, lubricin, and femoral cartilage thickness: a pilot study. ƵƌŽƉĞĂŶ :ŽƵƌŶĂů ŽĨ ƉƉůŝĞĚWŚǇƐŝŽůŽŐǇ.2016;116:1467Ͳ77. 93. MilaresLP,AssisL,SiqueiraA,ClaudinoV,DomingosH,AlmeidaT,etal.Effectivenessofanaquatic exercise program and lowͲlevel laser therapy on articular cartilage in an experimental model of osteoarthritisinrats.ŽŶŶĞĐƚŝǀĞdŝƐƐƵĞZĞƐĞĂƌĐŚ.2016;57:398Ͳ407. 94. NurH,SertkayaBS,TuncerT.Determinantsofphysicalfunctioninginwomenwithkneeosteoarthritis. ŐŝŶŐůŝŶŝĐĂůĂŶĚǆƉĞƌŝŵĞŶƚĂůZĞƐĞĂƌĐŚ.2017;10.1007/s40520Ͳ017Ͳ0784Ͳx. 95. vanderEschM,HollaJF,vanderLeedenM,KnolDL,LemsWF,RoordaLD,etal.Decreaseofmuscle strength is associated with increase of activity limitations in early knee osteoarthritis: 3Ͳyear results

153 Chapter6

from the cohort hip and cohort knee study. ƌĐŚŝǀĞƐ ŽĨ WŚǇƐŝĐĂů DĞĚŝĐŝŶĞ ĂŶĚ ZĞŚĂďŝůŝƚĂƚŝŽŶ. 2014;95:1962Ͳ8. 96. VanZantRS,BouillonLE.Strengthcycletraining:effectsonmuscularstrengthandaerobicconditioning. :ŽƵƌŶĂůŽĨ^ƚƌĞŶŐƚŚĂŶĚŽŶĚŝƚŝŽŶŝŶŐZĞƐĞĂƌĐŚ.2007;21:178Ͳ82. 97. Bastick AN, Runhaar J, Belo JN, BiermaͲZeinstra SM. Prognostic factors for progression of clinical osteoarthritisoftheknee:asystematicreviewofobservationalstudies.ƌƚŚƌŝƚŝƐZĞƐĞĂƌĐŚΘdŚĞƌĂƉǇ. 2015;doi:10.1186/s13075Ͳ015Ͳ0670Ͳx. 98. GarzonM,GaydaM,GarzonL,JuneauM,NigamA,LeoneM,etal.Biomechanicalanalysistodetermine the external power output on an immersible ergocycle. ƵƌŽƉĞĂŶ :ŽƵƌŶĂů ŽĨ ^ƉŽƌƚ ^ĐŝĞŶĐĞ. 2015;15:271Ͳ8. 99. Walters SJ, Bonacho dos Anjos HenriquesͲCadby I, Bortolami O, Flight L, Hind D, Jacques RM, et al. Recruitmentandretentionofparticipantsinrandomisedcontrolledtrials:areviewoftrialsfundedand published by the United Kingdom Health Technology Assessment Programme. D: KƉĞŶ. 2017;7:e015276. 100. GibsonAJ,ShieldsN.EffectsofAquaticTherapyandLandͲBasedTherapyversusLandͲBasedTherapy AloneonRangeofMotion,Edema,andFunctionafterHiporKneeReplacement:ASystematicReview andMetaͲanalysis.WŚǇƐŝŽƚŚĞƌĂƉǇĂŶĂĚĂ.2015;67:133Ͳ41. 101. Gusi N, TomasͲCarus P. CostͲutility of an 8Ͳmonth aquatic training for women with fibromyalgia: a randomizedcontrolledtrial.ƌƚŚƌŝƚŝƐZĞƐĞĂƌĐŚΘdŚĞƌĂƉǇ.2008;10:R24ͲR.

154 Chapter6 from the cohort hip and cohort knee study. ƌĐŚŝǀĞƐ ŽĨ WŚǇƐŝĐĂů DĞĚŝĐŝŶĞ ĂŶĚ ZĞŚĂďŝůŝƚĂƚŝŽŶ. 2014;95:1962Ͳ8. 96. VanZantRS,BouillonLE.Strengthcycletraining:effectsonmuscularstrengthandaerobicconditioning. :ŽƵƌŶĂůŽĨ^ƚƌĞŶŐƚŚĂŶĚŽŶĚŝƚŝŽŶŝŶŐZĞƐĞĂƌĐŚ.2007;21:178Ͳ82. 97. Bastick AN, Runhaar J, Belo JN, BiermaͲZeinstra SM. Prognostic factors for progression of clinical osteoarthritisoftheknee:asystematicreviewofobservationalstudies.ƌƚŚƌŝƚŝƐZĞƐĞĂƌĐŚΘdŚĞƌĂƉǇ. 2015;doi:10.1186/s13075Ͳ015Ͳ0670Ͳx. 98. GarzonM,GaydaM,GarzonL,JuneauM,NigamA,LeoneM,etal.Biomechanicalanalysistodetermine the external power output on an immersible ergocycle. ƵƌŽƉĞĂŶ :ŽƵƌŶĂů ŽĨ ^ƉŽƌƚ ^ĐŝĞŶĐĞ. 2015;15:271Ͳ8. 99. Walters SJ, Bonacho dos Anjos HenriquesͲCadby I, Bortolami O, Flight L, Hind D, Jacques RM, et al. Recruitmentandretentionofparticipantsinrandomisedcontrolledtrials:areviewoftrialsfundedand published by the United Kingdom Health Technology Assessment Programme. D: KƉĞŶ. 2017;7:e015276. 100. GibsonAJ,ShieldsN.EffectsofAquaticTherapyandLandͲBasedTherapyversusLandͲBasedTherapy AloneonRangeofMotion,Edema,andFunctionafterHiporKneeReplacement:ASystematicReview andMetaͲanalysis.WŚǇƐŝŽƚŚĞƌĂƉǇĂŶĂĚĂ.2015;67:133Ͳ41. 101. Gusi N, TomasͲCarus P. CostͲutility of an 8Ͳmonth aquatic training for women with fibromyalgia: a randomizedcontrolledtrial.ƌƚŚƌŝƚŝƐZĞƐĞĂƌĐŚΘdŚĞƌĂƉǇ.2008;10:R24ͲR. Summary

154

156 Summary

Cycling exercises on a stationary bike in a (heated) pool, called aquatic cycling, may prove to be useful as a therapeutic tool for individuals with knee osteoarthritis to reduce knee pain and improve physical functioning. However, this therapeutic approach for patients with mild to moderate knee osteoarthritis has not been investigated. Therefore, this dissertation aimed to develop and evaluate an aquatic cyclingprogrammethatistailoredtotheneedsofpatientswithkneeosteoarthritis.

Chapter1 This section provides an introduction to the central topics of this thesis: knee osteoarthritis and aquatic cycling. <ŶĞĞ ŽƐƚĞŽĂƌƚŚƌŝƚŝƐ is a chronic joint disease that burdensindividualswithkneepainanddisability.Bothincreasingageandobesityare the primary contributors to the rise in the prevalence of osteoarthritis. By 2020 osteoarthritis is expected to rank among the top five of the most disabling diseases worldwide.Manyoftheriskfactors(e.g.obesity,muscleweakness)andsymptomsof knee osteoarthritis can be lessened and improved by performing regular physical exercise. For individuals who have limited tolerance for weightͲbearing activities, waterͲbased exercise is frequently recommended. Aquatic immersion initiates physiologicalchangesthatarebeneficialinthemanagementofkneeosteoarthritis.For example,thebuoyancyofthewateroffloadsjoints,hydrostaticpressuresresultina massagingeffectthatreducespainsensationandwarmwatertemperaturesupportsa feeling of relaxation. More recently, aquatic cycling has emerged as a trendy fitness activity. ƋƵĂƚŝĐ ĐǇĐůŝŶŐ is a headͲout aquatic exercise activity performed on a stationarybikeinaswimmingpool.Participantsareimmersedinwaterlevelsrisingto betweennavelandchestheightwheretheyperformcyclinginseatedaswellasoutͲofͲ saddlepositions.Additionalexercisemovementscanbecombinedwithtypicalcycling sessionssuchasupperbodyandarmexercisesperformedagainstwaterresistance.

Chapter2 Thesecondchapterdescribesafeasibilitystudyofanaquaticexerciseprogrammefor patientswithkneeosteoarthritis.Tenpatientswithkneeosteoarthritisparticipatedin eightweeklysessionsconsistingofaquaticexercisesandtenminutesofaquaticcycling. The training programme was deemedsafe and feasible. In addition tohigh levels of attendanceandpositivepatientfeedback,individualsalsostatedthattheirkneepain improvedimmediatelyfollowingexercise.Participantsalsodemonstratedtheabilityto readilyexecutemostexerciseswhileshowingprogression.However,asopposedtothe seatedposition,aquaticcyclinginanoutͲofͲtheͲsaddlepositionwastoodemandingfor most participants. This finding suggests that more training is needed before outͲofͲ saddlepositionexerciseissafeandeffective.Feedbackfromfocusgroupssuggested high levels of patient satisfaction with the selection of exercises and that aquatic

157 cyclingtimecouldbeextendedbeyondtenminutes.Althoughseatedaquaticcycling was perceived as more tiring than landͲbased cycling, actual movements were perceivedaseasierandsmootherwhenperformedinwaterthanonland.However, participantsalsostatedthatseatedcyclingforanextendedperiodoftimemaybecome monotonous.Evaluationofoperationalaspectsofthisprogrammeshowedthataquatic cyclingcouldbereadilyincorporatedintocircuittraining.Seatedcyclingwaseasyto learnandonlyafewinstructionswereneeded,whichcreatedmoretimetosupervise patientsatotherworkstations.

Chapter3 This section presents a detailed description of a 12Ͳweek aquatic cycling exercise programme as well as the design of a randomised controlled trial to evaluate the efficacy of this programme in comparison with usual care for patients with knee osteoarthritis. The aquatic cycling programme consisted of twiceͲweekly 45Ͳminute sessionsconducted in a therapy pool and delivered to small groups of up to four patients. Each session combined seated cycling, outͲofͲsaddle position cycling, leg exercises and upper body exercises. Because there is a paucity of aquatic cycle research and no established guidelines specific to this therapeutic approach for patients with knee osteoarthritis exist, general exercise guidelines for osteoarthritis and evidence taken from landͲbased cycling programmes provided guidance with regardtoexerciseintensityandtrainingsetͲup.Thetrainingprogrammewasdesigned for patients with mild to moderate knee osteoarthritis. Inclusion criteria for participation in the training programme and randomised controlled trial were: knee pain between four and seven on a 10Ͳpoint numeric pain rating scale, a Kellgren/Lawrencescorebetweenonetothree(maximumscore:four)andindication forphysicaltherapyincombinationwithimpairmentsduetoosteoarthritis.Potential participantswithanycontraͲindicationforaquaticexercisetherapysuchassevereand unstablecardiorespiratorycoͲmorbidities,openwoundsorpatientsonawaitinglistfor totalkneesurgerywereexcluded.Moreover,allpotentialparticipantswithconditions that could limit the safe participation or bias symptoms of knee osteoarthritis (e.g. corticosteroid injection and/or hyaluronic acid injection in the last three resp. six month) were excluded. Primary outcome measurements included knee pain and physical function assessed with the Knee Injury and Osteoarthritis Outcome Score (KOOS). Secondary outcome measurements included KOOS subscales for symptoms, sport and diseaseͲspecific quality of life, patientͲreports on general disease severity andqualityoflife,physicalactivity,selfͲefficacy,andkinesiophobia.PerformanceͲbased outcomeswerethetimed“upandgo”test,thesixͲminutewalktestandlowerlimb muscle strength. Measurements were performed at baseline, 12Ͳweeks postͲ intervention,and24ͲweeksfollowͲup.

158 Summary

Chapter4 Chapter four provides a scoping review of the current state of the literature on the topic of aquatic cycling. SixtyͲthree publications were identified with this review providing a summary of programme setͲups of aquatic interventions as well as potential comparisons, core outcomes and involved participants. Detailed review of intervention parameters revealed that there is heterogeneity regarding the use and executionofaquaticcycling.Numerousvariableshavebeentestedrelatedtoaquatic cycling(e.g.typeofaquaticbike,watertemperatureorimmersionlevel,etc.).Although the experimental conditions differed notably across the studies, several shared characteristicscouldbeidentified.Mostresearchfocusedonthecomparisonbetween aquaticandlandcyclingorinvestigatedtheeffectsofdifferentexerciseconditions(e.g. watertemperature)duringaquaticcyclinginhealthy(mostlymale)participantsusinga crossͲover study design. In many of the studies cardiovascular parameters were investigated with results suggesting that cardiac demand of aquatic cycling is not differentcomparedwithlandͲbasedcycling.Outofallofthereviewedstudies,onlysix studies evaluated the effect of aquatic cycling interventions. These studies demonstratedthatanaquaticcyclingtrainingprogrammeisequallyeffectiveaslandͲ based cycling to improve cardiovascular fitness in healthy participants as well as in obeseparticipantsandpatientswithmultiplesclerosis.

Chapter5 Chapter five describes the results of the twoͲarm singleͲblind, parallelͲgroup, randomised controlled trial studying the effects of a 12Ͳweek aquatic cycling programmeonkneepainandphysicalfunctioninpatientswithmildtomoderateknee osteoarthritis.Effectsoftheaquaticcyclingprogrammewerecomparedwithapatient group receiving the usual care of the Early Osteoarthritis Outpatient Clinic of the Maastricht University Medical Centre+. Following the conclusion of the study trial period,participantsrandomisedtotheusualcaregroupwereinvitedtoattendtwelve weekly aquatic cycling sessions in a local indoor community pool. Enrolled into this study were 111 patients. However, nine patients withdrew their consent prior to completing baseline testing, whereas 19 patients were lost to followͲup. Since most group differences remained stable from postͲintervention to 24Ͳweeks followͲup, a singleestimateforthegroupdifferencesatbothtimepointscouldbeused.Significant improvements occurred for selfͲreported knee pain (B=8.16, p=0.014) and physical function (B=7.16, p=0.027) in the aquatic cycling group. Participants of the aquatic cyclinggroupalsoperformedthetimed“upandgo”testfaster(B=Ͳ0.91,p=0.001)and walkedmoremetresduringthesixͲminutewalktest(B=46.75,p=0.002)comparedwith theusualcaregroup.Furthermore,participantsoftheaquaticcyclinggroupreported decreasedfearofmovement(B=Ͳ3.84,p=0.002)andashortͲtermincreaseindiseaseͲ specificqualityoflife(B=13.03,p=0.001)comparedwiththeusualcaregroup.

159

Chapter6 Chaptersixprovidesabroadoverviewofthemainfindingsofthisthesis.Thissection briefly discusses the four essential aspects considered in both the development and evaluation of the aquatic cycling training programme: recruitment and dropouts, comparison of aquatic cycling with usual care, usual care compliance, and the effectiveness of the aquatic cycling exercise programme. Practical considerations on theprosandconsofaquaticcyclingandthoughtsonthefutureaccessibilityofaquatic cycling are described as well. The final comments in this dissertation include ideas regardingtheimportantnextstepsforthislineofresearch.

160

162 Samenvatting

Waterfietsenineenverwarmdzwembadismogelijkeenveelbelovendeoefenactiviteit voormensenmetartroseindeknieompijnteverlichtenenhetfysiekfunctionerente verbeteren.Echter,heteffectvanfietseninhetwater(Engels:aquaticcycling)isnog nooit eerder onderzocht bij patiënten met knieartrose. Het doel van dit proefschrift washetontwikkelenenevaluerenvaneenwaterfietsprogrammavoorpatiëntenmet lichtetotmatigeknieartrose.

Hoofdstuk1 Het eerste hoofdstuk introduceert de centrale onderwerpen van dit proefschrift: knieartrose en waterfietsen. <ŶŝĞĂƌƚƌŽƐĞ is een chronische gewrichtsaandoening die kniepijnenproblemenmethetfysiekefunctionerenveroorzaakt.Devergrijzingende toenamevanovergewichtindebevolkingzorgenervoordatsteedsmeermensenlast krijgenvan(knie)artroseenhierdoorzaldevraagnaargezondheidszorgvoorzieningen toenemen. Verwacht wordt dat artrose in 2020 tot de top vijf van de meest beperkende ziekten wereldwijd zal behoren. Veel risicofactoren (bijvoorbeeld overgewicht of spierzwakte) en symptomen van knieartrose kunnen worden verminderdenverbeterddoorregelmatigelichaamsbeweging.Voorpatiëntendieveel lasthebbenvanoefenenophetdrogekanhetoefeneninwatereenalternatiefzijn.De fysiologische veranderingen die optreden zodra een mens omgeven is van water hebbeneengunstigeffectopdesymptomenvanknieartrose.Hetdrijfvermogenvan het water ontlast de gewrichten, de hydrostatische druk resulteert in een massageͲ effectdathetpijngevoelvermindert,enwarmewatertemperaturenondersteuneneen gevoelvanontspanning.Indeafgelopenjarenheeftwaterfietsenzichontwikkeldtot eenpopulairefitnessactiviteit.tĂƚĞƌĨŝĞƚƐĞŶisfietseninwateropeenergometerdie vaststaat op de grond van een zwembad. Deelnemers zitten tot borsthoogte in het waterenfietseninzittendeenstaandehoudingen.Bovendienwordthetfietsenmet oefeningen voor het bovenlichaam tegen de waterweerstand gecombineerd in een typischewaterfietsͲsessie.

Hoofdstuk2 HettweedehoofdstukbeschrijfteenhaalbaarheidsstudievaneencircuitͲtraininginhet water(inclusiefwaterfietsen)bijpatiëntenmetknieartrose.Tienpatiëntennamendeel aan acht wekelijkse sessies, bestaande uit klassieke waterfitnessͲoefeningen en tien minuten waterfietsen. De circuitͲtraining werd als veilig en haalbaar beoordeeld. De opkomstvandemeestepatiëntenwasgoed,patiëntentolereerdendeoefeningenen rapporteerden een onmiddellijke pijnvermindering in de knie. De deelnemers waren goed in staat om de oefeningen uit te voeren en vertoonden vooruitgang in de moeilijkheidsgraad van de oefeningen. Echter, waterfietsen in een staande houding wasteveeleisendvoordemeestedeelnemers.Dezebevindingsuggereertdatermeer

163 trainingssessies nodig zijn voordat staand fietsen mogelijk wordt. Groepsinterviews toondenaandatdedeelnemerszeertevredenwarenoverdekeuzeenafwisselingvan deoefeningen.Metbetrekkingtothetwaterfietsenverklaardenpatiëntendatzijgraag langerdantienminutenhaddenwillenfietsen.Hoewelzittendfietseninhetwaterals vermoeiender werd ervaren dan fietsen op het land, werd de beweging ook als gemakkelijkerervarenenvoeldehetsoepeleraandandefietsbewegingophetdroge. Dedeelnemersbenadruktenechterdatenkelzittendfietsenvooreenlangereperiode te eentonig kan zijn. De evaluatie van de operationele aspecten toonde aan dat waterfietsen eenvoudig in een circuitͲtraining kan worden opgenomen, omdat er slechtseenpaarinstructiesnodigwarenendaardoormeertijdwasvoorpatiëntendie metandereoefeningeninhetcircuitbezigwaren.

Hoofdstuk3 Dit hoofdstuk geeft een gedetailleerde beschrijving van een 12Ͳweken durende waterfietsͲoefenprogramma en beschrijft de opzet van een gerandomiseerde studie omdeeffectenvanhetoefenprogrammatevergelijkenmetdegebruikelijkezorgvoor patiëntenmetknieartrose.Hetoefenprogrammawerdtweekeerperweekmetsessies van45minutengegevenaankleinschaligegroepenvanmaximaalvierpatiëntenineen therapiebad. In elke sessie werd zittend fietsen met staande fietshoudingen, beenoefeningen en oefeningen voor het bovenlichaam gecombineerd. Vanwege een gebrekaanvoorafgaandonderzoeknaardeeffectenvanwaterfietsͲinterventiesvoor patiënten met knieartrose, was de trainingsintensiteit en de opbouw gebaseerd op algemene oefenrichtlijnen voor artrose en de resultaten van fietsinterventies op het droge.Hetwaterfietsprogrammawerdontwikkeldvoorpatiëntenmetmildetotmatige knieartrose en de volgende inclusiecriteria voor de gerandomiseerde gecontroleerde studiewerdengehandhaafd:kniepijntussenvierenzevenopeennumeriekepijnschaal mettienpunten,eenKellgren/LawrenceͲscoretussenéénendrie(vierisdemaximum score) en een indicatie voor fysiotherapie in verband met de knieartrose. Potentiële deelnemers met een contraͲindicatie voor watertherapie zoals ernstige, onstabiele cardiorespiratoireaandoeningenenopenwondenofpatiëntenopeenwachtlijstvoor eengewrichtͲvervangendeknieoperatiewerdenuitgeslotenvandedeelname.Verder werdenpatiëntenuitgeslotenwaarbijeenveiligedeelnameaaneenoefenprogramma niet gegarandeerd was of waarbij de symptomen van de knieartrose mogelijk vertekend waren (bijvoorbeeld door recente injectie met corticosteroïden en / of hyaluronzuur in de afgelopen drie c.q. zes maanden). De primaire uitkomstmaten waren kniepijn en fysiek functioneren gemeten met de KOOS (Knee Injury and OsteoarthritisOutcomeScore)vragenlijst.SecundaireuitkomstmatenwarendeKOOSͲ vragenlijsten voor symptomen, sport en ziekteͲspecifieke kwaliteit van leven, en patiëntͲgerapporteerdescoresoverdealgemeneernstvandeziekteendealgemene kwaliteit van leven, fysieke activiteit, eigen effectiviteit en beweegangst. Het fysiek

164 Samenvatting functionerenwerdbovendien metde getimede"up and go" testendezes minuten wandeltest geëvalueerd. Verder werd de spierkracht van de onderste ledematen gemeten.

Hoofdstuk4 Hoofdstuk vier beschrijft een literatuuroverzicht waarin de huidige stand van zaken metbetrekkingtothetfietseninhetwaterwerdonderzocht.Drieënzestigpublicaties werdengeïdentificeerd.Hetonderzoekgeefteensamenvattingvandekenmerkenvan destudieszoalsdeopzetvandeinterventie,debelangrijksteresultaten,dekenmerken vandeelnemersendestudiedesigns.Destudiesvertoondenveelvariatieintermenvan externefactorenendeuitvoeringvanwaterfietsen.Zozijnindestudiesbijvoorbeeld verschillendewaterfietsengebruiktenzijnerveelverschillengevondenmetbetrekking tot de omgevingsfactoren zoals waterhoogte of watertemperatuur tijdens de interventies. Hoewel de experimentele omstandigheden in de studies duidelijk verschilden, werden er ook gedeelde kenmerken geïdentificeerd. De meeste studies waren gericht op de vergelijking tussen waterͲ en landfietsen of onderzochten de effecten van verschillende omstandigheden (bijvoorbeeld de watertemperatuur) tijdenswaterfietseningezonde(meestalmannelijke)deelnemersmetbehulpvaneen crossͲoverstudiedesign.Bovendienwerdencardiovasculaireparametersinveelvande studiesonderzocht,waarbijderesultatensuggereerdendatdecardialebelastingvan waterfietsenvergelijkbaarismetfietsenophetdroge.Slechtszesstudiesevalueerden heteffectvaneenwaterfietsͲoefenprogramma.Uitdezeinterventiestudiesbleekdat een waterfietsprogramma net zo effectief is als landͲfietsen om het uithoudingsͲ vermogen te verbeteren bij gezonde deelnemers, maar ook bij mensen met overgewichtenpatiëntenmetmultiplesclerose.

Hoofdstuk5 Hoofdstuk vijf beschrijft de resultaten van een tweearmige gerandomiseerde studie naardewerkzaamheidvaneen12Ͳwekendurendewaterfietsprogrammaopkniepijn en lichamelijk functioneren bij patiënten met lichte tot matige knieartrose. Het waterfietsprogramma werd vergeleken met de gebruikelijke zorg van de polikliniek voorvroegartrosevanhetMaastrichtUniversitairMedischCentrum+.Naheteindevan de trialperiode werden de mensen in de gebruikelijke zorggroep uitgenodigd voor waterfietsͲsessiesgedurende12wekenineenlokaalzwembad.Indezestudiewerden 111 patiënten gerandomiseerd, waarvan negen patiënten hun toestemming vóór de eerstemeetsessieterugtrokkenen19patiëntentijdensdeonderzoeksperiodestopten met deelname aan het onderzoek. Aangezien de meeste groepsverschillen stabiel bleven van de nameting tot aan de 24 weken followͲup, werd één enkele schatting gebruikt voor de groepsverschillen op beide meetmomenten. Significante verschillen

165 werden gevonden voor zelf gerapporteerde knieͲpijn (B=8.16, p=0.014) en fysiek functioneren (B=7.16, p=0.027) in het voordeel van de waterfietsgroep. Bovendien voerden de deelnemers van de waterfietsgroep de getimede "up and go" test significant sneller uit (B=Ͳ0.91, p=0.001) en liepen ze significant meer meters gedurende de zes minuten wandeltest (B=46.75, p=0.002). Verder had de waterfietsgroep een verminderde bewegingsangst (B=Ͳ3.84, p=0.002) en werd een kortdurende verbetering van ziekteͲspecifiekekwaliteit van leven (B=13.03,p=0.001) aangetoondinvergelijkingmetdegebruikelijkezorggroep.

Hoofdstuk6 Dithoofdstukgeefteenoverzichtvandebelangrijkstebevindingenvanditproefschrift, gevolgd door een discussie met betrekking tot vier essentiële aspecten in de ontwikkeling en evaluatie van het waterfietsprogramma: rekrutering en uitval van deelnemers, de vergelijking van waterfietsen met gebruikelijke zorg, gebruik van fysiotherapie in de controle groep, en het waterfietsprogramma. Verder worden praktischeoverwegingenoverdevoorͲennadelenvanwaterfietsenengedachtenover de (toekomstige) toegankelijkheid van waterfietsen beschreven. Ten slotte worden implicatiesvoortoekomstigonderzoekgepresenteerd.

166 werden gevonden voor zelf gerapporteerde knieͲpijn (B=8.16, p=0.014) en fysiek functioneren (B=7.16, p=0.027) in het voordeel van de waterfietsgroep. Bovendien voerden de deelnemers van de waterfietsgroep de getimede "up and go" test significant sneller uit (B=Ͳ0.91, p=0.001) en liepen ze significant meer meters gedurende de zes minuten wandeltest (B=46.75, p=0.002). Verder had de waterfietsgroep een verminderde bewegingsangst (B=Ͳ3.84, p=0.002) en werd een kortdurende verbetering van ziekteͲspecifiekekwaliteit van leven (B=13.03,p=0.001) aangetoondinvergelijkingmetdegebruikelijkezorggroep. Hoofdstuk6 Dithoofdstukgeefteenoverzichtvandebelangrijkstebevindingenvanditproefschrift, gevolgd door een discussie met betrekking tot vier essentiële aspecten in de ontwikkeling en evaluatie van het waterfietsprogramma: rekrutering en uitval van deelnemers, de vergelijking van waterfietsen met gebruikelijke zorg, gebruik van fysiotherapie in de controle groep, en het waterfietsprogramma. Verder worden Zusammenfassung praktischeoverwegingenoverdevoorͲennadelenvanwaterfietsenengedachtenover de (toekomstige) toegankelijkheid van waterfietsen beschreven. Ten slotte worden implicatiesvoortoekomstigonderzoekgepresenteerd.

166

168 Zusammenfassung

EinFahrradtrainingimWasserkönntealstherapeutischesMittelbeiPatientinnenund Patienten mit Kniearthrose eingesetzt werden. Dieser therapeutische Ansatz wurde jedochbishernochnichtfürPatientinnenundPatientenmitleichterbismittelschwerer Kniearthroseerforscht. Zieldieser Dissertation war esdaher, einauf die Bedürfnisse von Kniearthrotikern abgestimmtes Übungsprogramm auf einem aqualen Fahrrad zu entwickelnunddieEffektediesesTrainingszuevaluieren.

Kapitel1 Dieses Kapitel gibt eine Einführung in die zentralen Themen der vorliegenden Dissertation: Kniearthrose und aquales Fahrradtraining. <ŶŝĞĂƌƚŚƌŽƐĞ ist eine chronischeGelenkerkrankung,diesichbeidenmeistenPatientinnenundPatientenmit KnieschmerzenundkörperlichenEinschränkungenäußert.PrägendeMerkmaleunserer modernen Gesellschaft, wie eine steigende Lebenserwartung und die Zunahme von Übergewicht, tragen zum Anstieg der Prävalenz von Arthrose bei. Es wird davon ausgegangen, dass Arthrose im Jahr 2020 zu den fünf meist einschränkenden Krankheiten weltweit zählt. Viele der Risikofaktoren (z. B. Fettleibigkeit, Muskelschwäche usw.) und Symptome der Kniearthrose können durch regelmäßiges Training verbessert werden. Für Personen, die eine eingeschränkte Belastbarkeit für Sportaktivitäten vorweisen, wird häufig ein Training im Wasser empfohlen. Das EintaucheninsWasserlöstphysiologischeVeränderungenaus,diebeimTrainierenmit KniearthrosevonVorteilseinkönnen.DerAuftriebdesWassersentlastetdieGelenke. Zusätzlich führt der hydrostatische Druck einen Massageeffekt herbei, der das Schmerzempfindenreduziert.WeiterhinführteinewarmeWassertemperaturzueinem Gefühl der Entspannung. In jüngster Zeit hat sich das ĂƋƵĂůĞ &ĂŚƌƌĂĚĨĂŚƌĞŶ͕ oftmals auch mit dem englischen Begriff „aqua cycling“ betitelt, zu einem Trend im Fitnessbereich entwickelt. „Aqua cycling“ wird auf einem feststehenden Fahrrad, ähnlich eines Ergometers oder Spinningfahrrads, in einem Schwimmbecken durchgeführt.DieTeilnehmerinnenundTeilnehmersitzenaufdemFahrradundsind bis maximal Brusthöhe im Wasser eingetaucht. Die Übungen bestehen aus einer KombinationvonFahrradfahreninsitzenderundstehenderPositionundOberkörperͲ undArmübungengegendenWasserwiderstand.

Kapitel2 DaszweiteKapitelumfassteineMachbarkeitsstudieeinesaqualenZirkelprogrammsbei Patientinnen und Patienten mit Kniearthrose. Insgesamt zehn Patientinnen und Patienten nahmen an acht wöchentlichen Trainingseinheiten teil, die aus WassergymnastikübungenundzehnMinuten„aquacycling“bestanden.DieMehrheit derPatientinnenundPatientennahmanallenachtEinheitenteil.NachBeendigungder StudiewurdendieTeilnehmendeninFokusgruppenzumTrainingsprogrammbefragt. Das Ergebnis zeigt eine hohe Patientenzufriedenheit mit der Auswahl und

169

Zusammenstellung des Übungsprogramms. Weiterhin gaben die Patientinnen und Patienten,dasssichdieKnieschmerzenunmittelbarnachdemTrainingverbesserten. Die Patientinnen und Patienten konnten die meisten Übungen ohne Probleme ausführenundzeigtenFortschritteinBezugaufdenSchwierigkeitsgradbeifastallen Übungen. Das Radfahren in einer stehenden Position war jedoch für die meisten PatientinnenundPatientenkörperlichzuanspruchsvoll.DiesesErgebnisdeutetdarauf hin, dass ein differenziertes Training erforderlich ist, bevor die Ausübung stehender Positionen sicher und effektiv ist. Obwohl das aquale Fahrradfahren im Sitzen als ermüdenderwaralsdasFahrradfahrenanLand,wurdedieBewegungimWasserauch als leichter und angenehmer empfunden. Die Patientinnen und Patienten äußerten daher auch den Wunsch, die Übungszeit auf dem aqualen Fahrrad zu verlängern. GleichzeitiggabendiePatientinnenundPatientenan,dassRadfahrenimSitzenüber einenlängerenZeitraummonotonwerdenkönnte.DieBewertungderoperationellen AspektedesÜbungsprogrammszeigte,dassdasaqualeFahrradtrainingproblemlosin einaqualesZirkeltrainingintegriertwerdenkann.DasaqualeFahrradfahrenimSitzen war einfachzu erlernenundder Trainerin stand somit mehr Zeit fürdie individuelle BetreuungderTeilnehmendenananderenStationendesZirkeltrainingszurVerfügung.

Kapitel3 DiesesKapitelbeschreibtdieMethodikeinerrandomisiertenkontrolliertenStudiezur Evaluation eines 12Ͳwöchigen Übungsprogramms mit einem aqualen Fahrrad im Vergleich zu einer Kontrollgruppe, welche die üblichen Behandlungsempfehlungen einerFrüharthroseSprechstundeerhielt.DasTrainingmitdemaqualenFahrradwurde zweimalwöchentlichfür45MinuteninKleingruppenmitbiszuvierPatientinnenund Patienten in einem Therapiebad durchgeführt. In jeder Trainingseinheit wurde das Radfahren im Sitzen mit stehenden Positionen oder Beinübungen ergänzt sowie mit Übungen für die Arme und dem Oberköper kombiniert. Aufgrund der mangelnden EvidenzzumFahrradtrainingimWasserundderAnwendungdieserTrainingsformbei Kniearthrotikern, dienten Forschungsergebnisse bezüglich des körperlichen Trainings bei Arthrose und dem Ergometertraining an Land als Richtlinie zur Bestimmung der TrainingsintensitätunddemTrainingsaufbau.EinschlusskriterienfürdieTeilnahmeam Trainingsprogramm und derrandomisierten Studie waren: Knieschmerzenmiteinem Wert zwischen vier und sieben auf einer numerischen 10ͲPunkte Schmerzskala, ein Kellgren / LawrenceͲScore zwischen eins und drei (maximaler Wert: vier) und eine Indikation für eine physiotherapeutische Behandlung der Kniearthrose. Patientinnen und Patienten mit einer Kontraindikation für Bewegungstherapie im Wasser (z. B. kardiorespiratorischeNebenerkrankungenoderoffeneWunden)undPatientinnenund Patienten mit einer Indikation für ein künstliches Kniegelenk wurden von der Teilnahme ausgeschlossen. Darüber hinaus wurden Patientinnen und Patienten ausgeschlossen, deren körperliche Beeinträchtigungen eine sichere Teilnahme am

170 Zusammenfassung

Trainingverwehrten.EbenfallsvonderTeilnahmeausgeschlossenwarenPatientinnen undPatienten mit Erkrankungen (bspw. Rheuma) oderBehandlungen (z. B.Injektion von Kortikosteroiden und / oder Hyaluronsäure in den letzten drei bzw. sechs Monaten),dieSymptomederKniearthroseverschleiern.DieprimärenOutcomesder Studie waren Knieschmerzen und körperliche Funktionsfähigkeit gemessen mit dem KOOSͲFragenbogen (engl. Knee injury and Osteoarthritis Outcome Score). Sekundäre OutcomesumfasstendieKOOSͲSubskalenfürBeschwerden,Problememitsportlicher AktivitätundkrankheitsspezifischeLebensqualität,sowieFragebögenzurallgemeinen KrankheitsbelastungundLebensqualität,körperlicherAktivität,Selbstwirksamkeitund Bewegungsangst. PerformanceͲbasierte Messungen waren der „TimedͲUpͲandͲGo“Ͳ Test, der 6ͲMinutenͲGehtest und die Messung der Muskelkraft der unteren Extremitäten. Alle Messungen wurden zu Beginn der Studie, nach Ablauf der 12Ͳwöchigen Interventionsperiode und 24 Wochen nach Beginn der Studie durchgeführt.

Kapitel4 Das vierte Kapitel bietet eine Übersicht der wissenschaftlichen Literatur zum Thema aquales Fahrradtraining. Eine systematische Literaturrecherche identifizierte 63Publikationen, die anhand der folgenden Merkmale zusammengefasst wurden: Aufbau und Inhalt der Intervention und möglicher Vergleichsinterventionen, Probandenanzahlund–eigenschaften,sowiedieKernergebnissederStudien.Eszeigte sich, dass das aquale Fahrradtraining in den identifizierten Studien sehr heterogen angewendet wurde. Neben Unterschieden im Aufbau und Inhalt des Trainings erschwerteauchdieAusgestaltungvonexternenFaktoren(z.B.DesigndesFahrrads, WassertemperaturoderEintauchtiefederProbanden)dieVergleichbarkeitderStudien. Allerdings konnten auch gemeinsame Merkmale identifiziert werden. Die meisten Forschungsarbeiten konzentrierten sich auf den Vergleich zwischen WasserͲ und Landtraining oder untersuchten die Auswirkungen verschiedener Trainingsbedingungen (z. B. Wassertemperatur) während des Radfahrens im Wasser bei gesunden (meist männlichen) Teilnehmern, unter Verwendung eines CrossͲOverͲ Studiendesigns. In vielen Studien wurden zudem kardiovaskuläre Parameter untersucht.DieErgebnissedeutendaraufhin,dasssichdiekardialeBelastungwährend desaqualenFahrradfahrensnichtvonderBelastunganLandunterscheidet.Nursechs Studien untersuchten die Wirkung von mehrwöchigen Übungsprogrammen. Diese Studien zeigten, dass ein Trainingsprogramm mit dem aqualen Fahrrad genauso effektivistwiedasRadfahrenanLand,umdiekardiovaskuläreFitnessbeigesunden Teilnehmerinnen und Teilnehmern, sowie bei übergewichtigen Menschen und PatientinnenundPatientenmitMultiplerSklerosezuverbessern.

171

Kapitel5 Kapitel fünf beschreibt die Ergebnisse der randomisierten, zweiarmigen Parallelgruppenstudie, in der die Effekte des 12Ͳwöchigen Trainingsprogramms mit dem aqualen Fahrrad auf Knieschmerzen und körperliche Funktionsfähigkeit bei Patientinnen und Patienten mit leichter bis mittelschwerer Kniearthrose untersucht wurden.DasÜbungsprogrammwurdemitderStandardbehandlungderFrüharthroseͲ Sprechstunde des Universitätsklinikums Maastricht verglichen. Nach Abschluss der StudienteilnahmewurdendiePatientinnenundPatientenderKontrollgruppezuzwölf TrainingseinheitenaufdemaqualenFahrradineinlokalesSchwimmbadeingeladen. Insgesamt wurden 111 Patientinnen und Patienten randomisiert. Neun Patientinnen undPatientenzogenihreZustimmungvorderDurchführungderBaselineͲTestszurück. Im Verlauf der Studie beendeten 19 weitere Patientinnen und Patienten ihre Teilnahmevorzeitig. Die Analyse der Datenergab, dass der Unterschied der aqualen Trainingsgruppe mit derKontrollgruppezubeidenZeitpunktenderNachmessung,alsodirektimAnschluss an die Intervention und 24ͲWochen nach dem Start der Studienteilnahme, für die meisten Messwerte vergleichbar war. Somit konnte ein einziger statistischer Schätzwert für beide Zeitpunkte zur Ermittlung der Gruppenunterschiede berechnet werden. Die Analyse zeigte eine signifikante Verbesserungen der aqualen Trainingsgruppe hinsichtlich der Knieschmerzen (B=8,16; p=0,014) und der körperlichenFunktionsfähigkeit(B=7,16;p=0,027).DiePatientinnenundPatientender aqualen Trainingsgruppe absolvierten auch den"Timed Up and Go"ͲTest schneller (B=Ͳ0,91; p=0,001) und erzielten mehr Meter im 6ͲMinutenͲGehtest (B=46,75; p=0,002)alsdieKontrollgruppe.DesWeiterengabendieTeilnehmendenderaqualen Trainingsgruppe im Vergleich zur Kontrollgruppe eine verringerte Bewegungsangst (B=Ͳ3,84;p=0,002)ansowieeinekurzzeitigeVerbesserungderkrankheitsspezifischen Lebensqualität(B=13,03;p=0,001).

Kapitel6 KapitelsechsgibteinenumfassendenÜberblicküberdiewichtigstenErgebnissedieser Dissertation. Außerdem werden vier Aspekte erörtert, die hinsichtlich der InterpretationderStudienergebnisseunbedingtbeachtetwerdensollten:Rekrutierung und Ausfälle, der Vergleich des aqualen Übungsprogramms mit der üblichen Behandlung,sowiedieEinhaltungderBehandlungsempfehlungenderKontrollgruppe. WeiterhinwerdenVorͲundNachteiledesaqualenFahrradtrainingsbeschriebenund Hypothesen zur Beständigkeit dieses aktuellen Trendsports aufgestellt. Abschließend werden Ideen zur weiteren Erforschung dieser Bewegungsart als TherapieͲ und Sportmöglichkeitpräsentiert.

172

Kapitel5 Kapitel fünf beschreibt die Ergebnisse der randomisierten, zweiarmigen Parallelgruppenstudie, in der die Effekte des 12Ͳwöchigen Trainingsprogramms mit dem aqualen Fahrrad auf Knieschmerzen und körperliche Funktionsfähigkeit bei Patientinnen und Patienten mit leichter bis mittelschwerer Kniearthrose untersucht wurden.DasÜbungsprogrammwurdemitderStandardbehandlungderFrüharthroseͲ Sprechstunde des Universitätsklinikums Maastricht verglichen. Nach Abschluss der StudienteilnahmewurdendiePatientinnenundPatientenderKontrollgruppezuzwölf TrainingseinheitenaufdemaqualenFahrradineinlokalesSchwimmbadeingeladen. Insgesamt wurden 111 Patientinnen und Patienten randomisiert. Neun Patientinnen undPatientenzogenihreZustimmungvorderDurchführungderBaselineͲTestszurück. Im Verlauf der Studie beendeten 19 weitere Patientinnen und Patienten ihre Teilnahmevorzeitig. Die Analyse der Datenergab, dass der Unterschied der aqualen Trainingsgruppe mit derKontrollgruppezubeidenZeitpunktenderNachmessung,alsodirektimAnschluss Valorisationaddendum an die Intervention und 24ͲWochen nach dem Start der Studienteilnahme, für die meisten Messwerte vergleichbar war. Somit konnte ein einziger statistischer Schätzwert für beide Zeitpunkte zur Ermittlung der Gruppenunterschiede berechnet werden. Die Analyse zeigte eine signifikante Verbesserungen der aqualen Trainingsgruppe hinsichtlich der Knieschmerzen (B=8,16; p=0,014) und der körperlichenFunktionsfähigkeit(B=7,16;p=0,027).DiePatientinnenundPatientender aqualen Trainingsgruppe absolvierten auch den"Timed Up and Go"ͲTest schneller (B=Ͳ0,91; p=0,001) und erzielten mehr Meter im 6ͲMinutenͲGehtest (B=46,75; p=0,002)alsdieKontrollgruppe.DesWeiterengabendieTeilnehmendenderaqualen Trainingsgruppe im Vergleich zur Kontrollgruppe eine verringerte Bewegungsangst (B=Ͳ3,84;p=0,002)ansowieeinekurzzeitigeVerbesserungderkrankheitsspezifischen Lebensqualität(B=13,03;p=0,001).

172

174 Valorisationaddendum

„Knowledgeisofnovalueunlessyouputitintopractice.“AntonChekov Exercise is a cornerstone in the conservative management of knee osteoarthritis. Physicaltherapistsandscientistsaredriventoalleviatesymptomsandslowdownthe progression of knee osteoarthritis. In daily practice and in research we aim to “ŐĞƚ ƚŚĞŵ ƵƉ͕ ŵŽǀŝŶŐ ĂŶĚ ŽƵƚ ƚŚĞ ĚŽŽƌ͘͟1 Ideally, recommendations and results from physicaltherapyandresearchshouldbeintegratedandaccessibleinthedailylifeof patients.However,thetranslationofresearchoutcomestotherealworldneedssome careful thinking on how the results could add value to our society in the broadest sense.Thisprocessiscalledknowledgevalorisationandisdefinedasthe͞ƉƌŽĐĞƐƐŽĨ ĐƌĞĂƚŝŶŐ ǀĂůƵĞ ĨƌŽŵ ŬŶŽǁůĞĚŐĞ͕ ďǇ ŵĂŬŝŶŐ ŬŶŽǁůĞĚŐĞ ƐƵŝƚĂďůĞ ĂŶĚͬŽƌ ĂǀĂŝůĂďůĞ ĨŽƌ ƐŽĐŝĂů ;ĂŶĚͬŽƌ ĞĐŽŶŽŵŝĐͿ ƵƐĞ ĂŶĚ ďǇ ŵĂŬŝŶŐ ŬŶŽǁůĞĚŐĞ ƐƵŝƚĂďůĞ ĨŽƌƚƌĂŶƐůĂƚŝŽŶŝŶƚŽ ĐŽŵƉĞƚŝƚŝǀĞ ƉƌŽĚƵĐƚƐ͕ ƐĞƌǀŝĐĞƐ͕ ƉƌŽĐĞƐƐĞƐ ĂŶĚ ŶĞǁ ĐŽŵŵĞƌĐŝĂů ĂĐƚŝǀŝƚŝĞƐ͘͟ (adapted definitionbasedontheNationalValorisationCommittee2011:8)͘ In this chapter, we outline the potential relevance of our project for society. Furthermore,wedescribetheactionsthatweretakentodisseminateourknowledge todifferenttargetgroups.

Relevance Osteoarthritisisalreadythemostcommonjointdiseaseinolderadultsworldwide.2In addition,obesityandmuscleimpairments,butalsokneeinjuriesacceleratetheonset ofosteoarthritisatyoungerages.3Therefore,anadequatecaremanagementofknee osteoarthritisisandwillbeofgreatimportance.Theyoungerpatients(±45yearsof age) will suffer for a longer period of time from this chronic disease, which is likely accompaniedbytheneedformedicaltreatment.Asaresult,healthcareutilisationfor knee osteoarthritis will increase further. Moreover, indirect costs related to knee osteoarthritis, such as productivity loss or lost wages, will also increase since these younger patients still belong to the workforce.4 Because care management for knee osteoarthritis should be personalised and acceptable to patients,5 exercise therapy should meet the exercise preferences preferred by this younger generation of osteoarthritispatients.6Aquaticcyclingmightbeanattractiveoptionasitisatrending sportsactivityofferedbylocalswimmingpoolsandmodernsportsstudios.Thelatter facilities clearly meet the aesthetic dimension that is nowadays often related to exercise.7Furthermore,aquaticcyclingiseasytolearn,whichmightbeappealingto peoplewhodonothavemuchexperiencewithexercising.Inaddition,obesepatients might appreciate this type of aquatic activity, since the stable character of the bike preventsthemfromfloatingtoomuch.Furthermore,waterimmersionreducesbody exposure. This dissertation focused on the development and evaluation of a headͲout aquatic cyclingexerciseprogrammeforpatientswithmildtomoderatekneeOA.Sofar,the

175 evidenceontheeffectsofaquaticcyclingforpatientswithkneeosteoarthritisorother targetpopulationsisscarce.Wehopethatourprojectincreasedtheawarenessamong patients, healthcare providers, and aquatic fitness professionals to consider aquatic cyclingasasuitableandeffectiveexercisemodalityforthemselvesortheirpatientsor clients.

Targetgroups

ZĞƐĞĂƌĐŚĞƌƐ ThisPhDprojectcontributedtotheacademicfieldofaquaticexercisetherapy.Toour bestknowledge, our randomised trialwasthe first that evaluatedanaquatic cycling programmeforpatientswithkneeosteoarthritis.Researcherscanuseourresultsasa starting point for further research; possibly about the impact of aquatic cycling on various other outcomes relevant for patients with knee osteoarthritis or its effect among other patient groups. Hopefully, the detailed reporting of the intervention characteristicsofourowntrialandtheresultsfromourliteraturereviewwillstimulate fellow researchers to summarise their interventions comprehensively. This would increase our understanding of the effects of aquatic cycling and increase reproducibilityandcomparabilityofinterventionsandfindings.8 Furthermore, researchers should actively support the translation from scientific evidenceintoinformationthatisusefulandunderstandabletoaquatictherapistsand aquatic fitness instructors. The process of discovering new knowledge and its application in public health and clinical settings can take a long time9 and the translationfromthescientificknowledgeaboutaquaticcyclingintodailypracticewill needsometimeaswell.Inordertobeabletotransferourevidenceintodailypractice, we need to learn more about the effectiveness of aquatic cycling (for example in communitysettings).Sinceaquaticcyclingisanicheinthefieldofaquaticexerciseand aquatic therapy, the (relatively few) individuals and intermediaries with expertise in aquatic cycling, such as (local) health professionals, researchers, aquatic exercise professionals,educators,andopinionleadersinthefieldofaquaticfitnessandtherapy, should collaborate to increase our scientific knowledge about aquatic cycling and to transfertheresultsintodailypractice.10

ƋƵĂƚŝĐĞǆĞƌĐŝƐĞƉƌŽĨĞƐƐŝŽŶĂůƐ Aquaticprofessionals,includingaquaticexerciseinstructorsandaquatictherapists,are the most important target group to bring our scientific knowledge into (clinical) practice.TheinformationfromthisPhDprojectcanprovideguidanceinthesetͲupof aquaticcyclingclassesforpatientswithkneeosteoarthritis.Toincreasetheknowledge transfer, the evidence should be summarised and become available in sources frequentlyconsultedbyaquaticprofessionals.Forinstance,aquaticfitnessandtherapy

176 Valorisationaddendum associations could publish the information on their websites or incorporate the knowledgeintheircoursemanuals.Aquaticfitnessprofessionalshavetheopportunity tofollowatwoͲdaycoursethateducatesthemtobecomeanaquaticcyclinginstructor. The available knowledge from our study and fellow researchers regarding aquatic cyclingforpatientscouldbeimplementedinthesecoursesorcouldserveasabasisfor anadvancedaquaticcyclinginstructorcourse.Offeringtrainingcourses,whichallows potential users to observe and experience aquatic cycling with patients themselves beforeacquiringthistrainingmodalityandapplyingit,willstimulatethetransitionof knowledgeintopractice.Inordertostimulatethisdevelopment,thisdissertationwill besenttoseveralaquaticfitnessassociations.

ƋƵĂƚŝĐĨŝƚŶĞƐƐŝŶĚƵƐƚƌǇĂŶĚƉƌŽǀŝĚĞƌƐ The aquatic fitness industry is another possibility to disseminate research results. Manufacturers of aquatic bikes and organisations that provide aquatic training (e.g. rehabilitationcentres,largerphysiotherapycentres,andlocalswimmingpools)area groupthatmayreceivefinancialincentivesforestablishingaquaticcyclingprogrammes forpatientswithkneeosteoarthritis.Fortheseactivities,adisseminationplan,possibly with a more marketͲoriented approach, to disseminate our findings needs to be developed.11,12However,thedevelopmentofsuchadisseminationplanisbeyondthe scopeofthisdissertation.And,itshouldbenotedthatthereisariskthatourfindings could be overestimated or generalised for such marketing purposes. Currently, the fitness industry already promotes aquatic cycling with various benefits ranging from reducing cellulitis to improving coordination and supporting rehabilitation and treatmentofmusculoskeletaldisorders.Infact,noneofthesebenefitsarecompletely false since aquatic exercise, in general, has a positive influence on the aboveͲ mentioned conditions. However, most aquatic cycling classes, as provided in local public swimming pools, very likely do not fit the needs of the average knee osteoarthritispatient.Forexample,averyfastpedallingtempoorlongͲlastingoutͲofͲ saddleintervalsmightreducepatients’abilityforproperalignmentoftheknee,which canincreasetheriskofkneepain(forexamplebycausingirritationofthecollateral ligamentsorincreasedloadingofthekneejoint).

WĂƚŝĞŶƚƐǁŝƚŚŬŶĞĞŽƐƚĞŽĂƌƚŚƌŝƚŝƐ Thestrongestmotivationofmedicalresearchistoimprovethehealthofpatients,and our research was driven by the motivation to explore aquatic cycling as an exercise opportunityforpatientswithkneeosteoarthritis.Theevaluationofourtrialshoweda positive impact on the impairments of our intervention group. Furthermore, our feasibility study suggested that most patients accepted and enjoyed this type of exercise. Thus, participation might not only have resulted in improvements of symptoms,butmightalsohaveincreasedthepatients’motivationtostayactiveafter

177 the end of the trial. We hypothesize that participants gained a positive impression aboutexercise,whichisafacilitatorforfutureexerciseadherenceandmaintenance. Possibly, patients with mild symptoms or at a higher risk for developing knee osteoarthritis (e.g. overweight persons) might also benefit from aquatic cycling exercises and, depending on their exercise abilities, might be able to join regular aquatic cycling classes provided by sports instructors. However, from anecdotal evidencefromourtrialparticipants,wefeelthatpatientsneedmoreguidanceinthe developmentofanactivelifestyleandexerciseroutine.Ourpatientsreportedtypical barriers to physical activity and exercise like doubting the effectiveness, not being aware of adequate (lowͲcost) exercise opportunities, lacking support from health professionals,orlimitedfinancialcoverageofhealthservicesbyhealthinsurances.13,14 Physical therapists are commonly the primary healthcare practitioners to whom patients with knee osteoarthritis are referred for support in their osteoarthritis management.However,comparabletootherconservativetreatmentoptions,physical therapy is underͲutilised in the management of knee osteoarthritis.15 To increase utilisation of physical therapy in the management of knee osteoarthritis, 12 annual physicaltherapysessionsforosteoarthritisofthekneeorthehiphaverecentlybeen addedtothebasichealthcarepackageinTheNetherlands.Thisisanimportantstep towards the needs of osteoarthritis patients16 and will hopefully increase exercise participationamongthislargegroupofpatients.

Activitiesandproducts This PhD project resulted in various activities to increase awareness and knowledge regardingaquaticcyclingexerciseintheaboveͲmentionedtargetgroups.Resultsofour studiesareorwillbepublishedin(openaccess)journalsforhealthcareandexercise professionals. Furthermore, we presented our results at national and international conferences,nationalsymposia,andinformalhospitalmeetingstophysicaltherapists, aquaticexerciseprofessionals,andorthopaedicsurgeons.Nexttoourscientificoutput, undergraduateprofessionalsofphysicaltherapyandmovementscienceparticipatedin ourprojectduringtheirinternshipsattheMaastrichtUniversityMedicalCentre+oras partoftheirbachelorandmastertheses. Another activity of this project was the collaboration with „Maastricht Sport“, an organisation of the municipality of Maastricht that aims to support inhabitants of Maastrichttoachieveanactivelifestyle.Theorganisationisbasedinthebuildingofthe newlybuiltpublicswimmingpool,the“Geusseltbad”,whichopeneditsdoorsin2014; afewmonthsafterwehadstartedourtrial.Becausethe“Geusseltbad”offeredaquatic cyclingclasses,theresearchteamofthisPhDprojectapproached“MaastrichtSport” forcollaboration.ThisresultedinthesetͲupofanaquaticcyclinggroupforpatients withmusculoskeletalproblemsinthelowerextremitiesand/oralowfitnesslevel.The authorofthisdissertationprovidedthetrainingsessionsforthefirsttwoyears(2014to

178 Valorisationaddendum

2016), and the training was based on elements of the 12Ͳweek programme for the interventiongroup.Participantsofthecontrolgroupofourtrialwereinvitedtothis groupandcouldparticipatein12freetryͲoutsessions.Inaddition,participantsfrom theinterventiongroupandotherinterestedpeoplecouldjointhegroupaswellagainst payment of the regular admission fee. The group size grew from around five participants from our trial population by wordͲofͲmouth recommendation to approximately twelve participants every week. In total, 24 patients from the control group (out of 39 people who completed all three measurements and received the invitation) and ten people from the intervention group (out of 42 who followed the interventionandparticipatedinallthreemeasurements)participatedinthesesessions. After the last patient from our trial had finished the free tryͲout sessions (summer 2016), the “Geusseltbad” continued with offering these classes due tothe high demand of the participants. To date (April 2018), the classes are on the regular scheduleofthe“Geusseltbad”. Thus,theknowledgeofourresearchprojectcouldbeimplementedinthescheduleof aquaticfitnessactivitiesofthe“Geusseltbad”andprovidesanexerciseopportunitytoa population beyond our original trial population. Although the evidence of the effectivenessofthiscommunityapproachislacking,thepositiveresponseofthe‘new’ participantsispromising.However,theimplementationofourprogrammeshouldbe investigatedsystematicallytopreventgeneralisationofourtrialdataandtoserveasan example for other community sports facilities and research projects about the effectivenessofaquaticcycling.Asastartingpointandinspirationforthisfuturework, theauthorofthisdissertationaskedtheteamleaderfrom“MaastrichtSport”toreflect on the cooperation and to share some ideas for future collaborations. “dŚĞ ĐŽůůĂďŽƌĂƚŝŽŶŚĂĚ ĂƉŽƐŝƚŝǀĞ ŝŵƉĂĐƚ ŽŶ ƐĞǀĞƌĂů ĂƐƉĞĐƚƐ͘&ŝƌƐƚ͕ǁĞ ĐŽƵůĚ ĂƚƚƌĂĐƚĂ ŶĞǁ ƚĂƌŐĞƚ ŐƌŽƵƉ ƚŽ ŽƵƌ ĂƋƵĂƚŝĐ ĐǇĐůŝŶŐ ĐůĂƐƐĞƐ͕ ǁŚŝĐŚ ǁĞ ŚĂĚ ŶŽƚ ƚŚŽƵŐŚƚ ŽĨ ďĞĨŽƌĞ͘ /Ŷ ĂĚĚŝƚŝŽŶ͕ŽƵƌĂƋƵĂƚŝĐĨŝƚŶĞƐƐŝŶƐƚƌƵĐƚŽƌƐůĞĂƌŶĞĚŚŽǁƚŽĂĚũƵƐƚƚŚĞƌĞŐƵůĂƌůĞƐƐŽŶƐƚŽƚŚĞ ŶĞĞĚƐ ŽĨ ƉĞŽƉůĞ ǁŝƚŚ ŬŶĞĞ ŽƐƚĞŽĂƌƚŚƌŝƚŝƐ.” Furthermore, he explained that the participantsbenefittedfromapositiveexerciseexperience.“KŶĐĞƉĂƌƚŝĐŝƉĂŶƚƐƌĞĂůŝƐĞ ƚŚĂƚ ĞǆĞƌĐŝƐĞ ŝƐ ĨƵŶ͕ ƚŚĞǇ ŐĞƚ ŽƉĞŶͲŵŝŶĚĞĚ ĂŶĚ ŝŶƚĞƌĞƐƚĞĚ ŝŶ ŽƚŚĞƌ ĂĐƚŝǀŝƚŝĞƐ͘ &Žƌ ĞǆĂŵƉůĞ͕ƐŽŵĞƉĂƌƚŝĐŝƉĂŶƚƐĨƌŽŵƚŚĞĂƋƵĂƚŝĐĐǇĐůŝŶŐƚƌŝĂůƐƚĂƌƚĞĚǁŝƚŚĚƌǇͲůĂŶĚĞǆĞƌĐŝƐĞ ůŝŬĞƚĂŝĐŚŝ.”Futurecollaborationswithresearchersandhealthprofessionalsfromacute careshouldbeinstalledtoevaluatetheeffectivenessofaquaticcyclingandtocreatea guidanceandsupportnetworkhelpingpatientswithkneeosteoarthritistobecomeand to stay physically active. The team leader emphasised that “ƚŚĞ ĐŽůůĂďŽƌĂƚŝŽŶ ǁĂƐ Ă ƐƚĂƌƚŝŶŐƉŽŝŶƚƚŽĞǆƉĂŶĚĂĐŽŵŵƵŶŝƚǇƐƵƉƉŽƌƚƐǇƐƚĞŵĨŽƌŚŽƐƉŝƚĂůͲďĂƐĞĚĂŶĚƉƌŝŵĂƌǇ ĐĂƌĞ ƚŚĞƌĂƉǇ ĂĐƚŝǀŝƚŝĞƐ ŝŶ ƚŚĞ DĂĂƐƚƌŝĐŚƚ ĂƌĞĂ͘͟ For example, in collaboration with medical specialists, “Maastricht Sport” developed the exercise intervention „beweeg bewust“ (Engl. move deliberately; http://www.maastrichtsport.nl/maastrichtͲ sport/maastrichtͲsportͲverbindt/gezondheid/beweegbewust) to help patients moving fromacutecaremanagementthroughrevalidationintoanactivelifestylemaintenance.

179

We encourage the partners of this project to continue their collaboration and to developresearchͲpracticepartnerships17thatsupportthecollaborationacrosssectors and that speed up the translation of evidenceͲbased exercise interventions into practice.Wehopethatmanyaquaticinterventionswillbedeveloped,evaluatedand disseminated by this partnership and by other careͲtoͲlifestyle projects because aquaticexercise/therapysuitstheneedsofmanypatients.ͣtĞĨĞĞůƚŚĞƌĞŝƐĂďƌŝŐŚƚ ĨƵƚƵƌĞĨŽƌƚŚĞĨŝĞůĚ΀ŽĨĂƋƵĂƚŝĐƚŚĞƌĂƉǇ΁ĂƐĂďĞƚƚĞƌƐĐŝĞŶƚŝĨŝĐƵŶĚĞƌƐƚĂŶĚŝŶŐŽĨƚŚĞŵĂŶǇ ƉŚǇƐŝŽůŽŐŝĐ ďĞŶĞĨŝƚƐ ĞŵĞƌŐĞƐ͕ ƚŚĞƌĂƉĞƵƚŝĐ ƚĞĐŚŶŝƋƵĞƐ ĞǀŽůǀĞ͕ ŵŽƌĞ ƚŚĞƌĂƉŝƐƚƐ ĂƌĞ ĞĚƵĐĂƚĞĚŝŶƚŚĞƚĞĐŚŶŝƋƵĞƐĂŶĚƉƵďůŝĐĂŶĚƉƌŽĨĞƐƐŝŽŶĂůĂǁĂƌĞŶĞƐƐŽĨƚŚĞƐĞďƌŽĂĚĞŶƐ ƵƚŝůŝƐĂƚŝŽŶ͘ǁĞƚƚĞƌǁŽƌůĚƐŚŽƵůĚůĞĂĚƚŽĨĂƐƚĞƌĂŶĚŵŽƌĞĞĨĨŝĐŝĞŶƚŚĞĂůƚŚƌĞĐŽǀĞƌǇĨƌŽŵ ŝŶũƵƌǇŽƌƐƵƌŐĞƌǇ͕΀ĂŶĚ΁ŚĞĂůƚŚŵĂŝŶƚĞŶĂŶĐĞŝŶĂŶĂŐŝŶŐƉŽƉƵůĂƚŝŽŶ͙͞;ƌƵĐĞ͘ĞĐŬĞƌ͕ D͖ŶĚƌĞǁ:͘ŽůĞ͕DͿ͘18

180 Valorisationaddendum

REFERENCES

1. McArthurC,ZiebartC,PapaioannouA,CheungAM,LapradeJ,LeeL,etal."Wegetthemup,moving, andoutthedoor.Howdowegetthemtodowhatisrecommended?"Usingbehaviourchangetheory to put exercise evidence into action for rehabilitation professionals. ƌĐŚŝǀĞƐ ŽĨ KƐƚĞŽƉŽƌŽƐŝƐ. 2018; 13:7. 2. Litwic A, Edwards MH, Dennison EM, Cooper C. Epidemiology and burden of osteoarthritis. ƌŝƚŝƐŚ DĞĚŝĐĂůƵůůĞƚŝŶ.2013;105:185Ͳ99. 3. DeshpandeBR,KatzJN,SolomonDH,YelinEH,HunterDJ,MessierSP,etal.Thenumberofpersonswith symptomatickneeosteoarthritisintheUnitedStates:Impactofrace/ethnicity,age,sex,andobesity. ƌƚŚƌŝƚŝƐĂƌĞΘZĞƐĞĂƌĐŚ.2016;68:1743Ͳ50. 4. Gupta S HG, Laporte A, Croxford R, Coyte PC. The economic burden of disabling hip and knee osteoarthritis(OA)fromtheperspectiveofindividualslivingwiththiscondition.ZŚĞƵŵĂƚŽůŽŐǇ;KǆĨŽƌĚ͕ ŶŐůĂŶĚͿ.2005;44:1531Ͳ7. 5. Roos EM, Arden NK. Strategies for the prevention of knee osteoarthritis. EĂƚƵƌĞ ZĞǀŝĞǁƐ͗ ZŚĞƵŵĂƚŽůŽŐǇ.2016;12:92Ͳ101. 6. SalacinskiA,KrohnK,LewisS,HollandM,IrelandK,MarchettiG.Theeffectsofgroupcyclingongait and painͲrelated disability in individuals with mildͲtoͲmoderate knee osteoarthritis: a randomized controlledtrial.:ŽƵƌŶĂůŽĨKƌƚŚŽƉĂĞĚŝĐĂŶĚ^ƉŽƌƚƐWŚǇƐŝĐĂůdŚĞƌĂƉǇ.2012;42:985Ͳ95. 7. SchwierJ.DotherightthingsͲTrendsimFelddesSports.ĚƐǀͲ/ŶĨŽƌŵĂƚŝŽŶĞŶ.1998;13:7Ͳ13. 8. O'Neil J, McEwen D, Del Bel MJ, Jo D, Thevenot O, MacKiddie OS, et al. Assessment of the content reportingfortherapeuticexerciseinterventionsamongexistingrandomizedcontrolledtrialsonknee osteoarthritis.ůŝŶŝĐĂůZĞŚĂďŝůŝƚĂƚŝŽŶ.2018;10.1177/0269215518763714:269215518763714. 9. MorrisZS,WoodingS,GrantJ.Theansweris17years,whatisthequestion:understandingtimelagsin translationalresearch.:ŽƵƌŶĂůŽĨƚŚĞZŽǇĂů^ŽĐŝĞƚǇŽĨDĞĚŝĐŝŶĞ.2011;104:510Ͳ20. 10. Osterling KL, Austin MJ. The dissemination and utilization of research for promoting evidenceͲbased practice.:ŽƵƌŶĂůŽĨǀŝĚĞŶĐĞͲĂƐĞĚ^ŽĐŝĂůtŽƌŬ.2008;5:295Ͳ319. 11. ZiebartC,McArthurC,LeeL,PapaioannouA,LapradeJ,CheungAM,etal."Lefttomyowndevices,I don't know": using theory and patientͲreported barriers to move from physical activity recommendationstopractice.KƐƚĞŽƉŽƌŽƐŝƐ/ŶƚĞƌŶĂƚŝŽŶĂů.2018;10.1007/s00198Ͳ018Ͳ4390Ͳ3. 12. Wilson PM, Petticrew M, Calnan MW, Nazareth I. Disseminating research findings: what should researchersdo?Asystematicscopingreviewofconceptualframeworks./ŵƉůĞŵĞŶƚĂƚŝŽŶ^ĐŝĞŶĐĞ͗/^. 2010;5:91. 13. DobsonF,BennellKL,FrenchSD,NicolsonPJ,KlaasmanRN,HoldenMA,etal.Barriersandfacilitatorsto exercise participation in people with hip and/or knee osteoarthritis: synthesis of the literature using behaviorchangetheory.ŵĞƌŝĐĂŶ:ŽƵƌŶĂůŽĨWŚǇƐŝĐĂůDĞĚŝĐŝŶĞĂŶĚZĞŚĂďŝůŝƚĂƚŝŽŶ.2016;95:372Ͳ89. 14. KanavakiAM,RushtonA,EfstathiouN,AlrushudA,KlockeR,AbhishekA,etal.Barriersandfacilitators of physical activity in knee and hip osteoarthritis: a systematic review of qualitative evidence. D: KƉĞŶ.2017;7:e017042. 15. SeltenEM,VriezekolkJE,GeenenR,vanderLaanWH,vanderMeulenͲDillingRG,NijhofMW,etal. Reasons for Treatment Choices in Knee and Hip Osteoarthritis: A Qualitative Study. ƌƚŚƌŝƚŝƐ ĂƌĞ Θ ZĞƐĞĂƌĐŚ.2016;68:1260Ͳ7. 16. HurleyM,DicksonK,HallettR,GrantR,HauariH,WalshN,etal.Exerciseinterventionsandpatient beliefs for people with hip, knee or hip and knee osteoarthritis: a mixed methods review. ŽĐŚƌĂŶĞ ĂƚĂďĂƐĞ^ǇƐƚZĞǀ.2018;4:Cd010842. 17. HardenSM,JohnsonSB,AlmeidaFA,EstabrooksPA.Improvingphysicalactivityprogramadoptionusing integrated researchͲpractice partnerships: an effectivenessͲimplementation trial. dƌĂŶƐůĂƚŝŽŶĂů ĞŚĂǀŝŽƌĂůDĞĚŝĐŝŶĞ.2017;7:28Ͳ38. 18. Becker BE, Cole AJ. Comprehensive Aquatic Therapy. 3rd ed. Pullmann, WA: Washington State UniversityPublishing;2010.p.xiv

181

182

Acknowledgements

182

184 Acknowledgements

Often,aPhDprojectiscomparedtoamarathonrunͲitislongandweary.Tofinishthe raceyouneedtotrainhardandyouneedsupporterswhocarryyouthroughwhenit getsthough.Ithinkthisistrue,particularlybecausemyPhDwasmore‘TourdeFrance’ styleandIamaterriblecyclist.IbrokemyarmduringmountainbikingintheAustralian rainforest,gotalargegrazeinmyfaceaftertryingtoridemybikehandsͲfreeasachild, andpassedoutafteraspinningsession.IguessthisissubstantialevidencethatIcould not have reached the finishing line without ŵĂŶǇ great people who supported me alongtheway.So,grabacoffeeorteaandlookforwardtoyourhooray! First,Iwouldliketothankmysupervisionteam.MypromoterWƌŽĨ͘ƌ͘ZŽďĚĞŝĞ,my daily supervisor ƌ͘ /ůƐĞ DĞƐƚĞƌƐ, and my supervisors WƌŽĨ͘ ƌ͘ dŽŶ >ĞŶƐƐĞŶ, and ƌ͘ WŝĞƚĞƌŵĂŶƐ. The combination of your different backgrounds and networks enabled metosetͲupthetrialandforcedmetolookatthisprojectfromdifferentperspectives. Furthermore,Icouldswingbyforadviceatanytime,whichIappreciatedverymuch.I learned a lot from each of you and I would like to thank you for all the valuable discussions, your constructive feedback and your support and encouragement, especiallyduringtherecruitmentandinterventionperiod.

DearZŽď,Iamverythankfulthatyouheartenedmyenthusiasmforaquaticexercise. YouyoubroughtmeintocontactwithAquaKinetiqs(theaquaticbikemanufacturer) andthankstoyournetworkIwasabletodomyfeasibilitystudyinthetherapypoolof MaastrichtUniversityMedicalCentre+.Iamthankfulforthetrustyouplacedinmeand thatyoubelievedintheopportunitytocontinuetheaquaticcyclingjourneyasaPhD project.Withoutyou,thisspecialbiketourwouldhaveneverhappened. Dear /ůƐĞ, I want to thank you for your constant support and encouragement. You becameaverycommittedsupportridera,whichwasatremendoushelp,especiallyin difficult situations. I appreciated that you found the positive in everysituation. Your optimismalwayscheeredmeupandmotivatedmetogoon.Iamalsoverythankfulfor yourgoodandhelpfuladviceonmypapersandthefinalchaptersofthisdissertation. Youhelpedmetosharpentheviewfortheessentialswhileremindingmetoneverlose sightofthebiggerpicture. DeardŽŶ,Iamthankfulthatyouwerepartofthe‘aquacycling’team.Ivaluedyour rationalperspectiveonmyideasandyourclearcommentsonmypapers.Yourability toputmyideasandchallengesintoasimpleblackandwhiteperspectivemademany aspectslessoverwhelming(alsothankstoyourgreatsenseofhumour),forcedmeto

aInroadbicycleracingasupportrider,alsocalleddomestique,isariderwhoworksforthebenefitofhis teamandleader,ratherthantryingtowintherace.

185 reconsidercertainaspects,andmotivatedmetoworkhardertoconvinceyouofsome ofmyideas. DearWŝĞƚĞƌ,youcomplementedtheteamsowell,althoughthisprojectwasalittlebit outsideofyourdailybusiness.Yourcommentsoftenledtothefinishingtouchofthe papersandthisdissertation.Furthermore,youhelpedalotwiththerecruitmentofthe participantsandyounevergottiredofremindingyourcolleaguesabouttheaquatic cyclingstudy.Thankyouverymuchforyoureffortandsupport. Theterm‘aquaticcycling’impliesthatyouneedmorethanagreatsupervisionteam.I neededswimmingpoolsandassessmentfacilities,andofcourseaquaticbikes.Luckily, I could count on the support of the ĞƉĂƌƚŵĞŶƚ ŽĨ KƌƚŚŽƉĂĞĚŝĐ ^ƵƌŐĞƌǇ and the ĞƉĂƌƚŵĞŶƚ ŽĨ WŚǇƐŝŽƚŚĞƌĂƉǇ at DĂĂƐƚƌŝĐŚƚ hŶŝǀĞƌƐŝƚǇ DĞĚŝĐĂů ĞŶƚƌĞн, ƋƵĂ<ŝŶĞƚŝƋƐ, and DĂĂƐƚƌŝĐŚƚ ^ƉŽƌƚ and the 'ĞƵƐƐĞůƚďĂĚ. Some people I would like to thank in particular. tŝĞůtŝũŶĞŶ,youdidanamazingjobwiththerecruitmentofourparticipants.Thank you so much for explaining the trial over and over again to potential participants! Furthermore,Ivaluedourdiscussionsonconservativeosteoarthritismanagementand yourinputduringthedesignphaseofourtrial. 'ĞŽƌŐZŽŽǆ,Iamverygratefulforyoursupport.Thankstoyourpermissionwecould actuallyĚŽtheresearchonaquaticcycling.Youallowedustousethetherapypooland assessment facilities of the Department of Physiotherapy. Furthermore, I really appreciatedyourprofessionaladviceandthepersonalconversations. IwouldliketogiveaspecialthankstomyAͲteam(assessmentteam):'ĞƌƚsĂŶĚĞďŽƐĐŚ, ^ƚĞƉŚĂŶ<ůĞǇŶĞŶ,andZŽǇĚĞZŝĚĚĞƌ.Youdidjustgreat:explainingthequestionnaires and assessments hundreds of times and making sure to collect as much data as possible(despiteBiodexproblems,patientswhocamelateetc.). sŝĐƚŽŝƌĞ >ŽĨĨĞůĚ and /ůƐĞ >ƵŝũƚĞŶͲWŚŝůŝƉƉŝ͕ thank you for some extra support here and therewithregardtothecoordinationoftheassessmentsandforyourinterestinmy project. DĂƌŐŽƚ:ĂŶƐĞŶͲ:ĂĐŽď,thankyouforsupportingmewiththeexecutionofthetrainingin theveryearlybeginning. I would like to thank ZŝŬ ůŬĞŵĂĚĞ and ĞĞƐ ^ĐŚŽůƚŵĞŝũĞƌ from AquaKinetiqs for the goodcollaborationandourhonestdiscussionsonhowtoimprovetheAquaCruiserIITM.

186 Acknowledgements

,ĞŝŶWŽĞůů,thankyoutopavethewayforthecollaborationwithMaastrichtSportand theGeusseltbad. In addition to all the great people mentioned above, I could count on some extra supportriders,whichIamverygratefulfor. ĞŶhŵŵĞůƐfromAdelanteRehabilitation,thankyouforsharingyourthoughtsonthe trainingprogramme,thetryͲoutsessionsinthepoolandyourenthusiasmforthetopic. Furthermore,IenjoyedcollaboratingwithBachelorstudentsfromZuydUniversityof AppliedSciencesinHeerlenwhosupportedthedatacollectionandexecutionofthe trainingaspartoftheirinternshipsandthesisperiods:ĂƌĐǇhŵŵĞůƐ,ĞďŽƌĂŚdŚŝũƐƐĞŶ, tŝŵ ZŽǀĞƌƐ͕ dŚŽŵĂƐ tĞůůŝŶŬ, ZĞŵĐŽ WĂůŵĞŶ, DĂƌŝũŬĞ <ƵŝũƐ, >ŝŶĚĂ ůŽĞŵĞƌ and DŝĐŬ ƌĞƚƐ͘Manythanksforyourgreatsupport! ĂƌĐǇandtŝŵthankyouforyourcontinuingsupportasstudentassistantsduringyour Masters.TogetherwithtŝůůĞŵĞŵĞůŵĂŶƐyoudidagreatjobofenteringallthedata. ZĂĐŚĞůůĞǀĂŶĚĞŶƌĂŶĚ(TUEindhoven)and:ŽŽƉtĂŝũĞƌƐ(MaastrichtUniversity),your researchforyourMasterthesesresultedinsomevaluableandadditionalinsightson thetopicofaquaticcycling.Thankyouforyoureffortandtheinterestingknowledge exchange. Lastbutnotleastthisprojectwouldnothavebeenpossiblewithoutparticipants.Iam very grateful for your participation. Thank you so much for filling in all the questionnairesanddiaries,anddoingyourbestduringtheperformancetests.And,of course,thankyouforparticipatingintheaquaticcyclingtraining.Thesupervisionof thetrainingwasoneofthebestpartsofmywholePhDtrajectory. Nexttoallthefuninthepool,somedataanalysisandwritinghadtobedone,too. Thankfully,IhadgreatcoͲauthors,inadditiontomysupervisionteam,fromwhomI learnedalot. ƌ͘ŚƌŝƐƌƚƐ,thankyousomuchforyourenthusiasmaboutthetopicandyourgreat supervisionduringthefeasibilitystudy. ƌ͘ ĞŶ tĂůůĞƌ, :ŽŚĂŶ >ĂŵďĞĐŬ and ƌ͘ :ĞŶƐ ĂŶƐŝ, I am very thankful for the collaborationonthereviewpaper;Ilearnedalotfromallofyou. WƌŽĨ͘ƌ͘'ĞƌĂƌĚǀĂŶƌĞƵŬĞůĞŶ,thankyouverymuchforyourclearguidanceduringthe analysesofthetrialdata.IamstillsurprisedthatIwasabletodotheanalysesand, maybeevenmoresurprising,enjoyedit. dŝŶǇtŽƵƚĞƌƐ,thankyouformakingallthetextandtableslook‘pretty’.

187

Ialsowanttothankthemembersoftheassessmentcommitteefortakingthetimeto read my dissertation: WƌŽĨ͘ ƌ͘ ZŽď ^ŵĞĞƚƐ, WƌŽĨ͘ ƌ͘ ŶŶĞůŝĞƐ ŽŽŶĞŶ, WƌŽĨ͘ ƌ͘ :ŽŽƐƚ ĞŬŬĞƌ,ƌ͘ůďğƌĞ<ƂŬĞ,andƌ͘WŚŝůŝƉǀĂŶĚĞƌtĞĞƐ. Furthermore, I want to thank my PhD fellows and colleagues for their support. The dinners and social activitieswith the PhD fellows from͚zŽƵŶŐƉŝĚ͛werealwaysfun andrefreshing.Bigthankstothe͚ƉŝͲůĂĚŝĞƐ͛::ƵĚŝƚŚ,ƐƚŚĞƌ,sŝǀŝĂŶ,ŵŵǇůŽƵ,^ĂƌĂŚ,and ƵĚƌĞǇ who were always ready to support me. I am very happy that we are still in touch.SeeyousooninCopenhagen!? SpecialthankstoŵǇƌŽŽŵŝĞƐatMaastrichtUniversity,^ĂƌĂŚ,sŝǀŝĂŶandŵŵǇůŽƵ,for never complainingabout alltherecruitment calls, mywet bathingsuit drying in our officeandabighoorayforyourencouragement,helpfulfeedback,hugs,andchocolate. ^ĂƌĂŚ,thanksforallthegoodtalksonourwayhome,yourwillingnesstocometosome training sessions in the evening, and the fun times we had when having dinner together. ThankyoutomyPhDfellowsfromtheDepartmentofHealthPromotion:ƐƚƌŝĚ,<ĂƌŝŶ, <ƌŝƐƚŝŶĂ,andŽŵŝ.Ourlunchbreaks,coffeedates,andsportssessionswereimportant momentsforrechargingmybatteries. Furthermore,IwouldliketothankthePhDrepresentativeteamfortheirsupportin additiontoour‘reptasks’:'ŝůŝzĂƌŽŶand>ĂƚŝĨŝĂďŝĚŝ. ,ĂŶŶĞƌŝĞŬĞ ǀĂŶ ĚĞŶ ŽŽŵ, thank you for all your support, empathy and encouragement.YouareagreatPhDcoordinator. Moreover, I would like to thank all my colleagues from the ‘ƉŝĚ Ͳ ŵƵƐĐƵůŽƐŬĞůĞƚĂů ĚŝƐŽƌĚĞƌƐŐƌŽƵƉ’fortheirinterestinmyresearchandtheirhelpfuladvice.Inparticular,I wouldliketothankŽŶŶǇĚĞǁĂƌƚforherhelpwiththerandomisationandthegreat the support with all the administration that comes with a trial and the defence preparation. IalsowanttothankmycolleaguesfromƵǇĚhŶŝǀĞƌƐŝƚǇŽĨƉƉůŝĞĚ^ĐŝĞŶĐĞƐandZĞŐŝŽŶ ĂĐŚĞŶʹǁĞĐŬǀĞƌďĂŶĚforthegreatteamwork,yoursupportandyourunderstanding (especiallyinstressfulsituations). Finally,Iamverygratefulformyfriendsandfamilywhoaddedalittlesparkleandlots ofjoyfulandhappymomentstothisPhDadventure. ^Ăŵ,DŝĐŚĂĞů,ƌŵĞůůĞ,ĂŶĚ&ŝŶŶŝĂŶ.Mytimeinyourfamilywastheveryearlybeginning ofmyPhDresearchsinceIwrotemyproposalatyourkitchencounter.Thankyouso much for providing a wonderful AuͲPair experience in Australia that charged my batteriesforthenewadventuresinMaastricht.

188 Acknowledgements

DearWŚŝůŝƉƉ,youwillfindsomepͲvaluesinthisbook,butyouaredefinitelythebestP! ŝĂŶĂ,ŚƌŝƐƚŝĂŶ,ůĞŶĂ,ƌŝƐƚŝŶĂ,WŚŝůŝƉƉͲthankyouforallthegoodtimestogether. Dear/ƐĂand/ŶĞƐ,regardlesshowmanykilometresapartandnomatterhowbusyour lives are – we keep connected. Thank you for your interest and encouragement throughouttheyears.Ourfriendshipmeanssomuchtome. Dear‘meisjes’,^ƚĞƉŚŝĞ,ŶŝŬĂandŽŵŝ,ourfriendshipisoneofthebestthingsthat happened during my time at Maastricht University. Your understanding, encouragement,honesty,andsupportmadetoughtimeseasierandgoodtimeseven better.Thankyouand‘Stammtisch’forever! ^ŝůŬĞand<ĂƌŝŶ,myparanymphs.Iamgratefultobesupportedbysuchpositive,strong, lovingandcaringwomen.^ŝůŬĞ,weweresuchagoodthesisteamduringourBachelor andyoubecameapreciousfriend.IhaveknownforyearsthatIwouldaskyoutobe myparanymph!<ĂƌŝŶ,wheneverwemeetIfeelinspiredandmotivatedafterwards.It wasanhonourtobeyourparanymphandIamveryhappyyouareminenow.Ialso wanttothankyouforallthegreatsupportduringthepreparationofmythesisandthe defence. DearĞůŐŝƵŵĨĂŵŝůǇĂŶĚĨƌŝĞŶĚƐ,thankyouforthewarmwelcomeandyoursupport, especiallyduringSeattletimes.ZƵĚŝ,^ŽŶũĂ,:ĞůŬĂ,:ƵůŝĂ,>ĂƵƌĞŶƐ,^ŝĞŶ–thankyoufor makingmefeellikeapartofyourfamily. ŶŶŝŬĂ,youaresuchalovingandcaringsisterandIamverythankfulforourstrong connection.Thanksalotforthegreatcoverdesign!Iamlookingforwardtoalotmore qualitytimewithyou,Christian,Milanandourparents. DĂŵĂandWĂƉĂ,youneveraskedmetowintherace;instead,youvaluedallthesweat Iputinduringpracticeandalwaysremindedmewhatismostimportant:toenjoylife! Thankyouforallyourlove,supportandyourencouragement. DŝůĂŶ,youarethebestthingthathappenedtomeduringmyPhD.Youaremyspecial bluemindbͲlikebeinginthewater,youbringanimmeasurablesenseofpeacetomy heartandmind.Mercikesforeverything!

bTheterm‘bluemind’describesastateofwaterͲassociatedpeace;itis“amildlymeditativestate characterizedbycalm,peace,unity,andasenseofgeneralhappinessandsatisfactionwithlifeinthe moment."(WallaceJ.Nichols)

189

190

Abouttheauthor

190

192 Abouttheauthor

StefanieRewaldwasbornonJune3rd,1985inKassel,Germany.In2004,shereceived her Gymnasium diploma at the ‘Städtisches Gymnasium’ in Eschweiler, Germany. Subsequently, she studied Physiotherapy at Zuyd University of Applied Sciences in Heerlen,TheNetherlands.AfterobtainingherBachelor’sdegreein2008(cumlaude), she worked as a physiotherapist in a primary care practice, ‘Niessen & Stein’, in Eschweiler.Sincesheenjoyedsupervising(aquatic)exercisegroups,Stefaniedecided tocombineherclinicalworkwiththeMasterProgramme‘SportsandPhysicalActivity Interventions’ at Maastricht University, Maastricht, The Netherlands. For her Master thesis,sheinvestigatedthefeasibilityofanaquaticcircuitprogrammeforpatientswith knee osteoarthritis. The Master thesis was awarded with the ‘Catharina Pijls Award’ (2012).Afterhergraduationin2011,shetravelledtoBrisbane,Australiatosupporta working family of four as an AuͲPair. During her time in Australia, she worked on a researchproposalforarandomisedcontrolledtrialtofurtherexaminetheeffectsof aquaticcycling.In2012,theproposalwasawardedwithaNWOGraduateProgramme grant that enabled Stefanie to continue her research as a PhD candidate at the Department of Epidemiology within the Care and Public Health Research Institute (CAPHRI)atMaastrichtUniversity.In2015,Stefaniereceivedthe‘YoungInvestigators Award’ for her research on aquatic cycling at the international aquatic therapy congress“AquaLeuven”inLeuven,Belgium. During her PhD trajectory, Stefanie also worked as a physiotherapist (July 2014 to February 2015) at Maastricht University Medical Centre+ to support her colleagues whoconductedthedatacollectionforhertrial.Furthermore,Stefaniewasanactive memberoftheCAPHRIPhDrepresentativeteam(December2014toSeptember2016). ThisincludedbeingalowͲthresholdcontactforherfellowPhDcandidates,organising variousmeetings,editingthemonthlynewsletters,andactingasasoundingboardfor the PhD management of CAPHRI. In addition to her academic and clinical work, Stefaniealsoworkedasanaquaticcyclinginstructoratthe‘Geusseltbad’inMaastricht (October 2014 to September 2016). Here, she got the opportunity to transfer her scientificknowledge of aquatic cyclingintodailypractice.Moreover,Stefanie gained experienceinteachingundergraduatestudents.FromSeptember2015toMay2018, sheworkedasajuniorlectureratthePhysiotherapyDepartmentoftheZuydUniversity ofAppliedSciencesinHeerlen.Furthermore,shecombinedherteachingandresearch activitieswithajobasascientificstaffmemberofthe‘RegionAachen–Zweckverband’ inAachen,Germany(February2017toMarch2018).Shewasresponsibleforvarious eventsand initiatives focusing on occupationalhealth, the implementation of health innovationsintheAachenarea,andthepromotionofthelocalhealthcaresector. AsofJune2018,StefanieislivingwithherpartnerinCopenhagen,Denmark.

193

194

Listofpublications

194

196 Listofpublications

LISTOFPUBLICATIONS

Rewald S, Mesters I, Lenssen AF, Bansi J, Lambeck J, de Bie RA, Waller B. AquaticcyclingͲWhatdoweknow?AscopingreviewonheadͲoutaquaticcycling.Plos One.2017;12(5):e0177704. RewaldS,MestersI,EmansPJ,LenssenAF,WijnenW,deBieRA.EffectofaquaͲcycling on pain and physical functioning compared with usual care in patients with knee osteoarthritis: study protocol of a randomised controlled trial. BMC Musculoskelet Disord.2016;17:88. RewaldS,MestersI,EmansPJ,ArtsJJ,LenssenAF,deBieRA.Aquaticcircuittraining includingaquaͲcyclinginpatientswithkneeosteoarthritis:Afeasibilitystudy.JRehabil Med.2015;47(4):376Ͳ81. Rewald S, Mesters I, Emans PJ, Lenssen AF, de Bie RA. Feasibility of aqua cycling in patientswithkneeosteoarthritisͲApilotstudy.AnnRheumDis.2014;73:1226.

Submittedforpublication: RewaldS,MestersI,EmansPJ,ArtsJJ,LenssenAF,vanBreukelenG,deBieRA.Efficacy of aquatic cycling on knee pain and physical functioning in patients with knee osteoarthritis:arandomisedcontrolledtrial.

Conferenceproceedings Effect of aquaͲcycling on pain and physical functioning compared with usual care in patientswithkneeosteoarthritis:shorttermresultsfromaRCT(oralpresentation).3rd conference ofEvidenceͲBased Aquatic Therapy (Comprehensive Aquatic Therapy put intoPractice),Queretaro,Mexico,October2016. Knee OA and Aqua Cycling (oral presentation). Dutch symposium about knee osteoarthritis organised by Fy’net and Maastricht University Medical Centre+, Maastricht,TheNetherlands,June2016. Cycleawayfrompain?(posterpresentation).AnnualCAPHRIResearchDay,Maastricht, TheNetherlands,December2015. Aquaticcircuittrainingincludingaquacyclinginpatientswithkneeosteoarthritis–a feasibilitystudy(oralpresentation).2ndconferenceonEvidenceͲbasedAquaticTherapy (AquaLeuven),Leuven,Belgium,April2015.

197

Depijnonderdekniekrijgen.Aquacyclingenknieartrose(oralpresentation).Dutch symposium‘KennisinBedrijf’,HogeschoolZuyd,Heerlen,TheNetherlands,November 2013. Aquacyclingandkneeosteoarthritis–cycleawayfrompain?!Frompilottestingtoa RCT (poster presentation). VRA annual congress, Noordwijkerhout, The Netherlands, November2013. Design ideas for an aquatic cycling trial (oral presentation). Symposium Jagiellonian University,Krakow,Poland,November2012.

198 Life is like riding a bicycle. To keep your balance, you must keep moving.

Albert Einstein