Motor Function, Paratonia and Glycation Cross-Linked in Older People

Contact Drachten. Drachten. Paranimfen: Hans Drenth Fransisca Spa Fransisca Sjoukje Drenth in older people voor een borrel in een borrel voor voor 17 september. voor UITNODIGING Tevens bent u tussen bent Tevens [email protected] ZuidOostZorg/Sunenz, ZuidOostZorg/Sunenz, 16.00 en 18.00 welkom 16.00 en 18.00 welkom [email protected] [email protected] op de receptie ter plaatse op de receptie Broerstraat 5 te Groningen 5 te Broerstraat cross-linked and glycation Motor function, paratonia paratonia function, Motor Burgemeester Wuiteweg 140, Wuiteweg Burgemeester [email protected] [email protected] [email protected] Maandag 24 september om 12.45 om 12.45 Maandag 24 september van de Rijksuniversiteit Groningen, Groningen, de Rijksuniversiteit van verdediging van mijn proefschrift van verdediging Graag aanmelden voor de borrel via de borrel aanmelden voor Graag in de aula van het Academiegebouw Academiegebouw het in de aula van Aansluitend bent u van harte welkom welkom harte u van bent Aansluitend voor het bijwonen van de openbare de openbare van bijwonen het voor Hans Drenth in older peoplein older Motor function paratonia decline and

and their relation with Advanced Glycation End-Products Glycation with Advanced and their relation

Motor function, paratonia function,Motor paratonia and glycation cross-linked cross-linked glycation and

Hans Dretnh Motor Function, Paratonia and Glycation Cross-linked in Older People

Motor Function, Paratonia and Glycation Cross-linked in Older People

Motor function decline and paratonia and their relation with Advanced Glycation End-Products

Hans Drenth The work presented in this thesis was performed at the Research Group Healthy Ageing, Allied Health Care and Nursing, Hanze University of Applied Sciences, Groningen, the Netherlands, at the Research Institute SHARE of the Groningen Graduate School of Medical Sciences of the University Medical Center Groningen, University of Groningen, The Netherlands and at the Frailty in Ageing research group, Vrije Universiteit Brussel, Brussels, Belgium. The work described in this thesis was sponsored by Alzheimer Nederland (the Dutch Alzheimer Organisation) and ZuidOostZorg, organisation for Elderly Care, Drachten, the Netherlands.

Cover design Henri Groenewold Printed by Gildeprint Drukkerijen ISBN 978-94-034-0798-2 (printed version) ISBN 978-94-034-0797-5 (electronic version)

Acknowledgments The printing of this thesis was financially supported by - Research Group Healthy Ageing, Allied Health Care and Nursing, Hanze University of Applied Sciences, Groningen - University of Groningen - Research Institute SHARE, University Medical Center Groningen - University Medical Center Groningen - Scientific College Physical Therapy (WCF) of the Royal Dutch Society for Physical Therapy (KNGF) - Alzheimer Nederland (the Dutch Alzheimer Organisation) - Doctoral School of Life Sciences and Medicine, Vrije Universiteit Brussel - Dutch Society for Geriatric Physical Therapy (NVFG)

All rights reserved. No parts of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, without the prior written permission of the copyright owner.

Motor function, paratonia and glycation cross-linked in older people

Motor function decline and paratonia and their relation with Advanced Glycation End-Products

PhD thesis

to obtain the degree of PhD of the University of Groningen on the authority of the Rector Magnificus Prof. E. Sterken and in accordance with the decision by the College of Deans

and

to obtain the degree of PhD of Gerontological Sciences Vrije Universiteit Brussel

Double PhD degree

This thesis will be defended in public on

Monday 24 September 2018 at 12.45

Johannes Christiaan Drenth

born on 29 February 1968 in ‘s-Gravenhage, the Netherlands Supervisors Prof. C.P. van der Schans Prof. S.U. Zuidema Prof. I. Bautmans

Co-supervisors Dr. J.S.M. Hobbelen

Assessment Committee Prof. T. Hortobagyi Prof. D. Beckwée Prof. E. Boulanger Prof. O. Bruyere TABLE OF CONTENTS

Chapter 1: General introduction 7

Chapter 2: The Contribution of Advanced Glycation End-Product (AGE) 19 Accumulation to the Decline in Motor Function European Review of Aging and Physical Activity 2016, Mar 4; 13:3.

Chapter 3: Advanced Glycation End-Products are Associated with Physical 45 Activity and Physical Functioning in the Older Population Journals of Gerontology; Series A Biological Sciences and Medical Sciences 2018, Apr 28.

Chapter 4: Advanced Glycation End-Products are Associated with the 61 Presence and Severity of Paratonia in Early Stage Alzheimer Disease JAMDA 2017, Jul 1; 18(7): 636.e7-636.e12.

Chapter 5: Association between Advanced Glycation End-Products and 79 Functional Performance in Alzheimer’s Disease and Mixed Dementia International Psychogeriatrics 2017, Sep; 29 (9): 1525–1534.

Chapter 6: Psychometric Properties of the MyotonPRO in People with Paratonia 95 Gerontology 2017, Dec 22.

Chapter 7: Summary and General Discussion 119 Samenvatting 134 Dankwoord 138 Curriculum Vitae 140 Research Institute for Health Research SHARE 141

Chapter 1

General Introduction Chapter 1

BACKGROUND

Within the aging population, motor function decline such as reduced walking abilities and decline in activities of daily living are commonly observed 1–4. Most human physiologic systems decline during aging, independent of substantial disease effects, at an average linear loss rate of 0.34-1.28% per year between the ages of 30 and 70 years 5. Impaired motor function is a prominent characteristic of physical frailty and is associated with a wide range of adverse health consequences such as falls, disability, death, hospitalization, and institutionalization 6,7. Different mechanisms contribute to the age-related decline in motor function. One of these mechanisms that might be a factor in this decline are advanced glycation end-products (AGEs) which have been proposed as contributing to the age-related decline of the functioning of cells and tissues in normal aging. A specific age-related disease such as dementia is frequently accompanied by motor function decline which varies between the different dementia types8,9 . This decline in motor function precedes cognitive decline and progresses gradually over years3 . In early dementia stages, intentional movements become unstructured and clumsy followed by a decline in walking abilities indicated by the shortening of step length, diminished walking velocity, and unsteadiness 10–13. In the later stages, the patient becomes wheelchair bound or even bedridden due to paratonia, a specific manifestation of impaired motor function in people experiencing dementia14 . Paratonia was already being described beginning in 1828 15, but it was not until 2006 that a consensus definition of it was established16 . Despite this, paratonia is still fairly unknown, and the pathogenesis of paratonia is not well understood. It has been shown that patients in early stage dementia with diabetes mellitus (DM) have a significantly higher risk for the development of paratonia compared to those with dementia but without DM 17. Both Alzheimer’s disease (AD)18,19 and DM 20,21 are related to higher concentrations of AGEs, suggesting that AGEs could possibly be involved in the development of paratonia.

MOTOR FUNCTION

Motor function is defined as any movement or activity that uses motor neurons. Different motor activities derive from the coordinated activity of distributed motor networks within the central nervous system (CNS), extending via the peripheral nervous system to the musculoskeletal structures for generating movement 22. Motor impairment can be the result of damage to motor-related brain regions, and the location of the damage within CNS structures may lead to different presentations of motor function decline 3. Furthermore, motor function decline can also be caused by peripheral changes in the nervous system or skeletal muscles. Loss of skeletal muscle mass and muscle weakness is an important contributor to a decline in motor function and functional performance 23. Besides muscle

8 General introduction

atrophy from either disease or disuse, it has been suggested that loss of muscle contractile properties may possibly be due to biomechanical changes in the connective tissues 1 surrounding the muscle fibres (endomysium, perimysium, and epimysium) 23. Age-related intermolecular cross-linking processes in a muscle’s connective tissue leads to changes in its mechanical properties, causing loss of elasticity and increasing tissue stiffness 24.

PARATONIA

Paratonia is a distinctive form of hypertonia/movement stiffness that is observed in individuals experiencing dementia and has an estimated prevalence of 10% in the early stages and up to 90-100% with later stages of dementia 14,17. The severity of paratonia increases with the progression of the dementia and is associated with a further loss of functional mobility, severe contractures (see Figure 1), and pain.16 Due to paratonia, daily care, especially washing and dressing, becomes uncomfortable and painful. Severe paratonia, therefore, results in a substantial increase of the caretaker’s burden and a decrease of the quality of life in the advanced stages of dementia14 . However, in the early stages, paratonia already has a negative and significant impact on functional mobility, and this decline has been identified as a significant risk factor for falls for those with dementia17,25 .

Figure 1. Characteristic flexion posture in patient with more severe paratonia. Photo by José Verheijden.

Definition of paratonia

In 2006, the following operational definition of paratonia was established through an International Delphi procedure with known experts in the field 16.

9 Chapter 1

“Paratonia is a form of hypertonia with an involuntary variable resistance during passive movement. The nature of paratonia may change with progression of the dementing illness, meaning that active assistance (or “mitgehen”) is more common early in the course of degenerative dementias, whilst active resistance (or “gegenhalten”) is more common later in the course of the disease. The degree of resistance varies depending on the speed of movement (e.g., a low resistance to slow movement and a high resistance to fast movement). The degree of paratonia is proportional to the amount of force applied. Paratonia increases with progression of dementia. Furthermore, the resistance to passive movement is in any direction and there is no clasp-knife phenomenon”.

This definition enables differentiation between paratonia, Parkinsonian rigidity, and spasticity after a stroke. Contrary to paratonia, Parkinsonian (lead pipe) rigidity has a constant degree of resistance that is not influenced by the speed of the movement 26. Furthermore, in paratonia, there are no exaggerated tendon jerks (clasp-knife phenomenon) which is in contrast with spasticity 27.

Diagnosing Paratonia

The Paratonia Assessment Instrument (PAI) is the only valid and reliable instrument for assessing the presence or absence of Paratonia 28. The PAI is based on successively passive mobilisation of both shoulders towards ante-flexion/retro-flexion, elbows towards flexion/ extension, and combined hips/knees towards extension/flexion 28 (Figure 2). With the participant in a sitting or supine position, the examiner begins with a slow movement of the limb after which the movement is accelerated. Paratonia is diagnosed as being present when the following five criteria are all satisfied: (1) an involuntary variable resistance; (2) a degree of resistance that varies depending on the speed of the movement (e.g., low resistance to slow movements and high resistance to fast movements); (3) resistance to passive movement in any direction; (4) no clasp-knife phenomenon; and (5) resistance in two movement directions in the same limb or in two different limbs. The severity of paratonia is then scored by using a Modified Ashworth Scale for paratonia (MAS-P)29. During the PAI assessment, the assessor quantifies the muscle tone based on the resistance induced by the passive movements of the limbs. The MAS-P is based on a 5-point ordinal scale ranging from 0 to 4, meaning 0 = no resistance, 1= slight resistance, 2 = more marked resistance, 3 = considerable resistance, and 4 = severe resistance such that passive movement is impossible 29,30. A score of 0.5 is assigned in the event of active assistance.

Because of the inherent characteristics of paratonia (e.g., variability in movement resistance), accurate and objective measurement of paratonia severity proves to be challenging.

10 General introduction

Figure 2. Assessing paratonia with the PAI by conducting passive movement of the shoulders, elbows, and hips in flexion and extension 28.

Although MAS measurement is the worldwide standard for measuring the resistance to passive movements in clinical settings, extensive experience is necessary for the MAS to be reliable 30–32. An objective and accurate alternative for measuring muscle properties is available with the MyotonPRO. It is a quickly applicable, painless, and non-invasive handheld device that measures muscle properties such as tone, elasticity, stiffness, creep, and mechanical stress relaxation time, however, it still needs to be validated for people with paratonia.

ADVANCED GLYCATION END-PRODUCTS

Since the early 1980s, it has been proposed that the cross-linking of long-lived proteins mediated by advanced glycation end-products (AGEs) may contribute to the age-related decline of the functioning of cells and tissues in normal aging 33,34. AGEs formation occurs through the non-enzymatically reaction of monosaccharides with the amino groups of proteins, particularly the N-terminal amino groups and side chains of lysine and arginine. This modification, termed non-enzymatic glycosylation or the Maillard reaction, leads to a reversible, so called Schiff-base adduct. A Schiff base is a compound that has a carbon to nitrogen double bond where the nitrogen is not connected to hydrogen. The Schiff base subsequently experiences chemical rearrangement, known as the amadori rearrangement, and forms protein bound products that are more stable, or amadori products. Through subsequent oxidations and dehydrations, including free radical intermediates, a broad range of irreversible, heterogeneous, and sometimes fluorescent and yellow-brown products is formed, the so-called AGEs 18–20,34 (figure 3). AGEs formation may also be initiated by metal- catalyzed glucose auto-oxidation and lipid peroxidation 35.

11 Chapter 1

AGEs are spontaneously produced in human tissues as an element of normal metabolism which increases with aging and accelerates in hyperglycaemic environments 18,20,36,37. The increase of the level of free/unbound and protein bound AGEs in the blood circulation is also determined by an exogenous intake such as food 38. AGEs are removed from the body through enzymatic clearance and renal excretion.

Figure 3. AGE formation

With aging, there is an imbalance between the formation and natural clearance of AGEs that results in an incremental accumulation in tissues with slow turnover such as muscles, cartilage, tendons, eye lens, vascular media, and the dermis of the skin39,40 . Beside their role in AD and complications of DM, the accumulation of AGEs is a significant contributing factor in many age related diseases including renal failure, blindness, and cardiovascular diseases 18,21.

AGEs can be quantified in blood or tissue biopsies, however, due to their fluorescent properties, their presence in the skin can be noninvasively assessed using skin autofluorescence (SAF). Several types of AGEs have been described and can be categorized into fluorescent cross- linking such as Pentosidine; non-fluorescent cross-linking such as Glucosepane; fluorescent non-cross-linking such as Arginine-pyrimidine; and non-fluorescent non-cross-linking such as Carboximethyl-lysine 41–43. The cross-linking of long-lived proteins, particularly collagen, are responsible for an increasing proportion of insoluble extracellular matrix and thickening of tissue as well as increasing mechanical stiffness and loss of elasticity 24,36,44.

Non-cross-linking effects are exerted by the Advanced Glycation End-Products (AGEs) are binding of AGEs to the receptor for AGEs irreversible damaged proteins through binding (RAGE). RAGE is a multi-ligand member with sugars (figure 4). of the immunoglobulin superfamily of Cross-linking effects: cell surface molecules that is widely • thickening of tissue localized in a variety of cell lines including • increasing mechanical stiffness • loss of elasticity monocytes, endothelial, mesangial, neuronal, and muscle. The interaction Non-cross-linking effects (by pro-inflammatory processes and oxidative stress): with AGEs incites activation of intracellular • increasing mechanical stiffness signalling, gene expression, and production • loss of elasticity of pro-inflammatory cytokines (such as • neural function decline

12 General introduction

Interleukin (IL)-6, Tumor Necrosis factor (TNF)-alpha), and free radicals 45,46. At the peripheral (tissue) level, these inflammatory processes exhibit powerful proteolytic activity whereby 1 the collagen becomes more vulnerable and tissue stiffness increases24 At the central level (central nervous system), interaction between AGEs, Amyloid-beta, and hyper-phosphorylated tau-protein induce microglia and astrocytes to upregulate the production of reactive oxygen species, pro-inflammatory cytokines, and nitric oxide which affects neuronal function 18.

Figure 4. AGEs may form cross-links and/or adducts on collagen structures. This figure shows a collagen fibril and the formation of Glucosepane, which links Lysine and Arginine sidechains. The two amino acids can belong to separate molecules (top), forming an intermolecular cross-link, or to the same molecule (bottom).47

THE AIM OF THIS THESIS

The main objective of this thesis is to study whether AGE accumulation contributes to motor dysfunction in the aging population in general and to the pathogenesis of paratonia in particular. It will be the first step in unravelling the phenomenon of paratonia ona fundamental level. Another aim is to establish an objective method for quantifying the severity of paratonia by studying the psychometric properties of the MyotonPRO, a portable device that objectively measures muscle properties.

OUTLINE OF THIS THESIS

To identify a direct relationship between AGEs and motor function decline, a systematic review was conducted. This review revealed a lack of research into the relationship of AGEs and physical activity and function within the older population, leading to a specific

13 Chapter 1

investigation into this topic. The specific relationship between AGEs and paratonia and functional performance in the target group of dementia was then investigated. Finally, an easy and objective measurement tool for measuring paratonia severity was studied that could be used for further research into paratonia.

This thesis answers the following research questions; 1. Is there a direct relationship between circulating and/or tissue AGEs and motor function decline (Chapter 2)? 2. Are AGE levels associated with physical activity and physical functioning in older individuals (Chapter 3)? 3. Are AGE levels associated with the prevalence and severity of paratonia in patients with Alzheimer’s disease and mixed dementia Chapter( 4)? 4. Are AGE levels associated with a decline of functional performance in people with Alzheimer’s disease and mixed dementia Chapter( 5)? 5. Is the MyotonPRO a valid and reproducible tool for assessing paratonia severity? (Chapter 6)?

14 General introduction

REFERENCES

1. Arvanitakis Z, Wilson RS, Schneider JA, Bienias JL, Evans DA, Bennett DA. Diabetes mellitus and 1 progression of rigidity and gait disturbance in older persons. Neurology. 2004;63(6):996-1001. 2. Bellew JW, Symons TB, Vandervoort AA. Geriatric fitness: effects of aging and recommendations for exercise in older adults. Cardiopulm Phys Ther J. 2005;16(1):20-31. 3. Buchman AS, Bennett DA. Loss of motor function in preclinical Alzheimer’s disease. Expert Rev Neurother. 2011;11(5):665-676. 4. Rikkert MGO, Rigaud AS, van Hoeyweghen RJ, de Graaf J. Geriatric syndromes: medical misnomer or progress in geriatrics? Neth J Med. 2003;61(3):83-87. 5. Sehl ME, Yates FE. Kinetics of human aging: I. Rates of senescence between ages 30 and 70 years in healthy people. J Gerontol A Biol Sci Med Sci. 2001;56(5):B198-208. 6. Fried LP, Tangen CM, Walston J, et al. Frailty in older adults: evidence for a phenotype. journals Gerontol A, Biol Sci Med Sci. 2001;56(3):M146-56. 7. Sternberg SA, Schwartz AW, Karunananthan S, Bergman H, Clarfield AM. The identification of frailty: a systematic literature review. J Am Geriatr Soc. 2011;59(11):2129-2138. 8. Prehogan A, Cohen CI. Motor dysfunction in dementias. Geriatrics. 2004;59(11):53-54-60. 9. Scherder E, Eggermont L, Visscher C, Scheltens P, Swaab D. Understanding higher level gait disturbances in mild dementia in order to improve rehabilitation: “last in-first out”. Neurosci Biobehav Rev. 2011;35(3):699-714. doi:10.1016/j.neubiorev.2010.08.009. 10. Goldman WP, Baty JD, Buckles VD, Sahrmann S, Morris JC. Motor dysfunction in mildly demented AD individuals without extrapyramidal signs. Neurology. 1999;53(5):956-962. 11. Pugh KG, Lipsitz LA. The microvascular frontal-subcortical syndrome of aging. Neurobiol Aging. 2002;23(3):421-431. 12. Tanaka A, Okuzumi H, Kobayashi I, Murai N, Meguro K, Nakamura T. Gait disturbance of patients with vascular and Alzheimer-type dementias. Percept Mot Skills. 1995;80(3 Pt 1):735-738. doi:10.2466/pms.1995.80.3.735. 13. Pohjasvaara T, Mantyla R, Ylikoski R, Kaste M, Erkinjuntti T. Clinical features of MRI-defined subcortical vascular disease. Alzheimer Dis Assoc Disord. 2003;17(4):236-242. 14. Souren LE, Franssen EH, Reisberg B. Neuromotor changes in Alzheimer’s disease: implications for patient care. J Geriatr Psychiatry Neurol. 1997;10(3):93-98. 15. L.H. Friedlander. Arznygelahrtheit. Allg Lit zeiting. 1828;juni:257-258. 16. Hobbelen JSM, Koopmans RTCM, Verhey FRJ, Van Peppen RPS, de Bie RA. Paratonia: a Delphi procedure for consensus definition. J Geriatr Phys Ther. 2006;29(2):50-56. 17. Hobbelen JS, Tan FE, Verhey FR, Koopmans RT, de Bie RA. Prevalence, incidence and risk factors of paratonia in patients with dementia: a one-year follow-up study. Int Psychogeriatr. 2011;23(7):1051-1060. 18. Rahmadi A, Steiner N, Munch G. Advanced glycation endproducts as gerontotoxins and biomarkers for carbonyl-based degenerative processes in Alzheimer’s disease. Clin Chem Lab Med. 2011;49(3):385-391. 19. Munch G, Thome J, Foley P, Schinzel R, Riederer P. Advanced glycation endproducts in ageing and Alzheimer’s disease. Brain Res Brain Res Rev. 1997;23(1-2):134-143. 20. Ahmed N, Thornalley PJ. Advanced glycation endproducts: what is their relevance to diabetic complications? Diabetes, Obes Metab. 2007;9(3):233-245. 21. Gasser A, Forbes JM. Advanced glycation: implications in tissue damage and disease. Protein Pept Lett. 2008;15(4):385-391.

15 Chapter 1

22. Burke R. Motor Units: Anatomy, Physiology and Functional Organization. In: Brooks VB (Ed) Handbook of Physiology: The Nervous System. (II). Motor Control. Washington DC: American Physiological Society; 1981. 23. Bautmans I, Puyvelde K Van, Mets T. Sarcopenia and functional decline: pathophysiology, prevention and therapy. Acta Clin Belg. 2009;64(4):303-316. 24. Avery NC, Bailey AJ. Enzymic and non-enzymic cross-linking mechanisms in relation to turnover of collagen: relevance to aging and exercise. Scand J Med Sci Sports. 2005;15(4):231-240. 25. Harlein J, Dassen T, Halfens RJG, Heinze C. Fall risk factors in older people with dementia or cognitive impairment: a systematic review.J Adv Nurs. 2009;65(5):922-933. doi:10.1111/j.1365- 2648.2008.04950.x. 26. Kurlan R, Richard IH, Papka M, Marshall F. Movement disorders in Alzheimer’s disease: more rigidity of definitions is needed. Mov Disord. 2000;15(1):24-29. 27. Lance JW. Disordered muscle tone and movement. Clin Exp Neurol. 1981;18:27-35. 28. Hobbelen JSM, Koopmans RTCM, Verhey FRJ, Habraken KM, de Bie RA. Diagnosing paratonia in the demented elderly: reliability and validity of the Paratonia Assessment Instrument (PAI). Int Psychogeriatr. 2008;20(4):840-852. doi:10.1017/S1041610207006424. 29. Waardenburg, H., Elvers, J.W.H., van Vechgel, F., Oostendorp RAB. Is Paratonie Betrouwbaar te meten? Een betrouwbaarheid onderzoek voor het meten van paratonie met de MAS en de gemodificeerde tonusschaal van Ashworth. Ned Tijdschr voor Physiother. 1999;(2):30-35. 30. Meulenbroek T. Toelichtingsformulier MAS. http://www.meetinstrumentenzorg.nl/algemene meetinstrumenten aspx. Accessed January 25, 2017. 31. Leonard CT, Stephens JU, Stroppel SL. Assessing the spastic condition of individuals with upper motoneuron involvement: validity of the myotonometer. Arch Phys Med Rehabil. 2001;82(10):1416-1420. doi:10.1053/apmr.2001.26070. 32. Aarrestad DD, Williams MD, Fehrer SC, Mikhailenok E, Leonard CT. Intra- and interrater reliabilities of the Myotonometer when assessing the spastic condition of children with cerebral palsy. J Child Neurol. 2004;19(11):894-901. 33. Munch G, Westcott B, Menini T, Gugliucci A. Advanced glycation endproducts and their pathogenic roles in neurological disorders. Amino Acids. 2012;42(4):1221-1236. 34. Brownlee M. Advanced protein glycosylation in diabetes and aging.Annu Rev Med. 1995;46:223- 234. doi:10.1146/annurev.med.46.1.223. 35. Onorato JM, Jenkins AJ, Thorpe SR, Baynes JW. Pyridoxamine, an inhibitor of advanced glycation reactions, also inhibits advanced lipoxidation reactions. Mechanism of action of pyridoxamine. J Biol Chem. 2000;275(28):21177-21184. doi:10.1074/jbc.M003263200. 36. Monnier VM, Mustata GT, Biemel KL, et al. Cross-linking of the extracellular matrix by the maillard reaction in aging and diabetes: an update on “a puzzle nearing resolution.” Ann N Y Acad Sci. 2005;1043:533-544. 37. Uribarri J, Cai W, Peppa M, et al. Circulating glycotoxins and dietary advanced glycation endproducts: two links to inflammatory response, oxidative stress, and aging.journals Gerontol A, Biol Sci Med Sci. 2007;62(4):427-433. 38. Puyvelde K Van, Mets T, Njemini R, Beyer I, Bautmans I. Effect of advanced glycation end product intake on inflammation and aging: a systematic review. Nutr Rev. 2014;72(10):638-650. 39. Luevano-Contreras C, Chapman-Novakofski K. Dietary advanced glycation end products and aging. Nutrients. 2010;2(12):1247-1265. 40. Peppa M, Uribarri J, Vlassara H. Aging and glycoxidant stress. Hormones (Athens). 2008;7(2):123- 132. 41. Meerwaldt R, Graaff R, Oomen PH, et al. Simple non-invasive assessment of advanced glycation endproduct accumulation. Diabetologia. 2004;47(7):1324-1330.

16 General introduction

42. Biemel KM, Friedl DA, Lederer MO. Identification and quantification of major maillard cross- links in human serum albumin and lens protein. Evidence for glucosepane as the dominant compound. J Biol Chem. 2002;277(28):24907-24915. doi:10.1074/jbc.M202681200. 1 43. Shipanova IN, Glomb MA, Nagaraj RH. Protein modification by methylglyoxal: chemical nature and synthetic mechanism of a major fluorescent adduct.Arch Biochem Biophys. 1997;344(1):29- 36. doi:10.1006/abbi.1997.0195. 44. Kjaer M. Role of extracellular matrix in adaptation of tendon and skeletal muscle to mechanical loading. Physiol Rev. 2004;84(2):649-698. 45. Lindsey JB, Cipollone F, Abdullah SM, McGuire DK. Receptor for advanced glycation end-products (RAGE) and soluble RAGE (sRAGE): cardiovascular implications. Diab Vasc Dis Res. 2009;6(1):7- 14. 46. Schmidt AM, Yan SD, Yan SF, Stern DM. The biology of the receptor for advanced glycation end products and its ligands. Biochim Biophys Acta. 2000;1498(2-3):99-111. 47. Gautieri A, Redaelli A, Buehler MJ, Vesentini S. Age- and diabetes-related nonenzymatic crosslinks in collagen fibrils: candidate amino acids involved in Advanced Glycation End-products. Matrix Biol. 2014;34:89-95. doi:10.1016/j.matbio.2013.09.004.

Chapter 2

The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function

Hans Drenth, Sytse Zuidema, Steven Bunt, Ivan Bautmans, Cees van der Schans, Hans Hobbelen

European Review of Aging and Physical Activity2016, Mar 4; 13:3. Chapter 2

ABSTRACT

Diminishing motor function is commonly observed in the elderly population and is associated with a wide range of adverse health consequences. Advanced glycation end-products (AGEs) may contribute to age-related decline in the function of cells and tissues in normal aging. Although the negative effect of AGEs on the biomechanical properties of musculoskeletal tissuesand the central nervous system have been previously described, the evidence regarding the effect on motor function is fragmented, and a systematic review on this topic is lacking. Therefore, a systematic review was conducted from a total of eight studies describing AGEs related to physical functioning, physical performance, and musculoskeletal outcome which reveals a positive association between high AGEs levels and declined walking abilities, inferior ADL, decreased muscle properties (strength, power and mass) and increased physical frailty. Elevated AGEs levels might be an indication to initiate (early) treatment such as dietary advice, muscle strengthening exercises, and functional training to maintain physical functions. Further longitudinal observational and controlled trial studies are necessary to investigate a causal relationship, and to what extent, high AGEs levels are a contributing risk factor and potential biomarker for a decline in motor function as a component of the aging process.

20 The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function

BACKGROUND

In the aging population, decline in motor function such as reduced walking speed, poor balance, and loss of muscle strength are commonly observed phenomena 1–4. Most human physiologic systems regress with aging, independently of substantial disease effects, at an 2 average linear loss rate of 0.34-1.28% per year between the age of 30 and 70 years 5. The age related loss of skeletal muscle mass, which is accompanied by muscle weakness, is an important contributor to functional decline 6. More people over the age of 70 are having difficulties performing everyday functions because of motor function related problems 7. Impaired motor function is a prominent characteristic of physical frailty and is associated with a wide range of adverse health consequences such as falls, disability, death, hospitalization, and institutionalization 8,9. In addition to physical frailty, longitudinal studies suggest that a decreased level of motor function is associated with a decline of cognitive function and a subsequent development of both mild cognitive impairment and Alzheimer’s disease 3. A decline in motor function in older ages could be predicted by biomarkers which could allow for early interventions. Since the early 1980’s, it is proposed that the cross-linking of long-lived proteins mediated by advanced glycation end-products (AGEs) may contribute to the age-related decline of the functioning of cells and tissues in normal aging. AGEs could, therefore, be a potential biomarker and risk factor for the decline of motor function in the older population 10.

Advanced Glycation End-Products and Aging

The occurrence of AGEs is mediated by non-enzymatic condensation of a reducing sugar with an amino group. AGEs are spontaneously produced in human tissues as an element of normal metabolism which increases with aging 11–14. The increase of the level of free/ unbound and protein bound AGEs in the blood circulation is also determined by an exogenous intake such as food 15. AGEs are being removed from the body through enzymatic clearance and renal excretion. It has been proposed that, with aging, there is an imbalance between the formation and natural clearance of AGEs, leading to the accumulation of AGEs in tissues 16,17. In many age related diseases, the accumulation of AGEs is a significant contributing factor in degenerative processes, especially in renal failure, blindness, cardiovascular diseases, and complications of diabetes mellitus 13,18. Elevated AGEs levels in patients with diabetes mellitus (DM) is most likely due to an excessive elevation of glucose concentration which consequently accelerates the glycation of proteins 11,13,18. Furthermore, AGEs play a role in neurological disorders such as Alzheimer’s Disease(AD) 13,19. Interestingly, Hobbelen et al.20 ascertained that patients in early stage of dementia and with DM had a significantly higher risk for the

21 Chapter 2

development of muscle stiffness/hypertonia (defined as paratonia) compared to those with dementia but without DM (OR = 10.7, 95% CI: 2.2 - 51.7). In early stage dementia, paratonia already negatively and significantly impacts functional mobility such as walking speed. As the most common cause of dementia, Alzheimer’s disease, as well as DM, are related to higher serum concentrations of AGEs, the previous finding in the Hobbelen et al. study prompted the hypothesis that AGEs may partly or indirect be responsible for the development of paratonia. The pathogenesis of paratonia is unclear and central nervous system changes as well as peripheral biomechanical changes are hypothesized 20.

Several types of AGEs have been described and can be categorised into fluorescent cross-linking AGEs, non-fluorescent cross-linking AGEs, and non-cross-linking AGEs. The best chemically characterized AGEs are Pentosidine (fluorescent cross-linking) and Carboximethyl-lysine (CML, non-cross-linking) 21. Cross-linking AGEs are responsible for an increasing proportion of insoluble extracellular matrix and thickening of tissue as well as increasing mechanical stiffness and loss of elasticity12,22,23 . Cross-linking AGEs are considered as being involved in the pathophysiology of arthritis 24,25. In fact, the accumulation of AGEs in human articular cartilage increases cartilage stiffness and brittleness. Consequently, the cartilage becomes increasingly prone to damage and degeneration 26,27. In rheumatoid arthritis, elevated levels of cross-linking AG’s in serum and/or synovial fluids also appear to correlate with the levels of inflammatory markers and the disease activity24 . Another impact on the musculoskeletal system is within the skeletal bone where AGE-induced cross-links in the collagen matrix alter the mechanical properties of bone by increasing stiffness and fragility 27–29. Furthermore, significantly higher AGEs levels were reported in patients with osteoporosis, thereby increasing the risk of fractures 28,30.

Non-cross-linking effects are exerted by the binding of AGEs to the receptor forAGEs (RAGE). A wide variety of cells express RAGE, and the interaction with AGEs incites activation of intracellular signalling, gene expression, and production of pro-inflammatory cytokines (such as Interleukin (IL)-6, Tumor Necrosis factor (TNF)-alpha), and free radicals. At the peripheral (tissue) level, these inflammatory processes exhibit powerful proteolytic activity whereby the collagen becomes more vulnerable and tissue elasticity decreases31,32 . At the central level (central nervous system), interaction between AGEs, Amyloid-beta and hyperphosphorylated tau-protein induce microglia and astrocytes to upregulate the production of reactive oxygen species, pro-inflammatory cytokines, and nitric oxide which affects neuronal function 11,32.

Numerous reviews 16,33–39 on the relation between AGEs and motor function have been published, but the majority of these studies are narrative reviews and therefore showing

22 The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function

most often an indirect relationship. The negative effect of accumulating AGEsonthe biomechanical properties of peripheral musculoskeletal tissues and the central nervous system (CNS), for example accumulation in specific relevant motor-related brain regions, will plausibly have an impact on motor function. However, evidence regarding this subject is fragmented, and a systematic review on this topic is lacking. Therefore, the aim of this study 2 was to systematically review the literature for the direct relationship between circulating and/or tissue AGEs and motor function.

METHODS

Search Strategy

The Embase, Pubmed, Cinahl, and Web of science databases were searched from the earliest possible date through January 2016. Terms were employed both as free text words and mapped to subject headings terms with explosion when feasible. We utilized all possible relevant terminology regarding AGEs and motor function (decline). For detailed information on search terms, see Table 1. The search regarding terms for AGEs were combined with our search (with AND) regarding terms for motor function (decline). Search filters were established for studies on humans and the English, Dutch, or German language. Bibliographies from identified articles were reviewed for additional references.

Table 1. Detailed search terms Entry Keywords

AGEs “advanced glycosylation end products” (OR) “advanced glycosylation end products receptor (supplementary concept)” (OR) “advanced glycation end product(s)” (OR)“advanced glycation end product(s) receptor”(OR) “non-enzymatic glycation” (OR)“non-enzymatic glycosylation” (OR) “glycotoxins”.

AND motor function “musculoskeletal physiological phenomena” (OR) ”muscle fatigue” (OR) “muscle strength” (OR) “muscle rigidity” (OR) “muscle tonus” (OR) “muscle stiffness” (OR) “mechanical stiffness” (OR) “rigidity” (OR) “ physical endurance” (OR) “postural balance” (OR) “balance” (OR) “posture” (OR) “postural control” (OR) “articular range of motion” (OR) “psychomotor performance” (OR) “motor activity” (OR) “motor skills” (OR) “motor skills disorders” (OR) “coordination” (OR) “motor coordination” (OR) “motor control” (OR) “eye-hand coordination” (OR) “fine motor” (OR) “fine motor skills” (OR) “neuromuscular manifestations” (OR) “gait” (OR)“gait disorders” (OR) “walking” (OR) “locomotion” (OR) “falls (accidental)” (OR) “mobility limitation” (OR) “physical mobility” (OR) “physical performance” (OR) “physical activity” (OR) “activities of daily living” (OR) ”motor function” (OR) “motor function decline” (OR) “mobility impairment” (OR) “functional decline” (OR) “motor inhibition” (OR) “inhibition of action” (OR) “interlimb” (OR) “interlimb coordination”

23 Chapter 2

Inclusion Criteria and Outcome Variables

Studies were included providing that a direct relationship between measured AGEs levels and motor function was described. Motor function was defined as any movement or activity through the use of motor neurons in a broad sense (i.e. motor skills, motor control, coordination, locomotion etc.) and/or physical performance (i.e., activities of dailylife (ADL), walking, stair climbing, etc.) and/or musculoskeletal function (i.e., muscle strength or stiffness, range of motion etc.). Studies were excluded when merely the effects of AGEs on tissue level or their relation with specific diseases where described, without any direct outcome on motor function.

Data Extraction and Study Quality

Data were independently extracted by two reviewers. A standardized data extraction form was exploited to derive the following data from each eligible study: first author, year of publication, study design, study population (age, gender, and co-morbidity); the type of AGEs (e.g., cross-linking, non-cross-linking, and chemical characteristics); and the outcome of motor function and the strength of the relationship between AGEs and motor function (e.g., correlation coefficients, risk ratio’s etc.). The methodological quality of each incorporated study was assessed independently by two reviewers. All eligible studies turned out to be observational studies and were assessed with the 12 item EBRO Assessment Tool of Cohort Studies 40. This tool comprises eight validity questions, one question to decide whether to continue the checklist, two applicability items, and one item for statistical analysis. The two applicability items were disregarded as they merely answer questions to determine if the results are applicable to Dutch healthcare. Therefore, ten criteria remained (Table 2). The included studies were qualified as “good”, “moderate”, or “low”. To qualify as “good”, the study should rate “valid and applicable” and include sufficient statistical analyses (e.g., correction for potential confounders). A “moderate” qualification was assigned when the validity items rated “doubtful” and included sufficient statistical analyses or rated only “valid and applicable” for the validity items. Any other scenario would have given the study a “low” methodological qualification (Table 2). In the event of a discrepancy, the final score was decided by consensus.

Statistics

Risk ratios (RR), Odds ratio’s (OR), Hazard ratio’s (HR), correlation coefficients, and group differences (mean difference and 95% confidence interval) were presented as reported in the studies. The subjects’ ages were presented as reported in the studies or in means

24 The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function 48 No Yes Yes Yes Yes 2015 Tanaka et al. et Tanaka Good Yes Yes Yes Not Applicable Not Applicable 44 Yes Yes Yes Yes Yes 2 al., 2011 Momma et Momma et Good Yes Yes Yes Not Applicable Not Applicable 49 Yes Yes Yes Yes Yes De La Maza De La Maza et al., 2008 et Moderate No Yes No Not Applicable Not Applicable 43 Yes Yes Yes Yes Yes 2009 40

Dalal et al., Dalal et Good Yes Yes Yes Not Applicable Not Applicable 50 Yes Yes Yes Yes Yes 2015 Shah et al., Shah et Moderate No Yes No Not Applicable Not Applicable 47 Yes Yes Yes Yes Yes al., 2014 Whitson et Whitson et Good Yes Yes Yes Yes Yes 46 Yes Yes Yes Yes Yes 2012 Sun et al., Sun et Good Yes Yes Yes Yes Yes 45 Yes Yes Yes Yes Yes 2010 Semba et al., Semba et Good Yes Yes Yes Not Applicable Not Applicable of Cohort Studies tool Assessment with EBRO Quality of Observation studies Methodological To be eligible for further statistical analysis, the study results should rate “valid and applicable” or “doubtful” on question 9. To qualify “valid and applicable” “valid and applicable” 9. To qualify on question “doubtful” or “valid and applicable” rate results should study the analysis, further statistical be eligible for To items (1 t/m5 and 8). A validity remaining “yes” on the than 4 times a (6 and 7) more up questions on the 2 follow “Yes” score had to a longitudinal study “yes” less than 4 times on scored a study items (1 t/m5 and 8). When validity remaining scored on the was a “yes” given when 4 times was qualification “doubtful” up (6 and 7) follow regarding questions design the two had a cross-sectional When the study cancelled. was with the checklist further analysis items, the remaining disregarded. were Overall quality score Overall Correction for potential confounders: Odds ratio (OR), Odds confounders: for potential Correction Reduction (ARR), Mean Risk Risk (RR), Absolute Relative (HR) Ratio (MD), Hazard Difference Are the study results valid and applicable? valid results the study Are Have important confounders or prognostic factors been factors or prognostic confounders important Have in the design account into taken are they and identified or analysis? Can selective loss-to-follow-up sufficiently be excluded? be sufficiently loss-to-follow-up Can selective Is there sufficient long follow-up? long sufficient Is there Has exposure outcome been blinded? outcome Has exposure Is outcome clearly described and is the method for for clearly described and is the method Is outcome adequate? assessing the outcome Is the exposure clearly described and is the method for for clearly described and is the method Is the exposure adequate? assessing the exposure Can risk of bias sufficiently be excluded? be Can risk of bias sufficiently Table 2. Table Item clearly described? groups the comparing Are

25 Chapter 2

and standard deviations (SD) that were calculated from the available data. If association coefficients were not available, effects sizes (ES)41 were calculated to estimate the magnitude of the effect of AGEs on specific outcomes when the mean difference and standard deviation of compared groups were available or could be retrieved from confidence intervals.

The following formula were used:

SD = √ (number of subjects) * [upper limit - lower limit] / (2* 1.96)

2 2 ES = [mean 1 – mean 2]/SD pooled, where SD pooled = √ [(SD 1 + SD 2 ) / 2]

Cohen’s benchmarks were used to indicate small (ES = 0.20), medium (ES = 0.50), and large (ES = 0.80) effects size42 . Due to the heterogeneity across the studies, meta-analysis was not performed.

RESULTS

Search Results

The literature search revealed 1154 hits. After eliminating 208 duplicates and after an initial screening of titles and abstracts, 902 studies were excluded. Primary reason for exclusion were that the studies were describing an (indirect) effect only on tissue level (n = 223), were disease related (n = 291) or in the field of diet and weight (n = 16), all without outcome on movement, activity, performance and or function. The remaining excluded studies (n = 372) did not meet the inclusion criteria in general. A total of 44 full-text articles were retrieved for further analyses. Examination of the available reference lists of these 44 studies did not reveal additional studies. From the 44 potentially eligible studies, 36 studies were excluded because no relevant outcomes were described (n = 19), were narrative reviews n( = 11) or were conference abstracts (n = 4) and editorial comments (n = 2). Finally, eight studies were included for this review. A flowchart of the selection process is presented in Figure 1.

Study Quality

The study quality of six studies was scored as “good” 43–48. A moderate score was assigned to two studies 49,50 because potential confounders were not taken into account in the analysis (Table 2).

26 The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function

Figure 1. Flow Chart of selection process

27 Chapter 2

Data Extraction

Study design Of the eight included studies, six had a cross-sectional design43–45,48–50 , one had a longitudinal design 46, and one study employed a mixed longitudinal and cross-sectional design 47.

AGEs type The chemical characteristics of the reported AGEs was described in six studies 43,45–49. Carboxymethyl-Lysine (CML) was quantified in five studies 43,45–47,49 and Pentosidine in one study 48. In two studies the chemical characterisation of the AGEs were not specified, yet measured indirectly by skin fluorescence 44,50. For obtaining the AGEs levels, blood samples were used in five studies 43,45–48; tissue samples in one study 49; and the non-invasive AGE reader in two studies 44,50.

Motor function outcomes The outcomes for motor function from the eight retrieved studies were categorized into three categories; 1) Physical performance and functioning: walking abilities45,46 , activities of daily living (ADL)47 and upper extremity function 50; 2) Musculoskeletal: muscle strength, power and mass 43,44,48,49; 3) Physical frailty: in accordance with the Fried criteria 8 and defined as having three or more of the following characteristics: low physical activity, exhaustion, slow walking speed, decreased muscle strength, and unintentional weight loss 47. Study characteristics and an overview of the main findings and statistics from the studies are provided in Table 3.

Relationship of AGEs with Physical Performance and Functioning

Walking abilities Walking and the relationship with Carboxymethyl-Lysine (CML) was studied in two studies45,46 . In the cross-sectional study by Semba et al.45, walking speed was assessed in a representative sample of community dwelling elderly (n = 944). Slow walking speed was defined as the slowest quintile of walking speed which was <0.79 m/sec. Plasma CML was divided into quartiles with the cut-off at the highest quartile established at 424 ng/mL. Thestudy demonstrated that, after adjusting for covariates (age, education, smoking, cognition status, depression, and chronic disease), the subjects in the highest quartile of plasma CML were at greater risk of a slow walking speed (OR = 1.60, 95% CI: 1.02 - 2.52, P = 0.04) compared to those in the lower three quartiles of plasma CML. After an exclusion of subjects with

28 The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function

diabetes and adjusting for the same covariates, the study showed that participants in the highest quartile of plasma CML were at a higher risk for slow walking speed, (OR = 1.87, 95% CI: 1.15 - 3.04, P = 0.01). In the longitudinal study by Sun et al.46, walking speed and serum CML were measured in moderately to severely disabled community dwelling female subjects (n = 394). Walking disability was defined as the inability to walk or a walking speed 2 of <0.4 m/sec. Cut-off values according to plasma CML quartiles were 452.6, 558.8 and 689.1 ng/mL. The study showed that subjects in the highest quartile of CML were at a greater risk for developing a walking disability (HR = 1.68, 95% CI: 1.11 - 2.52, P = 0.03). Multivariable hazard models for different missing data scenario’s (for details, see Table 3) and adjusting for health characteristics (age, congestive heart failure, peripheral artery disease, diabetes mellitus type 2 (DM2), and renal insufficiency) revealed a significantly higher risk for women in the highest quartile of CML to develop a walking disability (HR’s between 1.46 and 1.63, all P <0.05).

Activities of daily living (ADL) The ability to perform ADL activities and the longitudinal relation with CML in older persons was described in the study by Whitson et al.47. Serum CML was measured in participants in the Cardiovascular Health study (n = 3373) and was divided into quintiles with the cut-off for the two highest quintiles established at 620 ng/mL. ADL disability was defined as difficulty (yes or no) in any of six ADL activities such as walking around the home, getting out of bed/ chair, dressing, bathing, eating, and toileting and was assessed at 6- or 12- month intervals for 14 years. AGE levels were compared between the groups with and without difficulty on the specific ADL activities. After adjusting for potential confounders (demographics, BMI, DM2, alcohol consumption, smoking status, total cholesterol, albumin, weight change, hypertension, heart disease, stroke, and claudication), a high level CML at baseline was associated with 10% (95% CI: 5% - 15%) higher incidence of a first ADL difficulty with a hazard ratio (HR) of 1.10 (95% CI: 1.05 - 1.15), P <0,01) per standard deviation (225 ng/mL CML). After further adjusting for arthritis, cognition, and kidney function, an increase in CML was associated with a 5% (95% CI: 1% - 10%) higher incidence of disability, HR of 1.05 (95% CI: 1.01 -1.11, P = 0.03) per standard deviation (225 ng/mL CML) in both males and females.

Upper extremity function Upper extremity function and the relation with AGEs was studied by Shah et al. 50 in groups with diabetes mellitus type 2 (DM2) (n = 26) and without DM2 (n = 26). Upper extremity disability was measured using the Disability of the Arm, Shoulder and Hand (DASH) self- report questionnaire. The DASH has 30 questions, including disability and pain. The scores were calculated for a range between 0% and 100%, where a higher number indicates more impairments. Skin tissue fluorescent AGEs levels were assessed utilizing an AGE reader. The

29 Chapter 2 (4) for (+ adjusted for (+ adjusted c c b c c (1) (2) (3) for (+ adjusted c c c b c c b c (+ adjusted for for (+ adjusted c c Risk for slow walking speed with high AGEs speed with high AGEs walking slow Risk for level: OR=1.60 (95%CI:1.02-2.52)* level: disability risk with high AGEs Walking HR= 1.68 (95%CI: 1.11-2.52)* HR= 1.63 (95%CI: 1.06-2.49)* level: AGEs and strength flexor Correlation r = 0,07 NS disability and extremity Upper Correlation level: AGEs r = 0,51* AGEs function and extremity Upper Correlation level: r =-0.44* elevation - humerothoracic r =-0.32 NS rotation external - glenohumeral ADL disability risk with high AGEs level: ADL disability risk with high AGEs HR= 1.08 (95%CI: 1.03-1.13)* Main findings and Outcome statistics: Main findings and Outcome DM2) OR=1.87 (95%CI:1.15-3.04)* HR= 1.10 (95%CI: 1.05-1.15)* HR= 1.05 (95%CI: 1.01-1.11)* HR= 1.24 (95%CI: 1.07-1.45)* HR= 1.10 (95%CI: 0.92-1.32) NS. HR= 1.06 (95%CI: 0.90-1.24) NS. HR= 0.91 (95%CI: 0.77-1.07) NS. arthritis, cognition, kidney disease) cognition, kidney arthritis, in level risk with high AGEs frailty Physical women: HR= 1.06 (95%CI: 0.91-1.23) NS. disease) cognition, kidney arthritis, HR= 1.56 (95%CI: 1.04-2.36)* HR= 1.54 (95%CI: 1.04-2.29)* HR= 1.46 (95%CI: 0.95-2.23) NS. disease) cognition, kidney arthritis, in men: level risk with high AGEs frailty Physical HR= 1.30 (95%CI: 1.14-1.48)* Walking speed (cut- Walking for slow value off <0.79m/s) walking disability Walking (inability or slow speed) for value (cut-off <0.4 walking slow m/sec) 1. Upper extremity disability 2. Upper extremity function Physical Physical performance and functioning outcome - - ADL disability frailty Physical characteristics: ≥3 of following handgrip strength 1. Low loss weight 2. Unintentional activity physical 3. Low 4. Exhaustion speed walking 5. Slow Shoulder flexor strength Musculoskeletal Musculoskeletal outcome Circulating plasma CML Circulating for high level value (cut-off = 424 ng/mL) serum CML Circulating for high level value (cut-off = 689.1 ng/mL) serum CML Circulating for high level value (cut-off = 620 ng/mL) (skin not defined AGE-type tissue autofluorescent AGEs) Cross-linking AGE /RAGE AGE Circulating/Tissue levels, Tissue type n = 26 (13M, 13F) Community dwelling older Community dwelling adults n = 944 (416M, 528F), aged 75±7 yrs severe to Moderately disabled community n = 394, aged F, dwelling 76±8 yrs older Community dwelling adults n = 3373 (1344M, 2029F) 78.1±4.8 yrs aged n = 26 (13M, DM2 patients 64,5±6,8 13F) aged Non DM 64,2±5,8 aged Population and co- gender n =, age, morbidity Cross-sectional Longitudinal study observational with 30 month follow-up Longitudinal observational with 14 year study ADL for follow-up disability. Cross-sectional frailty physical for components Cross- sectional Design 50 46 Study characteristics 45 47 Semba et al., Semba et 2010 Sun et al., 2012 Sun et Whitson et al., Whitson et 2014 Shah et al., 2015 Shah et Table 3. Table author/ First of publication year

30 The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function

c c

c 2 Group difference high vs. low AGEs level: vs. low high difference Group 20.1 (6.2) kg,18.2 (6.4) kg vs. 20.0 kg, -1.88 (SE=0.65)* 18.6 kg vs. Beta -1.31 (SE= 0.61)* Beta AGEs level: and strength grip Correlation r = -0,54* Association UMM and AGEs level: UMM and Association -0.11 NS r = -0.11 NS Beta AGEs level: LMM and Association -0.18* r = -0.21* Beta AGEs level: RSMI and Association -0.27* r = -0.18* Beta Group difference high vs low AGEs level: vs low high difference Group 1. Muscle (handgrip)strength: 41.7 (95%CI: 40.3- 43.1) kg vs. 44.5 (95%CI: 43.2- 45.9) kg* ES = 0.45 ES power: 2. Muscle (leg extension) 16.0 (95%CI: 14.9- 17.1) W/kg vs. 17.8 (95%CI: 16.6- 19.1) W/kg* = 0.44 ES - - idem + multivariate adjusted for potential confounders. (1) All confounders. for potential adjusted idem + multivariate - c Handgrip strength Handgrip strength 1. Upper extremity 1. Upper extremity muscle mass extremity 2. Lower muscle mass skeletal 3. Relative muscle mass index 1. Handgrip strength 2.Leg extension power multivariate adjusted for demographics, for demographics, adjusted multivariate b Circulating serum Circulating CML for high level value (cut-off = 0.68 mg/mL) muscle tissue Skeletal CML/RAGE Circulating serum Circulating pentosidine AGE-type not defined (skin not defined AGE-type tissue autofluorescent AGEs) Cross-linking for high level value (cut-off = 2.09–4.44 AF) n = 133, n =21 a Moderately to severe severe to Moderately disabled community n = 559, aged F, dwelling 76±8 yrs M, Healthy maintainers, - Weight n = 10, 41±4 yrs aged n = 7, gainers, - Weight 42±5 yrs aged n = 4, -Elderly, 67±2 yrs aged aged 66,8 ± 9,5 yrs aged Healthy M, aged 46 yrs 46 yrs M, aged Healthy (37-56) F, DM2 patients, - Grip strength analysis, analysis, - Grip strength n = 232 analysis, - Leg extension n = 138 Cross- sectional Cross- Cross-sectional Cross-sectional Cross- sectional 43 49 44 48 Median (interquartile range), ES: effect size, OR: odds ratio, HR: hazard risk, hazard HR: ratio, OR: odds size, effect ES: range), Median (interquartile AGE = advanced glycation end-product; CML = carboxymethyl-lysine; DM2 = diabetes mellitus type 2; F = female; M = male; NS = not significant; RAGE = receptor for advanced glycation glycation for advanced receptor RAGE = female; M = male; NS not significant; mellitus type 2; F = DM2 = diabetes carboxymethyl-lysine; CML = end-product; glycation = advanced AGE Momma et al., Momma et 2011 Tanaka et al., et Tanaka 2015 a Dalal et al., 2009 Dalal et al., et De La Maza 2008 end-products; yrs = years lower extremity muscle mass, RSMI: relative skeletal muscle mass index. * P< 0,05. women with missing data had multiple simulations to impute walking disability status prior to death. (2) Inverse probability weighting method. (3) All women with missing data treated as treated data women with missing method. (3) All weighting probability Inverse (2) to death. prior status walking disability to impute had multiple simulations with missing data women muscle mass, LMM: UMM: upper extremity death, disability prior to is, no walking that as censored, treated with missing data (4) All women death. disability prior to walking developing

31 Chapter 2

AGE reader autofluorescence (AF) was correlated to the DASH scores (r = 0.51, P = 0.009) indicating that higher AGE levels correlate with increased disability of arm, shoulder or hand. They also studied the three-dimensional humerothoracic and glenohumeral joint motion and found that AGE reader levels were negatively correlated to humerothoracic elevation (r = -0.44, P = 0.024), but not correlated to glenohumeral external rotation (r = -0.32,P = 0.13).

Relationship of AGEs with Musculoskeletal Outcome

Muscle strength and power The contribution of AGEs on the muscle properties strength and power was described in four studies 43,44,49,50. Dalal et al.43 examined handgrip strength and circulating CML concentrations in moderately to severely disabled, older females (n = 559). The quartile cut-offs for serum CML were 0.45, 0.55, and 0.68 mg/mL. Women in the highest quartile of CML experienced decreased mean grip strength compared with women in the lower three quartiles, 18.6 kg. vs. 20.0 kg, respectively, with a Beta of -1.88 (SE = 0.65, P = 0.004). After adjusting for covariates (age, race, BMI, cognition, depression, and DM2), a difference was reported of 18.2 (6.4) kg. vs. 20.1 (6.2) kg, respectively, with a Beta of -1.31 (SE = 0.61, P = 0.03). The evidence of tissue CML/RAGE in skeletal muscle (internal abdominal oblique) biopsies of healthy subjects (n = 21) differing in age and weight stability was studied in the cross-sectional study by de la Maza et al.49. The subjects (n = 21) were divided into a group of self-reported weight maintainers (WM) (n = 10); self-reported weight gainers (WG) (n = 7); and elderly (E) (n = 4). The study ascertained that handgrip strength was diminished in subjects with an increase of CML/RAGE in muscle tissue and reported a negative correlation (r = -0.54, P <0.05) between tissue CML and handgrip strength.

The relationship between AGEs and grip strength and leg extension power was described in the study by Momma et al.44. In their study, healthy adult males (n = 370) were divided into two groups, i.e., one group (n = 232) for analysing grip strength and another group (n = 138) for leg extension strength. Skin tissue fluorescent AGEs levels were assessed utilizing an AGE reader. The tertile cut-off ranges for the AGE reader autofluorescence (AF) were 1.28 - 1.84, 1.84 - 2.09 and 2.09 - 4.44. After adjusting for potential confounders (age, BMI, physical activity, smoking/drinking status, depression, education, occupation, total energy consumption, metabolic syndrome, DM2, kidney disease, and dietary intake of vitamin C), it was discovered that grip strength was 6.3% less for those in the highest tertile than for subjects in the lowest tertile of skin AF (P = 0.01). Leg extension power was 10 % lower for those in the highest tertile than for the lowest tertile of AF (P = 0.04). Shoulder flexor muscle strength and the relation with AGEs was described in the study by Shah et al.50 in groups with DM2 (n = 26) and without DM2 (n = 26). Skin tissue fluorescent AGEs

32 The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function

were assessed utilizing an AGE reader. The study revealed that AGE reader autofluorescence (AF) was not related to shoulder flexor muscle strength (r = 0.07, P = 0.7).

Muscle mass Loss of muscle mass and the relation with serum levels of Pentosidine was studied in the 2 study by Tanaka et al.48 In this study among postmenopausal women with DM2 (n = 133) upper and lower extremity muscle mass was measured by Dual-energy x-ray absorptiometry and also the relative skeletal muscle mass index (RSMI= appendicular skeletal muscle mass/height2) was calculated. They showed that serum Pentosidine levels were negatively correlated with muscle mass of the lower extremity (r = -0.21, P = 0.0017) and RSMI (r = -0.18, P = 0.039), but not with the muscle mass of the upper extremity (r = -0.11, P = 0.219). After correction for covariates (age, DM2 duration, serum creatinine, glycosylated haemoglobin (HbA1c) and insulin-like growth factor-1 (IGF-1)) they reported that serum Pentosidine levels were associated with RSMI with a Beta of -0.27 (P = 0.008) and muscle mass of the lower extremity with a Beta of -0.18 (P = 0.071).

Relationship of AGEs with physical frailty Physical frailty in relation to CML was reported by Whitson et al.47. Frailty was based on the five criteria previously described by Fried el al.8: less physical activity, self-reported exhaustion, decreased grip strength, slow walking speed, and unintentional weight loss. Participants with three or more criteria were classified as physically frail. In this cross-sectional study serum, CML was measured in older, healthy participants (n = 3373). Self-reported exhaustion was identified by evaluation of two statements of the Centre for Epidemiological Studies Depression Scale (CES-D) and was regarded positive if at least one condition was evident for three or more days during the previous week. The decreased physical activity criterion was positive if physical activity per week was <383 kcal/week for men and <270 kcal/week for women. Slow walking speed and inadequate grip strength were defined as the lowest 20% of the population. Finally, unintentional weight loss was defined as unintentional loss of 4.5 kg in the year prior to the current evaluation or unintentional weight loss of at least 5% of the previous year’s body weight 7. AGE levels were compared between groups with or without presence of the specific frailty components. Mean CML levels were significantly higher in men for low physical activity (P = 0.007), exhaustion (P <0.001) and decreased grip strength (P = 0.04), resulting in a significant association with the physical frailty phenotype (OR = 1.30, 95% CI: 1.14 - 1.48, P < 0.001). After adjusting for potential confounders (demographics, BMI, DM2, alcohol consumption, smoking status, total cholesterol, albumin, weight change, hypertension, heart disease, stroke, and claudication), a significant association remained (OR = 1.24, 95% CI: 1.07 -1.45, P = 0.04). The associations lost significance after further adjustment for arthritis, cognition, and kidney function. No significant associations between

33 Chapter 2

CML and low physical activity, exhaustion, and grip strength were determined for females. Furthermore, in men and women with a slow walking speed and unintentional weight loss, CML levels were higher, however, these differences in CML levels were not significant.

DISCUSSION

The results of this systematic review indicate that higher levels of AGEs are independently related to declined walking abilities, inferior ADL, decreased muscle properties (strength, power, mass) and increased physical frailty. The available literature on musculoskeletal outcomes 43,44,48,49 support the hypothesis that high AGEs levels are associated with a decline in muscle function. However, the correlations, Beta coefficients and calculated effect sizes indicate only a moderate relationship. It is known that AGEs can affect muscle function through a variety of pathways. In fact, AGEs can alter the biomechanical properties of muscle tissue, increasing stiffness and reducing elasticity through cross-linking and upregulated inflammation by RAGE binding and endothelial dysfunction in the intra-muscular microcirculation 43–46,49,51. This is also consistent with studies on sarcopenia in which decreased muscle mass and strength is explained by an overall increase in inflammatory burden 6. Examining the studies in this review that report decline in walking abilities, it is suggested by the authors that this decline is also attributed to the effects of AGEs on muscle tissue, thereby impairing muscle function 45,46. It has been considered that impaired muscle function - through AGEs-induced muscle damage – can contribute to decline in walking abilities and ADL and can also contribute to physical frailty. It remains ambiguous to what extend ADL involving the upper extremity are affected by high AGEs level. One small study 50 reports a relation with upper extremity disability, but further we could not identify which ADL are affected most by AGEs and if at all are related to upper extremity function. Thereby in a large study 47 the reported increased risk in inferior ADL is described as the association between baseline AGEs level and the time to first difficulty on any of the six self-reported ADL items and, although significant, the risk is low (HR = 1.10), therefore the practical importance is unclear. Furthermore our review shows that the decline of handgrip strength is positively associated with higher AGE levels in three studies 43,44,49. Interestingly, in the upper extremity shoulder muscles no association with high AGE levels is found in two studies 48,50. The underlying mechanism for this difference in the upper extremity remains unclear and further research is necessary on this topic.

We identified only 8 studies due to strict inclusion criteria for several reasons; 1) We restricted the inclusion to individual studies that reports a direct relationship between AGEs and functioning which enabled us to calculate effect sizes, association coefficients etc.

34 The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function

for direct comparison of the strength of the association. 2) To increase internal validity we were specifically interested in motor functioning so we excluded papers measuring solely an effect on tissue levels. 3) Also we did not include papers that only cover a specific disease (such as Diabetes) to increase external validity. As mentioned previously, AGEs have an effect on different types of human tissues 24,26,31,32. It must be considered that other AGEs- 2 induced effects, such as effects on the nervous system, ligaments, tendons or joint capsules, could contribute to the decline of physical performance and functions. Future research should, therefore, explore other possible underlying AGEs related pathways, not only on the pathophysiological changes on the tissue level, but also with a direct relation on motor function. The current systematic review shows that the relation between AGEs and motor function is not as strong as most narrative reviews suggest. Although the quality of the included studies was good to moderate, they employed a cross-sectional or observational design, therefore, a causal relationship cannot be inferred. A meta-analysis was not appropriate because of the heterogeneity of the studies.

It is known that, in addition to cognitive decline, accompanying decline in motor function is frequently reported in patients with dementia 52 and it is suggested that AGEs formation may explain many biochemical and neuropathological changes in the most common cause of dementia 13. However, the results from this review provide no conclusive evidence for a peripheral or a central level effect of AGEs on motor function in Alzheimer’s disease or other types of dementia. We were unable to locate any study describing the effect of AGEs on the CNS with a direct relation on motor function. It remains unclear if AGEs accumulate in specific relevant motor-related brain regions, having their effect on the complex inter- relationship between the distributed motor networks within the CNS as well as with the musculoskeletal structures for generating movement. Future research is required to determine the contribution of AGEs accumulation on the CNS with a direct relation on the decline in motor function.

It is important to realise that, in this review, decline in motor function was primarily associated with elevated CML levels. Association with circulating CML was determined in four studies 43,45–47, and a relation with tissue CML was found in one study 49. One study reported an association with Pentosidine 48 and two other studies with non-specified skin tissue fluorescent AGEs 44,50. It is suggested that fluorescent and non-fluorescent AGEs such as CML behave similarly and fluorescence may be employed as a marker for the total skin tissue AGEs pool53 . Although CML is a dominant AGE in blood circulation and correlates with other AGEs 14, it is possible that the association between AGEs and motor function outcome could be different if cross-linking AGEs such as Pentosidine were assessed.

35 Chapter 2

In five studies, only females 43,46,48 or males 44,49 were included, therefore, the reported decline in motor function with elevated AGEs cannot necessarily be extrapolated to the other gender. Previous longitudinal studies conducted in a large cohort suggest that the relation between serum AGEs and health outcomes may be gender specific as it was ascertained that elevated levels of serum AGEs were associated with all-cause mortality, cardiovascular disease, and coronary heart disease mortality only in women 54,55. In contradiction, Whitson et al.47 found a significant cross-sectional association between CML and physical activity, exhaustion, and muscle strength as components of physical frailty only among men and not among women. A possible explanation suggested by Whitson et al. is that aged cohorts may be subject to gender-specific survivor bias. They state that the relative significance of CML mediated factors in determining the risk of death is greatest for females prior to menopause but greater for men in advanced ages. If AGEs have more effect on women, higher-risk women may then be unlikely to participate in studies that measure CML late in life due to the negative health outcomes 47. On the other hand, they did find an association with men and women in their longitudinal study on ADL disability. In a comparable cohort, Semba et al.45 reported a strong association between men and women and do not report any significant gender differences. Therefore, this raises an additional question of whether or not the effect of AGEs could be gender specific. Future research should, therefore, include gender-based differences in the effects of AGEs.

The vast majority of participants included in this review were elderly people older than 64 years (n = 5433). Interestingly, in two studies44,49 , the participants were middle-aged between 37 and 56 years (n = 387). This indicates that the negative effect of AGEs on motor function already begins during midlife and, as AGEs levels increase with aging, could be an important factor in age related decline in motor function. A high AGEs level, as a biomarker, therefore, could predict a decline in motor function later in life. This could also imply that preventive interventions should start as soon as possible as part of healthy aging. In accordance with the results of this review, it would be interesting to investigate whether motor function can be improved by reducing AGEs levels. Intensive glycaemic control may be a method to decrease AGEs formation. CML levels correlate to dietary consumption 14, therefore, dietary intake is a possible factor that can be influenced. It is suggested that, in order to lower daily AGEs intake, foods rich in sugar and fat and those prepared by frying or grilling should be avoided 14,56. However, evidence of the harmful effects of long-term exposure to dietary AGEs is still inconclusive 15. Improvements of glycaemic control by regular physical exercise could also attenuate the formation and accumulation of AGEs11,21,57 . Elevated AGEs levels might be an indication to initiate (early) treatment such as dietary advice, muscle strengthening exercises, and functional training to maintain physical functions. However, literature regarding the effects of physical exercise on AGEs formation is minimal, and the

36 The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function

optimal exercise modalities remain ambiguous21 . Further research should focus on whether dietary modifications, exercise programs, or medication to reduce AGEs levels can prevent or counter decline in motor function.

CONCLUSION 2

This review indicates that decline in motor function is independently associated with high AGEs levels. Further longitudinal observational and controlled trial studies are necessary to investigate a causal relationship, and to what extent, high AGEs levels are a contributing risk factor and potential biomarker for decline in motor function as a component of the aging process.

37 Chapter 2

REFERENCES

1. Arvanitakis Z, Wilson RS, Schneider JA, Bienias JL, Evans DA, Bennett DA. Diabetes mellitus and progression of rigidity and gait disturbance in older persons. Neurology. 2004;63(6):996-1001. 2. Bellew JW, Symons TB, Vandervoort AA. Geriatric fitness: effects of aging and recommendations for exercise in older adults. Cardiopulm Phys Ther J. 2005;16(1):20-31. 3. Buchman AS, Bennett DA. Loss of motor function in preclinical Alzheimer’s disease. Expert Rev Neurother. 2011;11(5):665-676. 4. Rikkert MGO, Rigaud AS, van Hoeyweghen RJ, de Graaf J. Geriatric syndromes: medical misnomer or progress in geriatrics? Neth J Med. 2003;61(3):83-87. 5. Sehl ME, Yates FE. Kinetics of human aging: I. Rates of senescence between ages 30 and 70 years in healthy people. J Gerontol A Biol Sci Med Sci. 2001;56(5):B198-208. 6. Bautmans I, Puyvelde K Van, Mets T. Sarcopenia and functional decline: pathophysiology, prevention and therapy. Acta Clin Belg. 2009;64(4):303-316. 7. U.S. Department of Health and Human Services: Healthy People 2000 Review.; 1999. 8. Fried LP, Tangen CM, Walston J, et al. Frailty in older adults: evidence for a phenotype. journals Gerontol A, Biol Sci Med Sci. 2001;56(3):M146-56. 9. Sternberg SA, Schwartz AW, Karunananthan S, Bergman H, Clarfield AM. The identification of frailty: a systematic literature review. J Am Geriatr Soc. 2011;59(11):2129-2138. 10. Sprott RL. Biomarkers of aging and disease: introduction and definitions. Exp Gerontol. 2010;45(1):2-4. 11. Ahmed N, Thornalley PJ. Advanced glycation endproducts: what is their relevance to diabetic complications? Diabetes, Obes Metab. 2007;9(3):233-245. 12. Monnier VM, Mustata GT, Biemel KL, et al. Cross-linking of the extracellular matrix by the maillard reaction in aging and diabetes: an update on “a puzzle nearing resolution.” Ann N Y Acad Sci. 2005;1043:533-544. 13. Rahmadi A, Steiner N, Munch G. Advanced glycation endproducts as gerontotoxins and biomarkers for carbonyl-based degenerative processes in Alzheimer’s disease. Clin Chem Lab Med. 2011;49(3):385-391. 14. Uribarri J, Cai W, Peppa M, et al. Circulating glycotoxins and dietary advanced glycation endproducts: two links to inflammatory response, oxidative stress, and aging.journals Gerontol A, Biol Sci Med Sci. 2007;62(4):427-433. 15. Puyvelde K Van, Mets T, Njemini R, Beyer I, Bautmans I. Effect of advanced glycation end product intake on inflammation and aging: a systematic review. Nutr Rev. 2014;72(10):638-650. 16. Luevano-Contreras C, Chapman-Novakofski K. Dietary advanced glycation end products and aging. Nutrients. 2010;2(12):1247-1265. 17. Peppa M, Uribarri J, Vlassara H. Aging and glycoxidant stress. Hormones (Athens). 2008;7(2):123- 132. 18. Gasser A, Forbes JM. Advanced glycation: implications in tissue damage and disease. Protein Pept Lett. 2008;15(4):385-391. 19. Munch G, Westcott B, Menini T, Gugliucci A. Advanced glycation endproducts and their pathogenic roles in neurological disorders. Amino Acids. 2012;42(4):1221-1236. 20. Hobbelen JS, Tan FE, Verhey FR, Koopmans RT, de Bie RA. Prevalence, incidence and risk factors of paratonia in patients with dementia: a one-year follow-up study. Int Psychogeriatr. 2011;23(7):1051-1060. 21. Magelhaes PM, j. Appell H, Duarte JA. Involvement of advanced glycation end products in the pathogenesis of diabetic complications: the protective role of regular physical activity. Eur Rev Aging Phys Act. 2008;(5):17-29.

38 The Contribution of Advanced Glycation End-Product (AGE) Accumulation to the Decline in Motor Function

22. Kjaer M. Role of extracellular matrix in adaptation of tendon and skeletal muscle to mechanical loading. Physiol Rev. 2004;84(2):649-698. 23. Avery NC, Bailey AJ. Enzymic and non-enzymic cross-linking mechanisms in relation to turnover of collagen: relevance to aging and exercise. Scand J Med Sci Sports. 2005;15(4):231-240. 24. Takahashi M. Pentosidine, an advanced glycation endproduct, and arthritis.Curr Rheumatol Rev. 2006;2(4):319-324. 2 25. Verzijl N, DeGroot J, Ben ZC, et al. Crosslinking by advanced glycation end products increases the stiffness of the collagen network in human articular cartilage: a possible mechanism through which age is a risk factor for osteoarthritis. Arthritis Rheum. 2002;46(1):114-123. 26. Verzijl N, Bank RA, TeKoppele JM, DeGroot J. AGEing and osteoarthritis: A different perspective. Curr Opin Rheumatol. 2003;15(5):616-622. 27. Karim L, Tang SY, Sroga GE, Vashishth D. Differences in non-enzymatic glycation and collagen cross-links between human cortical and cancellous bone. Osteoporos Int. 2013;24(9):2441- 2447. 28. Saito M, Marumo K, M. S, K. M. Collagen cross-links as a determinant of bone quality: A possible explanation for bone fragility in aging, osteoporosis, and diabetes mellitus. Osteoporos Int. 2010;21(2):195-214. 29. Sroga GE, Vashishth D. Effects of bone matrix proteins on fracture and fragility in osteoporosis. Curr Osteoporos Rep. 2012;10(2):141-150. 30. Schwartz A V, A.V. S. Diabetes Mellitus: Does it Affect Bone? Calcif Tissue Int. 2003;73(6):515- 519. doi:10.1007/s00223-003-0023-7. 31. Lindsey JB, Cipollone F, Abdullah SM, McGuire DK. Receptor for advanced glycation end-products (RAGE) and soluble RAGE (sRAGE): cardiovascular implications. Diab Vasc Dis Res. 2009;6(1):7- 14. 32. Schmidt AM, Yan SD, Yan SF, Stern DM. The biology of the receptor for advanced glycation end products and its ligands. Biochim Biophys Acta. 2000;1498(2-3):99-111. 33. Abate M, Schiavone C, Salini V, Andia I. Management of limited joint mobility in diabetic patients. Diabetes, Metab Syndr Obes Targets Ther. 2013;6:197-207. 34. Atienzar P, Abizanda P, Guppy A, Sinclair AJ. Diabetes and frailty: An emerging issue. Part 1: Sarcopaenia and factors affecting lower limb function. Br J Diabetes Vasc Dis. 2012;12(3):110- 116. 35. Lebiedz-Odrobina D, Kay J. Rheumatic Manifestations of Diabetes Mellitus.Rheum Dis Clin North Am. 2010;36(4):681-699. 36. Milaneschi Y, Tanaka T, Ferrucci L, Y. M, T. T, L. F. Nutritional determinants of mobility. Curr Opin Clin Nutr Metab Care. 2010;13(6):625-629. 37. Semba RD, Nicklett EJ, Ferrucci L. Does accumulation of advanced glycation end products contribute to the aging phenotype? journals Gerontol A, Biol Sci Med Sci. 2010;65(9):963-975. 38. Toth C, Martinez J, Zochodne DW. RAGE, diabetes, and the nervous system. Curr Mol Med. 2007;7(8):766-776. 39. Wrobel JS, Najafi B. Diabetic foot biomechanics and gait dysfunction. J Diabetes Sci Technol. 2010;4(4):833-845. 40. EBRO. Dutch Cochrane Center. EBRO Assessment tool of Cohort Studies. available at http://dcc. cochrane.org/beoordelingsformulieren-en-andere-downloads. 41. Rosnow RL, Rosenthal R. Computing contrasts, effect sizes, and counternulls on other people’s published data: General procedures for research consumers. Pyschological Methods,. 1996;(1):331-340. 42. Cohen J. Statistical Power Analysis for the Behavioural Sciences. 2nd ed. Hillsdale: Erlbaum; 1988.

39 Chapter 2

43. Dalal M, Ferrucci L, Sun K, et al. Elevated serum advanced glycation end products and poor grip strength in older community-dwelling women. Journals Gerontol - Ser A Biol Sci Med Sci. 2009;64(1):132-137. 44. Momma H, Niu K, Kobayashi Y, et al. Skin advanced glycation end product accumulation and muscle strength among adult men. Eur J Appl Physiol. 2011;111(7):1545-1552. 45. Semba RD, Bandinelli S, Sun K, Guralnik JM, Ferrucci L. Relationship of an advanced glycation end product, plasma carboxymethyl-lysine, with slow walking speed in older adults: the InCHIANTI study. Eur J Appl Physiol. 2010;108(1):191-195. 46. Sun K, Semba RD, Fried LP, Schaumberg DA, Ferrucci L, Varadhan R. Elevated Serum Carboxymethyl-Lysine, an Advanced Glycation End Product, Predicts Severe Walking Disability in Older Women: The Women’s Health and Aging Study I. J Aging Res. 2012;2012:586385. 47. Whitson HE, Arnold AM, Yee LM, et al. Serum carboxymethyl-lysine, disability, and frailty in older persons: the Cardiovascular Health Study. journals Gerontol A, Biol Sci Med Sci. 2014;69(6):710- 716. 48. Tanaka K, Kanazawa I, Sugimoto T. Elevated Serum Pentosidine and Decreased Serum IGF-I Levels are Associated with Loss of Muscle Mass in Postmenopausal Women with Type 2 Diabetes Mellitus. Exp Clin Endocrinol Diabetes. 2015. 49. Maza MPD La, Uribarri J, Olivares D, et al. Weight increase is associated with skeletal muscle immunostaining for advanced glycation end products, receptor for advanced glycation end products, and oxidation injury. Rejuvenation Res. 2008;11(6):1041-1048. 50. Shah KM, Clark BR, McGill JB, Lang CE, Maynard J, Mueller MJ. Relationship Between Skin Intrinsic Fluorescence--an Indicator of Advanced Glycation End Products--and Upper Extremity Impairments in Individuals With Diabetes Mellitus. Phys Ther. 2015;95(8):1111-1119 9p. doi:10.2522/ptj.20140340. 51. Haus JM, Carrithers JA, Trappe SW, et al. Collagen, cross-linking, and advanced glycation end products in aging human skeletal muscle. J Appl Physiol. 2007;103(6):2068-2076. 52. Prehogan A, Cohen CI. Motor dysfunction in dementias. Geriatrics. 2004;59(11):53-54-60. 53. Meerwaldt R, Graaff R, Oomen PH, et al. Simple non-invasive assessment of advanced glycation endproduct accumulation. Diabetologia. 2004;47(7):1324-1330. 54. Kilhovd BK, Juutilainen A, Lehto S, et al. High serum levels of advanced glycation end products predict increased coronary heart disease mortality in nondiabetic women but not in nondiabetic men: a population-based 18-year follow-up study.Arterioscler Thromb Vasc Biol. 2005;25(4):815- 820. 55. Kilhovd BK, Juutilainen A, Lehto S, et al. Increased serum levels of advanced glycation endproducts predict total, cardiovascular and coronary mortality in women with type 2 diabetes: a population-based 18 year follow-up study. Diabetologia. 2007;50(7):1409-1417. 56. Prasad C, Imrhan V, Marotta F, Juma S, Vijayagopal P. Lifestyle and Advanced Glycation End Products (AGEs) Burden: Its Relevance to Healthy Aging. Aging Dis. 2014;5(3):212-217. 57. Couppé C, Svensson R, Grosset J, et al. Life-long endurance running is associated with reduced glycation and mechanical stress in connective tissue. Age (Omaha). 2014;36(4):9665. doi:10.1007/s11357-014-9665-9.

Chapter3

Advanced Glycation End-Products are Associated with Physical Activity and Physical Functioning in the Older Population

Hans Drenth, Sytse U. Zuidema, Wim P. Krijnen, Ivan Bautmans, Andries J. Smit, Cees van der Schans, Hans Hobbelen

Journals of Gerontology; Series A Biological Sciences and Medical Sciences 2018, Apr 28. Chapter 3

ABSTRACT

Background: Decline in physical activity and functioning is commonly observed in the older population and might be associated with biomarkers such as advanced glycation end-products (AGEs). AGEs contribute to age-related decline in the function of cells and tissues in normal aging and have been found to be associated with motor function decline. The aim of this study is to investigate the association between the levels of AGEs, as assessed by skin autofluorescence, and the amount of physical activity and loss of physical functioning in older participants.

Methods: Cross-sectional data of 5,624 participants aged 65 years and older from the LifeLines cohort study was used. Linear regression analyses were utilized to study associations between skin autofluorescence/AGE-levels (AGE reader), the number of physically active days (SQUASH), and physical functioning (RAND-36), respectively. A logistic regression analysis was used to study associations between AGE-levels and the compliance with the Dutch physical activity guidelines (SQUASH).

Results: A statistical significant association between AGE levels and the number of physically active days (β = -0.21, 95% confidence interval: -0.35 to -0.07, P = .004), physical functioning (β = -1.60, 95% confidence interval: -2.64 to -0.54, P = .003), and compliance with the Dutch physical activity guidelines (OR = 0.76, 95% confidence interval: 0.62 to 0.94, P = .010) was revealed.

Conclusions: This study indicates that high AGE levels may be a contributing factor as well as a biomarker for lower levels of physical activity and functioning in the older population.

44 Advanced Glycation End-Products are Associated with Physical Activity and Physical Functioning in the Older Population

INTRODUCTION

In the aging population, decline in physical activity and functioning is commonly observed. Physical activity is generally defined as any skeletal muscle effort resulting in more energy being used than when at rest; physical functioning is defined as being able to perform activities of daily life (ADL) 1,2. With aging, most human physiologic systems regress independently from substantial disease effects at an average linear loss rate of 0.34- 1.28% per year between the age of 30 and 70 years 3. Regular moderate physical activity 3 has an advantageous influence on health status and can reduce the risk on and improve the prognosis of chronic diseases such as diabetes mellitus (DM) and cardiovascular (CV) disease 4,5. A lower level and accelerated decline of physical functioning, such as gait, has been determined to predict the subsequent development of mild cognitive impairment and Alzheimer’s disease and can precede cognitive impairment by several years 6. A lack of physical activity is a known precipitating factor for the age-related loss of muscle mass (sarcopenia) leading to strength losses and physical disability 7.

A decline in motor function, such as decreased muscle properties, declined walking abilities and declined activities of daily living have been found to be associated with advanced glycation end-products (AGEs) in the aging population8 . AGEs accumulate in hyperglycaemic environments and contribute to the age-related decline of the functioning of cells and tissues in normal aging 9,10. In many age related diseases, the accumulation of AGEs is a significant contributing factor in degenerative processes, especially in renal failure, CV diseases, DM, and Alzheimer’s disease 9,11. The formation of AGEs is mediated by non- enzymatic condensation of a reducing sugar with proteins and is accelerated during not only glycaemic but also oxidative stress9,10,12 . It is suggested that AGEs alter organ properties including the biomechanical properties of muscle tissue which leads to impaired muscle function through collagen cross-linking and/or upregulated inflammation by the binding of AGEs to their receptor 13–15. Increasing levels of AGEs are also determined by the exogenous intake of AGEs that are spontaneously generated in standard diets 16. AGEs are removed from the body through enzymatic clearance and renal excretion. It has been proposed that, with aging, there is an imbalance between the formation and natural clearance of AGEs which results in an incremental accumulation in tissues with slow turnover such as muscles, cartilage, tendons, eye lens, vascular media, and the dermis of the skin while blood levels have fewer changes 17,18. AGEs can be biochemically quantified in blood or tissue biopsies but, due to their fluorescent properties, their presence in the dermis of the skin can be noninvasively assessed using skin autofluorescence (SAF).

45 Chapter 3

The AGEs induced tissue damage negatively affects motor function (e.g., muscle function, walking impairment) and may influence the amount of physical activity. While improvements of glycaemic control by regular physical activity or exercise are suggested to attenuate the formation and accumulation of AGEs it is currently unclear if the accumulation of AGEs is also a contributor to a loss of physical activity 10,19. Although, AGEs have been found to be associated with declined motor function, these studies are few in number, and none studied the association between AGEs and physical activity in a large sample.

The aim of this study is to investigate the association between AGE levels, as assessed by SAF, and the amount of physical activity and loss of physical functioning in older participants.

METHODS

Design and Study Population

The cross-sectional data from the LifeLines Cohort Study were used. In brief, the LifeLines Cohort Study is a large population-based cohort study and biobank that was established as a resource for research on phenotypic, genomic, and environmental factors interacting between the development of chronic diseases and healthy aging20 . Between 2006 and 2013, inhabitants of the northern part of the Netherlands were invited to participate. Eligible participants were invited to participate in the LifeLines Cohort Study through their general practitioner, unless the participating general practitioner considered thepatient not eligible based on the following criteria: severe psychiatric or physical illness, limited life expectancy (<5 years), insufficient knowledge of the Dutch language to complete a Dutch questionnaire. Participants visited one of the LifeLines research centers for a physical examination and additional measurements such as AGE assessment and cognition tests. They also completed extensive questionnaires. Baseline data were collected for 167,729 participants ranging in ages from 6 months to 93 years, with 7.6% being 65 years and older21 . For this study, we utilized the data of the LifeLines participants who were 65 years and older and who had complete SAF-AGE level measurements. All of the participants provided written informed consent. The LifeLines Cohort Study is conducted according tothe principles of the Declaration of Helsinki and is approved by the medical ethical committee of the University Medical Center Groningen, the Netherlands (M12.113965). Additional details on the LifeLines study were described previously 20.

46 Advanced Glycation End-Products are Associated with Physical Activity and Physical Functioning in the Older Population

Outcome Measures

Physical activity Data on physical activity were extracted from the LifeLines database, which was assessed with the Short Questionnaire to Asses Health-Enhancing Physical Activity (SQUASH) 22. The SQUASH is a valid and reliable instrument and contains questions about the amount of time a participant has spent on physical activity at work, housework, leisure activities, and sports activities. Each of the 11 physical activity items consists of three main questions: 3 the number of days spent per week, average time per day, and intensity. The total scores from the SQUASH are used to calculate the average number of physically active days per week and to estimate whether a participant complies with the Dutch Physical Activity (DPA) guidelines, meaning a desired moderately intensive activity for 30 minutes at least 5 days a week.

Physical functioning Data on physical functioning were extracted from the LifeLines database which was assessed with the physical functioning section of the RAND-36 questionnaire 23 that comprises 10 questions regarding daily activities such as walking, stair climbing, lifting groceries, washing, and dressing. End scores are established by transforming the raw scores into a scale ranging from 0 to 100. A high score represents that the participant can perform strenuous activities (such as sports). Participants with low scores are severely restricted in performing all activities including washing and dressing. The RAND-36 is a valid and reliable questionnaire with a high internal consistency 23.

AGE levels AGE levels were assessed by measuring SAF using an AGE Reader device (Diagnoptics, Groningen, the Netherlands). The AGE reader measures fluorescent skin tissue AGEs and is reported as being a reliable and valid instrument for the quantification of AGEs accumulation24 . The AGE reader is a desktop device that has a light source which illuminates the skin of the forearm and uses the fluorescent properties of AGEs to measure tissue accumulation of AGEs 24. The AGE reader software calculates the SAF as the ratio between the emission light and the excitation light, multiplied by 100, and expressed in arbitrary units (AU). An elevated SAF score corresponds to a high tissue AGEs level 24. All AGE reader measurements were performed with the participants in a seated position and the volar side of the forearm placed on top of the AGE reader. The measurements were performed on the skin without sweat, skin lotions, or visible skin abnormalities. The mean of three consecutive measurements was used. SAF values were not used in this study when skin reflection was below 10% because pigmentation influences SAF measurement thereby excluding people with a skin-type of IV- VI on the Fitzpatrick scale 25.

47 Chapter 3

Other variables Gender, age, history of DM, CV disease, chronic pulmonary disease, kidney disease, cancer, smoking status, and alcohol consumption 5,6,9,10,13,19,24,26–29 were assessed by questionnaire. Participants were regarded as having a history of cardiovascular disease if they reported having had a history of stroke, heart attack, thrombosis, hypertension, heart failure, or atherosclerosis. Pulmonary disease was defined as a history of asthma or chronical obstructive pulmonary disease. If the participants had stopped smoking, were currently smoking, or smoked in the past month, they were considered smokers. Alcohol consumption was classified as drinking alcoholic beverages less than 1 day per week or 1 day or more days per week. Cognitive function was measured with the Mini-Mental State Examination (MMSE) 30 which is an 11-item questionnaire with a score of 0 to 30 ( with higher scores representing better cognitive function). Glucose levels and body mass index (BMI)were determined as described in the LifeLines protocol20.

Statistical Analysis

Study population characteristics are categorized in tertile groups of SAF-AGE levels. Differences between SAF-AGEs tertiles (low SAF ≤ 2.19 AU, middle SAF: 2.19 > < 2.56 AU and high SAF ≥ 2.56 AU) were tested using the analysis of variance (ANOVA) for continuous and chi-square tests for categorical variables. To estimate the association of AGEs with physical activity (active days per week) and physical functioning (total score) multiple linear regression analysis was used. To estimate the association of AGEs with the binary outcome on compliance with the DPA guidelines multiple logistic regression analysis was used. Each analysis started with several known potential confounders; gender, age, DM, CV, pulmonary and kidney disease, cancer, MMSE, BMI, glucose level, smoking status, and alcohol consumption 5,6,9,10,13,19,24,26–29 , as well with physical activity in models with physical functioning as the response variable and vice versa. Because of the growing evidence of gender-differences in factors associated with physical activity and functioning31 and gender- based differences in the effects of AGEs 8, additional gender-AGE interaction analyses were performed on all models. Backward manual selection was utilized to identify statistically significant explanatory variables. During this process the variables AGE levels, gender and age were always retained. Missing data was handled through pairwise deletion. Testing for inflation factors indicated that multicollinearity was not of concern. Analyses were conducted using the SPSS software, version 22 for Windows, and a P value <0.05 was considered statistically significant in two-sided tests.

48 Advanced Glycation End-Products are Associated with Physical Activity and Physical Functioning in the Older Population

RESULTS

Study Population Characteristics

Out of a number of 167,729 participants in the LifeLines study, 12,685 (7.6%) were 65 years and older. A SAF-AGE-level measurement was performed for 5,925 participants (46.7%) of the older subpopulation in the LifeLines study. Unfortunately, 301 persons (5.1%) had to be excluded due to skin reflection value less than 10% which resulted in 5,624 participants 3 with a mean (SD) SAF-AGE level of 2.41 (0.48) for analysis. The number of participants with complete data on the response variables SQUASH and RAND-36 was 4,202 and 4,641 with both a mean age (SD) of 69.4 (4.2) years and a mean (SD) SAF-AGE level of 2.39 (0.47) and 2.40 (0.47), respectively. Missingness on the SQUASH and RAND-36 were 25% and 17%, respectively. These participants had a mean age (SD) of 71.0 (5.1) and 71.5 (5.3) years and a mean (SD) SAF-AGE level of 2.47 (0.52) and 2.49 (0.53) respectively (see Appendix

Table 1. Characteristics of the participants according to the tertiles of AGE levels Tertiles of AGE levels

n Low Middle High P Value SAF ≤ 2.19 AU SAF 2.19 > < 2.56 AU SAF ≥ 2.56 AU

Participants, n 5,624 1,874 1,874 1,876 AGE levels (SAF) 5,624 1.94 (0.19) 2.37 (0.10) 2.94 (0.37) Female, n (%) 3,054 1197 (63.9) 997 (53.2) 860 (45.8) <.001 Age, y 5,624 69 (3.9) 69.6 (4.3) 70.7 (4.9) <.001 Medical history (yes) Diabetes, n (%) 498 95 (5.1) 149 (8.0) 254 (13.5) <.001 CV disease, n (%) 2,445 746 (39.8) 787 (42.0) 912 (48.6) <.001 Kidney disease, n (%) 48 13 (0.7) 10 (0.5) 25 (1.3) .012 Pulmonary disease, n (%) 653 174 (9.3) 189 (10.1) 290 (15.5) <.001 Cancer, n (%) 815 254 (13.6) 271 (14.5) 290 (15.5) .222 Glucose, mmol/L 5,566 5.31 (0.85) 5.42 (1.08) 5.61 (1.28) <.001 BMI 5,620 26.76 (3.71) 27.16 (3.79) 27.72 (4.05) <.001 Smoking, n (%) 3,227 946 (50.5) 1076 (57.4) 1205 (64.2) <.001 Alcohol, ≥ 1 day a week, n (%) 3,041 1092 (58.3) 984 (52.5) 965 (51.4) .003 MMSE, score 0-30 a 5,585 27.80 (2.32) 27.62 (2.42) 27.41 (2.60) <.001 SQUASH Physical active days, score 0-7 a 4,202 4.94 (2.10) 4.71 (2.26) 4.53 (2.35) <.001 DPA guidelines (Yes), n (%) 3,743 1,325 (70.7) 1,253 (66.9) 1,165 (62.1) <.001 RAND-36. score 0-100 a 4,641 84.70 (16.49) 83.34 (17.93) 80.73 (20.27) <.001 AGE = advanced glycation end-product; AU = arbitrary units i.e., the output units of the AGE reader; BMI = body mass index; CV = cardiovascular; DPA = Dutch Physical Activity; MMSE = Mini Mental State Examination; SAF = skin autofluorescence (AGE reader). Data represent mean values (SD) unless indicated otherwise. a. High score indicates better performance

49 Chapter 3

for more details). Table 1 provides the characteristics of the participants categorized into groups according to the tertiles of SAF-AGE levels; 54.3% females and 45.7% males with a mean (SD) age of 69.8 (4.5) years. One-way ANOVA and chi-square tests revealed significant differences in means for all covariates among the SAF-AGE level tertile groups withthe exception of some specific cancer subgroups.

Association Between AGE Levels and Physical Activity (Number of Physical Active Days)

Participants, on average, were physically active for (SD) 4.73 (2.24) days per week.The high AGE level group showed less active days per week compared to the low-level group. The linear regression model showed that after correcting for all potentially confounding variables the number of physically active days was significantly associated with higher AGE levels (β = -0.19, 95% confidence interval [CI]: -0.34 to -0.05, P = .009). Table 2 shows, after adding gender-AGEs interaction to the model, that the number of physically active days was significantly associated with higher AGE levels (β = -0.30, 95% CI: -0.50 to -0.10, P = .003). The interaction term between AGEs and gender was not found significant, suggesting insufficient evidence for difference in the relationship of AGEs by genderon the number of physical active days. Backward selection on linear regression indicated that, after correcting for gender, age, CV disease, BMI, cognition (MMSE) and physical functioning (RAND-36), the number of active days was lower for participants with higher AGE levels (β = -0.21, 95% CI: -0.35 to -0.07, P = .004).

Association Between AGE Levels and Physical Activity (Compliance With theDPA Guidelines)

The percentage of participants who complied with the DPA guidelines was 66.6%. The mean (SD) AGE levels of the group that did and did not comply with the DPA guidelines were 2.38 (0.48) and 2.48 (0.53) AU, respectively. The logistic regression model showed, after correcting for all potentially confounding variables compliance with DPA guidelines was lower in participants with higher AGE levels (odds ratio = 0.76, 95% CI: 0.62 to 0.94,P = .013). Table 2 shows, after adding gender-AGEs interaction to the model, that the compliance with DPA guidelines was lower in participants with higher AGE levels (odds ratio = 0.74, 95% CI: 0.57 to 0.96, P = .025). The interaction term between AGEs and gender was not found significant, suggesting insufficient evidence for difference in the relationship of AGEs by gender on the compliance with the DPA guidelines. Backward selection on logistic regression indicated that, after correcting for gender, age, glucose, BMI, and physical functioning (RAND-36), compliance with DPA guidelines was lower in participants with higher AGE levels (odds ratio = 0.76, 95% CI: 0.62 to 0.94, P = .010).

50 Advanced Glycation End-Products are Associated with Physical Activity and Physical Functioning in the Older Population

Table 2. Association between AGE levels and Physical activity (SQUASH) Number of physical active days Compliance with the DPA guidelines (SQUASH)a (SQUASH)b

Unstandardized 95% CI P value Odds Ratio 95% CI P value

Beta lower upper lower upper limit limit limit limit

Model 1 (n = 4,177) Model 1 (n = 4,104) (Constant) 4.54 2.79 6.30 <.001 38.18 .021 3 AGE levels -0.30 -0.50 -0.10 .003 0.74 0.57 0.96 .025 Gender (male) -0.18 -0.87 0.51 .608 1.27 0.42 3.85 .675 Age (years) -0.01 -0.02 0.01 .554 0.97 0.95 0.99 .008 CV disease (yes) -0.25 -0.39 -0.11 <.001 0.93 0.75 1.15 .494 DM (yes) -0.02 -0.31 0.26 .877 0.95 0.64 1.41 .800 Pulmonary disease (yes) 0.16 -0.05 0.37 .134 1.29 0.93 1.78 .129 Kidney disease (yes) -0.14 -0.79 0.51 .676 1.35 0.45 4.07 .597 Cancer (yes) -0.13 -0.32 0.05 .162 1.07 0.79 1.43 .675 MMSE (0-30) 0.06 0.03 0.09 <.001 1.05 0.98 1.13 .191 Smoking (yes) -0.07 -0.21 0.07 .325 0.97 0.78 1.21 .776 Alcohol (yes) 0.13 -0.02 0.28 .090 1.04 0.83 1.31 .711 BMI -0.05 -0.07 -0.03 <.001 0.95 0.93 0.98 <.001 Glucose (mmol/L) -0.04 -0.11 0.04 .363 0.90 0.81 1.00 .049 Physical functioning 0.01 0.01 0.02 <.001 1.02 1.02 1.03 <.001 (RAND-36, 0-100) Gender*AGEs level 0.22 -0.06 0.50 .122 1.09 0.71 1.68 .699

Model 2 (n = 4,190) Model 2 (n = 4,172) (Constant) 4.43 2.73 6.14 <.001 181.43 <.001 AGE levels -0.21 -0.35 -0.07 .004 0.76 0.62 0.94 .010 Gender (male) 0.36 0.22 0.50 <.001 1.61 1.29 2.00 <.001 Age (years) -0.01 -0.02 0.01 .414 0.97 0.95 0.99 .007 CV disease (yes) -0.26 -0.40 -0.12 <.001 - - - - MMSE (0-30) 0.06 0.03 0.09 <.001 - - - - BMI -0.06 -0.08 -0.04 <.001 0.95 0.93 0.98 <.001 Glucose (mmol/L) - - - - 0.89 0.82 0.97 .006 Physical functioning 0.01 0.01 0.02 <.001 1.02 1.02 1.03 <.001 (RAND-36, 0-100)

AGE = advanced glycation end-product; BMI = body mass index; CI = confidence interval; CV = cardiovascular; DM = diabetes mellitus; DPA = Dutch Physical Activity; MMSE = Mini Mental State Examination. Model 1: Full model with gender*AGEs level interaction. Model 2: Obtained after removing statistically insignificant variables (retaining AGEs level and biological variables; age and gender). Missing response variables; SQUASH: 25%, RAND- 36: 17%, missing predictors; kidney disease:18%, alcohol:18%, other(range): 0% - 1%. a Linear regression analysis. b Logistic regression analysis.

51 Chapter 3

Association Between AGE Levels and Physical Functioning

Physical functioning, measured by the RAND-36, was lower for participants with higher AGE levels. The linear regression model showed that after correcting for all potentially confounding variables that physical functioning was lower in participants with higher AGE levels (β = -1.46, 95% CI: -2.51 to -0.40, P = .007). Table 3 shows that when adding the gender-AGEs interaction variable to the model, its size appeared not to be significant. Adding this interaction term into the model resulted also in becoming not statistically significant of

Table 3. Association between AGE levels and physical functioning (RAND-36) Physical functioning (RAND-36)

Unstandardized 95% CI P value Beta lower limit upper limit

Model 1 (n = 4,177) (constant) 160.99 149.16 172.82 <.001 AGE levels -1.20 -2.63 2.29 .100 Gender (male) 8.00 2.96 13.04 .002 Age -0.73 -0.85 -0.62 <.001 CV disease -3.69 -4.71 -2.67 <.001 DM -3.47 -5.55 -1.40 .001 Pulmonary disease -9.18 -10.70 -7.68 <.001 Kidney disease (yes) -4.26 -9.01 0.50 .079 Cancer (yes) -0.05 -1.42 1.32 .941 MMSE (0-30) 0.10 -0.11 0.31 .352 Smoking (yes) -0.94 -2.00 1.14 .800 Alcohol (yes) 3.32 2.25 4.38 <.001 BMI -1.19 -1.32 -1.06 <.001 Glucose (mmol/L) 0.43 -0.51 0.60 .878 Number of active days 0.74 0.52 0.96 <.001 Gender*AGEs level -0.54 -2.57 1.50 .604

Model 2 (n = 4,182) (constant) 165.01 156.31 173.87 <.001 AGE levels -1.60 -2.64 -0.54 .003 Gender (male) 6.46 5.45 7.48 <.001 Age -0.74 -0.85 -0.63 <.001 CV disease -3.64 -4.63 -2.65 <.001 DM -3.42 -5.16 -1.68 <.001 Pulmonary disease -9.24 -10.75 -7.73 <.001 Alcohol (yes) 3.32 2.25 4.38 <.001 BMI -1.19 -1.32 -1.06 <.001 Number of active days 0.74 0.52 0.96 <.001 AGE = advanced glycation end-product; BMI = body mass index; CI = confidence interval; CV= cardiovascular; DM = diabetes mellitus; MMSE = Mini Mental State Examination. Linear Regression analysis. Model 1: Full model with gender*AGEs level interaction. Model 2: Obtained after removing statistically insignificant variables (retaining AGEs level and biological variables; age and gender). Missing response variables; SQUASH: 25%, RAND-36: 17%, missing predictors; kidney disease: 18%, alcohol:18%, other(range): 0% - 1%

52 Advanced Glycation End-Products are Associated with Physical Activity and Physical Functioning in the Older Population

the association between AGEs and physical functioning (β = -1.20, 95% CI: -2.63 to 2.29, P = .100). However, without the interaction term, after the backward selection on linear regression correcting for age, gender, DM, CV- and pulmonary disease, alcohol status, BMI, and the number of physically active days (SQUASH), the association between AGEs and physical functioning was statistically significant, indicating that physical functioning was lower in participants with higher AGE levels (β = -1.60, 95% CI: -2.64 to -0.54, P = .003).

DISCUSSION 3

We found evidence that AGEs, as assessed by SAF, are associated with lower physical activity and physical functioning in older individuals. The revealed associations were consistently determined considering the presence of various independent variables for several measurements of physical activity and physical functioning. This study indicated that, in those individuals with one unit of AGE increase, the number of active days per week was 21% of a day less, and the risk of not complying with the DPA guidelines increased by 24%. Reference values for AGE levels in healthy people can be described as 0.024 x person’s age + 0.83 (R2 = 60%) 32. For those over 70 years, reference values are unknown but an enhanced increase in AGE levels may be expected because they may develop age related diseases 32. In individuals with early-stage dementia, an AGE level increase of 10 times as much as normal has been found after 1 year33 . AGE formation from a reversible to an irreversible end product usually takes weeks to months, but for AGE levels to increase by 0.3 AU is a process that generally takes 10 years in normal aging 10,32,34. Considering this, the revealed β coefficients on physical activity and functioning appear low, but as AGE formation is accelerated in age related diseases such as DM and AD, they may become relevant. Notwithstanding that this study provides evidence for a relationship between higher AGE levels and lower physical activity and functioning, other factors, such as intrinsic motivation, access to activities/exercise, or pain may play a role. Further research over several years is necessary to improve insight in the long-term effects of AGEs on physical activity and physical function in older people.

The results of this study provide partial evidence that AGE formation and accumulation contributes to motor function decline and consequently to decline in the amount of physical activity. Physical activity is in turn considered to be effective for maintaining healthor preventing functional decline and disability; however, the exact underlying physiological pathways remain unclear 35,36. Physical activity can be an intervention for reducing AGE formation in order to improve physical functioning and thus essential to healthy aging. Previous studies have shown that individuals who are regularly physically active have, on average, lower AGE levels than those that are hardly or not physically active 37–39. On the

53 Chapter 3

other hand, as mentioned previously, AGE accumulation is known to affect lung, cardiovascular, and musculoskeletal tissue which could have a negative impact on physical activity. Therefore, whether the association between high AGE levels and a decline in physical activity exists because of AGEs induced damage of relevant tissues, whether loss of physical activity influences AGEs accumulation, or both, remains to be determined. Due to the temporal bivariate relationship, we could not show this in this cross-sectional analysis since physical activity is also limited by declined physical functioning. Future research with follow-up assessments is necessary to study the size of effect of physical activity on AGE accumulation, adjusted for physical functioning impairment, as a proof of concept underpinning the causal relationship of AGEs and physical activity.

The relationship between AGEs and poorer physical functioning corresponds with studies describing the effect of AGEs on walking abilities and ADL and contributes to the increasing evidence that AGE accumulation is associated with functional decline8 . Although it has been suggested that AGE induced impaired musculoskeletal function is a contributor to functional decline 8,13, it must also be considered that AGE accumulation in the central nervous system(CNS) may hamper physical functioning. AGEs have been shown to be associated with less grey matter volume and to accumulate in brain tissue 26. Also, elevated levels of SAF were associated with a decline in cognitive performance26 . AGE accumulation in specific relevant motor-related brain regions may possibly affect the complex inter-relationship between the motor networks within the CNS as well as with the musculoskeletal structures. Future research is required to determine the contribution of AGEs accumulation on the CNS with a direct relationship on the decline in physical functioning.

Our analyses show a statistical significant gender effect on physical activity and physical functioning, which is in line with the scientific literature on this topic 31, 40. Although is it suggested that the effect of AGEs could be gender specific 8, we could not confirm this in our study. Future longitudinal research is necessary to study gender-based differences in the effects of AGEs on physical activity and physical functioning in depth.

Strength and Limitations

This study is one of the few to investigate the association between AGEs and physical activity and physical functioning in a large sample of older individuals. This study also has a number of limitations. First, this is a cross-sectional study; therefore, a causal relationship cannot be inferred. Second, outcome measures on physical activity and physical functioning were established by valid and reliable questionnaires and not with physical measurements in the LifeLines study. This may have resulted in an underestimation of our results. Third, although

54 Advanced Glycation End-Products are Associated with Physical Activity and Physical Functioning in the Older Population

the response to the invitation to participate was comparable in size to other large-scale population cohort studies 20, it cannot be completely ruled out that healthy, active, older people were preferentially included because of their capabilities to visit the research site. Expanding the population under study with participants who are less active would broaden the range of outcome and probably result in a larger effect. Future studies should take that into account and include frail and/or cognitive impaired people. Finally, a critical point on the current study pertains to the missing cases not completely ad random potentially causing some bias in the findings. The various analysis of the data based upon pair-wise as 3 well as list-wise deletion, the representativeness of the sample judged by the proportions and age range sustains the generalizability of our conclusions. Also, those missing SQUASH and RAND-36 were older than those not missing data, and also had a higher AGEs level. This might suggest that our results underestimate the true effect of AGE levels due to missing data. It seems, however, clear that the final word is to future research of similar or larger size cohorts to confirm our findings from settings in e.g. other countries. In conclusion, this study indicates that high AGE levels may be a contributing factor as well as a biomarker for lower physical activity and functioning in older adults. Further longitudinal observational and controlled intervention studies with physical activities are necessary to investigate a causal relationship.

55 Chapter 3

Appendix. Characteristics of the participants according to non-missing vs. missing data on physical activity (SQUASH) and functioning (RAND-36), respectively.

SQUASH RAND-36 non-missing missing P Value non-missing missing P Value Participants, n 4,202 1,422 4,641 983 AGE levels (SAF) 2.39 (0.47) 2.47 (0.52) <0.001 2.40 (0.47) 2.49 (0.53) <0.001 Female, n (%) 2,306 (54.9) 748 (52.6) 0.136 2,493 (53.7) 561 (57.1) 0.056 Age, years 69.4 (4.2) 71.0 (5.1) <0.001 69.4 (4.2) 71.5 (5.3) <0.001 Medical history (yes, %) Diabetes, n (%) 346 (8.2) 152 (10.7) 0.003 388 (8.4) 110 (11.2) 0.002 CV disease, n (%) 2,102 (50.0) 343 (24.1) <0.001 2,328 (50.2) 117 (11.9) <0.001 Kidney disease, n (%) 37 (0.9) 11 (0.8) 0.001 48 (1.0) - Pulmonary disease, n (%) 465 (11.1) 188 (13.5) 0.015 518 (11.2) 135 (13.7) 0.008 Cancer, n (%) 613 (14.6) 202 (14.2) 0.901 680 (14.7) 135 (13.7) 0.678 Glucose, mmol/L 5.41 (1.06) 5.55 (1.17) <0.001 5.42 (1.07) 5.71 (1.17) <0.001 BMI 27.06 (3.84) 27.66 (3.94) <0.001 27.11 (3.85) 27.71 (3.94) <0.001 Smoking, n (%) 2,507 (59.7) 720 (50.6) <0.001 2,757 (59.4) 470 (47.8) <0.001 Alcohol, ≥ 1 day a week, n (%) 2,764 (65.8) 277 (19.5) 0.990 3,039 (65.6) 2 (0.2) 0.308 MMSE, score 0-30 28.49 (1.33) 24.95 (3.06) <0.001 28.47 (1.34) 23.43 (2.31) <0.001 AGE = advanced glycation end-product; AU = arbitrary units i.e., the output units of the AGE reader; BMI = Body Mass Index; CV = cardiovascular; MMSE = Mini Mental State Examination; SAF = skin autofluorescence (AGE reader) Data represent mean values (SD) unless indicated otherwise.

56 Advanced Glycation End-Products are Associated with Physical Activity and Physical Functioning in the Older Population

REFERENCES

1. Caspersen CJ, Powell KE, Christenson GM. Physical activity, exercise, and physical fitness: definitions and distinctions for health-related research. Public Health Rep. 1985;100(2):126- 131. 2. Jette AM. Toward a common language for function, disability, and health. Phys Ther. 2006;86(5):726-734. doi:10.1093/ptj/86.5.726. 3. Sehl ME, Yates FE. Kinetics of human aging: I. Rates of senescence between ages 30 and 70 years in healthy people. J Gerontol A Biol Sci Med Sci. 2001;56(5):B198-208. doi: 10.1093/ gerona/56.5.B198 3 4. Lee I-M, Shiroma EJ, Lobelo F, Puska P, Blair SN, Katzmarzyk PT. Effect of physical inactivity on major non-communicable diseases worldwide: an analysis of burden of disease and life expectancy. Lancet (London, England). 2012;380(9838):219-229. doi:10.1016/S0140-6736(12)61031-9. 5. Warburton DER, Nicol CW, Bredin SSD. Health benefits of physical activity: the evidence.CMAJ . 2006;174(6):801-809. doi:10.1503/cmaj.051351. 6. Buchman AS, Bennett DA. Loss of motor function in preclinical Alzheimer’s disease. Expert Rev Neurother. 2011;11(5):665-676. doi: 10.1586/ern.11.57. 7. Bautmans I, Puyvelde K Van, Mets T. Sarcopenia and functional decline: pathophysiology, prevention and therapy. Acta Clin Belg. 2009;64(4):303-316. doi:10.1179/acb.2009.048. 8. Drenth H, Zuidema S, Bunt S, Bautmans I, van der Schans C, Hobbelen H. The Contribution of Advanced Glycation End product (AGE) accumulation to the decline in motor function. Eur Rev Aging Phys Act. 2016;13:3. doi:10.1186/s11556-016-0163-1. 9. Rahmadi A, Steiner N, Munch G. Advanced glycation endproducts as gerontotoxins and biomarkers for carbonyl-based degenerative processes in Alzheimer’s disease. Clin Chem Lab Med. 2011;49(3):385-391. 10. Ahmed N, Thornalley PJ. Advanced glycation endproducts: what is their relevance to diabetic complications? Diabetes, Obes Metab. 2007;9(3):233-245. doi: 10.1515/CCLM.2011.079. 11. Gasser A, Forbes JM. Advanced glycation: implications in tissue damage and disease. Protein Pept Lett. 2008;15(4):385-391. doi: 10.2174/092986608784246515. 12. Monnier VM, Mustata GT, Biemel KL, et al. Cross-linking of the extracellular matrix by the maillard reaction in aging and diabetes: an update on “a puzzle nearing resolution.” Ann N Y Acad Sci. 2005;1043:533-544. doi: 10.1196/annals.1333.061. 13. Semba RD, Nicklett EJ, Ferrucci L. Does accumulation of advanced glycation end products contribute to the aging phenotype? journals Gerontol A, Biol Sci Med Sci. 2010;65(9):963-975. doi: 10.1093/gerona/glq074. 14. Payne GW. Effect of inflammation on the aging microcirculation: impact on skeletal muscle blood flow control. Microcirculation. 2006;13(4):343-352. doi:10.1080/10739680600618918. 15. Avery NC, Bailey AJ. Enzymic and non-enzymic cross-linking mechanisms in relation to turnover of collagen: relevance to aging and exercise. Scand J Med Sci Sports. 2005;15(4):231-240. doi: 10.1111/j.1600-0838.2005.00464.x. 16. Puyvelde K Van, Mets T, Njemini R, Beyer I, Bautmans I. Effect of advanced glycation end product intake on inflammation and aging: a systematic review. Nutr Rev. 2014;72(10):638-650. doi: 10.1111/nure.12141. 17. Luevano-Contreras C, Chapman-Novakofski K. Dietary advanced glycation end products and aging. Nutrients. 2010;2(12):1247-1265. 18. Peppa M, Uribarri J, Vlassara H. Aging and glycoxidant stress. Hormones (Athens). 2008;7(2):123- 132.

57 Chapter 3

19. Magelhaes PM, j. Appell H, Duarte JA. Involvement of advanced glycation end products in the pathogenesis of diabetic complications: the protective role of regular physical activity. Eur Rev Aging Phys Act. 2008;(5):17-29. doi: 10.1007/s11556-008-0032-7 20. Scholtens S, Smidt N, Swertz MA, et al. Cohort Profile: LifeLines, a three-generation cohort study and biobank. Int J Epidemiol. 2015;44(4):1172-1180. doi:10.1093/ije/dyu229. 21. Klijs B, Scholtens S, Mandemakers JJ, Snieder H, Stolk RP, Smidt N. Representativeness of the LifeLines Cohort Study. PLoS One. 2015;10(9):e0137203. doi:10.1371/journal.pone.0137203. 22. Wendel-Vos GCW, Schuit AJ, Saris WHM, Kromhout D. Reproducibility and relative validity of the short questionnaire to assess health-enhancing physical activity. J Clin Epidemiol. 2003;56(12):1163-1169. doi: 10.1016/S0895-4356(03)00220-8. 23. van der Zee KI, Sanderman R. Het Meten van de Algemene Gezondheidstoestand Met de RAND- 36, Een Handleiding. Groningen; 2012. 24. Meerwaldt R, Graaff R, Oomen PH, et al. Simple non-invasive assessment of advanced glycation endproduct accumulation. Diabetologia. 2004;47(7):1324-1330. doi: 10.1007/s00125-004- 1451-2. 25. Na R, Stender IM, Henriksen M, Wulf HC. Autofluorescence of human skin is age-related after correction for skin pigmentation and redness. J Invest Dermatol. 2001;116(4):536-540. doi:10.1046/j.1523-1747.2001.01285.x. 26. Spauwen PJJ, van Eupen MGA, Kohler S, et al. Associations of advanced glycation end-products with cognitive functions in individuals with and without type 2 diabetes: the maastricht study.J Clin Endocrinol Metab. 2015;100(3):951-960. doi:10.1210/jc.2014-2754. 27. Hoonhorst SJM, Lo Tam Loi AT, Hartman JE, et al. Advanced glycation endproducts and their receptor in different body compartments in COPD. Respir Res. 2016;10(9):481-490. doi:10.1016/j.metabol.2014.06.006. doi: 10.1016/j.metabol.2014.06.006. 28. van Waateringe RP, Slagter SN, van der Klauw MM, et al. Lifestyle and clinical determinants of skin autofluorescence in a population-based cohort study.Eur J Clin Invest. 2016;46(5):481-490. doi:10.1111/eci.12627. 29. Piperi C, Adamopoulos C, Papavassiliou AG. Potential of glycative stress targeting for cancer prevention. Cancer Lett. 2017;390:153-159. doi:10.1016/j.canlet.2017.01.020. doi: 10.1016/j. canlet.2017.01.020. 30. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189-198. doi: 10.1016/0022-3956(75)90026-6. 31. Pollard TM, Wagnild JM. Gender differences in walking (for leisure, transport and in total) across adult life: a systematic review. BMC Public Health. 2017;17(1):341. doi:10.1186/s12889-017- 4253-4. 32. Koetsier M, Lutgers HL, de Jonge C, Links TP, Smit a J, Graaff R. Reference values of skin autofluorescence.Diabetes Technol Ther. 2010;12(5):399-403. doi:10.1089/dia.2009.0113. 33. Drenth H, Zuidema SU, Krijnen WP, Bautmans I, van der Schans C, Hobbelen H. Advanced Glycation End-Products Are Associated With the Presence and Severity of Paratonia in Early Stage Alzheimer Disease. J Am Med Dir Assoc. 2017. doi:10.1016/j.jamda.2017.04.004. 34. Brownlee M. Advanced protein glycosylation in diabetes and aging.Annu Rev Med. 1995;46:223- 234. doi:10.1146/annurev.med.46.1.223. 35. Cooper R, Kuh D, Hardy R. Objectively measured physical capability levels and mortality: systematic review and meta-analysis. BMJ. 2010;341:c4467. doi: 10.1136/bmj.c4467. 36. Vogel T, Brechat P-H, Lepretre P-M, Kaltenbach G, Berthel M, Lonsdorfer J. Health benefits of physical activity in older patients: a review. Int J Clin Pract. 2009;63(2):303-320. doi:10.1111/ j.1742-1241.2008.01957.x.

58 Advanced Glycation End-Products are Associated with Physical Activity and Physical Functioning in the Older Population

37. Simon Klenovics K, Kollarova R, Hodosy J, et al. Reference values of skin autofluorescence as an estimation of tissue accumulation of advanced glycation end products in a general Slovak population. Diabet Med. 2014;31(5):581-585. doi:10.1111/dme.12326. 38. Couppé C, Svensson R, Grosset J, et al. Life-long endurance running is associated with reduced glycation and mechanical stress in connective tissue. Age (Omaha). 2014;36(4):9665. doi:10.1007/s11357-014-9665-9. 39. Yoshikawa T, Miyazaki A, Fujimoto S. Decrease in serum levels of advanced glycation end- products by short-term lifestyle modification in non-diabetic middle-aged females. Med Sci Monit. 2009;15(6):PH65-73. 40. Mottram S, Peat G, Thomas E, Wilkie R, Croft P. Patterns of pain and mobility limitation in older 3 people: cross-sectional findings from a population survey of 18,497 adults aged 50 years and over. Qual Life Res. 2008;17(4):529-539. doi:10.1007/s11136-008-9324-7.

Chapter 4

Advanced Glycation End-Products are Associated with the Presence and Severity of Paratonia in Early Stage Alzheimer Disease

Hans Drenth, Sytse U. Zuidema, Wim P. Krijnen, Ivan Bautmans, Cees van der Schans, Hans Hobbelen

JAMDA 2017, Jul 1; 18(7): 636.e7-636.e12. Chapter 4

ABSTRACT

Objective: Paratonia, a distinctive form of hypertonia in dementia patients, causes lossof functional mobility in early stage dementia to severe contractures and pain in the late stages. The pathogenesis of paratonia is not well understood. Patients in early stage dementia with Diabetes Mellitus (DM) showed a significantly higher risk for the development of paratonia. Both Alzheimer disease (AD) and DM are related to higher concentrations of advanced glycation end-products (AGEs). The purpose of this study is to explore the association of AGEs with the prevalence and severity of paratonia in patients with Alzheimer disease.

Design: Observational longitudinal, 1-year follow-up cohort study with 3 assessments.

Setting: Day care centers for patients with dementia.

Participants: A total of 144 community-dwelling patients with early stage Alzheimer or Alzheimer/ vascular disease patients were recruited from 24 dementia day care centers in The Netherlands.

Measurements: The presence of paratonia (Paratonia Assessment Instrument), the severity of paratonia (Modified Ashworth Scale for paratonia), and AGE levels (AGE-reader).

Results: From the 144 participants (56.3% female and 43.7% male, with a mean [standard deviation] age of 80.7 [7.7] years), 118 participants were available for final follow-up. A significant association between AGE levels and the presence of paratonia (odds ratio 3.47, 95% confidence interval [CI]: 1.87 – 6.44, P <.001) and paratonia severity (β = 0.17, 95% CI: 0.11 - 0.23, P <.001) was determined. In participants who developed paratonia and those with persistent paratonia throughout the study the AGE levels (95% CI: - 0.38 to - 0.13, P <.001 and 95% CI: -0.46 to - 0.06, P =.012, respectively) and the severity of paratonia (95% CI: -0.60 to - 0.35, P <.001 and 95% CI: -0.38 to - 0.12, P <.001, respectively) significantly increased, whereas the AGE levels remained stable in those participants without paratonia. Notwithstanding, change in AGE levels was not significantly (P =.062) related to change in paratonia severity, mixed model analyses provided evidence for both a significant time and between participant effect of AGEs on paratonia severity.

Conclusions: This study suggests that elevated AGE levels are a contributing factor to paratonia and its severity and could be the result of peripheral biomechanical changes reducing elasticity and increasing stiffness. These results provide a new perspective on paratonia and gives rise to further research whether paratonia could be postponed or movement stiffness can be improved by reducing AGE levels.

62 Advanced Glycation End-Products are Associated with the Presence and Severity of Paratonia in Early Stage Alzheimer Disease

INTRODUCTION

A wide variety of movement disorders are reported in patients with dementia. Paratonia, a distinctive form of hypertonia, is one of these movement disorders and is characterized by an active unintentional resistance against passive movement 1. Paratonia has a prevalence of 10% in the early/mild stages, and up to 90-100% in later/severe stages of dementia. As dementia progresses, the severity of paratonia increases from actively assisting the passive movement (because of an inability to relax) and mild resistance towards severe high resistance and muscle tone which causes loss of mobility, severe contractures, and pain in the last stage of dementia2,3 . Resistance in paratonia is variable, in particular in the very early stages of dementia paratonia can fluctuate between no resistance, actively assisting, and 4 active resistance against passive movement 2–4. In early stage dementia, paratonia hampers functional mobility such as raising from a chair, walking and turning 2. The pathogenesis of paratonia is not well understood, and no effective interventions are available to prevent, postpone or combat paratonia.

It has been shown that patients in early stage dementia with Diabetes Mellitus (DM) have a significantly higher risk for the development of paratonia in comparison with those with dementia but without DM 1. Previous research also reported that DM is a risk factor for muscle rigidity in patients without dementia 5. Interestingly, both Alzheimer disease (AD) and DM are related to higher concentrations of advanced glycation end-products (AGEs), suggesting that AGEs could possibly be involved in the development of paratonia. The occurrence of AGEs is mediated by non-enzymatic condensation of a reducing sugar with an amino group 6–8. With aging, there is an imbalance between the formation and natural clearance of AGEs, which results in to an incremental accumulation in tissues 6–9.

As stated before, the pathogenesis of paratonia is not well understood. Obviously with dementia, central nervous system pathology plays a role; yet, peripheral biomechanical changes are also suggested. In this perspective, it is of interest to know that the cross-linking processes by AGEs are responsible for an increasing proportion of insoluble extracellular matrix and thickening of the tissues, thereby increasing mechanical stiffness and loss of elasticity 6,10,11. Non-cross-linking effects occur through the binding of AGEs to the receptor for AGEs (RAGE). RAGE is a multi-ligand member of the immunoglobulin superfamily of cell surface molecules that is widely localized in a variety of cell lines, including endothelial, neuronal, smooth muscle, mesangial and monocytes 12. AGE/RAGE binding subsequently incites activation of intracellular signalling, gene expression, and production ofpro- inflammatory cytokines and free radicals. At the peripheral (tissue) level, these inflammatory processes exhibit strong proteolytic activity by which the collagen becomes more vulnerable

63 Chapter 4

and tissue elasticity decreases 12,13. At the central level (central nervous system), AGE/RAGE interaction appears to affect neuronal function 7,13. The role of AGEs in the development of paratonia is currently unknown. The purpose of this study is to explore the association of AGEs with the prevalence and severity of paratonia in patients with AD.

METHODS

Study Design and Ethical Consideration

This study was a multicenter, longitudinal, 1-year follow-up cohort study with 3 assessments; at baseline, after 6 months and after 12 months. The medical ethical committee ofthe University Medical Centre Groningen approved the study (NL43641.042.13).

Study Population

Participants, recruited from 24 dementia day care centers in The Netherlands, were considered to be eligible for inclusion in the study when they satisfied the following criteria: (1) an established diagnosis of AD or Alzheimer/vascular dementia (AD/VaD) according to DSM-IV criteria 14; (2) Early stage dementia; a score of stage 5 or lower on the global deterioration scale 15 and (3) having a light colored (Caucasian) skin (skin pigmentation influence AGE measurements 16). Written informed consent was obtained from patients or their legal representatives. Participants were excluded if they had an established diagnosis of dementia other than type AD or AD/VaD or were using first generation psychotropic drugs as these drugs can possibly mimic paratonic rigidity.

Outcome Measures

Paratonia The presence of paratonia was assessed by trained and experienced physiotherapists by employing the Paratonia Assessment Instrument (PAI), a reliable and valid measurement based on a successively passive mobilization of both shoulders towards ante-flexion/retro- flexion, elbows towards flexion/extension, and combined hips/knees towards extension/ flexion17 . With the participant in a sitting position the examiner began with a slow movement of the limb, after which the movement was accelerated. Paratonia was diagnosed as being present when the following five criteria are all met: (1) an involuntary variable resistance; (2) a degree of resistance that varied depending on the speed of the movement (e.g., a low resistance to slow movements and a high resistance to fast movement); (3) resistance to

64 Advanced Glycation End-Products are Associated with the Presence and Severity of Paratonia in Early Stage Alzheimer Disease

passive movement in any direction; (4) no clasp-knife phenomenon; and (5) resistance in two movement directions in the same limb or two different limbs. We used a Modified Ashworth Scale, validated to assess paratonia severity (MAS-P) as described by Waardenburg et al. 18. In the MAS-P each category contains a tone and passive movement feature, thereby creating a more consistent scale for paratonia 18. During the PAI assessment, the assessor quantified the muscle tone based on the resistance induced by the passive movements of the limbs into a total of 12 MAS-P scores (the corresponding movement directions from the PAI). The MAS-P is based on a 5-point scale ranging from 0 to 4, meaning 0 = no increase in muscle tone or no resistance during passive movement, 1 = slight increase in muscle tone or slight resistance during passive movement, 2 = more marked increase in muscle tone or more marked resistance during passive movement, 4 3 = considerable increase in muscle tone or considerable resistance during passive movement, and 4 = severe resistance such that passive movement is impossible. A score of 0.5 was assigned in the event of active assistance. For further analyses the 12 scores were summarized as the average MAS-P value. The MAS-P shows acceptable reliability 18.

AGE levels AGE levels were assessed by the main researcher through skin autofluorescence (SAF) by using an AGE Reader device (Diagnoptics, Groningen, The Netherlands). The AGE reader measures fluorescent skin tissue AGEs which correlate with non-fluorescent AGEs 19,20. The AGE reader is a desktop device that has a light source which illuminates the skin of the forearm and uses the fluorescent properties of AGEs to measure tissue accumulation of AGEs19 . SAF is calculated (by the AGE reader software) as the ratio between the emission light and the excitation light, multiplied by 100 and expressed in arbitrary units. An elevated SAF (arbitrary units) score corresponds to a high tissue AGEs level 19. All AGE reader measurements were performed at room temperature in a standardized semi dark environment with the participants in a seated position and the volar side of the right forearm placed on top of the AGE reader. The measurements were performed on the skin without sweat, skin lotions, or visible skin abnormalities and with the assessor being blinded for PAI/MAS-P scores. The mean of 3 consecutive measurements was used. The AGE reader is reported to be a reliable and valid instrument for the quantification of AGEs accumulation 19,20.

Cognitive function and comorbidities The diagnosis of dementia was established by the local physician or general practitioner based on the DSM-IV criteria. Dementia severity was staged by key staff personal using the global deterioration scale, which identifies 7 stages from no dementia to severe dementia15 . Cognitive function was measured by experienced psychologists or physicians with the Mini-Mental State Examination (MMSE) 21. The MMSE is an 11-item questionnaire with a

65 Chapter 4

maximum score of 30 which (30 = no cognitive decline; 0 = very severe cognitive decline). The comorbidities (International Classification of Diseases, Ninth Revision, Clinical Modification classification 22) and use of medication (Anatomical Therapeutic Chemical Classification System classification23 ) were retrieved from the patients’ medical records and general practitioner files. If a participant used 5 or more medications, this was labeled as polypharmacy 23.

Statistical Analyses

Sample size calculation Two a priori sample size calculations were undertaken. First, a mixed model sample size calculation 24 was based upon a 0.5 correlation between repeated measurements, 0.3 variance of the random intercept, 0.3 residual variance, a true effect size of 0.5, a desired power of 80%, and a two-sided alpha of 0.05. This resulted in a total sample size of 152. Considering this sample size, an additional calculation for the binary PAI outcomes ina logistic regression model 25 was performed to detect a true odds ratio (OR) = 1.65 with true binary probability 0.3 and a desired power of 80%. Accounting for a potential 10% withdrawal rate, the required total number of participants was approximately 165.

Calculation of MAS-P scores The degree in which the 12 MAS-P scores can be reliably summarized as an average value was investigated by the two-way intra class correlation (ICC) coefficient (type: average, two-way, agreement). A cut-off ICC value of .80, which indicates a strong correlation 26, was desired for the average MAS-P score. Furthermore, to what degree the 12 MAS-P measurements have a dominating direction of variation was investigated by principal component analysis.

Baseline characteristics and differences in characteristics between tertiles of AGE levels AGE levels were categorized in tertile groups and 1-way analysis of variance tests were performed for continuous data and χ2 tests for frequency data, with post hoc analysis (Tukey for 1-way analysis of variance), if relevant, to analyse differences in characteristics between the 3 AGE level groups.

Association between AGE levels and paratonia presence and severity To investigate the association between AGE levels and the presence of paratonia, a generalized linear mixed model was used, estimated by restricted maximum likelihood, and taking the presence of paratonia (PAI) as a response at each of the 3 visits (baseline, 6 months, 12 months). To investigate the association between AGE levels and the severity of paratonia, a linear mixed model (LMM) was used, estimated by restricted maximum

66 Advanced Glycation End-Products are Associated with the Presence and Severity of Paratonia in Early Stage Alzheimer Disease

likelihood, and taking the severity of paratonia measured by MAS-P as a response at each of the 3 visits (baseline, 6 months, 12 months). In both models, the association was statistically controlled for the (fixed) effects of sex, age, polypharmacy, dementia duration, cognition (MMSE), chronic kidney disease, cardiovascular disease, cerebral vascular disease and DM 8,27 as well for random effects of the participants. Backward model selection was usedfor detecting statistically significant explanatory variables. During this process, the fixed effects of AGE levels and visits were always retained to explore for longitudinal effects. To further investigate the within participant association of AGEs with paratonia severity (MAS-P), the differences in paratonia severity (MAS-P) and in AGEs, respectively, between Visit 3 and Visit 1 were computed separately for participants without paratonia, with incident paratonia, and persistent paratonia. To investigate changes in AGE levels and in paratonia 4 severity (MAS-P), paired sample t-tests were performed. Change in paratonia proportion was analyzed by χ2 tests. The association of these within-participant differences between AGEs and paratonia severity (MAS-P) were further investigated by correlation and simple regression. Data were analyzed using R version v3.2.0 and SPSS version v 22 and taking a P value of < .05 as statistically significant.

RESULTS

A flowchart of the enrolment of participants is presented in figure 1. A total of 144 participants were finally included at baseline. After 1 year, 26 participants (18%) were lost to follow-up: 11 had deceased and 15 were transferred to unknown addresses or became too ill to be reassessed (Figure 1). For comorbidity and medication use of 11% of the patients (n = 16), information was not accessible in the medical records.

Baseline Characteristics and Differences in Characteristics Between Tertiles of AGE Levels

Table 1 illustrates baseline characteristics of the participants according to the tertiles of AGE levels; 56.3% female and 43.7% male, with a mean (standard deviation [SD]) age of 80.7 [7.7] years. Participants had a mean (SD) dementia duration since diagnosis of 29.8 (35.9) months and had a mean (SD) of 1.18 (1.1) comorbidities. Paratonia was diagnosed in 60 participants (41.7%). The ICC for paratonia severity (MAS-P) scores of visit 1 to 3e was high (> .90), and principal component analysis indicated strong evidence that the MAS-P has 1 underlying construct so that the 12 MAS-P item scores were allowed to be entered as individual averages in the analyses.

67 Chapter 4

Figure 1. Study Flow Chart

Association Between AGE levels and Paratonia Presence and Severity

The presence (PAI) and severity (MAS-P) of paratonia were both significantly more evident in the highest AGE level tertile compared to the lowest tertile. Participants in the highest AGE level tertile demonstrated a substantially greater number of comorbidities and showed a significantly higher prevalence of DM (table 1).

The mixed model analyses reveal a significant time ffe ect as well as a significant regression type of association between AGEs and paratonia and AGES and paratonia severity (MAS-P). Table 2 shows that paratonia is significantly associated with the AGE levels (OR = 3.47, 95% CI: 1.87 - 6.44, P <.001) and with the progression over a time span of 1 year (OR = 2.89, 95% CI: 1.41 - 5.90, P = .004).

68 Advanced Glycation End-Products are Associated with the Presence and Severity of Paratonia in Early Stage Alzheimer Disease

Table 1. Baseline Characteristics and Differences in Characteristics of the Participants According to the Tertiles of AGE Levels

Tertiles of AGE levels Total Low Middle High P Value SAF ≤ 2.5 AU SAF 2.5 AU >< 3.1 AU SAF ≥ 3.1 AU

Participants, n (%) 144 52 (36,1%) 40 (27,8%) 52 (36.1%) .78 Male, n (%) 63 (43.7%) 16 (30.8%) 19 (47.5%) 28 (53.8%) .51 Age, years 80.7 (7.7) 79.9 (8.9) 80.5 (7.3) 81.5 (6.6) .56 Dementia duration, months* 29.8 (35.9) 35.9 (46.6) 30.36 (29.4) 22.90 (24.9) .24 Co morbidities, n* 1.18 (1.1) 0.9 (1.0) 1.15 (1.0) 1.51 (1.2) .018† Cardio vascular disease, n (%)* 41 (32.0%) 14 (11.0%) 13 (10.0%) 14 (11.0%) .55 Cerebral vascular Disease, n (%)* 22 (17.2%) 6 (4.7%) 5 (3.9%) 11 (8.6%) .26 Diabetes, n (%)* 37 (28.9%) 11 (8.6%) 6 (4.7%) 20 (15.6%) .020†, .015‡ 4 Cancer, n (%)* 14 (11.0%) 2 (1.7%) 3 (2.3%) 9 (7.0%) .015† COPD, n (%)* 14 (11.0%) 4 (3.1%) 6 (4.7%) 4 (3.1%) .30 Chronic kidney disease, n (%)* 11 (8.6%) 4 (3.1%) 2 (1.7%) 5 (3,9%) .72 Systemic, n (%)* 8 (6.3%) 3 (2.3%) 3 (2.3%) 2 (1.7%) .70 Digestive tract,n (%)* 6 (4.7%) 1 (0.8%) - 5 (3.9%) .048‡ Polypharmacy (≥ 5 medications), n (%)* 71 (55.5%) 25 (19.5%) 17 (13.3%) 29 (22.7%) .36 MMSE, score 0-30 19.4 (5.4) 19.14 (5.1) 19.10 (5.4) 19.94 (5.8) .69 GDS, score 1-7 3.84 (1.0) 3.85 (1.0) 3.93 (1.0) 3.77 (1.0) .76 Paratonia, n (%) 60 (41.7%) 13 (25%) 18 (45%) 29 (55.8%) .001† MAS-P, score 0-4 0.40 (0.40) 0.27 (0.26) 0.43 (0.35) 0.52 (0.49) .003† AGEs = advanced glycation end-products; AU = arbitrary units (ie, the output units of the AGE reader); COPD = chronic obstructive pulmonary disease; GDS = Global Deterioration Scale; MAS-P = modified Ashworth scale for paratonia; MMSE = Mini Mental State Examination; SAF = skin autofluorescence (AGE reader). Data represent mean values (SD) unless indicated otherwise. *: Based on GP medical files n=128. †: High vs low tertile. ‡: High vs middle tertile.

Table 2. Association between AGE levels and Paratonia Presence (PAI) Paratonia presence Estimate Odds Ratio 95% CI P value lower limit upper limit (Intercept) -4.16 0.02 0.00 0.10 <.001 AGE levels 1.24 3.47 1.87 6.44 <.001 Visit2 0.08 1.08 0.56 2.08 .819 Visit3 1.06 2.89 1.41 5.90 .004 GLMM, generalized linear mixed model. GLMM was obtained after removing statistically insignificant variables. Response variable: PAI (the presence of paratonia “no/yes”). Explanatory variable: AGE levels.

69 Chapter 4

Table 3 indicates that, after correction for MMSE, paratonia severity is significantly associated with the AGE levels (β = 0.17, 95% CI: 0.11 - 0.23, P <.001) and with the progression over a time span of 1 year (β = 0.18, 95% CI: 0.12 - 0.25, P <.001). To further investigate the degree in which the latter is due to within participant association, we computed respectively the difference in paratonia severity (MAS-P) and in AGEs between visit 3 and visit 1. These within participant difference were further investigated on degree of association and the regression of AGEs on paratonia severity (MAS-P).

Table 3. Association between AGE levels and Paratonia Severity (MAS-P) Paratonia severity Estimate 95% CI P value lower limit upper limit (Intercept) 0.30 0.04 0.56 .026 MMSE -0.02 -0.03 -0.01 <.001 AGE levels 0.17 0.11 0.23 <.001 Visit2 -0.01 -0.08 0.05 .576 Visit3 0.18 0.12 0.25 <.001 LMM, linear mixed model. LMM was obtained after removing statistically insignificant variables. Response variable: MAS-P. Explanatory variable: AGE levels.

The 1-year development of AGE levels and paratonia presence (PAI) and severity (MAS-P) is presented in Table 4. One hundred and eighteen participants (82%) completed this study, of whom 84 showed no paratonia at baseline. After a 1-year follow-up, 32 participants developed paratonia, resulting in a 1-year incidence of 22.2 %. Of the remaining participants, 40 exhibited no paratonia, 35 having persistent paratonia and 11 participants varied in paratonia status throughout the study.

In participants with paratonia throughout the study the AGE levels and severity (MAS-P) of paratonia increased. Participants who developed paratonia showed the highest increase in AGE levels and paratonia severity (MAS-P). AGE levels remained stable in participants without paratonia after 1-year follow up.

The difference in AGE levels change in 1 year between participants who developed paratonia and participants without paratonia was significant (95% CI: -0.40 to -0.04, P = .016). There was no significant difference in AGE levels change between participants with persistent paratonia and without paratonia nor between participants with persistent paratonia and participants who developed paratonia (P = .052 and P = .99, respectively). Changes in AGE levels between visits 3 and 1 were not significantly correlated (r = .17,P = .062) with changes

70 Advanced Glycation End-Products are Associated with the Presence and Severity of Paratonia in Early Stage Alzheimer Disease

Table 4. One-Year Development of AGE Levels and Paratonia Presence (PAI) and Severity (MAS-P) Visit 1 Visit 2 Visit 3 Difference Visit 1-Visit 3 95% CI lower limit upper limit

All participants Paratonia (%) 41.7 43.8 § 56.8* MAS-P 0.40 (0.40) 0.40 (0.38) † 0.58 (0.41) * -0.29 -0.14 AGE levels 2.8 (0.7) 2.9 (0.7) * 3.0 (0.7) * -0.25 -0.08 Participants with paratonia throughout the year (n=35) MAS-P 0.64 (0.32) 0.60 (0.43) † 0.89 (0.42) * -0.38 -0.12 AGE levels 3.1 (0.8) 3.2 (0.7) 3.4 (0.6) * -0.46 -0.06 Participants without paratonia 4 throughout the year (n=40) MAS-P 0.16 (0.20) 0.19 (0.22) † 0.28 (0.22) * -0.19 -0.05 AGE levels 2.7 (0.7) 2.7 (0.7) 2.7 (0.7) Participants developing paratonia throughout the year (n=32) MAS-P 0.18 (0.20) 0.29 (0.27) § 0.65 (0.37) * -0.60 -0.35 AGE levels 2.7 (0.6) 2.8 (0.7) 3.0 (0.5) * -0.38 -0.13 AGE levels change (delta visit 3/ visit 1) ‡ -0.40 -0.04 AGE = advanced glycation end-product; MAS-P = modified Ashworth scale for paratonia. Data represent mean values (SD) unless indicated otherwise. §: Significantly different from visit 1 and visit 3. *: Significantly different from visit 1. †: Significantly different from visit 3. ‡: Significantly different from participants without paratonia. in paratonia severity (MAS-P) over this time period. In the group that developed paratonia the results were similar (r = .17, P = .344). In addition, the linear regression shows that change in AGE levels is not predictive for changes in paratonia severity (MAS-P) (β = 0.14, 95% CI: -0.007 to 0.296, P = .062) even after adjusting for baseline characteristics (β = 0.13, 95% CI: -0.44 to 0.299, P = .144) as shown in Table 5.

71 Chapter 4

Table 5. Association Between 1-Year Change in AGE Levels and change in Paratonia Severity (MAS-P) MAS-P change Estimate 95% CI P value lower limit upper limit (Intercept) -0.37 -1.42 0.68 .481 Sex -0.04 -0.21 0.13 .674 Age 0.01 -0.00 0.02 .145 Polypharmacy -0.02 -0.21 0.17 .865 Chronic kidney disease 0.12 -0.19 0.42 .452 Diabetes 0.05 -0.14 0.24 .602 Dementia duration 0.00 -0.00 0.00 .383 Cardio vascular disease 0.06 -0.13 0.25 .545 Cerebral vascular disease 0.03 -0.20 0.26 .776 MMSE -0.01 -0.03 0.00 .113 AGE levels change 0.13 -0.04 0.30 .144 Linear regression model. Corrected for possible explanatory variables. Response variable: MAS-P change (delta visit 3/ visit1). Explanatory variable: AGE levels change (delta visit 3/ visit 1).

DISCUSSION

It was ascertained that AGE levels are associated with the presence and severity of paratonia in patients in early stage AD and AD/VaD. During the course of 1 year, paratonia presence and severity as well as AGE levels increased.

This study reveals a statistically significant increase of paratonia severity (MAS-P) of 0.17 following a 1-year duration. The mixed model analyses revealed a significant association, but mixed models are an average of within participant differences over time and between participant differences at the different time (visit) points. Therefore, we decided to conduct more analyses (correlation and regression analyses on the change scores) to provide more clarity. The effect of time in the mixed model is a within effect. Data from the extra analysis provide some evidence for association between paratonia severity and AGEs within participants (correlation and simple regression show both borderline significance of .062). Therefore the significance of the effect (regression coefficient) from mixed models of AGEs on paratonia severity is caused by within participant association, and by between (cross-sectional per visit) participant association. The fact that theff e ect is small in size is most probably caused by a relative small time span of 1 year between visit 1 and visit 3. Changes in paratonia severity could be explained by the occurrence of new paratonia cases and due to slow aggravation of paratonia severity in those patients who already exhibited paratonia at baseline. Furthermore, because severe paratonia is related to profound

72 Advanced Glycation End-Products are Associated with the Presence and Severity of Paratonia in Early Stage Alzheimer Disease

limitations in functional mobility, as mentioned in the introduction, this relationship is of clinical importance. It has to be considered that our cohort existed of early stage dementia patients in which paratonia was less severe resulting in a limitation of range inMAS-P scores. The MAS has a tendency to cluster scores in the lower range, limiting the ability to discriminate between patients with low scores 28. The follow-up period of 1 year was probably too brief to detect a significant correlation between changes in AGE levels and changes in paratonia severity, however, it does provide an estimation on the progression speed on AGEs accumulation as well as the worsening of paratonia within patients with early stage dementia. In a range between 0 and 4 on the MAS-P, an increase of 0.17 in 1 year seems to be clinically moderate in size. However, with a typical clinical disease duration of 8 to 10 years 29, it becomes clinically relevant when these differences are extrapolated. 4 Further research with a longer follow up is necessary to investigate how changes in AGE levels are related to changes in paratonia severity during the course of AD, AD/VaD or other dementias.

The speed by which collagen tissue is stretched, influences the degree to which the tissue elongates. The more rapid a movement occurs, the stiffer the connective tissue behaves 30. One of the criteria for diagnosing paratonia is the perceived increase in resistance during passive movement acceleration by the assessor and is exactly what can be expected from cross-linking AGE accumulation in peripheral tissue. This could imply that in early stage dementia, 2 phenomena are responsible for the development of paratonia: a central cerebral factor because of the AD pathology and a peripheral factor because of AGE accumulation. It could be hypothesised that, in early stage AD, AGE accumulation is a greater factor initiating movement stiffness and that the central cerebral factor builds up or even accelerates paratonia. Future research is necessary to study this more in depth.

It is of importance to investigate whether paratonia could be postponed or movement stiffness could be improved by reducing AGE levels. Excessive elevation of glucose concentration, such as in DM, most likely accelerates the glycation of proteins7,8,27 . Intensive glycaemic control may be a method for decreasing AGEs formation. For this study, we did not request permission to obtain blood samples from our participants as we expected that this would hamper ethical approval. Therefore, as a consequence, glycaemia could not be controlled in this study.

AGEs are not only produced endogenously; they can also be ingested via food. Specific circulating AGE levels correlate to dietary consumption, especially in food that has been processed at very high temperatures, such as frying, broiling, grilling and roasting9 . Therefore, dietary intake is a possible factor that can be influenced31 . However, evidence of the harmful

73 Chapter 4

effects of long-term exposure to dietary AGEs are currently inconclusive 32. Regular physical exercise could also attenuate the formation and accumulation of AGEs 7,33,34, however, literature regarding the effects of physical exercise on AGEs formation is minimal, and the optimal exercise modalities remain ambiguous 33. Pharmacological approaches to prevent AGE formation or AGE accumulation can be divided into several classes as a function of their mechanism of action: AGE absorption inhibitors, AGE formation inhibitors, AGE cross–link breakers, RAGE antagonists, and AGE binders. Several of these pharmacological strategies with anti-AGEs effects are currently being studied, but results show conflicting evidence 35 and additional research is necessary to obtain an optimal combination of possible strategies. Studies investigating the AGE targeted interventions on paratonia could be considered as a proof of concept underpinning the causal relationship of AGEs and paratonia.

Strength and Limitations

This is the first study investigating the association between AGEs and paratonia. The primary strength of this study is its longitudinal design with 3 assessments. A prolonged study over several years in a larger cohort could result in improved insight into the long- term effects and the within participant relation of AGEs on the development of paratonia. In fact the AGE formation, from a reversible to an irreversible end product takes weeks to months 7,36. The (collagen) tissue accumulation and its potential impact on paratonia, is a process that generally costs several months to years. Further studies involving dementia patients in very early stages over a longer time period (e.g., several years) could help to enlighten this aspect. In addition, a causal relationship cannot be inferred and further longitudinal studies over several years are needed to investigate this. Another strength of this study is that patients were included spread across The Netherlands (urban and rural), which created a study sample representative for this population. This study also has a number of limitations. First, the calculated number of 152 participants was not reached. However, the study did comprise a reasonable number of 144 participants for baseline and 118 for follow up analysis. Second, it was not feasible to indicate which type of AGEs (e.g., cross-linking or non-cross-linking) was responsible for the findings because of the limitations of the AGE reader. Future fundamental research is necessary to further explore this in order to enable targeted interventions to postpone the development or worsening of paratonia in patients with dementia.

74 Advanced Glycation End-Products are Associated with the Presence and Severity of Paratonia in Early Stage Alzheimer Disease

CONCLUSION

This study suggests that elevated AGE levels are a contributing factor to the presence of paratonia as well as its severity in patients with early stage AD or AD/VaD. These results provide a new perspective on paratonia. Future research is necessary to investigate the effect of on AGE-targeted interventions to reduce or postpone the development of paratonia in dementia.

75 Chapter 4

REFERENCES

1. Hobbelen JS, Tan FE, Verhey FR, Koopmans RT, de Bie RA. Prevalence, incidence and risk factors of paratonia in patients with dementia: a one-year follow-up study. Int Psychogeriatr. 2011;23(7):1051-1060. 2. Hobbelen JSM, Koopmans RTCM, Verhey FRJ, Van Peppen RPS, de Bie RA. Paratonia: a Delphi procedure for consensus definition. J Geriatr Phys Ther. 2006;29(2):50-56. 3. Souren LE, Franssen EH, Reisberg B. Neuromotor changes in Alzheimer’s disease: implications for patient care. J Geriatr Psychiatry Neurol. 1997;10(3):93-98. 4. Beversdorf DQ, Heilman KM. Facilitory paratonia and frontal lobe functioning. Neurology. 1998;51(4):968-971. 5. Arvanitakis Z, Wilson RS, Schneider JA, Bienias JL, Evans DA, Bennett DA. Diabetes mellitus and progression of rigidity and gait disturbance in older persons. Neurology. 2004;63(6):996-1001. 6. Monnier VM, Mustata GT, Biemel KL, et al. Cross-linking of the extracellular matrix by the maillard reaction in aging and diabetes: an update on “a puzzle nearing resolution.” Ann N Y Acad Sci. 2005;1043:533-544. 7. Ahmed N, Thornalley PJ. Advanced glycation endproducts: what is their relevance to diabetic complications? Diabetes, Obes Metab. 2007;9(3):233-245. 8. Rahmadi A, Steiner N, Munch G. Advanced glycation endproducts as gerontotoxins and biomarkers for carbonyl-based degenerative processes in Alzheimer’s disease. Clin Chem Lab Med. 2011;49(3):385-391. 9. Uribarri J, Cai W, Peppa M, et al. Circulating glycotoxins and dietary advanced glycation endproducts: two links to inflammatory response, oxidative stress, and aging.journals Gerontol A, Biol Sci Med Sci. 2007;62(4):427-433. 10. Avery NC, Bailey AJ. Enzymic and non-enzymic cross-linking mechanisms in relation to turnover of collagen: relevance to aging and exercise. Scand J Med Sci Sports. 2005;15(4):231-240. 11. Kjaer M. Role of extracellular matrix in adaptation of tendon and skeletal muscle to mechanical loading. Physiol Rev. 2004;84(2):649-698. 12. Lindsey JB, Cipollone F, Abdullah SM, McGuire DK. Receptor for advanced glycation end-products (RAGE) and soluble RAGE (sRAGE): cardiovascular implications. Diab Vasc Dis Res. 2009;6(1):7- 14. 13. Schmidt AM, Yan SD, Yan SF, Stern DM. The biology of the receptor for advanced glycation end products and its ligands. Biochim Biophys Acta. 2000;1498(2-3):99-111. 14. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM- IV). Washington, DC; 1994. 15. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry. 1982;139(9):1136-1139. doi:10.1176/ ajp.139.9.1136. 16. Koetsier M, Nur E, Chunmao H, et al. Skin color independent assessment of aging using skin autofluorescence.Opt Express. 2010;18(14):14416-14429. 17. Hobbelen JSM, Koopmans RTCM, Verhey FRJ, Habraken KM, de Bie RA. Diagnosing paratonia in the demented elderly: reliability and validity of the Paratonia Assessment Instrument (PAI). Int Psychogeriatr. 2008;20(4):840-852. doi:10.1017/S1041610207006424. 18. Waardenburg, H., Elvers, J.W.H., van Vechgel, F., Oostendorp RAB. Is Paratonie Betrouwbaar te meten? Een betrouwbaarheid onderzoek voor het meten van paratonie met de MAS en de gemodificeerde tonusschaal van Ashworth. Ned Tijdschr voor Physiother. 1999;(2):30-35. 19. Meerwaldt R, Graaff R, Oomen PH, et al. Simple non-invasive assessment of advanced glycation endproduct accumulation. Diabetologia. 2004;47(7):1324-1330.

76 Advanced Glycation End-Products are Associated with the Presence and Severity of Paratonia in Early Stage Alzheimer Disease

20. Yamagishi S-I, Fukami K, Matsui T. Evaluation of tissue accumulation levels of advanced glycation end products by skin autofluorescence: A novel marker of vascular complications in high-risk patients for cardiovascular disease. Int J Cardiol. 2015;185:263-268. doi:10.1016/j. ijcard.2015.03.167. 21. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189-198. 22. National Center for Health Statistics. International Classification of Diseases,Ninth Revision, Clinical Modification (ICD-9-CM). http://www.cdc.gov/nchs/icd/icd9cm.htm. Accessed September 1, 2014. 23. World Health Organisation. The Anatomical Therapeutic Chemical Classification System with Defined Daily Doses (ATC/DDD). http://www.who.int/classifications/atcddd/en/. Accessed September 1, 2014. 24. Liu G, Liang KY. Sample size calculations for studies with correlated observations. Biometrics. 1997;53(3):937-947. 4 25. Hsieh FY, Bloch DA, Larsen MD. A simple method of sample size calculation for linear and logistic regression. Stat Med. 1998;17(14):1623-1634. 26. Evans JD. Straightforward Statistics for the Behavioral Sciences. Pacific Grove : Brooks/Cole Pub. Co.,; 1996. 27. Gasser A, Forbes JM. Advanced glycation: implications in tissue damage and disease. Protein Pept Lett. 2008;15(4):385-391. 28. Blackburn M, van Vliet P, Mockett SP. Reliability of measurements obtained with the modified Ashworth scale in the lower extremities of people with stroke. Phys Ther. 2002;82(1):25-34. 29. Bird TD. Alzheimer Disease Overview. In: Pagon RA, Adam MP, Ardinger HH, et al., eds. Seattle (WA); 1993. 30. Fratzl P. Collagen: Structure and Mechanics. New York: Springer US; 2008. 31. Engelen L, Stehouwer CDA, Schalkwijk CG. Current therapeutic interventions in the glycation pathway: evidence from clinical studies. Diabetes Obes Metab. 2013;15(8):677-689. doi:10.1111/ dom.12058. 32. Puyvelde K Van, Mets T, Njemini R, Beyer I, Bautmans I. Effect of advanced glycation end product intake on inflammation and aging: a systematic review. Nutr Rev. 2014;72(10):638-650. 33. Magelhaes PM, j. Appell H, Duarte JA. Involvement of advanced glycation end products in the pathogenesis of diabetic complications: the protective role of regular physical activity. Eur Rev Aging Phys Act. 2008;(5):17-29. 34. Couppé C, Svensson R, Grosset J, et al. Life-long endurance running is associated with reduced glycation and mechanical stress in connective tissue. Age (Omaha). 2014;36(4):9665. doi:10.1007/s11357-014-9665-9. 35. Nenna A, Spadaccio C, Lusini M, Ulianich L, Chello M, Nappi F. Basic and Clinical Research Against Advanced Glycation End Products (AGEs): New Compounds to Tackle Cardiovascular Disease and Diabetic Complications. Recent Adv Cardiovasc drug Discov. 2015;10(1):10-33. 36. Brownlee M. Advanced protein glycosylation in diabetes and aging.Annu Rev Med. 1995;46:223- 234. doi:10.1146/annurev.med.46.1.223.

Chapter 5

Association between Advanced Glycation End-Products and Functional Performance in Alzheimer’s Disease and Mixed Dementia

Hans Drenth, Sytse U. Zuidema, Wim P. Krijnen, Ivan Bautmans, Cees van der Schans, Hans Hobbelen

International Psychogeriatrics 2017, Sep; 29 (9): 1525–1534. Chapter 5

ABSTRACT

Background: People with Alzheimer’s disease (AD) experience, in addition to progressive loss of cognitive functions, a decline in functional performance such as mobility impairment and disability in activities of daily living (ADL). Functional decline in dementia is mainly linkedto the progressive brain pathology. Peripheral biomechanical changes by advanced glycation end- products (AGEs) have been suggested but have yet to be thoroughly studied.

Methods: A multi-center, longitudinal, one-year follow-up cohort study was conducted in 144 people with early stage Alzheimer’s disease or mixed Alzheimer’s/Vascular dementia. Linear mixed model analyses was used to study associations between AGE-levels (AGE reader) and mobility (Timed Up and Go), and activities of daily living (Groningen Activity Restriction Scale and Barthel index) respectively.

Results: A significant association between AGE levels and mobility (β = 3.57, 95% CI: 1.43 - 5.73) was revealed, however, no significant association between AGE levels and activities of daily living was found. Over a one-year time span, mean AGE levels significantly increased, and mobility and activities of daily living performance decreased. Change in AGE levels was not significantly correlated with change in mobility.

Conclusions: This study indicates that high AGE levels could be a contributing factor to impaired mobility but lacks evidence for an association with ADL decline in people with early stage Alzheimer’s disease or mixed dementia. Future research is necessary on the reduction of functional decline in dementia regarding the effectiveness of interventions such as physical activity programs and dietary advice possibly in combination with pharmacologic strategies targeting AGE accumulation.

80 Association between Advanced Glycation End-Products and Functional Performance in Alzheimer’s Disease and Mixed Dementia

BACKGROUND

People with Alzheimer’s disease (AD) experience, in addition to progressive loss of cognitive functions, a decline in functional performance such as mobility impairment and disability in activities of daily living (ADL) 1. In early stage AD, there are already decreases in step length and walking velocity and, in early vascular dementia (VaD), a small step gait, slow stepping, ataxic gait, and unsteadiness is found 2. Impairment in basic ADL (BADL) functions (i.e., bathing, toileting, feeding and dressing) in early dementia is more reliant on decline in motor function, whereas impairment in instrumental ADL (IADL) functions (i.e., housekeeping, cooking and finance management) is more reliant on cognition decline 3. Furthermore, it is known that specifically in the early stages initiating certain IADL (i.e., preparing a meal, finance management) is found to be more strongly impaired than their performance 4. Decline in functional performance contributes to an increase in care burden and a decrease in the quality of life 5. Lower level and accelerated decline of functional performance is 5 suggested to predict the subsequent development of mild cognitive impairment and AD and can precede cognitive impairment by several years6 . Ramakers et al. determined that, even five years prior to the dementia diagnosis, walking impairments were significantly higher in pre-clinical people with dementia compared to the control group 7. A decline in functional performance could be predicted with biomarkers; one of the proposed biomarkers is advanced glycation end-products (AGEs). AGE accumulation contributes to the age-related decline of the functioning of cells and tissues in normal aging 8. Interestingly, AD is related to higher concentrations of AGEs 8,9. AGEs, therefore, are a potential biomarker and an accompanying risk factor for the decline of functional performance in people with AD and mixed dementia (AD/VaD).

Advanced Glycation End-Products

Decline in functional performance in dementia is primarily associated with central mechanisms as a result of progressive brain pathology. Peripheral biomechanical changes have been suggested but have yet to be thoroughly studied. Studies in stroke patients show that immobility is associated with adaptive mechanical and morphological changes in muscle tissue in which muscles become stiffer10 . A review comprising eight studies suggests that, in participants without dementia, AGEs-induced muscle biomechanical changes contribute to a decline in walking abilities and in BADL as well as physical frailty 11. It was recently found that AGEs are associated with the presence and severity of paratonia, a distinctive form of hypertonia/movement stiffness in dementia, which suggests that peripheral biomechanical changes contribute to movement stiffness in early stage dementia 12.

81 Chapter 5

AGEs are formed by the non-enzymatic condensation of a reducing sugar with proteins or lipids and accumulate in hyperglycaemic environments. The accumulation of AGEs is an element of normal metabolism that accelerates in a wide variety of diseases and during normal aging 8,13. AGEs can be categorized into fluorescent cross-linking, non-fluorescent cross-linking, fluorescent non-cross-linking, and non-fluorescent non-cross-linking 14. AGEs tissue accumulation can be estimated non-invasively by skin autofluorescence (SAF) with an AGE reader by utilizing the fluorescent properties of specific AGEs that correlate with non- fluorescent AGEs 14.

The cross-linking of long-lived proteins, particularly collagen, are responsible for increasing mechanical stiffness and loss of elasticity 15. Non-cross-linking effects occur by the binding of AGEs to the receptor for AGEs (RAGE) that incites the production of pro-inflammatory cytokines and free radicals. At the central level, interaction between AGEs, Amyloid-beta, and tau-protein have been ascertained to affect neuronal function13 . At the peripheral level, this AGE/RAGE interaction affects collagen tissue and may play a role in sarcopenia (loss of muscle mass and strength) through upregulated inflammation and endothelial dysfunction in the intra-muscular microcirculation 16.

These peripheral and central effects of AGEs may have a direct or indirect influence on muscle function and functional performance decline; however, this has not been studied in people with dementia. Early detection could initiate interventions targeting AGE accumulation, such as physical activity programs and dietary advice possibly in combination with pharmacologic strategies 17–19 to attenuate functional decline, affording people with dementia longer independence. The aim, therefore, was to investigate the association between AGEs and functional performance in people with AD and mixed dementia (AD/ VaD).

METHOD

Design

The study was designed as a multi-center, longitudinal, one-year observational follow-up cohort study with three assessments: at baseline, after six months, and after 12 months.

Study Population

Participants, selected from 24 dementia day-care centers in the Netherlands, were considered to be eligible for inclusion when they satisfied four criteria: (1) an established

82 Association between Advanced Glycation End-Products and Functional Performance in Alzheimer’s Disease and Mixed Dementia

diagnosis of Alzheimer’s disease (AD) or mixed Alzheimer’s/Vascular dementia (AD/VaD) according to DSM-IV criteria 1; (2) a score of stage five or lower on the Global Deterioration Scale (GDS) 20; (3) able to walk ten meters (a walking aid was allowed); and (4) having a light-colored (Caucasian) skin due to the limitations of the AGE-reader device 14. Written informed consent was obtained from the participants or their legal representatives. The medical ethical committee of the University Medical Centre Groningen approved the study (NL43641.042.13). Design and study population are derived from the PARAGE study 12 and described more extensively.

Outcome Measures

Mobility The participant’s functional mobility was assessed with the Timed Up and Go (TUG). 5 It measures the amount of time a participant takes to stand up from a chair (with an approximate height of 46 cm) and walk 3 m, and then turn around a cone, walk back to the chair, and sit down. The TUG is a validated and reliable test for people with dementia. A score of 20 sec or more indicates the presence or increase of additional mobility problems21 . TUG measurements were obtained by experienced physiotherapists followed by the AGEs measurements assessed by the main researcher at each visit. Prior to the study, the physical therapists were trained by the main researcher on how to perform the measurements.

ADL To assess the broad construct of ADL, the outcome from the participants’ perspectives is measured with the Groningen Activity Restriction Scale (GARS)22 and from the day-care center staffs’ perspective with the Barthel index (BI) 23. The GARS has been ascertained as being valid for measuring disabilities in personal care. With the GARS, the participants are questioned about their capabilities in personal care on two subscales. The first subscale is regarding BADL (11 items), and the second subscale relates to IADL (7 items). The answers are rated on a four-point scale with 1 meaning no support and 4 meaning only with help. A lower score on the combined subscales indicates more ADL independence with 18 as the minimum score 24. The BI is a valid and reliable measurement for assessing a person’s ability of self-care. Ten items regarding BADL and mobility are rated by the participant’s caregiver based on the amount of assistance required to complete each activity. A higher score indicates more BADL independence with 20 as the maximum score 23. The questionnaires (GARS and BI) were administered by key staff personnel from the day-care centers within the same week that the AGEs measurements were taken. Before the study, the participating day-care staff was trained by the primary researcher to perform the measurements.

83 Chapter 5

AGE level AGE levels are measured with the AGE reader (Diagnoptics, Groningen, the Netherlands) that is a desktop device measuring fluorescent skin AGEs and is reported to be valid and reliable for the quantification of AGEs tissue accumulation 14. The skin of the forearm is illuminated by a light source in the AGE reader through a 1-cm2 hole that is guarded against surrounding light. Excitation light in the wavelength range of 300–420 nm (peak excitation ~350 nm) is projected onto the skin surface. The intensity of light emitted from the skin in the wavelength range of 420 and 600 nm is measured with a spectrometer. SAF is calculated as the ratio between the emission light and the excitation light using the AGE reader software and expressed in arbitrary units (AU). A high SAF score corresponds to a high tissue AGEs level14 . The measurements were performed on the forearm without sweat, skin lotions, or visible skin abnormalities and with the assessor being blinded for functional performances scores. All AGE reader measurements were performed at room temperature in a standardized semi dark environment with the participants in a seated position and the volar side of the right forearm placed on top of the AGE reader. The mean of three consecutive measurements was used for analysis.

Other variables Dementia characteristics were provided by the general practitioner (GP) or the local physician. Cognitive functioning was measured by experienced psychologists or physicians with the Mini-Mental State Examination (MMSE). The MMSE is an 11-item questionnaire with a maximum score of 30, which indicates no cognitive decline, and a minimum score of 0, which indicates a very severe cognitive decline 25. Dementia severity was categorized by key staff personnel from the day-care centers using the GDS that identifies seven clinically recognizable stages from normal (no dementia) to severe dementia 20. Paratonia was diagnosed by physical therapists with the Paratonia Assessment Instrument (PAI) at each visit. The PAI is a reliable and valid dichotomous assessment instrument with which an examiner can establish the presence (or absence) of paratonia by successively moving all four limbs passively in flexion and extension while the participant is in a sitting position 26.

The use of medication and the presence of comorbidities (ICD-9 classification) were retrieved from the participant’s medical records and GP files. The use of five or more medications was labeled as polypharmacy 27.

Statistical Analyses

Sample size calculation A mixed model sample size calculation 28 was based upon a 0.5 correlation between repeated measurements, 0.3 variance of the random intercept, 0.3 residual variance, a true

84 Association between Advanced Glycation End-Products and Functional Performance in Alzheimer’s Disease and Mixed Dementia

effect size of 0.5, a desired power of 80%, and a two-sided alpha of 0.05. This resulted in a total sample size of 152. Addressing an eventual 10% withdrawal resulted in a required total number of approximately 165 participants. Baseline characteristics are depicted by descriptive statistics and presented in Table 1.

Table 1. Baseline characteristics Total n = 144 FREQUENCY RANGE MIN – MAX

Male, n (%) 63 (43,7%) Age, years 80.7 (7.7) 53-98 Dementia duration, months a 29.8 (35.9) 1-252 Co-morbidities, n a 1.18 (1.1) 0-4 AD, n (%)a 107 (83.6%) Mixed AD/VaD, n (%) a 21 (16,4%) 5 AD or Mixed AD/VaD, n (%) b 16 (11.11%) CVD, n (%)a 41 (32.0%) CeVD (CVA, TIA), n (%)a 22 (17.2%) DM, n (%)a 37 (28.9%) Cancer, n (%)a 14 (11.0%) COPD, n (%)a 14 (11.0%) CKD, n (%)a 11 (8.6%) Systemic, n (%)a 8 (6.3%) Digestive tract,n (%)a 6 (4.7%) Polypharmacy (≥ 5 meds), n (%)a 71 (55.5%) MMSE, score 0-30 19.4 (5.4) 6-29 GDS, score 1-7 3.84 (1.0) 2-5 Paratonia, PAI Yes 60 (41.7%) TUG, seconds 17 (9.7) 7-65 GARS BADL, score 11-44 16.1 (5.7) 1-37 GARS IADL, score 7- 28 16.8 (6.3) 7-28 GARS total, score 18-72 32.9 (10.7) 18-65 BI, score 0-20 16 (3.5) 6-20 AGE levels, SAF (AU) 2.8 (0.7) 0.4-4.9 AGEs = advanced glycation end-products; AU = arbitrary units i.e., the output units of the AGE reader; BADL = basic activities daily living; BI = Barthel index; CeVD = cerebral vascular disease; CKD = chronic kidney disease; COPD = chronic obstructive pulmonary disease; CVA = cerebral vascular accident; CVD = cardio vascular disease; DM = diabetes mellitus; GARS = Groninger Activity Restriction Scale; GDS = Global Deterioration Scale; IADL = instrumental activities daily living; MMSE = Mini Mental State Examination; SAF = skin autofluorescence (AGE reader); TIA = transient ischemic attack; TUG = timed up and go. Frequency data represent mean values (SD) unless indicated otherwise. a Based on GP medical files n = 128. b Based on chart diagnoses provided by day-care staff personnel for study inclusion.

85 Chapter 5

Association between AGE levels and Mobility and ADL To investigate the association between AGE levels on mobility and ADL(BADL/IADL), linear mixed model analyses (LMM) was employed, estimated by restricted maximum likelihood taking the TUG, GARS, and BI measurements at each of the three visits as the response variable.

The models controlled statistically for the fixed effects of AGEs level, time (visit), gender, age, polypharmacy, dementia duration, cognition (MMSE), paratonia, chronic kidney disease (CKD), cardiovascular disease (CVD), cerebral vascular disease (CeVD), and diabetes mellitus (DM) 8. Participants were taken as random effects. Backward model selection was utilized to identify statistically significant explanatory variables. During this process, AGEs level and time were always retained in order to study the size of their effect.

To investigate change in mean over one year (between the last visit and baseline) on AGE levels and TUG, GARS, and BI, paired sample t-tests were performed. Pearson’s R was calculated to investigate the association between change in AGE levels and change in the previously described variables. To further explore specific longitudinal effects, a linear regression was used where the change of the described scores between visit three and one was used as a response variable and changes of AGEs level and baseline characteristics (described above) as explanatory variables.

Data were analyzed using R version 3.2.0 and SPSS version 22, taking a p-value < 0.05 as statistically significant.

RESULTS

From the 244 people with dementia approached to take part in the study, 87 were not included due to not satisfying the inclusion criteria; 13 participants withdrew informed consent prior to the baseline assessment. Finally, 144 participants were included at baseline. After one year, 26 participants (18%) were lost to follow-up: 11 deceased while 15 were transferred to unknown addresses or became too ill to be reassessed (Figure 1). For comorbidity and medication use in 11% of the participants (n = 16), no information was accessible in the medical records (information derived from the PARAGE study 12). The baseline characteristics are summarized in Table 1.

86 Association between Advanced Glycation End-Products and Functional Performance in Alzheimer’s Disease and Mixed Dementia

Figure 1. Study Flow Chart

Association Between AGEs and Mobility

Table 2 indicates that, after correction for age, polypharmacy, CKD, CVD, and MMSE, functional mobility (TUG) was significantly associated with the AGE levels (β = 3.57, P = 0.001, 95% CI: 1.43 - 5.73) and with the progression of dementia over a one-year time span (β = 3.73, P = 0.001, 95% CI: 1.46 - 5.91)

87 Chapter 5

Table 2. Association between AGEs and mobility (TUG) corrected for possible explanatory variables obtained after Backward model selection Mobility (TUG) Estimate P value 95% CI Lower Upper limit limit (Intercept) -14.33 0.196 -35.59 6.90 Age 0.42 0.002 0.16 0.68 Polypharmacy 4.75 0.020 0.88 8.60 CKD -8.75 0.012 -15.31 -2.19 MMSE -0.76 <0.001 -1.11 -0.40 AGE levels 3.57 0.001 1.43 5.73 Visit2* 2.41 0.026 0.30 4.50 Visit3* 3.73 0.001 1.46 5.91 AGE = advanced glycation end-product; CKD = chronic kidney disease; MMSE = Mini Mental State Examination. Response variable: Timed Up & Go (TUG) in seconds. Explanatory variable: AGE levels (SAF) AU. * visit effects with respect to baseline (Visit1).

Association Between AGEs and ADL

Table 3 indicates that, after removing covariates not being statistically significant, the AGE levels did not have a significant effect on ADL, however, the GARS (BADL/IADL) and BI were associated with the progression of dementia over one-year time span (β = 2.58, P < 0.001, 95% CI: 1.47 - 3.67, β = 1.98, P <0.001, 95% CI: 0.79 - 3.17 and β = -2.00, P <0.001, 95% CI: -2.58 to -1.42, respectively).

Table 3. Association between AGEs and ADL (GARS IADL and BADL and Barthel Index), corrected for possible explanatory variables obtained after backward model selection. GARS BADL GARS IADL Barthel Index Estimatea P value 95% CI Estimateb P value 95% CI Estimatec P value 95% CI Lower Upper Lower Upper Lower Upper limit limit limit limit limit limit

(Intercept) 12.63 <0.001 9.32 15.94 14.83 <0.001 11.51 18.16 23.27 <0.001 17.04 29.47 Age ------0.13 0.001 -0.21 -0.06 Polypharmacy 3.30 0.001 1.38 5.23 - - - - 0.25 <0.001 0.14 0.35 AGE levels 0.46 0.404 -0.62 1.54 0.68 0.219 -0.41 1.77 -0.49 0.110 -1.08 0.11 Visit2* 2.12 <0.001 1.06 3.17 1.16 0.044 0.02 2.29 -1.01 <0.001 -1.55 -0.45 Visit3* 2.58 <0.001 1.47 3.67 1.98 0.001 0.79 3.17 -2.00 <0.001 -2.58 -1.42 AGE = advanced glycation end-product; MMSE = Mini Mental State Examination. a Response variable: basic activities daily living (BADL) Groninger Activity Restriction Scale (GARS). b Response variable: instrumental activities daily living (IADL) Groninger Activity Restriction Scale (GARS). c Response variable: Barthel index (BI). Explanatory variable: AGE levels (SAF) AU. * visit effects with respect to baseline (Visit1).

88 Association between Advanced Glycation End-Products and Functional Performance in Alzheimer’s Disease and Mixed Dementia

The one-year development of AGE levels and functional mobility and ADL (BADL/IADL) is presented in Table 4. From the 118 participants (82%) who completed this study, the longitudinal data over one year indicates that there was a significant increase in the overall AGE levels as well as TUG and GARS scores and a significant decrease in BI scores. Changes in functional performance (TUG, GARS and BI) between baseline and visit three indicated no significant correlation with changes in AGEs level over one year. The linear regression models show that change in AGEs level is not predictive for changes in functional performance, not even after adjusting for baseline characteristics.

Table 4. One-year development of AGE levels and Mobility and ADL Visit 1 Visit 3 P value R P value BETA P value Visit 3-Visit 1 FREQUENCY RANGE FREQUENCY RANGE AGE levels 2.8 (0.7) 0.4-4.9 3.0 (0.7) 1.6-5.3 <0.001 - - - - TUG 17 (9.7) 7-65 20 (15) 6-103 <0.001 0.002 0.984 -1.44 0.487 5 GARS BADL 16.1 (5.7) 1-37 18.3 (6.9) 11-39 <0.001 0.053 0.569 -0.81 0.428 GARS IADL 16.8 (6.3) 7-28 18.6 (6.7) 7-28 <0.001 0.053 0.570 0.18 0.896 Barthel index 16 (3.5) 6-20 14.2 (4.5) 1-20 <0.001 0.011 0.908 0.86 0.186 AGE = advanced glycation end-product; BADL = basic activities daily living; GARS = Groninger Activity Restriction Scale; IADL = instrumental activities daily living; TUG = timed up and go. Frequency data represent mean values (SD) unless indicated otherwise. R = Pearson correlation coefficient between change (delta visit 3 visit 1) in AGE levels and change (delta visit 3 visit 1) TUG, GARS, Barthel index scores. Beta = Linear regression model between change (delta visit 3 visit 1) in AGE levels and change (delta visit 3 visit 1) TUG, GARS, Barthel index scores.

DISCUSSION

Over a one-year time span, the AGE levels significantly increased and mobility, BADL and IADL performance decreased. This study shows that AGE levels are significantly associated with functional mobility, however, not with BADL or IADL in people experiencing early stage AD and mixed Alzheimer’s/Vascular disease (AD/VaD).

Although, change in AGE levels was not significantly related to change in functional mobility, mixed model analyses revealed a significant combined time and between participant effect of AGEs on functional mobility. Participants with higher AGE levels scored higher on the TUG, indicating lower functional mobility. The encountered Beta effect means that, with every unit of AGE reader increase, the time to perform the TUG increases with 3.57 sec. The TUG measures functionality when transferring from sitting to standing, turning, and walking speed. A TUG score greater than 20 sec indicates mobility problems and slow walking speed that is associated with a wide range of adverse health consequences such as frailty, falls,

89 Chapter 5

disability, hospitalization, and institutionalization 21. It is suggested that every decrease of 0.1 m/sec in walking speed already increases the risk of these negative health outcomes 29. With a typical clinical AD duration of eight to ten years 30, extrapolating the Beta effect of 3.57 sec would become even more relevant.

The result from this study is consistent with studies describing the effect of AGEs on the decline of walking abilities and contributes to the increasing evidence that decline in functional mobility can be attributed to the effects of AGEs on muscle tissue 11. The loss of skeletal muscle mass and weakness is an important contributor to functional decline. Both muscle weakness and walking impairment are prominent characteristics of physical frailty31 , suggesting that high AGEs level are a contributing factor to physical frailty. Itremains ambiguous whether the association between high AGE levels and functional mobility decline exists because AGEs damage muscle tissue or whether loss of physical activity due to functional mobility decline influences AGEs accumulation. Future research is necessary to study this more in depth.

In accordance with the results of this study, it would be interesting to investigate whether decline in functional mobility can be attenuated by reducing AGEs levels. Excessive elevation of glucose concentration, such as in DM, most likely accelerates the glycation of proteins8,13 . Intensive glycaemic control may be a method to decrease AGEs formation. AGEs are not only produced endogenously, but are also spontaneously generated in standard diets 18. Therefore, dietary intake is a possible factor that can be influenced. In order to lower daily AGEs intake, it is suggested that foods rich in sugar and fat as well as those prepared by frying or grilling should be avoided 17. However, evidence of the harmful effects of long-term exposure to dietary AGEs are currently inconclusive 17. Additionally, regular physical activity has demonstrated a correlation with reduced glycation and AGE formation, however, the optimal exercise modalities still remain unclear 18. In addition, pharmacologic strategies to prevent AGE formation or AGE accumulation are being studied, but results show conflicting evidence and additional research is necessary19 .

An association between AGE levels and decline in ADL (BADL/IADL) was not determined. This is in contrast with a large cohort (n = 3,373) study by Whitson et al. that reported an association between serum AGE levels and the time to incident BADL disability in heathy participants over 14 years (HR = 1.10, 95% CI: 1.05 - 1.15)32 . Although it is likely that impaired muscle function - through AGEs-induced muscle damage – can contribute to impaired BADL, the results from the current study did not confirm this in people with early stage dementia. Our sample size was possibly too small and a study duration of one year too brief to detect a decline in ADL, and/or the participants were less ADL independent at baseline. Further

90 Association between Advanced Glycation End-Products and Functional Performance in Alzheimer’s Disease and Mixed Dementia

research with a longer follow up time is necessary to investigate if AGE levels are related to the deterioration of ADL during the course of AD, mixed dementia (AD/VaD), or other forms of dementia.

The strengths of this study are its longitudinal design with three assessments and that participating personnel were well trained in using the measurements. A study sample representative for this population was also created by including participants from rural and urban areas who were dispersed across the Netherlands. This study also has a number of limitations. First, the initial number of 152 participants fromour a priori sample size calculation could not be included. However, the study still comprised a reasonable number of 144 participants for baseline and 118 for follow up analysis. Second, due to the limitation of the AGE reader, it was not possible to indicate what types of AGEs (e.g. cross-linking or non-cross-linking) are responsible for our findings. Future fundamental research is necessary to further explore this. Finally, the follow-up period of one year is possibly 5 insufficient for detecting an association between changes in AGE levels and changein functional performance over time. Prolonging the study over several years and in a larger cohort could result in improved insight in the long-term effects of AGEs on functional performance. Further longitudinal studies over several years are needed to investigate a causal relationship. In conclusion this study indicates that high AGE levels could be a contributing factor to the decline in functional mobility in addition to the progressive brain pathology but lacks evidence for an association with ADL decline in people with early AD or mixed dementia. This result contributes to the increasing evidence that high AGE levels could affect functional mobility in the aging population. Future research is necessary into interventions such as physical activity programs and dietary advice targeting AGE accumulation possibly in combination with pharmacologic strategies to attenuate functional decline in those experiencing dementia.

91 Chapter 5

REFERENCES

1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM- IV). Washington, DC; 1994. 2. Scherder E, Eggermont L, Visscher C, Scheltens P, Swaab D. Understanding higher level gait disturbances in mild dementia in order to improve rehabilitation: “last in-first out”. Neurosci Biobehav Rev. 2011;35(3):699-714. doi:10.1016/j.neubiorev.2010.08.009. 3. Martyr A, Clare L. Executive function and activities of daily living in Alzheimer’s disease: a correlational meta-analysis. Dement Geriatr Cogn Disord. 2012;33(2-3):189-203. doi:10.1159/000338233. 4. Giebel CM, Sutcliffe C, Challis D. Hierarchical Decline of the Initiative and Performance of Complex Activities of Daily Living in Dementia. J Geriatr Psychiatry Neurol. 2017;30(2):96-103. doi:10.1177/0891988716686835. 5. Andersen CK, Wittrup-Jensen KU, Lolk A, Andersen K, Kragh-Sorensen P. Ability to perform activities of daily living is the main factor affecting quality of life in patients with dementia. Health Qual Life Outcomes. 2004;2:52. doi:10.1186/1477-7525-2-52. 6. Buchman AS, Bennett DA. Loss of motor function in preclinical Alzheimer’s disease. Expert Rev Neurother. 2011;11(5):665-676. 7. Ramakers IHGB, Visser PJ, Aalten P, et al. Symptoms of preclinical dementia in general practice up to five years before dementia diagnosis. Dement Geriatr Cogn Disord. 2007;24(4):300-306. doi:10.1159/000107594. 8. Rahmadi A, Steiner N, Munch G. Advanced glycation endproducts as gerontotoxins and biomarkers for carbonyl-based degenerative processes in Alzheimer’s disease. Clin Chem Lab Med. 2011;49(3):385-391. 9. Li J, Liu D, Sun L, Lu Y, Zhang Z. Advanced glycation end products and neurodegenerative diseases: mechanisms and perspective. J Neurol Sci. 2012;317(1-2):1-5. doi:10.1016/j.jns.2012.02.018. 10. Farmer SE, James M. Contractures in orthopaedic and neurological conditions: a review of causes and treatment. Disabil Rehabil. 2001;23(13):549-558. 11. Drenth H, Zuidema S, Bunt S, Bautmans I, van der Schans C, Hobbelen H. The Contribution of Advanced Glycation End product (AGE) accumulation to the decline in motor function. Eur Rev Aging Phys Act. 2016;13:3. doi:10.1186/s11556-016-0163-1. 12. Drenth H, Zuidema S, Krijnen W, Bautmans I, van der Schans C, Hobbelen H. Advanced Glycation End-products are Associated with the Presence and Severity of Paratonia in early stage Alzheimer’s Disease. J Am Med Dir Assoc. 2017. doi:10.1016/j.jamda.2017.04.004. 13. Ahmed N, Thornalley PJ. Advanced glycation endproducts: what is their relevance to diabetic complications? Diabetes, Obes Metab. 2007;9(3):233-245. 14. Meerwaldt R, Graaff R, Oomen PH, et al. Simple non-invasive assessment of advanced glycation endproduct accumulation. Diabetologia. 2004;47(7):1324-1330. 15. Avery NC, Bailey AJ. Enzymic and non-enzymic cross-linking mechanisms in relation to turnover of collagen: relevance to aging and exercise. Scand J Med Sci Sports. 2005;15(4):231-240. 16. Payne GW. Effect of inflammation on the aging microcirculation: impact on skeletal muscle blood flow control. Microcirculation. 2006;13(4):343-352. doi:10.1080/10739680600618918. 17. Puyvelde K Van, Mets T, Njemini R, Beyer I, Bautmans I. Effect of advanced glycation end product intake on inflammation and aging: a systematic review. Nutr Rev. 2014;72(10):638-650. 18. Magelhaes PM, j. Appell H, Duarte JA. Involvement of advanced glycation end products in the pathogenesis of diabetic complications: the protective role of regular physical activity. Eur Rev Aging Phys Act. 2008;(5):17-29.

92 Association between Advanced Glycation End-Products and Functional Performance in Alzheimer’s Disease and Mixed Dementia

19. Nenna A, Spadaccio C, Lusini M, Ulianich L, Chello M, Nappi F. Basic and Clinical Research Against Advanced Glycation End Products (AGEs): New Compounds to Tackle Cardiovascular Disease and Diabetic Complications. Recent Adv Cardiovasc drug Discov. 2015;10(1):10-33. 20. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry. 1982;139(9):1136-1139. doi:10.1176/ ajp.139.9.1136. 21. Ries JD, Echternach JL, Nof L, Gagnon Blodgett M. Test-retest reliability and minimal detectable change scores for the timed “up & go” test, the six-minute walk test, and gait speed in people with Alzheimer disease. Phys Ther. 2009;89(6):569-579. doi:10.2522/ptj.20080258. 22. Kempen GIJM, Doeglas DM ST. Het Meten van Problemen Met Zelfredzaamheid Op Verzorgend En Huishoudelijk Gebied Met de GARS: Een Handleiding. 1st ed. Groningen: Noordelijk Centrum voor Gezondheidsvraagstukken, Rijksuniversiteit Groningen; 1993. 23. Collin C, Wade DT, Davies S, Horne V. The Barthel ADL Index: a reliability study. Int Disabil Stud. 1988;10(2):61-63. 24. Metzelthin SF, Daniels R, van Rossum E, de Witte L, van den Heuvel WJA, Kempen GIJM. The psychometric properties of three self-report screening instruments for identifying frail older people in the community. BMC Public Health. 2010;10:176. doi:10.1186/1471-2458-10-176. 25. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the 5 cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189-198. 26. Hobbelen JSM, Koopmans RTCM, Verhey FRJ, Habraken KM, de Bie RA. Diagnosing paratonia in the demented elderly: reliability and validity of the Paratonia Assessment Instrument (PAI). Int Psychogeriatr. 2008;20(4):840-852. doi:10.1017/S1041610207006424. 27. World Health Organisation. The Anatomical Therapeutic Chemical Classification System with Defined Daily Doses (ATC/DDD). http://www.who.int/classifications/atcddd/en/. Accessed September 1, 2014. 28. Liu G, Liang KY. Sample size calculations for studies with correlated observations. Biometrics. 1997;53(3):937-947. 29. Graham JE, Ostir G V, Fisher SR, Ottenbacher KJ. Assessing walking speed in clinical research: a systematic review.J Eval Clin Pract. 2008;14(4):552-562. doi:10.1111/j.1365-2753.2007.00917.x. 30. Bird TD. Alzheimer Disease Overview. In: Pagon RA, Adam MP, Ardinger HH, et al., eds. Seattle (WA); 1993. 31. Sternberg SA, Schwartz AW, Karunananthan S, Bergman H, Clarfield AM. The identification of frailty: a systematic literature review. J Am Geriatr Soc. 2011;59(11):2129-2138. 32. Whitson HE, Arnold AM, Yee LM, et al. Serum carboxymethyl-lysine, disability, and frailty in older persons: the Cardiovascular Health Study. journals Gerontol A, Biol Sci Med Sci. 2014;69(6):710- 716.

Chapter 6

Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

Hans Drenth, Sytse U. Zuidema, Wim P. Krijnen, Ivan Bautmans, Cees van der Schans, Hans Hobbelen

Gerontology 2017, Dec 22. Chapter 6

ABSTRACT

Background: Paratonia is a distinctive form of hypertonia, causing loss of functional mobility in early stages of dementia to severe high muscle tone and pain in the late stages. For assessing and evaluating therapeutic interventions, objective instruments are required. Objective: Determine the psychometric properties of the MyotonPRO, a portable device that objectively measures muscle properties, in dementia patients with paratonia.

Methods: Muscle properties were assessed with the MyotonPRO by 2 assessors within one session and repeated by the main researcher after 30 min and again after 6 months. Receiver operating characteristic curves were constructed for all MyotonPRO outcomes to discriminate between participants with (n = 70) and without paratonia (n = 82). In the participants with paratonia, correlation coefficients were established between the MyotonPRO outcomes and the Modified Ashworth Scale for paratonia (MAS-P) and muscle palpation. In participants with paratonia, reliability (intraclass correlation coefficient) and agreement values (standard error of measurement and minimal detectable change) were established. Longitudinal outcome from participants with paratonia throughout the study (n = 48) was used to establish the sensitivity for change (correlation coefficient) and responsiveness (minimal clinical important difference).

Results: Included were 152 participants with dementia (mean [standard deviation] age of 83.5 [98.2]). The area under the curve ranged from 0.60 to 0.67 indicating the MyotonPRO is able to differentiate between participants with and without paratonia. The MyotonPRO explained 10 - 18% of the MAS-P score and 8 - 14% of the palpation score. Interclass correlation coefficients for interrater reliability ranged from 0.57 to 0.75 and from 0.54 to 0.71 for intrarater. The best agreement values were found for tone, elasticity, and stiffness. The change between baseline and 6 months in the MyotonPRO outcomes explained 8 - 13% of the change in the MAS-P scores. The minimal clinically important difference values were all smaller than the measurement error.

Conclusion: The MyotonPRO is potentially applicable for cross-sectional studies between groups of paratonia patients and appears less suitable to measure intraindividual changes in paratonia. Because of the inherent variability in movement resistance in paratonia, the outcomes from the MyotonPRO should be interpreted with care; therefore, future research should focus on additional guidelines to increase the clinical interpretation and improving reproducibility.

96 Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

INTRODUCTION

Paratonia is a well-defined characteristic movement disorder in patients experiencing dementia. Paratonia is defined as a form of hypertonia (high muscle tone with movement stiffness) with an involuntary variable resistance during passive movement. It hasan estimated prevalence of 10% in the early/mild stages and up to 90–100% in later/severe stages of dementia1,2 . Paratonia progresses from active assistance during passive movement in the early stages of dementia towards strong active resistance in the later stages. Notably, the degree of resistance varies depending on the speed of movement (e.g., low resistance to slow movement and high resistance to fast movement). Furthermore, the resistance to passive movement is in any direction, and there is no clasp-knife phenomenon. This definition enables differentiation between paratonia, Parkinsonian rigidity, and spasticity after stroke3 . Due to paratonia, daily care (e.g., washing and dressing) becomes uncomfortable and painful. Severe paratonia, therefore, results in a substantial increase of the caretaker’s burden and decreases the quality of life for those in the advanced stages of dementia 1,2. However, even in early stage dementia, paratonia already has a negative and significant impact on functional mobility 1. 6

The contractile component (hypertonia) in paratonia brings on worse muscle recovery and quicker muscle fatigue resulting in higher tone, stiffness and elasticity, and lower viscoelastic properties4,5 . Noncontractile components are also suggested to contribute to the movement resistance due to biomechanical and viscoelastic changes of connective tissue 6.

The Paratonia Assessment Instrument (PAI) is a valid and reliable instrument for assessing the presence of paratonia 7. The severity of paratonia is usually scored by using a Modified Ashworth Scale for paratonia (MAS-P) 8. In clinical settings, this scale is the worldwide standard as a rating scale to measure abnormality in tone or resistance to passive movements. Because the assessor must determine muscle tone by the perceived resistance during passive movement, extensive clinical experience is necessary for the (M)AS to be reliable 9–11. In addition, it remains ambiguous which of the muscle properties such as tone, biomechanical, or viscoelastic properties are perceived by the assessor12,13 .

Valid, reliable, objective and responsive instruments for measuring muscle properties are necessary for assessing and evaluating therapeutic interventions. To measure muscle properties such as tone, elasticity, stiffness, creep, and mechanical stress relaxation (MSR) time, the Myoton device is available. A Myoton measurement is an objective, quickly applicable, painless, and noninvasive method which has been validated and proved to be reliable for patients with stroke (mean age ranging from 54.7 to 67.5 years) 12–14, upper

97 Chapter 6

motor neuron disorders (mean age ranging from 47.5 to 54.7 years) 10,13, Parkinson disease (mean age ranging from 61.3 to 77.3 years) 15,16, and in healthy subjects (mean age ranging from 24.5 to 71.7 years) 17–21. Recently, the MyotonPRO has been studied in a small group (n = 16, age ranging from 70 to 98 years) of patients with paratonia and demonstrated low to high interrater reliability, moderate to high between series intrarater reliability, and poor to moderate between day intrarater reliability and agreement 22.

Because the MyotonPRO objectively measures 5 different muscle properties on a continuous scale, we hypothesize that this tool provides a measurement of muscle tone/stiffness and is a more comprehensive and accurate alternative to the MAS-P. Therefore, if the MyotonPRO is a proven valid and reliable tool for assessing paratonia severity, it may accelerate future research in this area. The aim of this study is to determine the psychometric properties of the MyotonPRO including construct validity with the PAI, the concurrent validity against MAS-P and muscle palpation, and reproducibility and sensitivity/responsiveness to change in dementia patients with paratonia.

METHODS

Design and Study Population

The study was designed as a multicenter, prospective study with 2 assessments within one session and repeated assessments after 30 min and after 6 months. A convenient sample of 14 nursing homes in the Netherlands was selected as recruitment facilities. Participants were considered to be eligible for inclusion when they were community-dwelling (visiting day-care center) and/or institutionalized dementia patients with an established diagnosis according to the DSM-IV criteria. Written informed consent was obtained from participants or their legal representatives. Because the prevalence and severity of paratonia increases with the progression of dementia, we included people with dementia equally distributed over 3 dementia stages; (1) early-stage dementia (Global Deterioration Scale [GDS] 23 score of 2, 3, or 4); (2) moderate-stage dementia (GDS score of 5); and (3) severe-stage dementia (GDS score of 6 or 7). This study was conducted to determine the psychometric properties of the MyotonPRO in people with paratonia. To diagnose paratonia, all 5 criteria of the PAI must be met. If for a specific participant not all criteria are met, there may be no or another tone disorder (spasticity or rigidity), and these participants were assigned to the nonparatonia group. Furthermore, participants were excluded from the study if their health was unstable. The medical ethical committee of the University Medical Centre Groningen approved the study (NL54144.042.15).

98 Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

Assessment Instruments

Myoton device The MyotonPRO (Myoton AS, Estonia) is a small, portable, handheld device for measuring mechanical muscle properties24 . Measurement consists of 3 main components: (1) exertion of mechanical impulse, (2) registration of co-oscillation, and (3) computation of parameters.

The tip of the 3-mm diameter probe is applied perpendicular to the skin surface above the muscle that is being measured. A constant pre-pressure (0.18 N) is applied, whereby the subcutaneous superficial tissues are slightly compressed. A brief (15 ms), low-force (0.4 N) mechanical impulse is then transmitted to the underlying muscle. The subsequent dampened oscillation of the muscle is recorded by an accelerometer and simultaneously quantifies and displays the following muscle properties25 :

Tone: Indicates the intrinsic tension state (oscillation frequency, HZ) of the muscle at rest. The higher the oscillation frequency, the greater the muscle tension, which increases by contraction 4. 6

Biomechanical properties: Elasticity (logarithmic decrement of the dampened oscillation), the muscle’s ability to recover its shape after being deformed. The smaller the value for decrement (expressed in arbitrary units), the smaller the dissipation of mechanical energy and the higher the elasticity of a tissue4 . Stiffness (expressed in N/m), the ability to resist an external force that is attempting to modify its shape. The higher the N/m value, the stiffer the muscle 4.

Viscoelastic properties: MSR time (ms), the time for a muscle to restore its shape from deformation after an external force is removed. Creep (scored in a so-called Deborah number) is the gradual elongation of a muscle over time when placed under a constant tensile stress. It is estimated by the MyotonPRO as the ratio of the relaxation andthe deformation time of the muscle. High values indicate high viscoelasticity 25.

PAI and MAS-P The PAI is conducted with a short movement examination in which the limbs of the participant are being moved (first slowly and then accelerated) in flexion and extension, with the participant in a comfortably seated position or, when bedridden, in asupine position. The PAI diagnoses the presence of paratonia when the following 5 criteria are all satisfied: (1) an involuntary variable resistance; (2) a degree of resistance that varies depending on the speed of the movement (e.g., a low resistance to slow movement and a

99 Chapter 6

high resistance to fast movement); (3) resistance to passive movement in any direction; (4) no clasp-knife phenomenon; and (5) resistance in two movement directions in the same limb or two different limbs 7.

With the MAS-P, muscle tone is judged by the perceived resistance during passive movement and rated on a 5-point scale ranging from 0 to 4, in which 0 = no resistance to passive movement, 1 = slight resistance during passive movement, 2 = more marked resistance to passive movement, 3 = considerable resistance to passive movement, and 4 = severe resistance, such that passive movement is impossible 8. A score of 0.5 was assigned in the event of active assistance.

Manual muscle tone palpation Manual muscle palpation is conducted to assess the muscle tone/stiffness at rest and is graded as normal, hypotonia, or hypertonia 26.

Clinical Global Impression A change in perceived muscle tone/movement stiffness is assessed with the Clinical Global Impression (CGI). The CGI overall comparison of the participant’s baseline condition with the current state of general movement stiffness is rated as 1 = very much improved, 2 = much improved, 3 = minimally improved, 4 = no change, 5 = minimally worse, 6 = much worse, or 7 = very much worse 27.

Study Procedures

For assessing muscle properties, the m. biceps brachii was used as this is an important muscle when assessing paratonia in an upper limb. An upper limb is also affected earlier by paratonia than a lower limb 1. Because of the importance in upper arm functionality, the m. biceps brachii was also used in other Myoton studies 11,12,15,16,19,21,22.

In all nursing homes, the participants were assessed in their own or in a separate room in order to reduce external stimuli. Assessment began with the PAI and MAS-P assessment by the main researcher who has extensive clinical experience in these measurements. Then, the main researcher assessed muscle properties with the MyotonPRO. To ensure maximum relaxation during these measurements, the participants were seated in a comfortable position with both arms resting on their lap with the elbow in 90-degree flexion supported by a standard pillow (Fig. 1) or, when in a supine position, with their elbows in 90-degree flexion resting comfortably on their stomach. In these positions, it was ensured that the device could be used within the recommended 0 - 100 degrees to the gravity vector with

100 Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

the probe perpendicular to the skin, and the muscle was not hanging 25. The MyotonPRO was placed respectively on the left and right m. biceps brachii in the middle of the muscle belly which was detected by visual inspection and palpation (at approximately 1/3 of the distance between the cubital fossa and the lateral tip of the acromion) and marked with a skin marker. The multiscan mode consisting of 10 single measurements with a 1-s interval was used, resulting in a mean and coefficient of variation (CV) for the 10 measurements (i.e., one measurement set) 17,19–21. In order to register satisfactory measurements, the CV of the parameters should be less than 3% 25. If this was not the case, then the measurement was repeated after corrective actions with a maximum of 3 re-measurements. If a high CV remained, the MyotonPRO manual suggests that this variability may also be caused by the participant’s neurological condition, reflecting the current condition of the muscle 25. The set with the lowest CV was then used for the analysis.

Immediately after and blinded from the main researcher, one of the local physiotherapists (n = 17) performed the MyotonPRO measurement (interrater reliability). Prior to the study, all participating physiotherapists were instructed and trained by the main researcher on the use of the MyotonPRO device and the testing protocol. In the same session, the local 6 physiotherapist graded the perceived tone/stiffness during palpation of the m. biceps brachii.

Figure 1. Illustration of measurement of the MyotonPRO

101 Chapter 6

For assessing intrarater reliability and agreement, the MyotonPRO measurement was repeated by the main researcher with the same procedure after 30 min, thereby creating a sufficient time interval to minimalize clinical change12 . After 6 months, the assessments with the PAI, MAS-P, and MyotonPRO were repeated by the main researcher with the same procedure. Simultaneously, the CGI was assessed by the participant’s professional caretaker because cognitive decline in dementia could hamper the validity of self-reporting clinical change.

Statistical Analyses

For the sample size calculation, we established a desired validity correlation coefficient of 0.7 with 95% confidence interval (CI) between 0.57 and 0.81, which resulted in a sample size of 50 for each subgroup (early-, moderate-, and severe-stage dementia). Taking into account a withdrawal percentage of 10%, a total of 165 participants are usually required to achieve the sample size.

Descriptive statistics were used to summarize the means and standard deviations (SD). Since visual inspection of Q-Q plots did not reveal any violations of normality, independent t tests were performed for continuous data and χ2 tests for frequency data in order to analyze differences in characteristics between participants with and without paratonia. Since the severity of paratonia is only determined if it is diagnosed, except for the analyses for construct validity, all other analyses were conducted for participants with paratonia throughout the study. Paratonia is usually generalized throughout the body; therefore, in order to analyze whether data of only one arm could be used, we compared differences between left and right MyotonPRO outcomes with paired t tests. Data were analyzed using SPSS version 22 and taking a P value <0.05 as statistically significant.

Construct Validity

For construct validity, we validated the MyotonPRO with the PAI. Receiver operating characteristic (ROC) curves were constructed by plotting the sensitivity and 1 − specificity for all possible cut-off points of MyotonPRO outcomes in order to discriminate between participants with and without paratonia. We expect that the outcomes from the MyotonPRO on tone and stiffness are higher, and elasticity and viscoelastic properties are lowerin participants with paratonia as the result of hypertonia (e.g., muscle in contracted state) 5. The area under the ROC curve (AUC) reflects the ability of the MyotonPRO scores to differentiate between participants with and without paratonia. The area under theROC curve ranged from 0.5 to 1; a higher score indicates better discrimination 28. Sensitivity and

102 Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

specificity proportions were estimated by applying Youden’s index in which we selected the optimal cut-off point of the ROC curve, where the calculated sensitivity + specificity is maximal 29, and 95% CI were calculated.

Concurrent Validity

Concurrent validity of the muscle properties measured with the MyotonPRO was determined using the Spearman correlation (ρ) test to establish correlation with the MAS-P scores and to establish correlation with muscle palpation of the m. biceps brachii. Correlation coefficients were interpreted as negligible (0 - 0.3), low (0.3 - 0.5), moderate (0.5 - 0.7), high (0.7 - 0.9), or very high (0.9 - 1.0) 30. For the analysis, we used the MAS-P score of the elbow extension as this represents the resistance of passive movement in the m. biceps brachii.

Reproducibility

Reproducibility concerns the degree to which repeated measurements with the MyotonPRO provide similar answers and can be divided into reliability and agreement. Measures of 6 agreement refer to the absolute measurement error, i.e. how close the scores on repeated measures are presented in the units of measurement of the corresponding MyotonPRO outcomes. Measures of reliability refer to the relative measurement error, i.e. the variation between subjects in relationship to the total variance of the measurements 31. Interrater and intrarater (test-retest after 30 min) reliabilities were analyzed through the intraclass correlation coefficient (ICC) using a 2-way mixed model with absolute agreement and single measures where ICC >0.75 is evaluated as being excellent, between 0.40 and 0.75 as being fair to good, and less than 0.40 as being poor 32. For agreement estimation, we used the standard error of measurement (SEM) and the minimal detectable change (MDC). The SEM provides the range within which a participant’s genuine score may fall. The SEM was estimated as SEM = SD × √(1 – ICC), where the SD is the pooled SD of the corresponding MyotonPRO outcome from test and retest 33. The SEM% indicates the relative amount of measurement error and was defined as SEM% = (SEM/mean) × 100, where the mean is that for the corresponding measurements from test-retest. The SEM% of <10% is suggested as being small 34. The MDC represents the magnitude of change necessary to exceed the measurement error of 2 repeated measures at a 95% CI and was calculated as MDC95 = SEM 35 × √2 × 1.96 . MDC95% was defined as MDC95% = (MDC/mean) × 100, where an MDC95% smaller than 10% is suggested to be excellent and an MDC95% smaller than 30% to be acceptable 12,36.

103 Chapter 6

Sensitivity to Change and Responsiveness

Two concepts are applied in the assessment of evaluative instruments. We endorse the recommendation that a distinction should be made between sensitivity and responsiveness. Sensitivity to change refers to the capacity of instruments to statistically measure change37 . Responsiveness addresses the detection of clinically relevant or important change 31. For sensitivity to change, the changes in the MyotonPRO and MAS-P measurements between baseline and 6 months were examined by the Pearson coefficient (r) between score changes. Responsiveness was determined by the minimal clinical important difference (MCID). The MCID was established with an anchor-based method 38 in which longitudinal change after 6 months in the MyotonPRO outcomes was related to an external criterion for important change (the “anchor”), being the CGI for change scale. The CGI “anchor” scores were used to categorize participants into 3 subgroups (CGI scores): “improved” (1 - 3), “no change” (4), and “worse” (5 - 7). To estimate the MCID, the mean change scores on the 5 MyotonPRO outcomes were calculated by subtracting each participant’s 6-month score from the baseline score. The mean change scores of the subgroup reported as being “worse” were used to determine the MCID. We tested for significance score changes between the anchor subgroups using one-way ANOVA.

RESULTS

From the 168 eligible participants, 16 were unable to participate due to various reasons (11 had to be excluded because they were restless, became agitated, or were too ill for assessment, and 5 died). At baseline, 152 participants (101 [66.4%] female and 51 [33.6%] male with a mean age [SD] of 83.5 [8.2]) were assessed, of which 70 (46.1%) were diagnosed with and 82 (53.9%) without paratonia by using the PAI. After 6 months, 41 participants (27%) were lost to follow-up (11 were excluded because of restlessness, agitation, or were too ill to be reassessed, 5 were transferred to unknown addresses, and 26 died). From the 111 remaining participants, 55 (49.5%) were diagnosed with and 56 (50. 5%) without paratonia. From the original 70 participants with paratonia at baseline, 48 (68.6%) remained and were included for longitudinal assessments. Figure 2 illustrates the flow and number of participants through the study. Baseline characteristics are summarized in Table 1.

Paired t test analysis between the left and right arm indicated no significant differences in MyotonPRO outcomes. We used the left m. biceps brachii for analysis because the left arm was mostly the nondominant arm in studies with healthy subjects. By using the data of the nondominant arm, the influence of preceding physical activity was suggested to be reduced22 .

104 Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

Construct Validity

The ROC analyses revealed that all MyotonPRO outcomes had sufficient ability to differentiate between participants with and without paratonia. In participants with paratonia tone, elasticity and stiffness were higher and viscoelasticity was lower. The AUC determined for tone was 0.66, P = 0.001; elasticity 0.61, P = 0.017; stiffness 0.60, P = 0.028; creep 0.67, P <0.001; and MSR time 0.65, P = 0.001. Sensitivity and specificity estimated for tone was 70% (95% CI: 63 - 77%) and 54% (95% CI: 46 - 62%), respectively; elasticity 66% (95% CI: 58 - 74%) and 56% (95% CI: 48 - 64%), respectively; stiffness 60% (95% CI: 52 - 68%) and 59% (95% CI: 51 - 67%), respectively; creep 63% (95% CI: 55 - 71%) and 61% (95% CI: 53 - 69%), respectively; and MSR time 60% (95% CI: 52 - 68%) and 59% (95% CI: 51 - 67%), respectively.

Concurrent Validity

Table 2 shows that a statistically significant correlation was ascertained between the MAS-P score and tone (ρ = 0.42, P <0.001), stiffness (ρ = 0.31, P = 0.009), creep (ρ = -0.39, P = 0.001), and MSR time (ρ = -0.38, P = 0.001), but not for elasticity (ρ = -0.23, P = 0.051). A 6 statistically significant correlation was also found between manual biceps palpation and tone (ρ = 0.37, P = 0.002), stiffness (ρ = 0.28, P = 0.019), creep (ρ = -0.31, P = 0.009), and MSR time (ρ = -0.33, P = 0.005), but not for elasticity (ρ = -0.13, P = 0.279).

Inclusion N= 152

To (baseline): construct validity No paratonia N= 82 (53.9%) vs. Paratonia N=70 (46.1%)

No paratonia N= 82 (53.9%) Paratonia N= 70 (46.1%)

T0 (baseline): concurrent validity T0 (baseline): inter-rater reliability T1 (retest after 30 min): intra-rater reliability and agreement

Dropped out

Paratonia N= 22 (31.4%) Paratonia N= 48 (68.6%)

T2 (6 months): sensitivity to change and responsiveness

Figure 2. Flow and number of participants throughout the study Figure 2. Flow and number of participants throughout the study

105 Chapter 6

Table 1. Baseline characteristics of participants without and with paratonia No Paratonia Paratonia P value n = 82 (53.9%) n = 70 (46.1%)

Female, n (%) 56 (68.3) 45 (64.3) ns Male, n (%) 26 (31.7) 25 (35.7) ns Age, years 83.5 (8.7) 83.5 (7.7) ns Early dementia (GDS 0-4), n (%) 37 (45.1) 14 (20) 0.001 Moderate dementia (GDS 5), n (%) 26 (31.7) 22 (31.4) 0.001 Severe dementia (GDS 6-7), n (%) 19 (23.2) 34 (48.6) 0.001 Alzheimer’s Disease, n (%) 40 (48.8) 32 (45.7) ns Vascular dementia, n (%) 14 (17.1) 16 (22.9) ns Mixed AD/VaD, n (%) 9 (11.0) 11 (15.7) ns Fronto-temporal, n (%) 5 (6.1) 2 (2.9) ns Parkinson/Lewy body, n (%) 1 (1.2) 2 (2.9) ns Otherwise specified, n (%) 13 (15.8) 7 (10.0) ns Dementia duration, months 50.8 (44.0) 61.3 (40.4) ns Co-morbidities, n 1.6 (1.1) 1.5 (1.1) ns CVD, n (%) 34 (41.5) 18 (25.7) 0.022 CeVD (CVA, TIA), n (%) 23 (28.0) 19 (27.1) ns DM, n (%) 16 (19.5) 16 (22.9) ns Internal, n (%) 15 (18.3) 7 (10.0) ns Cancer, n (%) 8 (9.8) 12 (17.1) ns COPD, n (%) 9 (11.0) 5 (7.1) ns Systemic, n (%) 5 (6.1) 8 (11.4) ns Digestive tract,n (%) 3 (3.7) 4 (5.7) ns Musculoskeletal, n (%) 2 (2.4) 3 (4.3) ns Parkinson’s disease, n (%) - 2 (2.9) ns Depression, n (%) 1 (1.2) 1 (1.4) ns Epilepsy, n (%) - 1 (1.4) ns Polypharmacy (≥ 5 meds), n (%) 53 (64.6) 47 (67.1) ns Psychotropics, n (%) 8 (9.8) 14 (20.0) 0.068 MAS-P* 0.4 (0.5) 1.9 (1.1) <0.001 Muscle palpation* hypotonia, n (%) 10 (12.3) 9 (12.9) ns normal, n (%) 65 (80.2) 47 (67.1) ns hypertonia, n (%) 6 (7.4) 14 (20) 0.023 Tone (Hz)* 12.14 (1.34) 13.18 (1.92) <0.001 Elasticity (log decrement)* 1.81 (0.36) 1.67 (0.36) 0.017 Stiffness (N/m)* 234.30 (25.44) 244.70 (36.03) 0.046 Creep (Deborah number)* 1.71 (0.24) 1.54 (0.29) <0.001 MSR time (ms)* 27.51 (3.92) 25.05 (4.58) 0.001 AD = Alzheimer disease; CeVD = cerebral vascular disease; COPD = chronic obstructive pulmonary disease; CVA = cerebral vascular accident; CVD = cardio vascular disease; DM = diabetes mellitus; GDS = Global Deterioration Scale; MSR = mechanical stress relaxation; ns = not statistically significant; TIA = transient ischemic attack; VaD = vascular dementia. Data represent mean values (SD) unless indicated otherwise. *: Data from the left arm.

106 Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

Table 2. Concurrent validity with MAS-P and with perceived tone/ stiffness by manual muscle palpation MyotonPRO MAS-P arm P value Palpation BB P value Spearman Rho Spearman Rho Tone (Hz) 0.42 <0.001 0.37 0.002 Elasticity (log decrement) -0.23 0.051 -0.13 0.279 Stiffness (N/m) 0.31 0.009 0.28 0.019 Creep (Deborah number) -0.39 0.001 -0.31 0.009 MSR time (ms) -0.38 0.001 -0.33 0.005 BB = m. biceps brachi; MSR = mechanical stress relaxation.

Reproducibility

Table 3 shows that the inter- and intrarater reliabilities were fair to good across the MyotonPRO outcomes (ICC: 0.57 - 0.75 and ICC: 0.54 - 0.71, respectively). Intrarater agreement, expressed as the SEM in the units of measurement of the MyotonPRO for tone was 1.17 (Hz), elasticity 0.18 (log decrement), stiffness 21.60 (N/m), creep 0.20 (Deborah number), and MSR time 3.00 (ms). The SEM% for tone was 8.9%, elasticity 11%, stiffness

8.9%, creep 12.6%, and MSR time 11.8%. The MDC95 for tone was 3.24 (HZ), elasticity 0.50 6 (log decrement), stiffness 59.87 (N/m), creep 0.54 (Deborah number), and MSR time 8.29

(ms). The MDC95% for tone was 24.6%, elasticity 29.9%, stiffness 24.7%, creep 34.6%, and MSR time 32.8%.

Sensitivity to Change

A statistically significant correlation was found between baseline and increase inMAS-P scores and increase in tone (r = 0.29, P = 0.043) and decrease in elasticity (r = -0.36, P = 0.011), creep (r = -0.35, P = 0.014), and MRS time (r = -0.34, P = 0.017), but not for increase in stiffness (r = 0.02, P = 0.878) after 6 months (Fig. 3).

Responsiveness

Per subgroup based on the CGI anchors, 29 participants with paratonia were reported with having no change, 19 were reported as worsening, and none of the participants improved. Table 4 shows that the MCID in the units of measurement of the MyotonPRO (SD) for tone was 0.56 (1.86), elasticity 0.11 (0.45), stiffness 1.79 (39.02), creep 0.12 (0.31), and MSR time 1.69 (4.37). One-way ANOVA revealed significant differences in the changes in the MyotonPRO outcomes for tone (P = 0.005), elasticity (P = 0.032), creep (P = 0.001), and MSR time (P = 0.005) among participants who were reported to have worsened, suggesting sufficiently discriminative change categories. There was no statistically significant difference in MAS-P change (P = 0.173) among the change categories.

107 Chapter 6 (%) 95 3.24 (24.6) 0.50 (29.9) 0.54 (34.6) 8.29 (32.8) 59.87 (24.7) MDC 1.17 (8.9) 0.18 (11.0) 21.60 (8.9) 0.20 (12.6) 3.00 (11.8) SEM (%) ICC (95% CI) 0.71 (0.57 – 0.80) 0.60 (0.43 – 0.74) 0.59 (0.41 – 0.70) 0.54 (0.35 – 0.68) 0.57 (0.38 – 0.70) Intrarater reliability Intrarater 1.68 (0.32) 1.58 (0.29) 13.12 (1.77) 25.56 (4.53) T1 rater 1 (SD) T1 rater 239.13 (31.27) ICC (95% CI) 0.70 (0.55 – 0.80) 0.64 (0.48 – 0.76) 0.75 (0.63 – 0.84) 0.57 (0.39 – 0.71) 0.62 (0.45 – 0.74) Interrater reliability Interrater 1.71 (0.41) 1.60 (0.28) 13.00 (1.86) 25.74 (4.30) T0 rater 2 (SD) T0 rater 240.49 (34.78) 1.67 (0.36) 1.54 (0.29) 13.18 (1.92) 25.05 (4.58) T0 rater 1 (SD) T0 rater 244.70 (36.03) reliability and agreement and intrarater inter- Reproducibility: MyotonPRO Table 3. Table Tone (Hz) Tone Elasticity (log decrement) (log decrement) Elasticity Stiffness (N/m) Creep (Deborah number) number) (Deborah Creep MSR time (ms) ICC = intra class correlation; MDC = minimal detectable change; MSR = mechanical stress relaxation; SD = standard deviation; SEM = standard error of measurement. standard = SEM deviation; standard SD = relaxation; stress MSR = mechanical change; MDC = minimal detectable class correlation; ICC = intra 30 minutes. after T0 = baseline; T1 retest

108 Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

Table 4. Mean change scores categorized by CGI Anchor Mean change scores (SD)

CGI Improved No change Worse Anchor n 0 29 19 MyotonPRO Tone (Hz) - 0.95 (1.67) -0.56 (1.86)* Elasticity (log decrement) - -0.21 (0.51) 0.11 (0.45)* Stiffness (N/m) - 7.93 (39.99) 1.79 (39.02) Creep (Deborah number) - -0.21(0.34) 0.12 (0.31)* MSR time (ms) - -2.81 (5.53) 1.69 (4.37)* MAS-P - 0.28 (0.89) -0.05(0.64) MSR = mechanical stress relaxation; MAS-P = modified Ashworth scale for paratonia * P <0.05, statistically significant.

Figure 3. Scatterplots. Correlation between change in MAS-P and change in MyotonPRO outcomes after 6 months Figure 3. Scatterplots. Correlation between change in MAS-P and change in MyotonPRO outcomes after 6 months

109 Chapter 6

DISCUSSION

This study was conducted to assess the psychometric properties of MyotonPRO in order to investigate if this device is an objective and more accurate alternative to theMAS-P in patients with paratonia. With the MAS-P, muscle tone is measured by the perceived resistance during passive movement. The MyotonPRO exerts a local passive movement of the muscle and is able to measure a construct of 5 different parameters of tone and stiffness. We regard the MyotonPRO a feasible tool. Measurement time takes approximately 10 s to about 1 min if it needs to be repeated up to 3 times because of unsatisfactory CV values. The participating physiotherapist found the MyotoPRO easy to use after some practicing. We assessed the validity, reproducibility, sensitivity to change, and responsiveness of the MyotonPRO in a cohort of dementia patients.

Validity

We found evidence that the MyotonPRO is able to differentiate between people with and without paratonia. Muscle tone, elasticity (lower decrement), and stiffness were significantly higher, and viscoelasticity (creep and MSR time) was significantly lower in people with paratonia which is to be expected when a muscle is not relaxed or it contracts 5. Paratonia is an active resistance to passive movement caused by the inability to relax or unintentionally contracted muscles. In view of the sensitivity and specificity values found in this study, the MyotonPRO cannot yet be recommended to diagnose paratonia. Although significant, the correlation coefficients indicate a small correlation between the MyotonPRO outcomes and the MAS-P, which are similar between the MyotonPRO outcomes and manual muscle palpation. The indication of muscle tone by manual palpation proved challenging in our cohort as many participants had low biceps muscle mass, which makes the results questionable. Revealed correlations with the MAS-P are in concordance with the validity study using MAS measurements in stroke survivors 14. Leonard et al. 10 found higher correlation coefficients with the MAS in their study with people with spasticity but with the notification that variables were reduced by clustering MAS scores into fewer categories. Possible explanations for the low correlation with the MAS-P can be given from the fact that MAS measurements have the tendency to cluster scores in the lower range, limiting its ability to discriminate between patients 39. In addition, while the main researcher has experience in using the MAS-P, some subjectivity cannot be ruled out. Notwithstanding that a MAS measurement is the criterion for clinical muscle tone assessment, one could contend that it is not a suitable “golden” standard and hypothesize that the MyotonPRO is more accurate than the MAS-P.

110 Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

Reproducibility

In the current study, inter- and intrareliability for the MyotonPRO was evaluated as fair to good. The best agreement was found for tone, elasticity, and stiffness with small SEM% and acceptable MDC95%. A person exhibits genuine change when the change between repeated 31 measurements is larger than the MDC95 . Viscoelastic properties exceed the SEM% and

MDC95% and, therefore, require large values to detect real change and are not sensitive for detecting small relevant changes. In the study by van Deun et al.22 , the reliability coefficients were higher in people with paratonia, but these were within session reliability values. On the other hand, their between-days reliability coefficients were lower. Resistance in paratonia is known to be variable. In particular, in the early stages of dementia, paratonia can fluctuate between no resistance, actively assisting, and active resistance against passive movement2,3 . We also found that, especially in people with advanced dementia, keeping the participants from moving during the tests was challenging, and the measurements were often repeated because of unsatisfactory CV values. Bias induced by this variability cannot be completely ruled out but is an undeniable and essential clinical presentation of dementia patients with paratonia. Although it is suggested that a series of 10 taps is sufficient for Myoton 6 measurement 21, van Deun et al. 22 found that adding a second series of taps improved reproducibility in patients with paratonia. Future research on diurnal variation is necessary to establish whether paratonia (and thus MyotonPRO measurements) vary over time. This is important for increasing the clinical interpretation and improving reproducibility.

Sensitivity to Change and Responsiveness

The MyotonPRO is able to record change over time in patients with paratonia. Monitoring change over time is important for studying intervention effects, and the MDC and MCID can be used for the interpretation of the MyotonPRO outcome to determine whether the observed change is real or meaningful. To be able to distinguish meaningful or important change from the measurement error, the MCID should be at least as large as the MDC 38. In this study, all of the MyotonPRO outcomes exceeded this threshold, meaning that, if a person has a change score as large as the MCID, it cannot be certain that this change is not due to a measurement error. Peripheral biomechanical changes through cross-linking processes by advanced glycation end-products (AGEs) are suggested as contributing to the perceived resistance in paratonia6 . Cross-linking of intramuscular collagen tissue is associated with increased muscle stiffness and reduced viscoelastic properties 40, which were detected by the MyotonPRO. On the other hand, the MyotonPRO measured higher elasticity in the paratonia group which is not to be expected if there was AGE-induced biomechanical stiffness and elasticity loss. As

111 Chapter 6

mentioned previously, the MyotonPRO registers higher tone, elasticity and stiffness, and lower viscoelasticity in contracted versus relaxed muscles 5. This could indicate that the peripheral biomechanical changes were less prominent and overruled by the inability to relax or hypertonia in this cohort. It might also be hypothesized that AGE induced effects on collagen tissue results in a mechanism to maintain some elastic properties in low quality muscles. Couppé et al. 41 found in their study that collagen concentration was reduced in tendon tissue, whereas cross-linking AGEs concentration was elevated in older aged men versus younger aged men but found no difference in biomechanical properties. Further research is necessary to study this more in depth.

Strengths and Limitations

This is the first study to assess all MyotonPRO properties and study sensitivity to change and responsiveness after 6 months in dementia patients with paratonia. This study also has a number of limitations. First, we were not able to use a true objective standard for determining criterion validity. Laboratory techniques for objectively assessing muscle properties, such as magnetic resonance elastography or ultrasound imaging, are not clinically feasible and often too stressful in this patient group. Second, we experienced difficulties in assessing muscle tone by manual palpation due to the low muscle mass of the m. biceps brachii. Due to this, it was often difficult to determine the exact location of the muscle belly. Validating the MyotonPRO against the MAS and muscle palpation, therefore, might have caused an underestimation of our results. Third, we did not control for variables that could influence muscle properties such as body weight/BMI, room and body temperature, blood flow, alcohol consumption, and the degree of preceding physical activity 22, which could have influenced the results.

Clinical Implications

Accurate and objective measurement of paratonia severity proves challenging. Because the MDC values surpass the MCID, the responsiveness of detecting small relevant changes in this population is complicated. Hence, at this moment, we cannot recommend the MyotonPRO to be used to evaluate paratonia progress over time. The MyotonPRO is more precise and objective than MAS-P measurements and is potentially suitable to evaluate therapeutic interventions, cross-sectional between groups, but because of the inherent characteristics of paratonia (e.g., variability in movement resistance), the outcomes from the MyotonPRO should be interpreted with care. Multiple measurements during a period of time might adjust for the diurnal variation of paratonia.

112 Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

CONCLUSION

The MyotonPRO is potentially applicable to quantify paratonia severity in cross-sectional and controlled intervention studies between groups of people with paratonia. Nevertheless, it appears less suitable to measure clinically relevant intra-individual changes in paratonia. Because of the inherent variability in movement resistance in paratonia, the outcomes from the MyotonPRO should be interpreted with care. Therefore, future research should focus on additional guidelines for MyotonPRO measurements for increasing the clinical interpretation and improving reproducibility in dementia patients with paratonia.

113 Chapter 6

REFERENCES

1. Hobbelen JS, Tan FE, Verhey FR, Koopmans RT, de Bie RA. Prevalence, incidence and risk factors of paratonia in patients with dementia: a one-year follow-up study. Int Psychogeriatr. 2011;23(7):1051-1060. 2. Souren LE, Franssen EH, Reisberg B. Neuromotor changes in Alzheimer’s disease: implications for patient care. J Geriatr Psychiatry Neurol. 1997;10(3):93-98. 3. Hobbelen JSM, Koopmans RTCM, Verhey FRJ, Van Peppen RPS, de Bie RA. Paratonia: a Delphi procedure for consensus definition. J Geriatr Phys Ther. 2006;29(2):50-56. 4. Gavronski G, Veraksits A, Vasar E, Maaroos J. Evaluation of viscoelastic parameters of the skeletal muscles in junior triathletes. Physiol Meas. 2007;28(6):625-637. doi:10.1088/0967- 3334/28/6/002. 5. Schneider S, Peipsi A, Stokes M, Knicker A, Abeln V. Feasibility of monitoring muscle health in microgravity environments using Myoton technology. Med Biol Eng Comput. 2015;53(1):57-66. doi:10.1007/s11517-014-1211-5. 6. Drenth H, Zuidema S, Krijnen W, Bautmans I, van der Schans C, Hobbelen H. Advanced Glycation End-products are Associated with the Presence and Severity of Paratonia in early stage Alzheimer’s Disease. J Am Med Dir Assoc. 2017. doi:10.1016/j.jamda.2017.04.004. 7. Hobbelen JSM, Koopmans RTCM, Verhey FRJ, Habraken KM, de Bie RA. Diagnosing paratonia in the demented elderly: reliability and validity of the Paratonia Assessment Instrument (PAI). Int Psychogeriatr. 2008;20(4):840-852. doi:10.1017/S1041610207006424. 8. Waardenburg, H., Elvers, J.W.H., van Vechgel, F., Oostendorp RAB. Is Paratonie Betrouwbaar te meten? Een betrouwbaarheid onderzoek voor het meten van paratonie met de MAS en de gemodificeerde tonusschaal van Ashworth. Ned Tijdschr voor Physiother. 1999;(2):30-35. 9. Meulenbroek T. Toelichtingsformulier MAS. http://www.meetinstrumentenzorg.nl/ algemenemeetinstrumenten aspx. Accessed January 25, 2017. 10. Leonard CT, Stephens JU, Stroppel SL. Assessing the spastic condition of individuals with upper motoneuron involvement: validity of the myotonometer. Arch Phys Med Rehabil. 2001;82(10):1416-1420. doi:10.1053/apmr.2001.26070. 11. Aarrestad DD, Williams MD, Fehrer SC, Mikhailenok E, Leonard CT. Intra- and interrater reliabilities of the Myotonometer when assessing the spastic condition of children with cerebral palsy. J Child Neurol. 2004;19(11):894-901. 12. Chuang L-L, Lin K-C, Wu C-Y, et al. Relative and absolute reliabilities of the myotonometric measurements of hemiparetic arms in patients with stroke. Arch Phys Med Rehabil. 2013;94(3):459-466. doi:10.1016/j.apmr.2012.08.212. 13. Chuang L, Wu C, Lin K. Reliability, validity, and responsiveness of myotonometric measurement of muscle tone, elasticity, and stiffness in patients with stroke. Arch Phys Med Rehabil. 2012;93(3):532-540. doi:10.1016/j.apmr.2011.09.014. 14. Rydahl SJ, Brouwer BJ. Ankle stiffness and tissue compliance in stroke survivors: a validation of Myotonometer measurements. Arch Phys Med Rehabil. 2004;85(10):1631-1637. 15. Marusiak J, Kisiel-Sajewicz K, Jaskolska A, Jaskolski A. Higher muscle passive stiffness in Parkinson’s disease patients than in controls measured by myotonometry. Arch Phys Med Rehabil. 2010;91(5):800-802. doi:10.1016/j.apmr.2010.01.012. 16. Ratsep T, Asser T. Changes in viscoelastic properties of skeletal muscles induced by subthalamic stimulation in patients with Parkinson’s disease. Clin Biomech (Bristol, Avon). 2011;26(2):213- 217. doi:10.1016/j.clinbiomech.2010.09.014. 17. Agyapong S, Aird L, Bailey L, Mooney K, Mullix J, Warner M, Samuel D SM. Interrater reliability of muscletone, stiffness and elasticity measurements of rectus femoris and biceps brachii in healthy young and older males. Work Pap Heal Sci. 2013;1(4).

114 Psychometric Properties of the MyotonPRO in Dementia Patients with Paratonia

18. Aird L, Samuel D, Stokes M. Quadriceps muscle tone, elasticity and stiffness in older males: reliability and symmetry using the MyotonPRO. Arch Gerontol Geriatr. 2012;55(2):e31-9. doi:10.1016/j.archger.2012.03.005. 19. Mooney K, Warner M SM. Symmetry and within-session reliability of mechanical properties of biceps brachii muscles in healthy young adult males using the Myoton PRO device. Work Pap Heal Sci. 2013;1(3). 20. Bailey L, Samuel D, Warner M SM. Parameters Representing Muscle Tone, Elasticity and Stiffness of Biceps Brachii in Healthy Older Males: Symmetry and Within-Session Reliability Using the Myoton PRO. J Neurol Disord. 2013;1(1). 21. Agyapong-Badu S, Warner M, Samuel D, Stokes M. Measurement of ageing effects on muscle tone and mechanical properties of rectus femoris and biceps brachii in healthy malesand females using a novel hand-held myometric device. Arch Gerontol Geriatr. 2016;62:59-67. doi:10.1016/j.archger.2015.09.011. 22. Van Deun B, Hobbelen JSM, Cagnie B, Van Eetvelde B, Van Den Noortgate N, Cambier D. Reproducible Measurements of Muscle Characteristics Using the MyotonPRO Device: Comparison Between Individuals With and Without Paratonia. J Geriatr Phys Ther. 2016. doi:10.1519/JPT.0000000000000119. 23. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry. 1982;139(9):1136-1139. doi:10.1176/ ajp.139.9.1136. 24. Myoton. Myoton. http://www.myoton.com/technology/. Accessed January 25, 2017. 6 25. Myoton. MyotonPRO User Manual. 6th ed. London: Myoton; 2013. 26. Hoppenfeld S. Physical Examination of the Spine and Extremities. 1st ed. London: Appleton Century Crofts.; 1976. 27. Schneider LS, Olin JT, Doody RS, et al. Validity and reliability of the Alzheimer’s Disease Cooperative Study-Clinical Global Impression of Change. The Alzheimer’s Disease Cooperative Study. Alzheimer Dis Assoc Disord. 1997;11 Suppl 2:S22-32. 28. de Vet HCW, Ostelo RWJG, Terwee CB, et al. Minimally important change determined by a visual method integrating an anchor-based and a distribution-based approach. Qual Life Res. 2007;16(1):131-142. doi:10.1007/s11136-006-9109-9. 29. Youden WJ. Index for rating diagnostic tests. Cancer. 1950;3(1):32-35. 30. Mukaka MM. Statistics corner: A guide to appropriate use of correlation coefficient in medical research. Malawi Med J. 2012;24(3):69-71. 31. Terwee CB, Bot SDM, de Boer MR, et al. Quality criteria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol. 2007;60(1):34-42. doi:10.1016/j. jclinepi.2006.03.012. 32. Fleiss J. Reliability of Measurement. The Design and Analysis of Clinical Experiments. New York: Wiley and sons; 2007. 33. Portney L WM. Statistical Measures of Reliability. Foundations of Clinical Research: Applications to Practice. 2nd ed. Upper Saddle River, New Jersey: Prentice Hall; 2000. 34. Flansbjer U-B, Holmback AM, Downham D, Patten C, Lexell J. Reliability of gait performance tests in men and women with hemiparesis after stroke. J Rehabil Med. 2005;37(2):75-82. doi:10.1080/16501970410017215. 35. de Vet HC, Terwee CB, Ostelo RW, Beckerman H, Knol DL, Bouter LM. Minimal changes in health status questionnaires: distinction between minimally detectable change and minimally important change. Health Qual Life Outcomes. 2006;4:54. doi:10.1186/1477-7525-4-54. 36. Smidt N, van der Windt DA, Assendelft WJ, et al. Interobserver reproducibility of the assessment of severity of complaints, grip strength, and pressure pain threshold in patients with lateral epicondylitis. Arch Phys Med Rehabil. 2002;83(8):1145-1150.

115 Chapter 6

37. Liang MH. Longitudinal construct validity: establishment of clinical meaning in patient evaluative instruments. Med Care. 2000;38(9 Suppl):II84-90. 38. Copay AG, Subach BR, Glassman SD, Polly DWJ, Schuler TC. Understanding the minimum clinically important difference: a review of concepts and methods. Spine J. 2007;7(5):541-546. doi:10.1016/j.spinee.2007.01.008. 39. Blackburn M, van Vliet P, Mockett SP. Reliability of measurements obtained with the modified Ashworth scale in the lower extremities of people with stroke. Phys Ther. 2002;82(1):25-34. 40. Haus JM, Carrithers JA, Trappe SW, et al. Collagen, cross-linking, and advanced glycation end products in aging human skeletal muscle. J Appl Physiol. 2007;103(6):2068-2076. 41. Couppé C, Hansen P, Kongsgaard M, et al. Mechanical properties and collagen cross-linking of the patellar tendon in old and young men. J Appl Physiol. 2009;107(3):880-886.

116

Chapter 7

Summary and General Discussion Chapter 7

Decline in motor function is a common problem in the aging population and is associated with a wide range of adverse health consequences. Different mechanisms contribute to the age-related decline in motor function. One of these that may possibly be a factor in this decline are advanced glycation end-products (AGEs). Since the early 1980s, it has been proposed that the cross-linking of long-lived proteins mediated by AGEs contributes to the age-related decline of the functioning of cells and tissues in normal aging. In addition, the decline in motor function is a consequence of age related diseases such as dementia. Focussing on dementia, motor problems are well described and are considered as an integral element of the diagnosis, however, they are less studied than the cognitive aspects. This is especially the case with paratonia, a specific form of progressive hypertonia and movement stiffness that is experienced by those individuals with dementia. Paratonia was already being described beginning 1828, however, it was not until 2006 that a consensus definition of it was established 1. Despite this, paratonia is still fairly unknown. Although patients, clinicians, and caregivers are being confronted daily with the deterioration in motor function due to paratonia, only minimal attention is still being paid to it in current scientific literature.

The focus of this thesis was to gain insight into paratonia and general motor dysfunction by studying the potential role of AGEs, which have been described as causing mechanical stiffness and loss of elasticity in muscle tissue through collagen cross-linking and/or inflammatory processes. The results of this thesis give rise for further research into motor function decline in the aging population and paratonia in particular.

The main aims of this thesis were to study the contribution of AGEs to the decline in motor function in the aging populationChapters ( 2 and 3) and to study the contribution of AGEs to the pathogenesis of paratonia (Chapter 4) and functional performance in people with dementia Chapter( 5). In addition, this thesis answers the question of whether the MyotonPRO is a valid and reproducible tool that can be used to objectively quantify paratonia severity in people with dementia Chapter( 6).

SUMMARY OF MAIN FINDINGS

Chapter 2 describes the results from the systematic review that was conducted in order to gain insight in the relationship between AGEs and motor function decline. Several studies on the relationship between AGEs and motor function have been published, however, the majority of these studies most often showed an indirect relationship. The review identified eight studies of moderate-to-good quality with a direct relationship between AGEs and motor function. The results indicated that higher levels of AGEs are independently related

120 Summary and General Discussion

to decreased walking abilities, inferior activities of daily living (ADL), diminished muscle properties, and increased physical frailty in a predominantly older population.

Chapter 3 describes the results of a cross-sectional study in 5,624 participants aged 65 years and older from the LifeLines cohort database into the direct relationship between AGE levels and motor dysfunction. In this study, we found evidence that AGEs can be an independent additional explaining factor of decreased physical activity and physical functioning.

Chapters 4 and 5 describe the results of the PARAGE (Paratonia and AGE) study. The PARAGE study was a one-year follow-up study with three measurements (baseline, after six months, and after 12 months) in 144 community dwelling individuals with early stage Alzheimer’s disease (AD) or mixed dementia.

Chapter 4 describes the contribution of AGEs to paratonia. The results show that high levels of AGEs are associated with paratonia and suggest that elevated AGE levels are a contributing factor to both its presence and severity. These findings provide anew perspective on paratonia. In early stage dementia, the AGE induced biomechanical muscle tissue stiffness and elasticity loss might be the resistance that is perceived during passive movement examination (e.g., with the paratonia assessment instrument). This could mean that, besides a central cerebral factor, because of the dementia pathology, a biomechanical 7 peripheral factor is involved in the development of paratonia.

Chapter 5 describes the contribution of AGEs to functional mobility decline and impaired activities of daily living (ADL). The study reveals a significant association between high AGE levels and a decline in functional mobility. This finding is consistent with studies describing the effect of AGEs on the decline of walking abilities and contributes to the increasing evidence that decline in functional mobility can be attributed to the effects of AGEson muscle tissue. Although impaired muscle function - through AGEs-induced muscle damage – can probably contribute to impaired ADL, the results from the PARAGE study did not confirm this in people with early stage dementia. This could be because the sample size of 144 participants was possibly too small and a study duration of one year too brief to detect a decline in ADL, and/or the participants were less ADL independent at baseline.

Chapter 6 describes the study into the MyotonPRO, a noninvasive portable device, for measuring paratonia severity as an alternative to the modified Ashworth scale for paratonia (MAS-P). The MyotonPRO was studied regarding its validity, reproducibility and sensitivity/ responsiveness to change in dementia patients with paratonia. With the MAS-P, muscle tone is measured subjectively by the perceived resistance during passive

121 Chapter 7

movement. The MyotonPRO exerts a local passive movement of the muscle and is able to objectively measure a construct of five different parameters of tone and stiffness. In this study, in dementia patients with (n=70) and without paratonia (n=82), muscle properties were assessed with the MyotonPRO by two assessors within one session and repeated by the main researcher after 30 minutes and again after six months. The results show that accurate and objective measurement of paratonia severity proves to be challenging. The MyotonPRO measurements are more precise and objective than MAS-P measurements and is potentially suitable to evaluate therapeutic interventions, cross-sectional between groups. However, because of the inherent characteristics of paratonia (e.g., variability in movement resistance) the outcomes from the MyotonPRO should be interpreted with care.

GENERAL DISCUSSION

The following section discusses the implications of the findings and the strengths and limitation of this thesis. In addition, clinical implications and future considerations are being discussed. The section will be completed with a number of final remarks.

Contribution of AGEs to Motor Function Decline

The results from this thesis contribute to the increasing evidence that AGE induced impaired skeletal muscle function, through collagen cross-linking and intramuscular inflammation, contributes to motor function decline in the aging population. AGE formation and accumulation contributes to motor function decline and to a consequential decline in physical activity. In addition, decline in physical activity is suggested to increase AGE formation and accumulation. High AGE levels can be considered as a catalyst for the deterioration of motor function and, in this way, a vicious circle is created (see Figure 5). Therefore, high AGE levels can be regarded as a biomarker and risk factor for a decline in motor function that has a subsequent negative influence on age-related diseases.

Given the mutual inter-relationship between the different structures involved in generating movement, the AGE induced damage can relate to these structures individually or even simultaneously. Therefore, besides the peripheral effect on musculoskeletal structures, AGE induced damage on peripheral neural structures could also contribute to impaired motor function by hampering neuro-muscular and/or sensory signalling processes. It also has to be considered that AGE accumulation in the central nervous system (CNS) may affect motor function. AGE accumulation in specific relevant motor-related brain regions might have an effect on the complex inter-relationship between the motor networks within the CNS for generating movement. It is even conceivable that the effect of AGEs on both central and

122 Summary and General Discussion

Hyper glycaemia

Oxidative stress Ageing AGE formation

Physical activity decline AGE accumulation

- Diabetic complications - Cardiovascular disease Motor function decline - Alzheimer’s disease

Functional performance / ADL impairment

- Walking impairment - Muscle strength and mass loss - Paratonia

Figure 5. Vicious circle of the relationship between AGEs and motor function decline peripheral tissue levels might even augment the decline in motor function. Future research 7 is necessary to study this in depth.

Contribution of AGEs to Paratonia

With the novel finding that AGEs are associated with the development of paratonia, a new piece is added to the complex puzzle and contributes to the unravelling of this specific motor problem in those individuals with dementia. The most innovative finding in this thesis is that paratonia appears to be developing on a peripheral, skeletal-muscular level next to the proposed central, cerebral level. Considering the evidence, the findings confirm the hypothesis that, in early stage dementia, the slight to more marked resistance perceived during passive movement is caused by biomechanical changes because of AGE accumulation. It could be hypothesised that, in these early stages, AGE accumulation is a greater factor initiating movement stiffness and that the central cerebral factor builds up or even accelerates paratonia. Here also arises the above-described vicious circle (Figure 5) in which as a result of the progressive dementia process, motor function becomes impaired and, as a consequence, physical activity decreases, which in turn contributes to the accumulation of AGEs. This AGE accumulation again results in the progression of paratonia, and so on. Although the central factor, the cerebral damage caused by the dementia

123 Chapter 7

process, seems to be the most obvious cause of paratonia in the late stages of dementia, we still have little insight into these central mechanisms on paratonia. Severe paratonia, as evidenced by high resistance during movement or active opposition, is especially seen in the later stages of dementia. In these severe stages of dementia, paratonia has been associated with the return of primitive neonatal reflexes 2,3. Also, clinicians and caregivers who deal in daily practice with patients suffering from paratonia have observed that movement stiffness and active opposition (described as gegenhalten 4) during ADL can vary and may occur at random moments. Factors such as aggression, agitation, anxiety, pain, and confusion, but also sudden external stimuli (e.g., light, sound) are noticed as influencing paratonia. AGEs may possibly be involved in the central factor causing paratonia because interaction between AGEs, Amyloid-beta, and tau-protein have been described to affect neuronal function 5. When these neuronal functions are affected in motor function related areas, they may possibly contribute to the pathogenesis of paratonia.

The findings in this thesis suggests that both peripheral and central, AGE induced, damage contribute to the pathogenesis of paratonia. Therefore, targeting AGE accumulation is imperative.

Clinical Implications for the General Aging Population

To prevent AGE induced tissue damage, the inhibition of the formation and accumulation of AGEs might be a factor to reduce or postpone motor function decline in the aging population. Approaches to prevent AGE formation or AGE accumulation can be divided into: pharmacologic, dietary, and exercise strategies. Pharmacologic strategies on AGE formation inhibition and cross-link breaking are being studied, however, results show conflicting evidence, therefore, none of the results can currently be used clinically 6. A restricted intake of AGEs is a possible factor to reduce the AGE burden in the human body. AGEs are present in foods that have undergone the chemical Maillard-reaction that occurs in frying, grilling, broiling, and roasting. A diet low on AGEs is shown to decrease the levels of circulating AGEs and may possibly prevent inflammation and age related diseases, however, evidence of the harmful effects of long-term exposure to dietary AGEs is still inconclusive7 .

The glycation process is accelerated by the excessive elevation of glucose concentration and oxidative stress. Removing these accelerators through intensive glycaemic control and reducing oxidative stress may be the key-method for decreasing AGEs formation. Glucose intake can be influenced by nutrition, and physical activity and exercise has demonstrated that it reduces glycation and oxidative stress and thereby consequently reducing AGE formation 8,9.

124 Summary and General Discussion

The negative effect of AGEs on motor function already begins during midlife and, as AGE levels increase with aging, therefore, an elevated AGEs level as a biomarker could predict a decline in motor function later in life. Elevated AGE-levels can be detected at an early stage fairly easily and non-invasively by using an AGE-reader. This could occur, for example, during regular check-ups at the general practitioner or possibly as a self-assessment. In this way, an increased risk of AGEs-related diseases such as cardio-vascular diseases and the risk of motor function decline can also be detected. Preventive interventions on AGE reduction, therefore, should begin as soon as possible as part of healthy aging.

Clinical Implications for Dementia Patients With and Without Paratonia

The studies in this thesis involving individuals with paratonia show that, in the early stages of dementia, the variability in movement resistance is frequently present. This is in accordance with previous paratonia studies. This variability suggests that paratonia can be influenced. It is suggested that the phenomenon of active assistance (described as mitgehen 3) or the inability to relax could be seen as a precursor of active resistance or severe paratonia 3. It could be hypothesized that, when this active assistance is observed together with an increased AGE level, the risk of developing paratonia increases. Targeting AGE accumulation early on, preventing peripheral musculoskeletal tissue damage, and even influencing damage in the CNS might be a method to combat paratonia. 7

As mentioned earlier, the effects of paratonia hamper motor function already in the early stages but, in the late stages of dementia, the effects of paratonia are devastating. Daily, clinicians and caregivers of people with dementia who experience paratonia must deal with the consequences of paratonia. In the Netherlands, practice-based interventions such as stabilizing cushions, special bed mattresses, and the concept of Passivity in Daily Life (PDL) are used, however, scientific evidence of any effects is lacking. Several years ago, passive movement therapy was the primary physiotherapeutic intervention but proved to be ineffective and even harmful for patients with paratonia in the severe stages of dementia 10. These findings resulted in physiotherapists being no longer, or much less, involved with motor problems in individuals with dementia in health care institutions (nursing homes and small-scale living accommodations).

In the Netherlands, the elderly population in these health care institutions are structurally physically inactive 11. Given the fact that still no effective interventions are available to prevent or postpone paratonia, the lack of physical activity and reduced involvement of the physiotherapists in these health care institutions is an undesirable development. Because of the positive effects of exercise on the formation of AGEs, next to the effect on physical

125 Chapter 7

functioning and performance, the physiotherapist should be involved in this frail group of elderly people with dementia. In this way, tailor-made exercise programs can be made and motor function decline (e.g., paratonia) could be prevented or postponed.

Future Considerations

The results of the studies presented in this thesis gives rise to several in depth future considerations.

Motor function decline in pre-clinical dementia The most common cause of dementia is Alzheimer’s disease (AD) which is usually associated with a progressive decline in cognitive function, especially memory loss. AD is characterized by amyloid plaques and neurofibrillary tangles that are present in high numbers in the grey matter of the affected brain. AGEs have been shown to be associated with lower grey matter volume and to be accumulating in brain tissue and are suggested to explain many biochemical and neuropathological changes in AD 5,12,13. In addition to cognitive decline, accompanying decline in motor function is frequently reported and is one element of the dementia diagnosis. It has been ascertained that, several years prior to the dementia diagnosis, motor function decline is already observed in pre-clinical people with dementia compared to the control group 14. Several authors suggest that a decline in motor function and/or a lower level of physical activity are risk factors for future AD15–19 . However, examining the literature of the potential role of AGEs in AD combined with the findings on the role of AGEs on motor function raises an interesting question: Is motor function decline a first sign of dementia or is it a risk factor for getting dementia? Either way, as already mentioned, motor function decline resulting in decline in physical activity increases AGE accumulation that subsequently has a negative effect on motor function. Because this motor function decline can precede cognitive impairment by several years, it is important to detect these high AGE levels as early as possible in order to prevent further decline in motor function. It could even be hypothesized that lowering AGE levels may possibly contribute to postponing or slowing down the dementia process. Therefore, future research is necessary to study the long-term effect of AGEs on the different types of dementia.

The early motor function decline in mild cognitive impairment and early dementia presents itself, among other things, with slow, unsteady, and stiff movements that could be the result of biomechanical changes caused by AGEs. In this context, it would be interesting to study if this movement stiffness is already present in the general older population without dementia or is this movement stiffness specific for people with dementia where AGEs accumulation is accelerated, and this movement stiffness is (an early form of) paratonia. This could also

126 Summary and General Discussion

contribute to unravelling the peripheral and/or central component of paratonia. Hence, it would be important to study the presence of movement stiffness and the relationship with high AGE levels in older people without dementia.

Re-evaluating paratonia The results from this thesis gives rise to re-evaluate the current state regarding paratonia. For example, differentiation between the broad spectrum of severe hypertonia to active opposition during passive movement and movement stiffness due to biomechanical muscular changes may possibly be necessary. This is especially important for future studies into effective interventions. Severe hypertonia and active opposition may require other interventions than peripheral biomechanical changes.

For diagnosing paratonia, the paratonia assessment instrument (PAI) is the only valid measurement available where the severity of the perceived resistance during the PAI is scored with the modified Ashworth scale for paratonia (MAS-P). Still, diagnosing paratonia proves to be challenging and experience is necessary, especially for detecting slight movement resistance during assessment with the PAI. Therefore, future research should also study and develop instruments that are more objective to diagnose paratonia.

The results of the MyotonPRO study may provide an impetus to study whether the 7 MyotonPRO is suitable for the diagnosis of paratonia. In the study with the MyotonPRO, it was established that it was able to discriminate participants with and without paratonia. Because of the sensitivity and specificity values that were found, the MyotonPRO cannot yet be recommended for diagnosing paratonia. Using medical imaging technology such as functional magnetic resonance imaging (fMRI) in future studies in people with dementia could provide more clarity on which brain regions are involved in paratonia. Further studies with the MyotonPRO, possible in combination with fMRI technology, could lead to an objective instrument to diagnose paratonia. Using the MyotonPRO as an alternative for the subjective MAS-P has potential; however, future studies should focus on additional guidelines for MyotonPRO measurements, for example, multiple measurements during a period of time similar to blood pressure measurement. This could possibly adjust for the diurnal variation of paratonia and increase the clinical interpretation and improve reproducibility. Until the time that objective measuring instruments are available to measure the presence and severity of paratonia, it is recommended that, for subsequent studies involving paratonia, multiple outcome measures are measured on different domains (e.g., functional, behavioural, care burden) together with the PAI.

127 Chapter 7

AGE targeting A combined approach of dietary and physical exercise possibly in combination with future drug therapy might be the most effective way of targeting AGE formation, also because the multidisciplinary approach in glycaemic control has proven to be effective in diabetes patients. Future studies on targeting AGE levels, therefore, should include these multidisciplinary interventions.

Regular physical activity is considered to be effective for maintaining health or preventing disability in general, however, the underlying physiological pathways remain unclear. The results from this thesis provide indications for AGE accumulation reduction to contribute to health maintenance. The exact way in which AGEs play a role has yet to be further investigated. For example, it remains unclear what these physical activities should entail and at what frequency and at what intensity level they should be done in order to reduce AGE accumulation. It is, therefore, important that future research examines the modalities of physical activity and exercises that significantly reduce AGE levels in the aging population in general and specifically for people with dementia. This may include low or high intensity strength and/or endurance exercises or maybe simple, everyday physical activities. This would help to provide specific, customized advice and exercises which is especially important because combating central AGE accumulation may require different exercise modalities than combating peripheral AGE accumulation. Even cross-linking AGEs may require a different approach than non-cross-linking AGE.

Strengths and Limitations

Motor function, and specifically paratonia, in people with dementia has only been minimally studied while the consequences have such a significant impact on the quality of life of the patient and the caregiver. A major strength of this thesis is that, with the PARAGE and MyotonPRO study, we had the opportunity to longitudinally study a large cohort of people with dementia in different settings throughout the Netherlands. The limitation of the majority of studies in this thesis studying the relationship between motor function and AGEs is that a causal relationship cannot be inferred. The results demonstrate that high AGE levels are associated with several motor function impairments in the aging population and with paratonia and decline in functional performance in people with dementia. Future studies should focus on investigating AGE targeted interventions on motor function, which could be considered as a proof of concept underpinning the causal relationship of AGEs and decline in motor function.

128 Summary and General Discussion

FINAL REMARKS

Glycation, the spontaneous non-enzymatic reaction of sugars with proteins and lipids that results in AGEs is a topic of increasing importance in human health. There is growing evidence that AGEs play an important role in aging and age-related diseases. This thesis shows that there are strong indications that AGE induced protein damage, collagen cross- links, and inflammatory processes leading to musculoskeletal tissue impairment as well as peripheral and central neural function loss contributes to the age-related decline in motor function.

With the indications from this thesis that AGEs are involved in the pathogenesis of paratonia, we finally have something with which to pursue further research in combating paratonia, at least in the early stages of dementia, which might possibly prevent severe paratonia. The results of this thesis are a starting point for further research into the relationship between AGEs and paratonia to keep people with dementia independent longer and improve quality of life and daily care of patients suffering from paratonia.

Finally, it is beneficial to reflect on the fact that the population of Western countries is aging. This is consequently accompanied by having a greater number of chronic diseases and places extensive burdens on health care finances 20. There is growing evidence that preventive 7 measures such as exercise and healthy diet as part of a healthy lifestyle reduce the risk on and improve the prognosis of chronic diseases. The results from this thesis contribute to this by providing evidence for the role of AGEs as a risk factor for motor function decline and, as a consequence, decline in functioning and risk on age-related diseases. Therefore, it is important that preventive measures on lowering AGEs begin as soon as possible. This also means that healthcare and health insurance companies should invest in these preventive, relatively low-cost interventions.

129 Chapter 7

REFERENCES

1. Hobbelen JSM, Koopmans RTCM, Verhey FRJ, Van Peppen RPS, de Bie RA. Paratonia: a Delphi procedure for consensus definition. J Geriatr Phys Ther. 2006;29(2):50-56. 2. Vahia I, Cohen CI, Prehogan A, Memon Z. Prevalence and impact of paratonia in Alzheimer disease in a multiracial sample. Am J Geriatr Psychiatry. 2007;15(4):351-353. doi:10.1097/ JGP.0b013e31802ea907. 3. Beversdorf DQ, Heilman KM. Facilitory paratonia and frontal lobe functioning. Neurology. 1998;51(4):968-971. 4. Kleist, K. Gegenhalten (motorischer Negativismus) Zwanga greifen und Thalamus Opticus. Monatschr Psychiat Neurol. 1927;65:317. 5. Rahmadi A, Steiner N, Munch G. Advanced glycation endproducts as gerontotoxins and biomarkers for carbonyl-based degenerative processes in Alzheimer’s disease. Clin Chem Lab Med. 2011;49(3):385-391. 6. Nenna A, Spadaccio C, Lusini M, Ulianich L, Chello M, Nappi F. Basic and Clinical Research Against Advanced Glycation End Products (AGEs): New Compounds to Tackle Cardiovascular Disease and Diabetic Complications. Recent Adv Cardiovasc drug Discov. 2015;10(1):10-33. 7. Puyvelde K Van, Mets T, Njemini R, Beyer I, Bautmans I. Effect of advanced glycation end product intake on inflammation and aging: a systematic review. Nutr Rev. 2014;72(10):638-650. 8. Ahmed N, Thornalley PJ. Advanced glycation endproducts: what is their relevance to diabetic complications? Diabetes, Obes Metab. 2007;9(3):233-245. 9. Magelhaes PM, j. Appell H, Duarte JA. Involvement of advanced glycation end products in the pathogenesis of diabetic complications: the protective role of regular physical activity. Eur Rev Aging Phys Act. 2008;(5):17-29. 10. Hobbelen JHSM, Tan FES, Verhey FRJ, Koopmans RTCM, de Bie RA. Passive movement therapy in severe paratonia: a multicenter randomized clinical trial. Int psychogeriatrics. 2012;24(5):834- 844. doi:10.1017/S1041610211002468. 11. Hiemstra A. Factsheet Belang van Bewegen Voor Ouderen. Ede; 2016. 12. Li J, Liu D, Sun L, Lu Y, Zhang Z. Advanced glycation end products and neurodegenerative diseases: mechanisms and perspective. J Neurol Sci. 2012;317(1-2):1-5. doi:10.1016/j.jns.2012.02.018. 13. Spauwen PJJ, van Eupen MGA, Kohler S, et al. Associations of advanced glycation end-products with cognitive functions in individuals with and without type 2 diabetes: the maastricht study.J Clin Endocrinol Metab. 2015;100(3):951-960. doi:10.1210/jc.2014-2754. 14. Ramakers IHGB, Visser PJ, Aalten P, et al. Symptoms of preclinical dementia in general practice up to five years before dementia diagnosis. Dement Geriatr Cogn Disord. 2007;24(4):300-306. doi:10.1159/000107594. 15. Camicioli R, Howieson D, Oken B, Sexton G, Kaye J. Motor slowing precedes cognitive impairment in the oldest old. Neurology. 1998;50(5):1496-1498. 16. Buchman AS, Bennett DA. Loss of motor function in preclinical Alzheimer’s disease. Expert Rev Neurother. 2011;11(5):665-676. 17. Buracchio T, Dodge HH, Howieson D, Wasserman D, Kaye J. The trajectory of gait speed preceding mild cognitive impairment.Arch Neurol. 2010;67(8):980-986. doi:10.1001/archneurol.2010.159. 18. Wang L, Larson EB, Bowen JD, van Belle G. Performance-based physical function and future dementia in older people. Arch Intern Med. 2006;166(10):1115-1120. doi:10.1001/ archinte.166.10.1115.

130 Summary and General Discussion

19. Aggarwal NT, Wilson RS, Beck TL, Bienias JL, Bennett DA. Motor dysfunction in mild cognitive impairment and the risk of incident Alzheimer disease. Arch Neurol. 2006;63(12):1763-1769. doi:10.1001/archneur.63.12.1763. 20. W.H.O. World Report on Ageing and Health 2015. Geneva; 2015.

131

Samenvatting Dankwoord Curriculum vitae Dissertations of research institute SHARE Samenvatting

SAMENVATTING

Motorfunctie, paratonie en versuikering onderling verbonden bij ouderen

Door de dubbele vergrijzing (er komen steeds meer ouderen en die ouderen worden ook steeds ouder) zien we ook een toename aan leeftijd gerelateerde problemen, immers “de ouderdom komt met gebreken “. Kenmerkend bij veroudering is dat het bewegen (de motorische functies) verandert en minder soepel wordt. Verschillende mechanismen dragen bij aan de afname van de motorische functies. Eén van de mechanismen die een rol hierbij zou kunnen spelen zijn beschadigde eiwitten door suikers (in het Engels advanced glycation end-products of kortweg AGE’s). Sinds het begin van de jaren tachtig wordt al beschreven dat deze AGE’s bijdragen aan de achteruitgang van het functioneren van cellen bij normale veroudering en betrokken zijn bij aandoeningen als suikerziekte en de ziekte van Alzheimer. Verschillende ziekten hebben ook een effect op het bewegen, bijvoorbeeld dementie. Bij dementie zijn stoornissen in motorische functies wel beschreven en zijn onderdeel van de diagnose, alleen wordt dit minder bestudeerd dan bijvoorbeeld het geheugen. Vooral paratonie, een specifieke vorm van verhoogde spierspanning en bewegingsstijfheid bij dementie, wordt weinig bestudeerd. Ondanks dat paratonie al werd beschreven in 1828 is paratonie nog steeds vrij onbekend. In 2006 is er een nieuwe consensus definitie vastgesteld waarmee onderzoek naar paratonie beter mogelijk werd. Hoewel patiënten, artsen en zorgverleners dagelijks worden geconfronteerd met de achteruitgang van het bewegen als gevolg van paratonie, wordt er in de huidige wetenschappelijke literatuur nog steeds beperkt aandacht aan besteed. Eerder onderzoek heeft ons geleerd dat paratonie veel voorkomt en dat mensen met dementie en suikerziekte een bijna 11 keer grotere kans hebben om paratonie te ontwikkelen dan mensen met dementie die geen suikerziekte hebben.

De focus van dit proefschrift was om inzicht te krijgen in paratonie en de algemene achteruitgang van motorische functies door de mogelijke rol van AGE’s te bestuderen. Van AGE’s is bekend dat ze mechanische stijfheid en verlies van elasticiteit in spierweefsel veroorzaken door dwarsverbindingen in het bindweefsel te vormen en/ of ontstekingsprocessen. De resultaten van dit proefschrift geven een aanzet tot verder onderzoek naar het afnemen van motorische functies bij de vergrijzende bevolking en naar paratonie in het bijzonder.

De belangrijkste doelstellingen van dit proefschrift waren het bestuderen van de bijdrage van AGE’s aan de achteruitgang van motorische functies bij ouderen hoofdstuk( 2 en 3), de bijdrage van AGE’s aan de ontwikkeling van paratonie (hoofdstuk 4) en de achteruitgang in functionele activiteiten bij mensen met dementiehoofdstuk ( 5). Bovendien beantwoordt

134 Samenvatting

dit proefschrift de vraag of de MyotonPRO een geschikt instrument is om de ernst van paratonia objectief te meten bij mensen met dementiehoofdstuk ( 6).

SAMENVATTING VAN DE BELANGRIJKSTE BEVINDINGEN

Hoofdstuk 2 beschrijft de resultaten van de literatuurstudie die werd uitgevoerd om inzicht te krijgen in de relatie tussen AGE’s en achteruitgang van motorische functies. Verschillende artikelen over de relatie tussen AGE’s en motorische functies zijn gepubliceerd, maar de meerderheid van deze artikelen laten een indirecte relatie zien. Deze literatuurstudie vond acht artikelen met een directe relatie tussen AGE’s en achteruitgang van motorische functies. De resultaten laten zien dat hoge concentraties van AGE’s gerelateerd zijn aan verslechtering van het lopen, achteruitgang in activiteiten van het dagelijks leven (ADL), verminderde spiereigenschappen en een toegenomen lichamelijke kwetsbaarheid bij ouderen.

Hoofdstuk 3 beschrijft de resultaten van het LifeLines bevolkingsonderzoek, bij 5.624 deelnemers van 65 jaar en ouder naar de directe relatie tussen AGE’s en stoornissen in motorische functies. Uit deze studie blijkt dat een verhoogde AGE’s concentratie een aanvullende verklarende factor is voor de vermindering van lichamelijke activiteit en functioneren.

Hoofdstuk 4 en 5 beschrijven de resultaten van de PARAGE (Paratonie en AGE) studie. De PARAGE-studie was een studie van 1 jaar met 3 metingen (baseline, na 6 en 12 maanden) bij 144 thuiswonende mensen in het vroege stadium van de ziekte van Alzheimer of gemengde dementie.

Hoofdstuk 4 beschrijft de bijdrage van AGE’s aan paratonie. De resultaten tonen aan dat hoge concentraties van AGE’s een verband hebben met paratonie en dat deze verhoogde AGE’s concentraties een bijdragende factor kunnen zijn aan zowel de aanwezigheid als de ernst van paratonie. Deze bevindingen geven een nieuw inzicht in paratonie. In het vroege stadium van dementie, zou de door AGE’s veroorzaakte spierweefselstijfheid de weerstand kunnen zijn die wordt waargenomen tijdens het passief bewegen zoals tijdens het bewegingsonderzoek met de paratonia assessment instrument (PAI). Dit zou kunnen betekenen dat, naast een factor vanuit de hersenen, vanwege de schade door de dementie, een beschadiging in de spier zelf betrokken is bij de ontwikkeling van paratonie.

Hoofdstuk 5 beschrijft de bijdrage van AGE’s aan de afname van de functionele mobiliteit (opstaan, lopen, draaien en gaan zitten) en de activiteiten van het dagelijks leven (ADL). De studie laat een significant verband zien tussen een hoge AGE’s concentratie en afname van

135 Samenvatting

de functionele mobiliteit. Deze bevinding komt overeen met onderzoeken die het effect beschrijven van AGE’s op de achteruitgang van het lopen en draagt bij aan het toenemende bewijs dat achteruitgang in functionele mobiliteit kan worden toegeschreven aan de effecten van AGE’s op spierweefsel. Hoewel verminderde spierfunctie - door AGE’s veroorzaakte spierschade - waarschijnlijk kan bijdragen aan de afname in de ADL, bevestigden de resultaten van de PARAGE-studie dit niet bij mensen in het vroege stadium van dementie. Dit kan zijn omdat het aantal van 144 deelnemers mogelijk te klein was en een studieduur van een jaar te kort om een afname in ADL te vinden en/of de deelnemers waren al minder ADL-onafhankelijk bij aanvang van de studie.

Hoofdstuk 6 beschrijft de studie naar de MyotonPRO, een patiëntvriendelijk, draagbaar apparaat om de ernst van paratonia te meten als alternatief voor de modified Ashworth-scale for paratonia (MAS-P). De MyotonPRO werd onderzocht of deze geschikt en betrouwbaar is om de ernst van paratonie te meten bij dementiepatiënten met paratonie. Met de MAS-P wordt de spierspanning subjectief gemeten door de waargenomen weerstand tijdens passieve beweging. De MyotonPRO oefent een plaatselijke passieve beweging van de spier uit en is in staat objectief vijf verschillende eigenschappen van spierspanning en -stijfheid te meten. In deze studie werd bij mensen met dementie, met en zonder paratonie, de spiereigenschappen met de MyotonPRO gemeten door twee beoordelaars en herhaald door de hoofdonderzoeker na 30 minuten en opnieuw na zes maanden. De resultaten laten zien dat een nauwkeurige en objectieve meting van de ernst van paratonia lastig blijkt. De MyotonPRO is nauwkeuriger en objectiever dan metingen met de MAS-P en lijkt geschikt om mensen met paratonie onderling te vergelijken. Vanwege de typische karakteristieken van paratonie (zoals veranderingen in bewegingsweerstand) moeten de uitkomsten van de MyotonPRO voorzichtig worden geïnterpreteerd.

Hoofdstuk 7 geeft een samenvatting van de belangrijkste bevindingen en beschrijft uitgebreid de algehele discussie.

Weefselschade door de versuikering van eiwitten is een onderwerp dat steeds meer aandacht krijgt in de gezondheidszorg. Er is toenemend bewijs dat AGE’s een belangrijke rol spelen bij het verouderingsproces zelf en bij leeftijd gerelateerde ziekten. Dit proefschrift laat zien dat er sterke aanwijzingen zijn dat door AGE’s veroorzaakte weefselstijfheid en ontstekingsprocessen in de spier, leidt tot stoornissen in de spierfunctie en hierdoor bijdraagt aan de afname van de motorische functies bij ouderen.

De vorming van AGE’s draagt bij aan de achteruitgang van de motorische functies en daaruit voortvloeiend aan de achteruitgang van lichamelijke activiteiten. Bovendien zou

136 Samenvatting

de achteruitgang van lichamelijke activiteit de vorming van AGE’s weer verhogen. Op deze manier ontstaat als het ware een vicieuze cirkel. Hoge AGE-niveaus kunnen daarom worden gezien als een indicatie van en risicofactor voor verslechtering van het bewegen, wat vervolgens een negatieve invloed heeft op leeftijdsgebonden ziekten.

Deze vicieuze cirkel van hoge AGE’s > stoornissen in de motorfunctie > verminderde lichamelijke activiteiten > toename van AGE’s is ook van toepassing op de bijdrage van AGE’s aan paratonie (bewegingsstijfheid) bij mensen met dementie. De bevindingen in dit proefschrift bevestigen het idee dat, in een vroeg stadium van dementie, de lichte tot meer uitgesproken bewegingsstijfheid wordt veroorzaakt door beschadigingen in de spier als gevolg van AGE’s. Er kan worden verondersteld dat, in het beginstadium, AGE’s een factor zijn die bijdragen aan de bewegingsstijfheid en dat de factor hersenbeschadiging bij dementie de paratonia verder versterkt of zelfs versnelt.

Met de aanwijzingen uit dit proefschrift dat AGE’s betrokken zijn bij het ontwikkelen van paratonie hebben we eindelijk iets in handen om verder onderzoek te doen naar de bestrijding van paratonie in de vroege stadia van dementie. Waarbij mogelijk ook ernstige paratonia voorkomen kan worden. De resultaten van dit proefschrift zijn een startpunt voor verder onderzoek naar de relatie tussen AGE’s en paratonie om mensen met dementie langer onafhankelijk te houden en de kwaliteit van leven en de dagelijkse zorg voor patiënten met paratonie te verbeteren.

Tot slot is het goed om na te denken over het feit dat de bevolking van de westerse landen steeds ouder wordt. Dit gaat gepaard met meer chronische ziektes met bijbehorende financiële gevolgen. Er is groeiend bewijs dat preventieve maatregelen als lichaamsbeweging en gezonde voeding, als onderdeel van een gezonde levensstijl, het risico verminderen op en de prognose verbeteren van chronische ziekten. De resultaten van dit proefschrift dragen hieraan bij door bewijs te leveren voor de rol van AGE’s als een risicofactor voor achteruitgang van de motorische functies en, als gevolg daarvan, een achteruitgang in het bewegend functioneren en risico op leeftijd gerelateerde ziekten. Daarom is het belangrijk dat preventieve maatregelen om AGE’s te verlagen zo snel mogelijk van start gaan.Dit betekent ook dat zorg- en zorgverzekeraars moeten investeren in deze preventieve, relatief goedkope interventies.

137 Dankwoord

DANKWOORD

Toen ik in 2013 begon aan dit promotieonderzoek had ik geen idee wat deze reis allemaal te weeg zou brengen. Tijdens mijn masterstudie kwam ik in aanraking met het doen van wetenschappelijk onderzoek en merkte dat ik me hierin graag verder wilde ontwikkelen. Ik was daarom ook erg blij dat ik de mogelijkheid kreeg om dit promotieonderzoek te gaan doen, met een onderwerp wat mij als geriatrie fysiotherapeut nauw aan het hart ligt. In mijn omgeving kreeg ik destijds geregeld de vraag of ik wel wist waar ik aan begon; promotieonderzoek en dat met een jong gezin, volledige baan en diverse andere bezigheden. Nu ik aan het eind van dit project ben besef ik dat het veel tijd en energie heeft gekost, maar ook dat ik er ontzettend veel voor terug gekregen heb. Als mens ben ik gegroeid, heb ontzettend veel geleerd en vele interessante mensen ontmoet. De reis is achter de rug en wat een avontuur is het geweest. Deze bestemming is nu bereikt, maar mijn reis begint nog maar net, laat maar komen.

Het doen van wetenschappelijk onderzoek, het schrijven van artikelen en de totstandkoming van dit proefschrift was niet mogelijk geweest zonder de hulp en ondersteuning van vele mensen.

In de eerste plaats wil ik mijn vaste begeleidingsgroep, Prof. dr. Cees van der Schans, dr. Hans Hobbelen, Prof. dr. Sytse Zuidema en Prof. dr. Ivan Bautmans bedanken voor de geweldige begeleiding, steun en geduld. Jullie hebben mij het vertrouwen gegeven dat ik dit kon en als ik jullie hulp nodig had stonden jullie altijd klaar. Door de ongelofelijke hoeveelheid kennis die jullie bezitten en manier van begeleiden had ik een heus “dream-team” om mij heen. Hans, daarnaast bedank ik jou ook voor de nodige morele ondersteuning, je relativeringsvermogen en het hart onder de riem wat ik toch af en toe wel nodig had.

Anke Huizenga, bestuurder ZuidOostZorg, ben ik veel dank verschuldigd. Dankzij jouw visie, te willen investeren in wetenschap en de ruimte die ik kreeg werd het mij mogelijk gemaakt om dit promotietraject te doen. Daarnaast heb je mijn ambities in wetenschap altijd ondersteund.

Alle zorgmedewerkers, verpleegkundigen, psychologen, specialisten ouderengeneeskunde, huisartsen, collega (geriatrie) fysiotherapeuten, deelnemende cliënten en hun familie/ mantelzorgers van de volgende instellingen die aan mijn onderzoeken hebben meegewerkt ben ik veel dank verschuldigd; Axioncontinu (IJsselstein; Isselwaerde, Utrecht; Koningsbruggen en de Ingelanden), De Hoven (Delfzijl; Vliethoven), Dementienetwerk TINZ (Friesland), Dignis (Zuidlaren; de Enk),

138 Dankwoord

Elkander/KwadrantGroep (Veenwouden; Talma Hus), Friese Staten (Sneek; Tinga State, Stiens;Hayema State), ICare (Assen; Boshof, Ceresstaete en Kloosterakker, Borger; De Herik, Beilen; Altingerhof, Gieten; Dekelhem, Westerbork; Altingerhoes), Interzorg (Assen; Anholt, Nieuw Graswijk en de Wijde Blik), Laurens (Rotterdam; Arcadia, Blijvenburg, de Hofstee en Stadzicht), Liberein (Enschede; Ariënsstaete), Livio (Enschede; Broekheurnerstede, Haaksbergen; het Saalmerink), Patyna (Bolsward; Bloemkamp, Joure; de Flecke, Sneek; Ielanen en Frittemahof), Zonnehuisgroep Noord (Leek; Vredewold, Zuidhorn; Zonnehuis Oostergast), Zorgboerderij BoerenBlij (Haskerhorne), Zorggroep Noorderboog (Meppel; Reggersoord en Argusvlinder, Steenwijk; Zonnekamp), Zorggroep Sint Maarten (Heerenveen; Mariënbosch), ZorgSpectrum (Houten; het Haltna Huis, Nieuwegein; Vreeswijk, Vianen; Hof van Batenstein), ZuidOostZorg (Appelscha; Stellinghaven, Drachten; NeiBertilla, Rispinge, de Wiken en Sunenz, Gorredijk; de Miente en Voltawerk, Haulerwijk; Sinnehiem, Oosterwolde; Stellinghaven).

Prof. dr. Andries Smit wil ik bedanken voor zijn kennis met betrekking tot advanced glycation end-products en de AGE-reader.

Jaron Brinkhuizen, bedankt voor je hulp bij de LifeLines databases.

Mijn collega fysiotherapeuten van ZuidOostZorg, die begrepen dat mijn ambities verder gingen dan de fysiotherapie, ben ik dankbaar voor hun steun en begrip dat ik niet altijd op de werkvloer aanwezig kon zijn.

Medewerkers van de onderzoeksgroep Healthy Ageing, Allied Health Care and Nursing; Judith, Ineke en Trudy, naast jullie secretariële ondersteuning hebben jullie mij, ondanks dat ik als externe promovendus weinig op locatie was, altijd het gevoel gegeven dat ik erbij hoorde. Wim bedankt voor je hulp in de wondere wereld van de statistiek.

Vrienden, vriendinnen en familie, ik dank jullie voor de niet afnemende belangstelling en steun tijdens de afgelopen jaren.

Mijn moeder, ze was apetrots toen ik met promoveren begon. Helaas heeft ze de promotie niet meer mogen meemaken, ze had het prachtig gevonden.

Tot slot wil ik de drie belangrijkste personen in mijn leven bedanken. Sjoukje en mijn dochters Ilse en Ellen. Ik weet dat jullie het heel gewoon vinden, maar zonder jullie steun, bemoediging, begrip en liefde was dit hele traject voor mij onmogelijk geweest.

139 Curriculum vitae

CURRICULUM VITAE

Hans Drenth werd geboren op 29 februari 1968 in Den Haag en verhuisde in 1974 naar Friesland. Na het behalen van zijn middelbareschooldiploma in 1986 aan het Drachtster Lyceum begon hij aan de inservice-opleiding voor verpleegkundige in ziekenhuis “de Tjongerschans” in Heerenveen en behaalde zijn getuigschrift voorbereidende periode opleiding verpleegkunde in 1987. Hans ontdekte dat hij liever fysiotherapeut wilde worden en begon in datzelfde jaar aan de opleiding fysiotherapie aan de academie voor fysiotherapie in Leeuwarden. In 1991 behaalde hij zijn bachelor diploma fysiotherapie. Na werkzaam te zijn geweest in verschillende particuliere praktijken ging hij als fysiotherapeut werken in toenmalig verpleeghuis “Bertilla” in Drachten. In 1999 rondde hij zijn post-bachelor opleiding fysiotherapie in de geriatrie in Utrecht af. In 2012 behaalde Hans zijn master in de geriatrie-fysiotherapie in Utrecht en werd tijdens deze studie zijn belangstelling voor wetenschappelijk onderzoek gewekt. In datzelfde jaar kwam hij als onderzoeker bij het lectoraat Healthy Ageing, Allied Health Care and Nursing aan de Hanzehogeschool Groningen. In 2013 begon zijn promotieonderzoek met het PARAGE-onderzoek vanuit dit lectoraat, UMC/RU Groningen en de Vrije Universiteit Brussel. Op dit moment is Hans werkzaam bij ZuidOostZorg, centra voor ouderenzorg in Friesland, als geriatrie fysiotherapeut en coördinator wetenschap en onderzoek.

Hans is een fervent muziekliefhebber, is getrouwd met Sjoukje de Boer en heeft twee dochters, Ilse (20) en Ellen (15) en woont in Drachten, Friesland.

140 Dissertations of research institute SHARE

DISSERTATIONS OF RESEARCH INSTITUTE SHARE

This thesis is published within the Research Institute SHARE (Science in Healthy Ageing and healthcaRE) of the University Medical Center Groningen / University of Groningen. Further information regarding the institute and its research can be obtained from our internet site: http://www.share.umcg.nl/

More recent theses can be found in the list below. ((co-) supervisors are between brackets)

2018

Metting, EI Development of patient centered management of asthma and COPD in primary care (prof T van der Molen, prof R Sanderman, dr JWH Kocks)

Suhoyo Y Feedback during clerkships: the role of culture (prof JBM Kuks, prof J Cohen-Schotanus, dr J Schönrock-Adema)

Veen HC van der Articulation issues in total hip arthroplasty (prof SK Bulstra, dr JJAM van Raay, dr IHF Reininga, dr I van den Akker-Scheek)

Elsenburg LK Adverse life events and overweight in childhood, adolescence and young adulthood (prof AC Liefbroer, dr N Smidt)

‘t Hoen EFM Practical applications of the flexibilities of the agreement on trade-related aspects of intellectual property rights; lessons beyond HIV for access to new essential medicines (prof HV Hogerzeil, prof BCA Toebes)

Stojanovska V Fetal programming in pregnancy-associated disorders; studies in novel preclinical models (prof SA Scherjon, dr T Plösch)

141 Dissertations of research institute SHARE

Eersel MEA van The association of cognitive performance with vascular risk factors across adult life span (prof JPJ Slaets, dr GJ Izaks, dr JMH Joosten)

Rolfes L Patient participation in pharmacovigilance (prof EP van Puijenbroek, prof K Taxis, dr FPAM van Hunsel)

Brandenbarg D The role of the general practitioner in the care for patients with colorectal cancer (prof MY Berger, prof GH de Bock, dr AJ Berendsen)

Oldenkamp M Caregiving experiences of informal caregivers; the importance of characteristics of the informal caregiver, care recipient, and care situation (prof RP Stolk, prof M Hagedoorn, prof RPM Wittek, dr N Smidt)

Kammen K van Neuromuscular control of Lokomat guided gait; evaluation of training parameters (prof LHV van der Woude, dr A den Otter, dr AM Boonstra, dr HA Reinders-Messelink)

Hornman J Stability of development and behavior of preterm children (prof SA Reijneveld, prof AF Bos, dr A de Winter)

Vries, YA de Evidence-b(i)ased psychiatry (prof P de Jonge, dr AM Roest)

Smits KPJ Quality of prescribing in chronic kidney disease and type 2 diabetes (prof P denig, prof GJ Navis, prof HJG Bilo, dr GA Sidorenkov)

Zhan Z Evaluation and analysis of stepped wedge designs; application to colorectal cancer follow-up (prof GH de Bock, prof ER van den Heuvel)

142 Dissertations of research institute SHARE

Hoeve Y ten From student nurse to nurse professional (prof PF Roodbol, prof S Castelein, dr GJ Jansen, dr ES Kunnen)

Ciere Y Living with chronic headache (prof R Sanderman, dr A Visser, dr J Fleer)

Borkulo CD van Symptom network models in depression research; from methodological exploration to clinical application (prof R Schoevers, prof D Borsboom, dr L Boschloo, dr LJ Waldorp)

For more 2018 and earlier theses visit our website

143

Contact Drachten. Drachten. Paranimfen: Hans Drenth Fransisca Spa Fransisca Sjoukje Drenth in older people voor een borrel in een borrel voor voor 17 september. voor UITNODIGING Tevens bent u tussen bent Tevens [email protected] ZuidOostZorg/Sunenz, ZuidOostZorg/Sunenz, 16.00 en 18.00 welkom 16.00 en 18.00 welkom [email protected] [email protected] op de receptie ter plaatse op de receptie Broerstraat 5 te Groningen 5 te Broerstraat cross-linked and glycation Motor function, paratonia function, paratonia Motor Burgemeester Wuiteweg 140, Wuiteweg Burgemeester [email protected] [email protected] [email protected] Maandag 24 september om 12.45 om 12.45 Maandag 24 september van de Rijksuniversiteit Groningen, Groningen, de Rijksuniversiteit van verdediging van mijn proefschrift van verdediging Graag aanmelden voor de borrel via de borrel aanmelden voor Graag in de aula van het Academiegebouw Academiegebouw het in de aula van Aansluitend bent u van harte welkom welkom harte u van bent Aansluitend voor het bijwonen van de openbare de openbare van bijwonen het voor Hans Drenth in older peoplein older Motor function paratonia decline and and their relation with Advanced Glycation End-Products Glycation with Advanced and their relation Motor function, paratonia function,Motor paratonia and glycation cross-linked cross-linked glycation and

Hans Dretnh Motor Function, Paratonia and Glycation Cross-linked in Older People