Knee Joint Distraction Compared with Total Knee Arthroplasty a RANDOMISED CONTROLLED TRIAL

 KNEE Knee joint distraction compared with total knee arthroplasty A RANDOMISED CONTROLLED TRIAL

J. A. D. van der Woude, Aims K. Wiegant, Knee joint distraction (KJD) is a relatively new, knee-joint preserving procedure with the R. J. van Heerwaarden, goal of delaying total knee arthroplasty (TKA) in young and middle-aged patients. We S. Spruijt, present a randomised controlled trial comparing the two. P. J. E m ans , Patients and Methods S. C. Mastbergen, The 60 patients ≤ 65 years with end-stage knee osteoarthritis were randomised to either F. P. J. G. Lafeber KJD (n = 20) or TKA (n = 40). Outcomes were assessed at baseline, three, six, nine, and 12 months. In the KJD group, the joint space width (JSW) was radiologically assessed, From UMC Utrecht, representing a surrogate marker of cartilage thickness. Utrecht, The Netherlands Results In total 56 patients completed their allocated treatment (TKA = 36, KJD = 20). All patient reported outcome measures improved significantly over one year (p < 0.02) in both groups.  J. A. D. van der Woude, MD, At one year, the TKA group showed a greater improvement in only one of the 16 patient- PhD, Resident in Orthopaedic Surgery, Limb and Knee related outcome measures assessed (p = 0.034). Outcome Measures in Rheumatology- Reconstruction Unit, Department of Orthopaedic Osteoarthritis Research Society International clinical response was 83% after TKA and 80% Surgery after KJD. A total of 12 patients (60%) in the KJD group sustained pin track infections. In the  R. J. van Heerwaarden, MD, PhD, Orthopaedic Surgeon, KJD group both mean minimum (0.9 mm, standard deviation (SD) 1.1) and mean JSW (1.2 mm, Limb and Knee Reconstruction Unit, Department of SD 1.1) increased significantly (p = 0.004 and p = 0.0003). Orthopaedic Surgery  S. Spruijt, MD, Orthopaedic Conclusion Surgeon, Limb and Knee Reconstruction Unit, In relatively young patients with end-stage knee osteoarthritis, KJD did not demonstrate Department of Orthopaedic inferiority of outcomes at one year when compared with TKA. However, there is a high Surgery Maartenskliniek Woerden, incidence of pin track infection associated with KJD. Woerden, The Netherlands.

 K. Wiegant, MD, PhD, Cite this article: Bone Joint J 2017;99-B:51–8. Resident in Orthopaedic Surgery, Rheumatology and In end-stage osteoarthritis (OA) of the knee a TKAs performed in patients aged under 65 Clinical Immunology total knee arthroplasty (TKA) is generally indi- years, a burden of revision arthroplasty may  S. C. Mastbergen, PhD, 1 9-10 Associate Professor, cated. This intervention for OA was first per- well be generated. Rheumatology and Clinical Immunology formed in the 1970s and is now generally On this basis, joint-preserving surgery is an  F. P. J. G. Lafeber, PhD, regarded as standard treatment.1,2 However, attractive option, but arthroscopic debride- Professor of Experimental Rheumatology, Rheumatology joint arthroplasty is not without drawbacks, ment, one of the most popular forms, is and Clinical Immunology, UMC Utrecht, Utrecht, The with up to 20% of patients dissatisfied with becoming less supported by modern evi- Netherlands. their TKA and up to 44% reporting continuing dence.11,12 Osteotomies have a place in uni-  P. J. Emans, MD, PhD, pain, a third of whom reported extreme persis- compartmental disease and the KineSpring Orthopaedic Surgeon, 3-6 Department of Orthopaedics tent pain in one study. Patients younger than (Moximed Inc., Hayward, California) has MUMC Maastricht, Maastricht, The Netherlands. 55 years have an almost fivefold increase in the recently been introduced for medial compart- lifetime risk of revision surgery compared with ment disease, although long-term evidence of Correspondence should be sent 7 13,14 to F. P. J. G. Lafeber; email: those aged 75 years or older. One of the main its efficacy is yet to become available. [email protected] reasons for this is aseptic loosening, a charac- Knee joint distraction (KJD) has been pro- ©2017 The British Editorial teristic of implant wear, which is responsible posed as a temporising technique for sympto- Society of Bone & Joint 7,8 Surgery for more than 30% of revisions. Since wear matic, arthritic knees in younger patients and doi:10.1302/0301-620X.99B1. increases with both time and activity, a young comprises a six- to eight-week period of dis- BJJ-2016-0099.R3 $2.00 patient with relatively higher functional traction of the knee by means of an external Bone Joint J demands of the implant can experience signifi- fixator. A prospective, uncontrolled study of 2017;99-B:51–8. Received 17 March 2016; cantly more wear than their older, more seden- 20 patients aged below 60 years at five-year Accepted after revision 13 September 2016 tary counterparts. In addition, with 40% of follow-up reported clinical improvement and

VOL. 99-B, No. 1, JANUARY 2017 51 52 J. A. D. VAN DER WOUDE, K. WIEGANT, R. J. VAN HEERWAARDEN, S. SPRUIJT, P. J. EMANS, S. C. MASTBERGEN, F. P. J. G. LAFEBER

All surgery was performed by one of three surgeons (RJvH, SS or PJE). Knee arthroplasty was performed using the Genesis II posterior stabilised system (Smith & Nephew, Warsaw, Indiana) with fixation using GentaPala- cos cement (Heraeus, Hanau, Germany). Routine post- operative rehabilitation and venous thrombo-embolism prophylaxis was provided. Distraction surgery was performed using the Triax proof-of-concept external distraction device (Stryker, Kalamazoo, Michigan), based around two tubes with internal 45 kg springs (Fig. 1). Eight fixation pins were used to provide two points of fixation for each tube, each side of the knee. The knee was distracted by 2 mm intra-operatively and distracted a further 1 mm during the following three days, achieving total distraction of 5 mm. This was verified radiologically on post-operative

Fig. 1 day four and patients were then discharged on crutches. Ini- tially patients were permitted to transmit up to 20% body Example of the bilateral external fixation frame used for knee joint distraction treatment (left) and an example of a radiograph (right) during weight through the operated limb, which increased to full distraction treatment with 5 mm of distraction. weight bearing within six weeks. Radiological evaluation of the distraction and clinical assessment of the pin sites evidence of cartilage repair. If these results were to be gen- was made at three weeks post-operatively in the outpatient erally applicable, the possibility arises of delaying the clinic. The fixator was removed at six weeks. Patients were requirement of TKA in young patients.15-18 prescribed low molecular weight heparin for the duration This study is a randomised controlled trial comparing of fixation and for three weeks thereafter. outcomes of KJD with TKA at one year. A total of 60 patients were recruited to the study, with four withdrawing consent once randomised to TKA, leav- Patients and Methods ing a population of 56 patients (Fig. 2). The demographic This study was conducted in two centres between 2011 and characteristics of the participants are given in Table I.22 2014. Randomisation was conducted in a 2:1 ratio in favour Initial baseline scores were taken at randomisation, and of TKA in blocks of six by a sealed envelope method. The again the day before surgery. Questionnaires were further ratio was a condition of the institutional review board’s completed at three, six, nine and 12 months post- favourable opinion. Group size calculation was based on a operatively. non-inferiority hypothesis. The sample size was based on The WOMAC and knee injury and osteoarthritis out- non-inferiority in the primary outcome measure, the Western come score (KOOS)23 were used to score clinical improve- Ontario and McMaster Universities Osteoarthritis Index ment, normalised to a 100-point scale with 100 being the (WOMAC),19 with a 5% type I error, and with a power of best condition possible. The intermittent and constant 80%.20 A 15% margin was allowed for loss to follow-up. A osteoarthritis pain score (ICOAP)24 for the knee was the difference in WOMAC score of more than 15 points (stand- secondary clinical outcome parameter (0 to 100, 0 repre- ard deviation (SD) 16.7) compared with the TKA group was senting no pain). A visual analogue scale for pain (VAS- considered clinically relevant.21 This resulted in group sizes pain; 0 mm to 100 mm, 0 representing no pain and 100 of 40 patients for TKA and 20 for KJD. Neither patient nor severe pain) was the tertiary clinical outcome parameter. surgeon was blinded to treatment. The EuroQol (EQ)-5D-3L25 was used to assess improve- Inclusion criteria were OA of the knee and eligibility for ment of quality of life. The obtained questionnaire was TKA according to routine standard of care, age ≤ 65 years, transformed to an EQ-5D index score (0 to 1, 1 being best). functionally intact knee ligaments, ≥ 120° flexion at the The Short Form (SF)-36 health survey26 was used to meas- knee and body mass index (BMI) < 35 kg/m2. Patients with ure the health status of the patients. The SF-36 items were primary patellofemoral OA were excluded, as were those converted to physical (PCS) and mental (MCS) component with a history of inflammatory or septic arthritis, inability summary scores. At baseline, three, six, nine and 12 months to tolerate external fixation, an already ankylosed knee, the KOOS/WOMAC, ICOAP, EQ-5D-3L and VAS-pain post-traumatic fibrosis due to a fracture of the tibial pla- were recorded. At baseline, six, and 12 months the SF-36 teau, previous surgery on the same knee within the past six was recorded. months, or previous joint arthroplasties. The study was Radiographs of the KJD group were obtained at baseline granted ethical approval (No 10/359/E) and was registered and 12 months post-operatively. The knee images were on the Netherlands National Trial Register (NTR2809). standardised weight-bearing, semi-flexed (7º to 10º flexion) All patients provided written informed consent before posterior-anterior views with a reference aluminum step enrolment. wedge close to the knee within the field of exposure. Images

THE BONE & JOINT JOURNAL KNEE JOINT DISTRACTION COMPARED WITH TOTAL KNEE ARTHROPLASTY 53

Enrolment

- Randomisation rate Randomisation 2:1 (TKA:KJD) - Stratified in blocks of 6

Allocation

- Allocated to intervention TKA (n = 40) - Allocated to intervention KJD (n = 20) - Received allocated intervention (n = 36) - Received allocated intervention (n = 20) - At MKW (n = 30 - At MKW (n = 17) - At MUMC (n = 6) - At MUMC (n = 3) - Not received allocated intervention (n = 4) - Not received allocated intervention (n = 0) Reason: withdrawal after inclusion

Follow-up

- Loss to follow-up at 1 yr (n = 0) - Loss to follow-up at 1 yr (n = 0)

Analysis

- Analysed clinical outcome (n = 36) - Analysed clinical outcome (n = 20) - Analysed structural outcome (n = 17) Reason: refused radiograph (n = 1) underwent TKA (n = 1) poor image quality (n = 1)

Fig. 2

Flowchart showing the numbers of excluded patients, as well as allocation of the randomised treatment and the analysed patients per treatment arm (TKA, total knee arthroplasty; KJD, knee joint distraction; MKW, Maarten- skliniek Woerden; MUMC, Maastricht University Medical Center). were evaluated using knee images digital analyses software surgery, with their scores at the last KJD follow-up used for (Image Sciences Institute, Utrecht, The Netherlands) to last observation carried forward completion of the dataset. derive mean and minimum joint space width (JSW) of the Clinical outcome. An improvement in WOMAC score was knee, by a single individual blinded to the identity of the seen in both groups, with a mean increase of 30 points participant.27 (SD 17) in the KJD group (p < 0.001) and 36 (SD 19) in the Statistical analysis. Two-sided paired t-tests were used to TKA group (p < 0.001) (Fig. 3). The WOMAC subscales evaluate within-group change between follow-up and base- for pain, stiffness, and function showed a similar trend line scores. Between-group change was evaluated using (Table II). At an individual level, 16 patients (80%) in the two-sided independent-samples t-tests. A p-value < 0.05 KJD group and 30 (83%) in the TKA group could be des- was considered statistically significant. SPSS v.22 software ignated as clinical responders by the Outcome Measures in (IBM, Armonk, New York) was used to perform statistical Rheumatology-Osteoarthritis Research Society Interna- analyses. tional responder criteria,28 defined as an increase of > 50% in WOMAC pain or function scales with > 20 points of Results improvement in either category, or an increase of > 20% of The mean follow-up of the TKA group was 12 months, and WOMAC pain and function scales with a ten-point was also 12 months for the KJD group. One patient in the improvement in each.28 The change in total WOMAC score KJD group continued to have disabling pain and functional between baseline and one year follow-up was not signifi- impairment and elected to undergo TKA within one year of cantly different between groups (p = 0.273) (Fig. 3).

VOL. 99-B, No. 1, JANUARY 2017 54 J. A. D. VAN DER WOUDE, K. WIEGANT, R. J. VAN HEERWAARDEN, S. SPRUIJT, P. J. EMANS, S. C. MASTBERGEN, F. P. J. G. LAFEBER

Table I. Baseline characteristics22

Mean (SEM) TKA KJD (n = 36) (n = 20) Male gender (n, %) 13/36 (36) 9/20 (45) Height (cm) 173 (1.7) 175 (2.2) Weight (kg) 88.3 (2.3) 83.9 (3.6) Body mass index (kg/m2) 29.4 (0.6) 27.4 (0.9) Affected knee (left) (%) 17/36 (47) 6/20 (30) Age at surgery (yrs) 55.2 (1.0) 54.9 (1.8) Kellgren-Lawrence grade22 (median) 3 4 Grade 0 (n, %) 0 (0) 0 (0) Grade 1 (n, %) 0 (0) 0 (0) Grade 2 (n, %) 9 (25) 1 (5) Grade 3 (n, %) 21 (58) 8 (40) Grade 4 (n, %) 6 (17) 11 ( 55) Previous surgery Operation (n) Anterior cruciate ligament reconstruction (n) 5 3 High tibial osteotomy (n) 4 4 Arthroscopy: 17 13 Partial meniscectomy (n) 4 7 Arthroscopic joint lavage (n) 13 6 Open meniscectomy (n) 7 6 Perichondriumplasty (n) 1 0 SEM, standard error of the mean; TKA, total knee arthroplasty; KJD, knee joint distraction

80 100 70 p = 0.272 TKA p < 0.001 60 80 50 KJD p < 0.001 40 60 30

WOMAC total WOMAC 20 Δ 40 10 WOMAC (total score) WOMAC 0 20 -10 036912 TKA KJD Mths Fig. 3

Western Ontario and McMaster Universities Osteoarthritis index (WOMAC) total score. Dotted line (left) represents the knee joint distraction (KJD) group (n = 20), solid line represents the total knee arthroplasty (TKA) group (n = 36). Mean values and standard error of the mean are shown. The p-values show statistical difference byt-testing of values at one-year follow-up compared with pre-treatment values. Mean change of WOMAC total score (right). For both groups (mean: dash) and for every individual patient (squares), no statistical significant difference in clinical efficacy was observed

The total KOOS improved significantly from baseline at Knee flexion in the KJD group initially fell before return- one year in both treatment groups (p < 0.001) but was not ing to baseline (125°) by six months. In the TKA group, significantly different between groups (p = 0.065) (Fig. 4). there was a decline of 4° from 124° at baseline to 120° at The five subscales of the KOOS all improved signifi- one year (p = 0.013) (Fig. 5). cantly over baseline at one year (p < 0.001) (Table II). Sim- A total of 18 patients in the KJD group had radiographs ilar results were seen for the ICOAP, the SF-36 PCS, the available at baseline and one year, as one went on to undergo VAS-pain score and the EQ-5D (Table II). The TKA group TKA and one refused imaging. One patient’s radiographs alone showed significant mean change in the KOOS quality were of insufficient quality for analysis. In the remaining 17 of life subscale (p = 0.021) (Table II). patients, the mean JSW of the most affected compartment

THE BONE & JOINT JOURNAL KNEE JOINT DISTRACTION COMPARED WITH TOTAL KNEE ARTHROPLASTY 55

Table II. Knee injury and osteoarthritis outcome score (KOOS), Western Ontario and McMaster Universities Osteoarthritis index (WOMAC), intermittent and constant osteoarthritis pain score (ICOAP), visual analogue score (VAS), EuroQol (EQ)-5D and Short Form (SF)-36 scores pre- and post- operatively

n Total knee arthroplasty Knee joint distraction

Score (SD) Baseline 12 mths Change Baseline 12 mths Change KOOS (0 to 100) Pain 42 (14) 82 (17)* 40 (23) 43 (11) 74 (21)* 31 (19) Symptoms 45 (16) 75 (15)* 30 (19) 45 (15) 68 (20)* 23 (17) ADL 47 (12) 84 (15)* 36 (18) 51 (11) 81 (16)* 30 (16) Sport/Rec 20 (13) 61 (27)* 42 (28) 22 (16) 52 (31)* 30 (27) QOL 27 (13) 67 (23)*† 39 (25) 28 (11) 51 (24)* 23 (24) Total 36 (12) 74 (17)* 37 (20) 38 (9) 65 (20)* 27 (18) WOMAC (0 to 100) Pain 46 (16) 85 (15)* 39 (22) 48 (13) 80 (20)* 32 (20) Stiffness 45 (20) 71 (23)* 26 (28) 38 (16) 67 (29)* 29 (25) Function 47 (12) 84 (15)* 36 (18) 51 (11) 81 (16)* 30 (16) Total 47 (13) 83 (16)* 36 (19) 49 (11) 80 (18)* 30 (17) ICOAP (100 to 0) Constant 55 (21) 12 (16)* -43 (27) 46 (18) 12 (14)* -34 (23) Intermittentnt 55 (17) 16 (16)* -40 (22) 47 (15) 17 (14)* -30 (21) Combined 55 (18) 14 (16)* -41 (24) 46 (16) 15 (13)* -32 (21) VAS (0 to 100) Pain 71 (17) 18 (19)* -53 (28) 65 (19) 28 (26)* -36 (2) EQ-5D (0 to 1) Index score 0.52 (0.3) 0.87 (0.1)* 0.35 (0.3) 0.58 (0.3) 0.76 (0.3)* 0.18 (0.3) SF-36 PCS 32 (7) 47 (9)* 15 (10) 33 (9) 44 (10)* 11 ( 8 ) MCS 54 (10) 53 (10) -1 (13) 54 (8) 53 (8) -1 (11) Mean flexion (°) Knee 124 (7) 120 (11)* -5 (11) 122 (9) 125 (9) 2 (11) † * change within group significant at the p < 0.05 level † change between groups significant at the p < 0.05 level SD, standard deviation; ADL, activities of daily life; QOL, quality of life; PCS, physical component score; MCS, mental component score

100 70 p = 0.065 60

80 TKA p < 0.001 50 40 # 60 30 # KJD p < 0.001 total KOOS

Δ 20

KOOS (total score) KOOS 40 10 0

20 -10 036912 TKA KJD Mths

Fig. 4

Knee injury and osteoarthritis outcome score (KOOS) total. Dotted line (left) represents the knee joint distraction (KJD) group (n = 20), solid line represents the total knee arthroplasty (TKA) group (n = 36). Mean values and standard error of the mean are shown. The p-values show statistical difference byt-testing of values at one-year follow-up compared with pre-treatment values. # indicates a significant difference between TKA and KJD. Mean change of KOOS total score (right). For both groups (mean: dash) and for every individual patient (squares); no statistical significant difference in clinical efficacy was observed.

increased significantly from 1.9 mm (SD 2.1) towards 3.2 antibiotics, due to a Staphylococcus aureus sepsis diag- mm (SD 2.1) at one year (p < 0.001). In line with this, the nosed from blood culture. One other patient demonstrated minimum JSW increased 0.9 mm (SD 1.1), from 0.6 mm a pyrexia of unknown origin after removal of the frame, (SD 1.2) at baseline to 1.5 mm (SD 1.1) at one year (p = 0.004) and was also prescribed intravenous antibiotics. One (Fig. 6). The mean JSW of the least affected compartment patient in the KJD group exhibited a post-operative foot increased from 7.1 mm (SD 2.0) towards 7.4 mm (SD 2.1) at drop, successfully treated with an ankle-foot orthosis. one year (p = 0.422). Five patients in the TKA group experienced post- In the KJD group 12 patients (60%) had single or operative stiffness and required manipulation under anaes- multiple pin site infections, ten of which were treated with thesia. One patient had a myocardial infarction six days oral antibiotics. One patient was admitted to hospital post-operatively and was treated by percutaneous coronary shortly after removal of the frame for intravenous intervention and subsequent pacemaker implantation.

VOL. 99-B, No. 1, JANUARY 2017 56 J. A. D. VAN DER WOUDE, K. WIEGANT, R. J. VAN HEERWAARDEN, S. SPRUIJT, P. J. EMANS, S. C. MASTBERGEN, F. P. J. G. LAFEBER

130 ns This is the first randomised study to compare KJD with * TKA. In one prospective uncontrolled study, 20 patients eligible for TKA underwent KJD for eight weeks.15-17 The 120 results at one year were comparable with this study; the p = 0.013 marginally increased outcome scores in this study may 29 110 relate to a lower mean BMI in our cohort. A number of studies of the outcome of TKA have

Knee flexion (˚) employed the same outcome measures as this study, with 100 comparable outcomes at one year.4,30,31 Though we saw fewer pin site infections in this study 0 than in the previous one, not least due to diligent patient 036912 Mths and staff education, the rate remained high and two septi- 33,33 Fig. 5 caemic events occurred. Pin site infection rates of 27% and deep infection rates of 16% have been reported in lit- Overview of knee flexion. Dotted line represents the knee joint distrac- 34,35 tion (KJD) group (n = 20), solid line represents the total knee arthroplasty erature, with 4% developing chronic osteomyelitis. (TKA) group (n = 36). Mean values and standard error of the mean are Our deep infection rate was, therefore, comparable. Possi- given. For the TKA group flexion was still significantly reduced compared with pre-treatment values, which was not observed for KJD. ble reasons for this include weight-bearing, high patient * indicates a significant difference between KJD and TKA. mobility and peri-articular placement, all suggested in the literature to play a part in pin site infection.36 This under- lines the paramount importance of meticulous pin insertion and site care.37 Coated pins may inhibit biofilm formation, thus reducing these infections, and newer devices and tech- 3 niques are under development.38,39 p < 0.001 Consideration must also be given to the risk of deep infec- Mean JSW tion in the context of a joint which will likely go on to require arthroplasty. For this reason, we sited all pins well away 2 from the synovial cavity of the joint.18 Registry follow-up of participants in a prospective study showed that three patients underwent TKA within five years of KJD, experienc-

JSW MAC (mm) JSW MAC ing no prosthetic joint infection and gaining equivalent out- Min JSW p = 0.004 1 comes to their non-KJD counterparts. The patient from the current study who received a TKA six months after KJD did not demonstrate any problems. As reported elsewhere, this patient had an abnormal posterior slope and should not have 0 been considered eligible for inclusion on this basis.40 01Our study has its limitations. We did not include sham surgery or placebo control group. As well as ethical and

Fig. 6 logistic implications, they have themselves been shown to exhibit a placebo effect.41 Findings in this study may thus The left panel shows representative standardised weight-bearing radiographs of a representative patient at baseline (upper radiograph) and overestimate effects of both surgical treatments. We also one year after treatment (lower radiograph). The right panel shows the included some patients with isolated, primarily medial mean and minimum joint space width (JSW) on weight-bearing radiographs of the most affected compartment (MAC) of the knee joint dis- compartment disease; a more valid question in these traction (KJD) group (n = 17). Mean values and standard error of the patients may have been how KJD compared with unicom- mean are given. The p-values show statistical difference by t-testing of values compared with pre-treatment values. partmental knee arthroplasty. Another limitation is the small follow-up. Previous studies describe 80% five-year and 65% ten-year survival and so our results should be taken in this context.17,18 One might speculate whether patients for whom KJD was considered initially successful, but experienced a gradual recur- Discussion rence of symptoms, would benefit from repetition of the This study demonstrates equivalence of clinical efficacy treatment. between KJD and TKA in the short-term. In both groups, Our sample size assumed a relatively large effect size and all patient reported outcome measures improved signifi- hence, should it have been smaller, the study may have been cantly. Only one, KOOS subscale quality of life, showed underpowered to detect it. In addition, despite the recruit- significantly greater improvements in the TKA group than ment of patients deemed to have end-stage arthritis, our in KJD. Recovery of flexion was better in KJD than TKA. randomisation process was compromised by four patients

THE BONE & JOINT JOURNAL KNEE JOINT DISTRACTION COMPARED WITH TOTAL KNEE ARTHROPLASTY 57 who refused to accept TKA, after they had been assigned to 8. Swedish Knee Arthroplasty Register. Annual report 2010. http://www.myk- that half of the trial. nee.se/pdf/114_SKAR2010_Eng1.0.pdf (date last accessed 21 September 2016). We found that in relatively young patients with knee 9. Naudie DD, Ammeen DJ, Engh GA, Rorabeck CH. Wear and osteolysis around total knee arthroplasty. J Am Acad Orthop Surg 2007;15:53–64. osteoarthritis who may otherwise undergo TKA, KJD 10. Kurtz SM, Lau E, Ong K, et al. Future young patient demand for primary and revision offers comparable clinical benefit at one year but comes at joint replacement: national projections from 2010 to 2030. Clin Orthop Relat Res the expense of superficial and deep infective complications. 2009;467:2606–2612. The long-term benefit of KJD is still to be determined fur- 11. Katz JN, Brownlee SA, Jones MH. The role of arthroscopy in the management of knee osteoarthritis. Best Pract Res Clin Rheumatol 2014;28:143–156. ther. As KJD preserves the knee joint, it represents a prom- 12. Finnikin SJ. Arthroscopic surgery for knee pain: where is the shared decision mak- ising therapeutic option in postponing a first TKA ing? BMJ 2016;354:4388. potentially preventing revision surgery later in life. 13. Brinkman JM, Lobenhoffer P, Agneskirchner JD, et al. Osteotomies around the knee: patient selection, stability of fixation and bone healing in high tibial osteotomies. J Bone Joint Surg [Br] 2008;90-B:1548–1557. Take home message: 14. London NJ, Smith J, Miller LE, Block JE. Midterm outcomes and predictors of - KJD offers comparable clinical benefit at one year, in rela- clinical succes with KineSpring knee implant system. Clin Med Insights Arthritis Mus- tively young patients with knee osteoarthritis who may other- culoskelet Disord 2013;6:19–28. wise undergo TKA 15. Intema F, Van Roermund PM, Marijnissen AC, et al. Tissue structure modifica- - As KJD preserves the knee joint, it represents a promising therapeutic tion in knee osteoarthritis by use of joint distraction: an open 1-year pilot study. Ann Rheum Dis 2011;70:1441–1446. option in postponing a first TKA, potentially preventing revision surgery 16. Wiegant K, van Roermund PM, Intema F, et al. Sustained clinical and structural later in life. benefit after joint distraction in the treatment of severe knee osteoarthritis. Osteoar- thritis Cartilage 2013;21:1660–1667. Supplementary material 17. van der Woude JAD, Wiegant K, van Roermund PM, et al. Five-year follow-up of A table showing the patient reported outcome meas- knee joint distraction; clinical benefit and cartilaginous tissue repair in an open ures scores for three, six and nine months after treat- uncontrolled prospective study.Cartilage 2016 August 26. (Epub ahead of print) ment can be found alongside the online version of this 18. Wiegant K, van Roermund PM, van Heerwaarden R, et al. Total knee prosthesis after joint distraction treatment. J Surg Surgical Res 2015;1:66–71. paper at http://www.bjj.boneandjoint.org.uk 19. Bellamy N. Outcome measurement in osteoarthritis clinical trials. J Rhematol Suppl Author contributions: 1995;43:49–51. J. A. D. van der Woude: Gathered and analysed all the data, Wrote the initial 20. Wiegant K, van Heerwaarden RJ, van der Woude JAD, et al. Knee joint distrac- draft of the manuscript and managed the study. tion as an alternative surgical procedure for patients with osteoarthritis considered K. Wiegant: Designed the study and revised the manuscript. for high tibial osteotomy or for a total knee prosthesis: rationale and design of two R. J. van Heerwaarden: Designed the study and revised the manuscript. randomized controlled trials. Int J Orthod 2015;2:155–159. S. Spruijt: Designed the study and revised the manuscript. P. J. Emans: Revised the manuscript. 21. Escobar A, Quintana JM, Bilbao A, et al. Responsiveness and clinically important S. C. Mastbergen: Designed the study and revised the manuscript. differences for the WOMAC and SF-36 after total knee replacement. Osteoarthritis F. P. J. G. Lafeber: Conceived the study, Ensured the accuracy of the data and the Cartilage 2007;15:273–280. analysis, Wrote and revised the manuscript. 22. Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957;16:494–502. This study was funded by ZonMw (The Netherlands Organisation for Health Research and Development). The study sponsor had no involvement in the 23. De Groot IB, Favejee MM, Reijman M, et al. The Dutch version of the Knee Injury study design, collection, analysis and interpretation of data, in writing of the and Osteoarthritis Outcome Score: a validation study. Health Qual Life Outcomes manuscript, and in the decision to submit the manuscript for publication. 2008;16:1–11. Although none of the authors has received or will receive benefits for personal 24. Hawker GA, Stewart L, French MR, et al. Understanding the pain experience in or professional use from a commercial party related directly or indirectly to the hip and knee osteoarthritisean OARSI/OMERACT initiative. Osteoarthritis Cartilage subject of this article, benefits have been or will be received but will be directed 2008;16:415–422. solely to a research fund, foundation, educational institution, or other non- 25. The EuroQol Group. EuroQol-a new facility for the measurement of health-related profit organization with which one or more of the authors are associated. quality of life. Health Policy 1990;16:199–208. This article was primary edited by P. Page and first proof edited by G. Scott. 26. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30:473–483. 27. Marijnissen AC, Vincken KL, Vos PA, et al. Knee Images Digital Analysis (KIDA): References a novel method to quantify individual radiographic features of knee osteoarthritis in detail. Osteoarthritis Cartilage 2008;16:234–243. 1. No authors listed. British Orthopaedic Association, Knee replacement: a guide to good practice. http://almacen-gpc.dynalias.org/publico/KNEE%20REPLACEMENT.pdf (date 28. Escobar A, Gonzalez M, Quintana JM, et al. Patient acceptable symptom state last accessed 16 November 2016). and OMERACT-OARSI set of responder criteria in joint replacement. Identification of cut-off values. Osteoarthritis Cartilage 2012;20:87–92. 2. Carr AJ, Robertsson O, Graves S, et al. Knee replacement. Lancet 2012;379:1331– 1340. 29. Liljensøe A, Lauersen JO, Søballe K, Mechlenburg I. Overweight preoperatively impairs clinical outcome after knee arthroplasty: a cohort study of 197 patients 3–5 3. Robertsson O, Dunbar M, Pehrsson T, Knutson K, Lidgren L. Patient satisfaction after knee arthroplasty: a report on 27,372 knees operated on between 1981 and 1995 years after surgery. Acta Orthop 2013;84:392–397. in Sweden. Acta Orthop Scand 2000;71:262–267. 30. Winther SB, Foss OA, Wik TS, et al. 1-year follow-up of 920 hip and knee arthro- 4. Bourne RB, Chesworth BM, Davis AM, Mahomed NN, Charron KD. Patient sat- plasty patients after implementing fast-track. Acta Orthop 2015;86:78–85. isfaction after total knee arthroplasty: who is satisfied and who is not? Clin Orthop 31. Skou ST, Roos EM, Laursen MB, et al. A randomized, controlled trial of total knee Relat Res 2010;468:57–63. replacement. N Engl J Med 2015;373:1597–1606. 5. Dunbar MJ, Richardson G, Robertsson O. I can’t get no satisfaction after my total 32. W-Dahl A, Toksvig-Larsen S, Lindstrand A. No difference between daily and knee replacement: rhymes and reasons. Bone Joint J 2013;95-B:148–152. weekly pin site care: a randomized study of 50 patients with external fixation. Acta 6. Wylde V, Hewlett S, Learmonth ID, Dieppe P. Persistent pain after joint replace- Orthop Scand 2003;74:704–708. ment: prevalence, sensory qualities, and postoperative determinants. Pain 33. Egol KA, Paksima N, Puopolo S, et al. Treatment of external fixation pins about 2011;152:566–572. the wrist: a prospective, randomized trial. J Bone Joint Surg [Am] 2006;88-A:349– 7. Australian Orthopaedic Association. National Joint Replacement Registry, Hip 354. and knee arthroplasty: annual report 2010. https://aoanjrr.sahmri.com/annual- 34. Iobst C, Liu R. A systematic review of incidence of pin track infections associated reports-2010 (date last accessed 21 September 2016). with external fixation. J Limb Lengthening Reconstr 2016;2:6–16.

VOL. 99-B, No. 1, JANUARY 2017 58 J. A. D. VAN DER WOUDE, K. WIEGANT, R. J. VAN HEERWAARDEN, S. SPRUIJT, P. J. EMANS, S. C. MASTBERGEN, F. P. J. G. LAFEBER

35. Giannoudis PV, Papakostidis C, Roberts C. A review of the management of open 39. Qu H, Knabe C, Radin S, Garino J, Ducheyne P. Percutaneous external fixator fractures of the tibia and femur. J Bone Joint Surg [Br] 2006;88-B:281–289. pins with bactericidal micron-thin sol-gel films for the prevention of pin tract infec- 36. Kazmers NH, Fragomen AT, Rozbruch SR. Prevention of pin site infection in exter- tion. Biomaterials 2015;62:95–105. nal fixation: a review of the literature. Strategies Trauma Limb Reconstr 2016;11:75– 40. van der Woude JA, van Heerwaarden RJ, Spruijt S, et al. Six weeks of contin- 85. uous joint distraction appears sufficient for clinical benefit and cartilaginous tissue 37. Bibbo C, Brueggeman J. Prevention and management of complications arising repair in the treatment of knee osteoarthritis. Knee 2016;23:785–791. from external fixation pin sites. J Foot Ankle Surg 2010;49:87–92. 41. Wartolowska K, Judge A, Hopewell S, et al. Use of placebo controls in the eval- 38. Shirai T, Watanabe K, Matsubara H, et al. Prevention of pin tract infection with iodine-supported titanium pins. J Orthop Sci 2014;19:598–602. uation of surgery: systematic review. BMJ 2014;348:3253.

THE BONE & JOINT JOURNAL