Findings

A meta‐analysis by Brosseau et al. (level I) included 14 studies (952 patients), after excluding many other studies on the topic which lacked sufficient quality of evidence.² The included studies used CPM from 5 to 20 hours per day, for a duration of 18 hours to 2 weeks postoperatively after primary TKA. They performed a pooled analysis comparing patients receiving postoperative physiotherapy combined with CPM, against those receiving physiotherapy only (nine studies included). Eight of the studies initiated CPM on the first postoperative day, and the other study started it on the second day.

Active knee flexion at two weeks following surgery (pooled results from four studies, 286 patients) was compared, and showed that CPM was associated with a weighted mean difference (increase) of 4.30° (95% confidence interval [CI]: 1.96–6.63). They also reported a clinically important benefit in association with CPM, with a relative difference in active knee flexion of 22–25% over the period from 3 to 14 days postoperatively. One of the included studies reported postoperative time to achieve 90° of knee flexion, and found that the CPM group achieved this an average of 4.7 days earlier than the control group (9.1 days vs 13.8 days).³ Active knee extension was also compared, but no statistically significant difference was found. Similarly, no significant differences were found between the groups for passive flexion or extension at follow‐up to six months. The requirement for MUA was used as a marker for knee stiffness, and this was also compared in this study (pooled results from three studies, all began CPM within 24 hours). CPM was associated with a significant reduction in MUA (risk ratio [RR] = 0.12; 95% CI: 0.03–0.53); however, they questioned the clinical importance (5–18% relative difference).²

Another case‐control study by Trzeciak et al. (level III) evaluated 101 TKAs in 93 patients, and assigned them into two groups.⁴ The study group received CPM and active exercises, whereas the control group had only conventional physiotherapy. CPM was started on day one postoperatively, for two hours per day, with an initial ROM of 0–40°. ROM was increased as tolerated, by a mean of 10° per day, until discharge. On day 10 postoperatively, the patients were assessed, demonstrating no significant difference in mean ROM between the two groups: CPM group: 83° ±14°; control group: 77° ±21°. They reported that there may be some subjective improvement in pain level, joint stiffness, and function associated with CPM, but were not able to demonstrate any statistically significant differences. This study focused on the early postoperative outcomes, as it only showed follow‐up at 10 days postoperatively, which may be too early to make a diagnosis of arthrofibrosis. Nonetheless, early progression in ROM and function is important in its prevention, and therefore these early outcomes may be related to arthrofibrosis risk.

There is debate about when CPM should be initiated to provide most benefit. A level III study by Daluga et al. evaluated the outcome of requirement for MUA after TKA over a three‐year period, with a change in the CPM protocol halfway through.⁵ The requirement for MUA was based on failure to meet flexion goals (70° before discharge – early MUA, failure to surpass 65–75° at early follow‐up – intermediate MUA, failure to achieve 80–85° at three‐month follow‐up – late MUA). During the first half, CPM was commenced on day 3 postoperatively, and during the second half of the study, CPM was commenced on day one, and continued until discharge (length of stay not reported). The MUA rate remained constant at 12% throughout the period despite the change in CPM protocol, suggesting no difference between starting CPM at day one or day three.

Additionally, there is no consensus about duration and method of CPM, and most studies do not evaluate this. The aforementioned meta‐analysis stated that a comparison was made between short and long duration of CPM, and also between small and large range CPM, but that no statistically significant differences were found in relation to any outcome measures and the results were not shown.²

Many other studies included CPM as part of their postoperative rehabilitation protocol for all patients. At the time of writing, there are no other available studies that compare CPM against a control group, reporting arthrofibrosis as an outcome. Further work is needed to reinforce the evidence of the benefit of CPM in prevention of arthrofibrosis, but the limited evidence currently available suggests that CPM may lead to early improvements in ROM, and therefore may have a preventative effect.²

Resolution of clinical scenario

This patient should be assessed by the physiotherapists and early movement encouraged. CPM may be included as part of this protocol if appropriate.