Background Total knee replacement (TKR) is a common intervention for patients with end-stage osteoarthritis of the knee. Post-surgical management may include cryotherapy. However, the effectiveness of cryotherapy is unclear. Objectives To evaluate the acute (within 48 hours) application of cryotherapy following TKR on pain, blood loss and function. Search methods We searched the Cochrane Database of Systematic Reviews, CENTRAL, DARE, HTA Database, MEDLINE, EMBASE, CINAHL, PEDro and Web of Science on 15th March 2012. Selection criteria Randomised controlled trials or controlled clinical trials in which the experimental group received any form of cryotherapy, and was compared to any control group following TKR indicated for osteoarthritis. Data collection and analysis Two reviewers independently selected trials for inclusion. Disagreements were discussed and resolved involving a third reviewer if required. Data were then extracted and the risk of bias of trials assessed. Main outcomes were blood loss, visual analogue score (VAS) pain, adverse events, knee range of motion, transfusion rate and knee function. Secondary outcomes were analgesia use, knee swelling, length of hospital stay, quality of life and activity level. Effects of interventions were estimated as mean differences (MD), standardised mean differences (SMD) or given as risk ratios (RR), with 95% confidence intervals (CI). Meta-analyses were performed using the inverse variance method and pooled using random effects. Main results Eleven randomised trials and one controlled clinical trial involving 809 participants met the inclusion criteria. There is very low quality evidence from 10 trials (666 participants) that cryotherapy has a small benefit on blood loss (SMD -0.46, 95% CI, -0.84 to -0.08), equivalent to 225mL less blood loss in cryotherapy group (95% CI, 39 to 410mL). This benefit may not be clinically significant. There was very low quality evidence from four trials (322 participants) that cryotherapy improved visual analogue score pain at 48 hours (MD = -1.32 points on a 10 point scale, 95% CI, -2.37 to -0.27), but not at 24 or 72 hours. This benefit may not be clinically significant. There was no difference between groups in adverse events (RR = 0.98, 95% CI, 0.28 to 3.47). There is low quality evidence from two trials (107 participants) for improved range of motion at discharge (MD 11.39 degrees of additional flexion, 95% CI 4.13 to 18.66), but this benefit may not be clinically significant. There was no difference between groups in transfusion rate (RR 2.13, 95% CI 0.04 to 109.63), and knee function was not measured in any trial. No significant benefit were found for analgesia use, swelling or length of stay. Outcomes measuring quality of life or activity level were not reported. Authors' conclusions Potential benefits of cryotherapy on blood loss, postoperative pain, and range of motion may be too small to justify its use, and the quality of the evidence was very low or low for all main outcomes. This needs to be balanced against potential inconveniences and expenses of using cryotherapy. Well designed randomised trials are required to improve the quality of the evidence.
