Effects of physical therapy modalities for early postoperative pain following total knee arthroplasty: a systematic review and network meta-analysis.

The trials varied widely in how the modalities were delivered: number of sessions, when treatment started, how long it lasted, how intense it was, and what else was given alongside. The authors say this makes the pooled results harder to compare. All the different control conditions, including no treatment, placebo and usual care, were lumped into one comparison group, which the authors admit may distort the indirect comparisons. They could not separate them without breaking the network apart. Follow-up was short. Pain was measured within about two weeks of surgery, so this says nothing about longer-term pain, function or quality of life. Trials used different pain scales and the results were combined as standardised mean differences, which are unitless. The authors say this is less intuitive for clinicians and patients than expressing pain on a 0 to 10 scale. The authors did not compare the pain reductions against a threshold patients would consider meaningful, so it is unclear whether the statistically significant differences mattered in practice. Most included trials were rated as having some concerns for risk of bias, largely because participants and staff could not be fully masked, which can push effects up or down. Only studies published in English or Chinese were included, and 23 of the 42 trials were from China, so the findings may not carry over to other healthcare systems and rehabilitation practices. TENS, neuromuscular electrical stimulation and acupuncture were each supported by only a handful of small trials, so their estimates are the least stable and the authors say they should be read with caution.

The authors declared that financial support was received for this work or its publication, in the form of a grant from the Fundamental Research Funds for the Central Universities (Exercise Rehabilitation Science Laboratory). No other funding source or industry involvement is stated in the text supplied, and no commercial sponsor is reported as having designed the study or written the manuscript.

Do not read this as proof that cryotherapy is the best treatment for pain after knee replacement. SUCRA values only describe how often a modality ranked highly within this particular network of trials. The evidence for nearly every comparison was rated low or very low certainty, and the authors did not check whether the pain reductions were large enough for patients to notice.