Does Tourniquet Use in TKA Increase Postoperative Pain? A Systematic Review and Meta-analysis

Although tourniquets are commonly used during TKA, that practice has long been surrounded by controversy. Quantifying the case for or against tourniquet use in TKA, in terms of patient-reported outcomes such as postoperative pain, is a priority. The purpose of this study was to meta-analyze the available randomized trials on tourniquet use during TKA to determine whether use of a tourniquet during TKA (either for the entire procedure or some portion of it) is associated with (1) increased postoperative pain; (2) decreased ROM; and (3) longer lengths of hospital stay (LOS) compared with TKAs performed without a tourniquet. We completed a systematic review and meta-analysis using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guidelines to assess the impact of tourniquet use on patients after TKA. We searched the following databases from inception to February 1, 2015, for randomized controlled trials meeting prespecified inclusion criteria: PubMed, Embase, and Cochrane Central Register of Controlled Trials. Postoperative pain was the primary outcome. Secondary outcomes were postoperative ROM and LOS. The initial search yielded 218 studies, of which 14 met the inclusion criteria. For our primary analysis on pain and ROM, a total of eight studies (221 patients in the tourniquet group, 219 patients in the no-tourniquet group) were meta-analyzed. We also performed a subgroup meta-analysis on two studies that used the tourniquet only for a portion of the procedure (from osteotomy until the leg was wrapped with bandages) and defined this as half-course tourniquet use (n = 62 in this analysis). The Jadad scale was used to ascertain methodological quality, which ranged from 3 to 5 with a maximum possible score of 5. Statistical heterogeneity was tested with I and chi-square tests. A fixed-effects (inverse variance) model was used when the effects were homogenous, which was only the case for postoperative pain; the other endpoints had moderate or high levels of heterogeneity. Publication bias was assessed using a funnel plot, and postoperative pain showed no evidence of publication bias, but the endpoint of LOS may have suffered from publication bias or poor methodological quality. We defined the minimum clinically important difference (MCID) in pain as 20 mm on the 100-mm visual analog scale (VAS). We found no clinically important difference in mean pain scores between patients treated with a tourniquet and those treated withou