Effects of cold therapy on ecchymosis and pain after subcutaneous heparin injection: a meta-analysis and systematic review

This systematic review and meta-analysis aimed to compare the efficacy of three local physical cooling techniques (ice pack compression, vapor cooling spray, and pre-cooled needles) on pain and ecchymosis after subcutaneous anticoagulant injection, re-examine prior controversies with the latest evidence, and provide evidence-based references for clinical nursing.

Five electronic databases (PubMed, Embase, Cochrane Library, Web of Science, CINAHL) were systematically searched for randomized controlled trials (RCTs) comparing local cold therapy with routine care. Primary outcomes included post-injection pain intensity, ecchymosis incidence, and ecchymosis size.

Ten randomized trials, including seven parallel-group and three randomized within-participant trials, enrolled 807 unique participants. Local cold therapy reduced immediate pain intensity after heparin injection (SMD = -0.64, 95% CI: -0.81 to -0.47, P < 0.00001). The analgesic effects did not differ significantly among cooling modalities (P = 0.65). Cold therapy increased the odds of being ecchymosis-free at 48 h (OR = 2.77, 95% CI: 1.57-4.88, P = 0.0004) and reduced ecchymosis size at 48 h (SMD = -3.03, 95% CI: -4.95 to -1.10, P = 0.002). At 72 h, cold therapy increased the odds of being ecchymosis-free (OR = 4.17, 95% CI: 1.31-13.29, P = 0.02), but did not significantly reduce ecchymosis size (SMD = -0.76, 95% CI: -1.89 to 0.37, P = 0.19). Heterogeneity was substantial for ecchymosis size at 48 h (I2 = 98%); therefore, the magnitude of this pooled effect should be interpreted cautiously.

Local cold therapy may reduce immediate injection pain and early ecchymosis after subcutaneous heparin injection. Evidence at 72 h was outcome-dependent: ecchymosis-free status favored cold therapy, whereas ecchymosis size did not differ significantly. The findings should be interpreted cautiously because of high heterogeneity, outcome-specific risk of bias, and inconsistent reporting of intervention and measurement procedures.

https://www.crd.york.ac.uk/PROSPERO/view/CRD420261278725, identifier CRD420261278725.