ABSTRACT Aims Skin temperature, including absolute temperature (at bony prominence areas under long‐term compression) and relative temperature (the difference between bony prominence and adjacent control area), may serve as early warning indicators for PI. However, the optimal indicator remains unclear. This meta‐analysis therefore synthesises evidence on their association with PI risk to identify the best indicator and evaluate its early‐warning accuracy. Design Systematic review and meta‐analysis. Methods We included prospective cohort studies of adult patients investigating longitudinal associations between skin temperature and subsequent PI development. We pooled standardised mean difference (SMD) and odds ratios, complemented by summary receiver operating characteristic (SROC) curve analysis. The overall quality of evidence was evaluated using the GRADE method. Data Sources We researched PubMed, Embase, CINAHL, Cochrane Library (CENTRAL), Wanfang and CNKI databases from inception to September 25, 2024. Results After screening 1354 titles and abstracts, ten studies comprising 1742 participants were included in the final synthesis. No significant difference in absolute temperature (combined SMD) was found between the PI and non‐PI groups (seven studies included). In addition, decreased relative temperature (< −0.1°C) was associated with a 16‐fold increased likelihood of PI (95% CI 6.38–40.19, I 2 = 79.4%) (three studies included), with the SROC curve analysis showing an AUC of 0.776. According to GRADE, the evidentiary certainty was very low for AT and low for RT. Conclusions Relative temperature is significantly related to the risk of PI, supporting its role as a promising early warning indicator. Future studies should establish a standardised measurement protocol to facilitate its clinical application. Implications for the Profession Monitoring skin temperature changes holds promise as a non‐invasive tool for early warning of PI risk. However, the amount and quality of available evidence limit our confidence in these findings, underscoring the need for further research before a definitive conclusion can be drawn. Reporting Method This study followed PRISMA guidelines. Patient or Public Contribution No patient or public contribution. Trial Registration PROSPERO CRD42024550099
Xu et al. (Mon,) studied this question.