A pilot observational study of the association of 24-hour mortality with the subjective assessment of the forearm skin temperature and moisture compared to other bedside indicators of illness severity

Background: Although the association of peripheral skin temperature with infection, serious illness and death have been recognised for centuries, few studies have explicitly compared this finding with other bedside indicators of illness severity. This study compared subjectively assessed dorsal forearm skin temperature and moisture with other indicators of illness severity. Methods: Non-interventional observational study of acutely ill medical patients admitted to a low-resource Ugandan hospital, which examined the association of subjectively assessed dorsal forearm skin temperature and other bedside findings with death within 24 h. Results: While in hospital 653 patients had 2,104 observations; the dorsal forearm skin was subjectively felt to be abnormally hot or cold at 239 observations, and this finding was associated with 24-hour mortality (odds ratio 4.48, 95% CI 1.89-10.46); this increased risk of death was comparable to the increased mortality risk associated with tachypnoea, hypoxia, and a Shock Index >1.0, but considerably lower than that associated with a Kitovu Fast Triage score >0. When stratified according to both temperature and wetness, 'cold and wet' and 'hot and wet' skin were associated with higher early warning scores. Cold or hot forearm skin had a specificity for 24-hour mortality of 0.83, but a sensitivity of only 0.34; therefore, its absence does not rule-out the chance of imminent death. Conclusion: Touching and feeling the skin temperature and moisture is a valuable clinical sign, which can be rapidly determined at the bedside. However, although it has high specificity, its sensitivity for imminent death is low.