Construction of a Risk Prediction Model for the Occurrence of Acute Skin Failure in Critically Ill Patients: A...

The risk factors for acute skin failure (ASF), a serious complication of the skin, are not fully understood.

This study was designed to explore the risk factors for ASF in critically ill patients and construct a clinical prediction model.

Intensive care unit patients were prospectively observed and assigned into two groups: with and without ASF. A logistic regression model was constructed, and its predictive power and clinical utility were evaluated.

Of the 204 eligible patients enrolled as participants, 58 (28.43%) developed ASF. Sepsis, vasoactive drugs, and age were shown to be risk factors for ASF, whereas peripheral perfusion index ratio and albumin level were shown to be protective factors. The area under the receiver operating characteristic curve was 0.83. The maximum Youden index of the model was 0.39 (specificity: 0.87, sensitivity: 0.77). The Hosmer-Lemeshow test (p = .20) and calibration curve showed good fitness and predictive utility with respect to the model.

The developed model effectively predicts ASF risk, allowing for the early identification of high-risk patients. Identifying risk factors such as sepsis, vasoactive drugs, and age and considering protective factors such as peripheral perfusion index and albumin levels may help optimize care plans. Clinical staff should pay special attention to these factors and their impact on skin health in critically ill patients.