Development of a predictive model for septic shock-associated acute skin failure using readily available clinical variables
摘要
Acute skin failure (ASF) is an understudied complication in patients with septic shock, and existing predictive models lack disease-specific variables. This study aimed to investigate the characteristics of cutaneous manifestations in patients with septic shock and to identify potential biomarkers and predictive factors for cutaneous deterioration.
MethodsThis retrospective cohort study analysed 154 adult patients with septic shock, defined according to the Sepsis 3.0 criteria, who were admitted to a tertiary intensive care unit (ICU) between September 2020 and September 2022. Based on cutaneous manifestations observed during hospitalisation, the patients were stratified into two groups: the ASF group and the non-ASF group. Clinical characteristics, therapeutic interventions and laboratory parameters were evaluated. Significant univariate predictors were included in multivariable logistic regression. A p-value below 0.05 was considered statistically significant.
ResultsThis study enrolled a total of 154 patients with septic shock, among whom 49 developed ASF, yielding an incidence rate of 31.8%. In the multivariate analysis, four independent predictors were identified: maximum norepinephrine (NE) dose (odds ratio [OR] = 2.47, p = 0.051), NE duration (OR = 1.19, p = 0.054), central venous oxygen saturation (ScvO₂) (OR = 0.97, p = 0.042) and absence of oedema (OR = 0.18, p = 0.008). The model achieved an area under the curve of 0.803 (95% CI: 0.730–0.876) with 72.3% sensitivity and 72.4% specificity at the optimal cut-off.
ConclusionsThe validated prediction model identifies patients with septic shock who are at high risk for ASF using four readily available clinical parameters: maximum NE dose, NE duration, ScvO₂ and absence of oedema, helping clinicians provide early warning and thereby reduce associated skin complications.
Clinical trial numberNot applicable.