Beyond conventional risk scales: temporal patterns and multidimensional predictors of early deep tissue pressure injury in ICU - a retrospective cohort study

Abstract

Background Deep tissue pressure injury (DTPI) is a severe form of pressure-induced tissue damage. While immobility and malnutrition are established risk factors, additional contributors such as haemodynamic instability, oedema and sedation may influence its onset in critically ill patients. Aim To evaluate temporal patterns and clinical predictors of early versus late DTPI in ICU patients, with emphasis on nutritional risk, oedema, sedation and routinely assessed clinical parameters. Study Design This retrospective cohort study was conducted in the adult ICUs of a tertiary-level university hospital between November 2024 and June 2025. A total of 211 patient records were reviewed. DTPIs were defined according to international guidelines and classified as early (onset <= 9 days) or late (onset > 9 days) based on the median onset time. Demographic, clinical, haemodynamic and treatment-related variables were analysed using bivariate comparisons, logistic regression and receiver operating characteristic curve analysis. Results Of 211 patients, 109 (51.7%) developed early-onset DTPI and 102 (48.3%) developed late-onset DTPI. Late DTPI was associated with longer hospital (median: 39.0 [26.75-68.00] days) and ICU stays (median: 14.0 [5.00-23.25] days), more oedema (28.4%) and higher sedative use (68.6%). Logistic regression identified shorter ICU stay, lower NRS-2002 score, oedema and sedation as independent predictors of early DTPI. ROC analysis showed limited discriminative power for the Braden score (AUC = 0.597), age (AUC = 0.581) and ICU length (AUC = 0.653). NRS-2002, oedema and sedation improved model discrimination (AUC = 0.60-0.70) and net clinical benefit. Conclusions Standard clinical risk scales alone may be insufficient to capture early DTPI risk during the initial days of ICU admission. Nutritional risk, oedema and sedation were independent predictors, whereas single biomarkers and conventional scales alone showed limited utility. These findings underscore the importance of integrating multidimensional clinical parameters with routine risk assessment to inform earlier nursing interventions and enhance DTPI prevention in ICU patients. Relevance to Clinical Practice Nursing managers and ICU clinicians should recognise that patients classified as low risk by the Braden Scale may still develop deep tissue pressure injury. Incorporating multidimensional risk assessment approaches that include nutritional status, oedema and sedative use alongside traditional scales can enhance early detection and prevention. Closer skin monitoring within the first 7-10 days of ICU admission, particularly for older patients, is essential to improve patient safety and outcomes.