Development and external validation of machine learning-based models to predict patients with cellulitis developing sepsis during hospitalisation.

Objective: Cellulitis is the most common cause of skin-related hospitalisations, and the mortality of patients with sepsis remains high. Some stratification models have been developed, but their performance in external validation has been unsatisfactory. This study was designed to develop and compare different models for predicting patients with cellulitis developing sepsis during hospitalisation.

Platelet-to-lymphocyte ratio predicts the duration of glucocorticoid therapy in the treatment of cutaneous adverse drug reactions.

Background: Glucocorticoid (GC) remains the mainstay of treatment for cutaneous adverse drug reactions (cADRs) but has been associated with side effects, emphasizing the importance of precisely managing the duration of high-dose GC treatment. Although the platelet-to-lymphocyte ratio (PLR) has been proven to be closely related to inflammatory disorders, its ability to predict the timing of GC dose reduction (Tr) during cADRs treatment remains obscure.

Methods: Hospitalized patients diagnosed with cADRs treated with glucocorticoids were analyzed in the present study to evaluate the association between PLR values and Tr values using linear, locally weighted scatter plot smoothing (LOWESS) and Poisson regression.