Clinical characteristics of probable invasive pulmonary aspergillosis in the Intensive Care Unit: an 8-year retrospective, observational, single-center study

Background The incidence of invasive pulmonary aspergillosis (IPA) in intensive care unit (ICU) patients is rising. Given the challenges in obtaining histological confirmation, clarifying the clinical features of possible IPA (PIPA) and identifying risk factors for mortality are crucial to improving diagnosis and treatment. Methods We retrospectively analyzed clinical and treatment data from 240 PIPA patients, collecting survival status at 12 weeks post-diagnosis. Patients were divided into survival and non-survival groups for systematic comparison. Univariate and multivariate Cox regression analyses were performed to identify independent predictors of 12-week mortality, which were then used to develop a prognostic model. Results Among the 240 PIPA patients in the study, 71 (29.58%) survived and 169 (70.42%) did not. The non-survival group had higher rates of hematological malignancies/ Hematopoietic Stem Cell Transplantation (HSCT), more chronic kidney disease (CKD), higher APACHE II (acute physiological assessment and chronic Health assessment II) scores, elevated serum galactomannan (GM) levels, and increased inflammatory markers [platelet count, C-reactive protein (CRP), procalcitonin (PCT)]. In contrast, solid organ transplantation (SOT) was more common in the survival group, which also had higher serum albumin levels and arterial partial pressure of oxygen/fraction inspired oxygen (PaO2/FiO2) ratio. More survivors received antifungal therapy, including voriconazole and inhaled amphotericin B, and had significantly longer treatment duration. Multivariate Cox regression showed that APACHE II score (hazard ratio [HR] = 1.06; 95% confidence interval [CI]: 1.03–1.09), serum GM level (HR = 1.21; 95% CI: 1.09–1.36), and length of antifungal therapy (LAT) (HR = 0.95; 95% CI: 0.93–0.97) were independent predictors of 12-week mortality. A prognostic model incorporating these factors achieved an area under the receiver operating characteristic curve (AUC) of 0.784 and demonstrated good calibration, internal validation, and clinical utility. This model provides a reliable tool for risk stratification and outcome prediction in critically ill PIPA patients in ICU settings. Conclusions An elevated APACHE II score, increased serum GM level, and shorter length of antifungal therapy were independently associated with higher 12-week mortality in ICU patients with PIPA. Larger studies are needed to validate these findings.