A retrospective analysis of mortality predictors in the geriatric population admitted to a tertiary care medical intensive care unit, department of internal medicine

KINACI, E., A Retrospective Analysis of Mortality Predictors in the Geriatric Population Admitted to a Tertiary Care Medical Intensive Care Unit, Department of Internal Medicine, Specialty Thesis, Van, 2025. Objective: This study aims to identify the risk factors associated with adverse clinical outcomes by evaluating the demographic data, laboratory findings, and therapeutic approaches of patients aged 65 and older treated in the Internal Medicine Intensive Care Unit of Van Yüzüncü Yıl University Faculty of Medicine Hospital between 2022 and 2023. The primary focus of the research is to investigate the predictive value of routine hemogram parameters—specifically mean platelet volume (MPV), neutrophil and lymphocyte levels, and the neutrophil-tolymphocyte ratio (NLR)—as inflammatory markers in anticipating undesirable clinical events. Materials and Methods: In this retrospective study, the clinical records of 160 geriatric patients (aged ≥65 years) treated in the internal medicine intensive care unit between 2022 and 2023 were evaluated. Patients with a length of stay ≤3 days or a history of malignancy were excluded. Demographic characteristics, comorbidities, and laboratory findings at the time of ICU admission and discharge (or mortality) were retrospectively screened. For prognostic assessment, alongside routine parameters, the neutrophil-to-lymphocyte ratio (NLR), mean platelet volume (MPV), and their temporal changes during the ICU stay (∆NLR and ∆MPV) were calculated. Statistical analyses included intergroup comparisons, univariate and multivariate logistic regression to evaluate the prognostic impact of variables, and ROC curve analysis to determine cut-off values. Results: The mean age of the 160 included patients was 76.5±7.3 years (53.1% male), with gastrointestinal bleeding being the most common reason for ICU admission (25.6%). The overall mortality rate was 34.4%. Compared to survivors, non-survivors exhibited significantly lower lymphocyte counts and albumin levels, alongside markedly higher baseline neutrophilto-lymphocyte ratios (NLR), dynamic changes in NLR (ΔNLR) and mean platelet volume (ΔMPV), as well as elevated urea, creatinine, uric acid, and C-reactive protein (CRP) levels (p<0.05). The prevalence of chronic kidney disease and the requirement for hemodialysis were also notably higher in the mortality group. Multivariate logistic regression analysis identified hemodialysis requirement, elevated NLR, ΔNLR, ΔMPV, low albumin levels, and prolonged V length of ICU stay as independent risk factors for mortality. ROC curve analyses revealed that ΔMPV (AUC: 0.736, cut-off: 0.25) and albumin (AUC: 0.724, cut-off: 2.45 g/dL) were the strongest predictors of mortality, followed by CRP (AUC: 0.673, cut-off: 61.5) and NLR (AUC: 0.654, cut-off: 24). Conclusion: Dynamic changes in the neutrophil-to-lymphocyte ratio (NLR) and mean platelet volume (MPV) during ICU stay (ΔNLR and ΔMPV) are identified as independent prognostic markers for predicting mortality in geriatric intensive care patients. The routine monitoring of these cost-effective and easily accessible hematological parameters can provide a practical advantage in early risk stratification and the individualization of clinical management. Nevertheless, large-scale, prospective, multicenter studies are required to validate these findings. Keywords: Geriatrics, Intensive care unit, Mortality, Neutrophil-to-lymphocyte ratio, Mean platelet volume, Prognostic biomarkers.