EARLY HYPOPHOSPHATEMIA AND PROLONGED MECHANICAL VENTILATION IN ELDERLY ICU PATIENTS: A RETROSPECTIVE COHORT STUDY

Introduction: Electrolyte disturbances are common in critically ill older adults and may influence respiratory muscle function and weaning outcomes. Hypophosphatemia has been implicated in longer durations of mechanical ventilation (MV ≥3 days), but data focusing specifically on elderly, invasively ventilated patients remain limited. To evaluate the association between early hypophosphatemia and the duration of mechanical ventilation(MV) in elderly intensive care unit (ICU) patients and to assess the relative contribution of concurrent electrolyte abnormalities. Methods: This single-center retrospective cohort study included elderly patients admitted to the ICU who required invasive MV. Serum electrolyte levels obtained within the first 24 hours of ICU admission were analyzed. Patients were categorized according to MV duration (<3 days vs. ≥3 days). Univariable and multivariable logistic regression analyses were performed to identify factors associated with MV ≥3 days. Correlation and receiver operating characteristic (ROC) analyses were also conducted where appropriate. Results: Electrolyte disturbances were highly prevalent, with more than 80% of patients exhibiting at least one abnormality at admission. Hypophosphatemia was more frequently observed in patients requiring MV ≥3 days compared with those ventilated for shorter durations. Lower serum phosphate levels were significantly associated with longer MV duration and demonstrated a moderate negative correlation with ventilator days. In multivariable analysis adjusting for age, sex, illness severity, and admission diagnosis, hypophosphatemia showed a borderline association with MV ≥3 days. Among other electrolytes, chloride and magnesium differed between groups in univariable analyses, whereas sodium, potassium, and calcium did not. Conclusions: In elderly ICU patients, early disturbances in serum phosphate levels were associated with MV ≥3 days, even when values remained within or near the conventional reference range. Although hypophosphatemia did not emerge as a statistically significant independent predictor after multivariable adjustment, the observed associations suggest potential clinical relevance. Prospective multicenter studies incorporating longitudinal electrolyte measurements, frailty assessment, and nutritional data are warranted to clarify causal relationships and optimize ventilatory management in this growing population.