Predictors of mortality in severe community acquired pneumonia in an intensive care unit of Sri Lanka

Background: Severe community-acquired pneu-monia (SCAP) carries high mortality. This study describes the predictors of mortality in SCAP.
Method: A prospective cross-sectional study was conducted in the Medical Intensive Care Unit (MICU) of National Hospital of Sri Lanka (NHSL) recruiting patients with SCAP admitted from July to December 2023. Disease outcome, complications and organ support received were documented.
Results: Among 56 participants, 36 (64.3%) were males and 33 (58.9%) were ≥50 years of age.  Septic shock requiring vasopressors was seen in 36 (69.6%), acute kidney injury (AKI) in 24 (42.9%) and 40 (71.4%) required mechanical ventilation (MV). There were 20 (35.7%) deaths.
A higher number of non-survivors had respiratory failure (100% vs 55.6%, p<0.0001), PaO2/FiO2 ratio<250 (100% vs 77.8%, p=0.02) and required MV (100% vs 55.6%, p<0.0001) compared to survivors. Length of MV was higher in non-survivors than survivors (8 vs 2.5 days p<0.001).
History of chronic renal failure (OR 6.214, 95% CI 1.84 - 20.99, p=0.003), presence of AKI (OR 9.0, 95% CI 2.54 - 31.80, p=0.001) and need for renal replacement therapy (OR 13.44, 95% CI 3.07 -58.71, p=0.001) were associated with mortality. Patients who had septic shock and required vasopressors (OR 15.2, 95% CI 1.833 - 126.07, p=0.01) had a higher risk of death. In a multivariable logistic regression model, AKI independently predicted mortality (adjusted OR 5.3, 95% CI 1.26 - 28.28, p<0.02).
Conclusion: Presence of respiratory failure, AKI and septic shock predicts mortality in patients with SCAP.