The prevalence of acute kidney injury in patients with community-acquired pneumonia who required mechanical ventilation

Abstract

BACKGROUND: Community-acquired pneumonia (CAP) is a common reason for intensive care unit (ICU) admission and sepsis. Acute kidney injury (AKI) is a frequent complication of community-acquired pneumonia and is associated with increased short- and long-term morbidity and mortality and healthcare costs. OBJECTIVE: Describe the prevalence of AKI in patients with CAP requiring mechanical ventilation and evaluate its association with inhospital mortality. DESIGN: Retrospective cohort SETTING: Intensive care unit PATIENTS AND METHODS: We included patients with CAP on mechanical ventilation. Patients were categorized according to the development of AKI in the first 24 hours of ICU admission using the Kidney Disease Improving Global Outcomes (KDIGO) classification from no AKI, stage 1 AKI, stage 2 AKI, and stage 3 AKI. MAIN OUTCOME MEASURES: The primary outcome was hospital mortality. Secondary outcomes were ICU mortality, hospital and ICU length of stay, ventilation duration, tracheostomy, and renal replacement therapy requirement. RESULTS: Of 1536 patients included in the study, 829 patients (54%) had no AKI while 707 (46%) developed AKI. In-hospital mortality was 288/829 (34.8%) for patients with no AKI, 43/111 (38.7%) for stage 1 AKI, 86/216 (40%) for stage 2 AKI, and 196/380 (51.7%) for stage 3 AKI ( P <.0001). Multivariate analysis revealed that stages 1, 2, or 3 AKI compared to no AKI were not independently associated with in-hospital mortality. Older age, vasopressor use; decreased Glasgow coma scale, PaO 2 /Fio 2 ratio and platelet count, increased bilirubin, lactic acid and INR were associated with increased mortality while female sex was associated with reduced mortality. CONCLUSION: Among mechanically ventilated patients with CAP, AKI was common and was associated with higher crude mortality. The higher mortality could not be attributed alone to AKI, but rather appeared to be related to multi-organ dysfunction. LIMITATIONS: Single-center retrospective study with no data on baseline serum creatinine and the use of estimated baseline creatinine distributions based on the MDRD (Modification of Diet in Renal Disease)equation which may lead to an overestimation of AKI. Second, we did not have data on the microbiology of pneumonia, appropriateness of antibiotic therapy or the administration of other medications that have been demonstrated to be associated with AKI.