Back
Acute kidney injury (AKI) is highly prevalent among COVID-19 patients, with incidence rates exceeding 50% in recent surveys. It is associated with poor outcomes and elevated mortality rates. This study aims to explore variations in patient outcomes based on the type of hospital governance—public versus private—in a large cohort of AKI patients from two institutions in Brazil.
We conducted a retrospective analysis at two major medical institutions in Rio de Janeiro, Brazil, from February 2020 to April 2022. We included patients aged 16 and above admitted to intensive care units (ICUs). Patients with end-stage kidney disease requiring dialysis were excluded. We analyzed data regarding demographics, comorbidities, clinical presentation, ICU resource utilization, and AKI incidence.
During the study period, we analyzed 9,112 patients after exclusions. Among these, 2,333 tested positive for COVID-19 via RT-PCR. Within this cohort, 1,292 were admitted to private hospitals and 1,041 to public hospitals. There was no difference in median age between hospitals (65 vs. 63 years), and there was a predominance of male patients in private hospitals (65.9% vs. 54.6%, p<0.01). The prevalence of diabetes was higher in private hospitals (35.2% vs. 31.9%, p<0.05), and there was no difference in hypertension and chronic kidney disease between the two settings. The median SAPS3 score was higher in private hospitals (49 [43–57] vs. 46 [40–54], p<0.001). AKI incidence did not differ significantly between the hospital types (80.4% in private vs. 78.8% in public). However, stage III AKI was more frequent in public hospitals (40.3% vs. 37.4%), as well as kidney replacement therapy requirement (24.8% vs. 20.5%, p<0.05). The utilization rates of ICU resources, such as high-flow nasal cannula, tracheostomy, parenteral nutrition, and extracorporeal membrane oxygenation, were significantly higher in private hospitals. Overall mortality was higher in public hospitals (46.7% vs. 31.3%, p<0.001). Among the 1,859 AKI patients, mortality rates were 54.1% in public hospitals and 45.9% in private hospitals. In multivariate analyses, admission to a public hospital was independently associated with increased mortality (OR 4.44, 95% CI 3.36 - 5.93), even after adjusting for demographics, comorbidities, and severity of presentation. However, after excluding patients who were directly referred from state emergency units to the public hospital ICUs—most of whom were already critically ill and on mechanical ventilation (comprising 315 patients)—and excluding patients admitted to the ICU on the day of emergency visit in the private hospital (119 patients), the mortality rates were comparable (24.4% in public vs. 23.6% in private hospitals, p=0.722).
Our study reveals significant disparities in outcomes and resource utilization between public and private hospitals, although AKI prevalence did not markedly differ between them. ICU resource utilization was higher in private settings, potentially reflecting disparities in access to advanced therapies.
Interestingly, when accounting for patients directly referred to public ICUs from state emergency units and those admitted to the ICU on the day of their emergency visit at a private hospital, mortality rates aligned.
This finding highlights a selection bias in the patient population and the severity of their presenting conditions.
Further research is needed to explore these relationships further and to substantiate these observations.