A prospective study of the demographics, management and outcome of patients with acute kidney injury in Cape Town, South Africa

Aim To study the demographics and outcome of acute kidney injury (AKI) at Groote Schuur Hospital, Cape Town, South Africa. Methods and findings A prospective observational study of AKI fulfilling the Kidney Disease: Improving Global Outcomes definition, from 8 July 2012 to 8 July 2013. Ethics approval was granted by the University of Cape Town Human Research Ethics Committee. Consent was waived because patient data was de-identified and patient management was not adversely affected by the study. A clerking sheet was used for data collection. Patients were reassessed after 3 months. Main outcomes were renal recovery and 3 month mortality. Descriptive statistics and multivariate logistic regression were carried out for risk factors. Over this period there were 10,750 hospital admissions and 366 patients with AKI giving an incidence of 3.4%. Median age was 44 years (IQR 14-82) and 214 (58.5%) were male, with 152 (41.5%) female. Most, 265 (72.4%), had community acquired AKI. Common underlying comorbidities were hypertension (n = 152, 41.5%), diabetes mellitus (n = 65, 17.8%) Human immunodeficiency virus (HIV) (n = 75, 20.6%), heart disease (n = 58, 16.1%), and chronic kidney disease (n = 37, 10.1%). Renal biopsies were performed in 36 (9.8%) patients. In total, 202 (55.2%) patients were in the intensive care unit, and of the whole study population 204 (55.7%) were dialysed. Those admitted to ICU who required dialysis amounted to 145 (39.6%). The overall 3 month mortality was 38.8%. Among the 145 patients dialysed in ICU, there were 71 deaths (49%) at 3 month follow up. Of the 119 patients with follow up serum creatinine, 95 (79.8%) had full renal recovery, and 4 (3.4%) had end-stage renal disease. On multivariate analysis, mechanical ventilation was associated with 3 month mortality (OR 2.46, p-value 0.019, 95% CI 1.41-4.03). Sepsis had a borderline significant association (OR 1.83, P-value 0.066, 95% CI 1.02-3.27), as did prolonged time to dialysis (OR 1.93, p-value 0.08, 095% CI 0.93-4.03). HIV status did not affect outcome. The main study limitations were the large numbers of patients with AKI stage 3, reflecting the fact that the institution is a tertiary referral centre and that patients with earlier stages of AKI tended not to be referred. Another study limitation was the low number of patients who were available for follow up for 3 month serum creatinine. Conclusions The incidence of AKI in the population studied is 3.4% of hospital admissions and carries a high mortality risk, most significant in mechanically ventilated patients. Sepsis and late dialysis initiation may carry a risk of mortality, but HIV infection did not affect outcome. Follow up of patients at least 3 months after an episode of AKI is essential to detect and appropriately manage those with incomplete renal recovery. In this study 36 patients underwent a kidney biopsy, and in many of these the results guided patient management. This study demonstrates finally that it remains imperative that clinicians actively pursue underlying causes of acute decline in renal function, including urine analysis, renal ultrasonography and if indicated and safe, a renal biopsy.