TO021PREVALENCE, AETIOLOGY AND OUTCOME FROM COMMUNITY ACQUIRED ACUTE KIDNEY INJURY (AKI) IN SUB-SAHARA AFRICA: RESULTS OF THE MALAWI AKI STUDY (MAKIST)

Abstract

Introduction and Aims: Worldwide epidemiological data on AKI comes predominantly from developed settings. Here, AKI is common causing significant morbidity and mortality in both the short and long term. However, 85% of the world's population lives in developing settings, and therefore the greatest impact of AKI is likely in the poorest parts of the world. It is thought that young patients are dying unnecessarily from AKI, hence gathering data into the causes and consequences of AKI in this setting has become a global healthcare priority. There has been no previous prospective study of community acquired AKI in sub-Sahara Africa. Our aim was to establish the prevalence, aetiology, and outcome of community acquired AKI in general medical admissions to a central hospital in Blantyre, Malawi. Methods: All medical admissions to QECH aged 14 or more were included. Patients were screened for AKI at admission with measurement of serum creatinine and urine output. Demographic data, primary cause for admission, medical/drug history were recorded. Patients with serum creatinine above reference range (>90μmol/l in women, and >104μmol/l in men) were reviewed by the nephrology service and a workup undertaken into the nature and cause of renal injury. Data was collected between April and July 2015. Patients with renal injury were reviewed daily and followed up for a minimum of 1 week or until hospital discharge. Hospital outcome was recorded in all. AKI was defined as per KDIGO, and causes categorised according to the AKIN STOP criteria. Results: 943 patients were admitted during the study period; 892 were enrolled in the study. Patients were young (median age 37, IQR 30-52), with a slight male predominance (56.1% M), and high prevalence of HIV (42.2%). 188 (21.1%) admissions had evidence of kidney disease; in 153 (17.2% of admissions) this was AKI. Of those with AKI, 93 (60.8%) had KDIGO stage 3. Acute dialysis was indicated in 21 (13.7%), and provided in 8 (5%). The predominant STOP cause across all stages of AKI was sepsis and hypoperfusion (87.6%). The most common admission diagnoses in these patients were acute gastroenteritis (30; 19.6%), acute febrile illness with no clear source (21; 13.7%) and non-tuberculous lower respiratory tract infection (16; 10.5%). In 70% of these cases, the AKI was multifactorial with toxin use a secondary contributor in all but one. The predominant toxins were NSAIDs (43.1%) and tenofovir (43.1%). Inpatient mortality was 13% (n=91) in patients without kidney disease, and 41.8% (n=61) in those with AKI (p <0.0001). Mortality increased with AKI stage (stage 1 and 2 30%, stage 3 49.5%, p=0.017). The unadjusted odds ratio (OR) for death was 4.63 (3.12-6.85; p < 0.0001) when comparing AKI vs. no kidney disease, and 4.27 (2.83-6.42; p <0.0001) when adjusted for age, sex, HIV, diabetes and hypertension. OR for death increased with AKI stage, adjusted OR 2.54 (1.36-4.77; p = 0.004) in stage 1 and 2 and 5.80 (3.56-9.44; p <0.0001) in stage 3 AKI, both compared to patients without kidney disease. Independent risk factors for developing any stage of AKI were HIV, diabetes, age > 40, and tenofovir use. Conclusions: AKI is common in general medical admissions in Malawi and leads to significantly increased inpatient mortality. The predominant causes are volume and infection related, both potentially treatable with relatively simple means. Effective interventions are urgently required to reduce preventable deaths from AKI in young patients in the developing world.