Diagnostic accuracy of semiquantitative point of care urine albumin to creatinine ratio and urine dipstick analysis in a primary care resource limited setting in South Africa

Abstract

BACKGROUND: The prevalence of chronic kidney disease (CKD) is predicted to rise over the next few decades. In resource-limited settings access to central laboratory services is limited. Point-of-care (POC) urine dipstick testing offers the potential to detect markers of kidney damage (albuminuria) as well as markers of other disease processes. We evaluated the diagnostic accuracy of the semi-quantitative albumin-creatinine ratio (ACR) Sysmex UC-1000 POC urine dipstick system as well as the extent of other abnormal dipstick findings in urine. METHODS: 700 participants from a rural area in South Africa were screened for albuminuria. A spot urine sample was used to measure POC and central laboratory ACR. We determined the sensitivity, specificity, positive predictive value and negative predictive value of the POC ACR, and recorded dipstick parameters. RESULTS: The prevalence of albuminuria was 11.6% (95%CI; 9.3-14.2). Those with albuminuria had higher mean diastolic (82 vs 79 mmHg, p = 0.019) and systolic (133 vs 128 mmHg, p = 0.002) blood pressures and a higher proportion of diabetes mellitus (17.6 vs 4.9%, p < 0.001). The sensitivity of the POC ACR system was 0.79, specificity 0.84, positive predictive value 0.39 and negative predictive value 0.97. The sensitivity improved to 0.80, 0.85, 0.85 and 0.89 in those with elevated blood pressure, diabetes mellitus, HIV positive status, and those 65 years and older, respectively. Abnormalities other than albuminuria were detected in 240 (34.3%) of the samples; 88 (12.6%) were positive for haematuria, 113 (16.1%) for leucocytes, 66 (9.4%) for nitrites and 27 (3.9%) for glycosuria. CONCLUSION: Our study shows that POC ACR has good negative predictive value and could be used to rule out albuminuria when screening for CKD. Additionally, a high proportion of participants had other urine abnormalities detected with dipsticks which may reflect kidney disease or co-morbid untreated genitourinary pathology such as urinary tract infections or endemic schistosomiasis with important implications for CKD.