Accuracy of Urinalysis Dipstick Techniques in Predicting Significant Proteinuria in Pregnancy

Abstract

The accuracy of dipstick urinalysis with a 1+ threshold in the prediction of significant proteinuria is poor and therefore of limited usefulness to the clinician. Accuracy may be improved at higher thresholds (greater than 1+ proteinuria), but available data are sparse and of poor methodological quality. Therefore, it is not possible to make meaningful inferences about accuracy at higher urine dipstick thresholds. There is an urgent need for research in this area of common obstetric practice.

Urinary Dipstick Proteinuria Testing: Does Automated Strip Analysis Offer an Advantage Over Visual Testing?

“…The sensitivity for detection of proteinuria on random urine samples using a visual reading method (56.0%) was consistent with published values for visual reading tests (55%), 5 but still at the low end of the reported range for automated testing (53.9% vs. 41% to 100%, respectively). [6][7][8][9] Our specificities were high for both visually read (95.3%) and automated dipstick (97.3%) testing compared with published values (84% for visual and 37% to 100% for automated testing).…”

“…[6][7][8][9] Our specificities were high for both visually read (95.3%) and automated dipstick (97.3%) testing compared with published values (84% for visual and 37% to 100% for automated testing). [5][6][7][8][9] We are aware of two studies that directly compared the diagnostic test properties of visual dipstick testing with automated testing. A strength of our study is that we assessed a broad spectrum of high-risk patients with and without significant proteinuria (diagnosed by a random urinary PrCr of ≥ 30 mg/mmol), consistent with published recommendations.…”

“…The final study sample (N = 109) had greater than 80% power to detect a 20% difference in sensitivity between visual and automated strip testing, a difference compatible with the published literature (two-sided alpha, error 0.05; sensitivity of visual method assumed to be 55%). 5,7 This study was part of a large quality improvement project of proteinuria assessment in our tertiary maternity hospital. 12,13 We included only women who provided nondilute urine samples with urinary creatinine concentration ≥ 3 mmol/L in our primary analysis.…”

“…In a review of six studies involving 1738 women and using a cut-off of ≥ 1+ for detection of significant proteinuria, measured by either 24-hour urine collection or random PrCr, sensitivity was 55%, with specificity of 84%. 5 Diagnostic test performance of visually read dipstick proteinuria measurement may be improved with use of an automated test strip reader. In theory, automation may also reduce subjectivity; nevertheless, published sensitivities (41%, 6 82%, 7 90%, 8 and 100% 9 ) and corresponding specificities (100%, 81%, 86%, and 37%) for automated reading have varied widely, even when the prevalence of proteinuria in the study populations was similar (i.e., 45% 7 and 48% 6 ).…”

Urinalysis vs urine protein–creatinine ratio to predict significant proteinuria in pregnancy

“…Specificities also vary widely. [2][3][4][5] Differences in methodology contribute to this large discrepancy. For example, automated dipstick urinalysis is more specific for predicting 24-h proteinuria than is ward dipstick urinalysis.…”