Protein-creatinine ratio and albumin-creatinine ratio for the diagnosis of significant proteinuria in pregnant women with hypertension: Systematic review and meta-analysis of diagnostic test accuracy.

Abstract Background The gold standard for assessment and diagnosis of significant proteinuria in pregnancy has been by 24-hour urine collection and analysis. Determining fast, accurate methods to identify clinically significant proteinuria would aid diagnosis of pre-eclampsia. The objective of this study was to determine the accuracy of spot protein-creatinine ratio (PCR) and albumin-creatinine ratio (ACR) measurements compared with 24-hour urine collection for the identification of clinically significant proteinuria in women with hypertensive disorders of pregnancy. Methods Search strategies were developed for electronic databases from inception to 1st October 2020. Data were assessed for methodological quality using the QUADAS-II checklist for risk of bias and quality of the evidence using GRADE. Meta-analysis was performed where there were at least four studies presenting data for the same comparison (test and threshold). This is an update of the review for NICE guideline NG133 (published June 2019) and includes additional data. Results Twenty-nine studies were included. PCR measurements (28 studies) showed high sensitivity (91%) and specificity (89%) at a threshold of 30 mg/mmol (n = 3577). Higher thresholds (>60 mg/mmol) increased specificity, but reduced sensitivity. At a threshold of PCR 30 mg/mmol, diagnostic accuracy improved for sensitivity and specificity (93% for both) in studies where the first morning void was excluded (n = 1868). Data available (4 studies) for ACR supports ruling out of significant proteinuria at less than 2 mg/mmol, though evidence was limited by paucity of data and wide confidence intervals around the result. Conclusions PCR and ACR have high accuracy compared to the gold standard 24-hour urine collection.