The Protein-to-Creatinine Ratio for the Prediction of Significant Proteinuria in Patients at Risk for Preeclampsia: a Meta-Analysis

Objective: To investigate the diagnostic accuracy of the protein-to-creatinine ratio from random urine collections to confirm the presence of proteinuria in women being evaluated for preeclampsia. Study design: Eligible studies, published between January 1966 and April 2010, were retrieved through general bibliographic databases. Accuracy of the protein-to-creatinine ratio was estimated compared with a 24-hour urine collection. Pooled estimates of diagnostic measures were calculated. A random-effects bivariate model was employed. Results: Twenty-four trials with 3,186 aggregate participants met inclusion criteria. Pooled sensitivities and specificities were 91.0% (95%CI 87.0 - 93.9) and 86.3% (95% CI 78.4 - 91.7) respectively. Pooled positive likelihood ratio was 6.7 (95% CI 4.1, 10.9) and pooled negative likelihood ratio 0.10 (95%CI 0.07, 0.16). Meta-regression analysis found that test accuracy was not affected by any of the co-variables explored. Conclusion: A random urine protein-to-creatinine ratio provides useful evidence to rule out the presence of significant proteinuria in patients at risk for preeclampsia. It appears that a cut-off value of > 0.30 is associated with the best accuracy. Condensation: The protein-to-creatinine ratio from a random urine sample provides useful evidence to rule out the presence of significant proteinuria in patients at risk for preeclampsia.