Objective: The purpose of this study was to determine the value of the protein/creatinine ratio in prediction of 24-hour urine total protein among women with suspected preeclampsia.
Study design: Women who were evaluated for suspected preeclampsia at >or=24 weeks of gestation were studied prospectively if there was no concurrent diagnosis of chronic hypertension, diabetes mellitus, or preexisting renal disease. A protein/creatinine ratio was obtained, which was followed by the initiation of a 24-hour urine evaluation. Positive and negative predictive values and sensitivity and specificity of the protein/creatinine ratio for significant (>or=300 mg) and severe proteinuria (>or=5000 mg) that were based on 24-hour urine total protein were calculated.
Results: A total of 220 women were evaluated; 43.2% of the women were black, and 80% of the women had government insurance. Mean maternal and gestational ages were 26.1 years and 36.5 weeks, respectively. Significant and severe proteinuria on 24-hour urine evaluation were identified in 76.4% and 8.2% of cases, respectively. Regression analysis of protein/creatinine ratio and 24-hour urine total protein level showed a poor correlation (r(2)=0.41). Receiver operator characteristic analysis revealed an area under the curve of 0.80, but the shoulder value of 390 mg/g carried a high false-negative rate (45.2%). With a more conservative cutoff value, a protein/creatinine ratio of >or=300 mg/g had a poor negative predictive value (47.5%), a specificity for significant proteinuria (55.8%), with a positive predictive value of 85.5%, and a sensitivity of 81%. For severe proteinuria, a protein/creatinine ratio of >or=5000 mg/g had a poor positive predictive value (61.9%) and sensitivity (72.2%), with a negative predictive value of 97.5%, and a specificity of 96.0%.
Conclusion: Protein/creatinine ratio does not exclude adequately the presence of significant proteinuria or predict severe proteinuria and should not be used as an alternative to 24-hour total protein evaluation.