Urinary calcium in asymptomatic primigravidas who later developed preeclampsia

To assess the efficacy of calciuria as a diagnostic test for the prediction of preeclampsia.

We asked young healthy primigravidas from the prenatal clinic in our hospital to collect a 24-hour urine sample at 17-20 weeks' gestation. Urinary calcium was measured and expressed as calciuria per 24 hours (mg/24 hours), urinary excretion of calcium per 24 hours on a basis of body weight (mg/kg/24 hours), and as the calciuria-creatinuria ratio (mg/mg). For each test, sensitivity, specificity, predictive values, and relative risk (RR) were calculated. The test with the best overall performance was determined by comparing receiver operating characteristic curves. Sixty-nine patients completed the study until the end of the puerperium, and 15 of them were diagnosed with preeclampsia.

The most efficient test was the urinary excretion of calcium per 24 hours based on body weight. Using 3.4 mg/kg/24 hours as a cutoff point, we obtained a sensitivity of 80% (95% confidence interval [CI] 59.8-100), specificity 64.8% (95% CI 52.1-77.5), positive predictive value 38.7% (95% CI 21.6-55.8), negative predictive value 92.1% (95% CI 83.5-100), and RR 4.9 (95% CI 1.5-15.8).

In young, apparently healthy primigravidas, a low urinary excretion of calcium per kilogram of body weight per 24 hours before the end of the first half of gestation is a risk factor for preeclampsia, with an acceptable sensitivity and high negative predictive value, but with a positive predictive value no better than chance.