The Prognostic Significance of Proteinuria Severity in Pregnancy: A Retrospective Cohort Study of Maternal and Neonatal Outcomes

Objective: To investigate the impact of proteinuria severity on obstetric and neonatal outcomes and to assess the predictive value of 24 h urinary protein excretion, both alone and within a multivariable model, for adverse pregnancy outcomes. Methods: This retrospective cohort study included 203 pregnant women with proteinuria who were classified into mild (≥0.3 g/day and <3.0 g/day, n = 50), severe (≥3.0 g/day and <5.0 g/day, n = 67), and massive (≥5.0 g/day; n = 86) groups based on 24 h urine protein levels. Maternal and neonatal outcomes were compared between these groups. Correlation analysis, receiver operating characteristic (ROC) curve analysis, and multivariable logistic regression were used to evaluate the predictive value of proteinuria for obstetric complications and identification of increased risk of early delivery. The AUC values of the proteinuria-only model and the multivariable model were compared using the DeLong test, as both models were derived from the same dataset and therefore represented correlated ROC curves. Results: The incidence of obstetric complications was significantly higher in the severe (68.7%) and massive (81.4%) proteinuria groups compared with the mild group (32.0%; p<0.001). Increasing proteinuria severity was associated with earlier gestational age at delivery, lower birth weight, and higher rates of fetal growth restriction (all p < 0.001). The 24 h proteinuria level demonstrated moderate predictive ability for obstetric complications (AUC 0.73; 95% CI 0.66–0.80). A multivariable model including nephrotic-range proteinuria (≥3 g/day) and gestational age at diagnosis showed improved discriminatory performance compared with proteinuria alone (AUC 0.81; 95% CI 0.75–0.88). The model based on continuous 24 h proteinuria yielded an AUC of 0.73 (95% CI, 0.66–0.80) for identifying pregnancies at increased risk of obstetric complications. The multivariable model showed a numerically higher AUC of 0.81 (95% CI, 0.73–0.86); however, the difference between the two AUCs was not statistically significant according to the DeLong test (z = 0.82, p = 0.41). Conclusions: The severity of maternal proteinuria is associated with a higher likelihood of adverse maternal and neonatal outcomes, and higher proteinuria levels appear to show a graded association with increasing risk. A multivariable model integrating proteinuria with key clinical parameters demonstrated moderate discriminatory ability for obstetric complications, may support a more holistic approach to risk stratification in clinical practice.