The effect of ethnicity on the performance of protein-creatinine ratio in the prediction of significant proteinuria in pregnancies at risk of or with established hypertension: an implementation audit and cost implications.

Introduction: The replacement of 24-h urine collection by protein-creatinine ratio (PCR) for the diagnosis of preeclampsia has been recently recommended. However, the literature is conflicting and there are concerns about the impact of demographic characteristics on the performance of PCR.

Material And Methods: This was an implementation audit of the introduction of PCR in a London Tertiary obstetric unit. The performance of PCR in the prediction of proteinuria ≥300 mg/day was assessed in 476 women with suspected preeclampsia who completed a 24-h urine collection and an untimed urine sample for PCR calculation. Multivariate logistic regression was used to assess the independent predictors of significant proteinuria.

Results: In a pregnant population, ethnicity and PCR are the main predictors of ≥300 mg proteinuria in a 24-h urine collection. A PCR cut-off of 30 mg/mmol would have incorrectly classified as non-proteinuric, 41.4% and 22.9% of black and non-black women, respectively. Sensitivity of 100% is achieved at cut-offs of 8.67 and 20.56 mg/mmol for black and non-black women, respectively. Applying these levels as a screening tool to inform the need to perform a 24-h urine collection in 1000 women, would lead to a financial saving of €2911 in non-black women and to an additional cost of €3269 in black women.

Conclusions: Our data suggest that a move from screening for proteinuria with a 24-h urine collection to screening with urine PCR is not appropriate for black populations. However, the move may lead to cost-saving if used in the white population with a PCR cut-off of 20.5.