Abstract
OBJECTIVE
To assess the urine protein/creatinine ratio in urine samples of pregnant women with hypertension in regard to: 1) the presence of significant variation at different periods of the day; 2) the differences if they exist, to identify the most reliable period of the day for sampling; and 3) whether the first sample, obtained when the patient arrives at the clinic, correlates with the same accuracy, with the 24-hour proteinuria.
DESIGN
Cross-sectional study.
PLACE
Obstetrics Emergency Department, Hospital de Clínicas de Porto Alegre, a teaching hospital in Porto Alegre, Brazil.
POPULATION
Seventy-five women with hypertension with 20-week gestation or over.
METHODS
Urine samples for determination of the protein/creatinine ratio were obtained on arrival (first specimen) and every 6 hours thereafter, totaling four samples in 24 hours. Four sampling periods were established: 1) from 8 am to 2 pm, 2) from 2 pm to 8 pm, 3) from 8 pm to 2 am, and 4) from 2 am to 8 am. The protein/creatinine ratio in the four different day periods were compared with the 24-hour proteinuria obtained simultaneously. The results were analyzed by the Spearman correlation and the receiver-operator characteristic (ROC) curve.
RESULTS
The urine protein/creatinine ratio is strongly correlated (Spearman correlation equal to 0.8 or greater) with the 24-hour proteinuria at all four periods of the day (p<0.001), as well as the first sample obtained on arrival (p=0.003). These findings were corroborated by the ROC curve in which the values of four day periods and that of the first sample were equal to or greater than 0.930.
CONCLUSION
In hypertensive pregnant women, the single voided urine sample protein/creatinine ratio, irrespective of sampling time, is strongly correlated with the 24-hour proteinuria, as is the sample obtained on arrival.
Collapse