Prospective 3D ultrasonographic evaluation of immediate postpartum urine retention volume in 100 women who delivered vaginally

Our objective was to determine the reproducibility and validity of urine retention volumes measured by 3D ultrasonography (3D-US) 2 h postpartum. This was a longitudinal prospective study of 100 women who delivered vaginally. Each woman underwent five successive urine retention measurements with the Bladder Scan 2 h after vaginal delivery, followed by transurethral catheterization to determine the real volume. The reproducibility of only the last four scan values was analyzed with an intraclass correlation coefficient, and that of determinations 4 and 5 with a Bland-Altman plot. The validity of the device was similarly evaluated using the mean four-scan and real values. Pearson's correlation coefficient was calculated and a least-trimmed squares robust linear-regression model was used. All tests were two-sided with a 5% alpha risk. Results showed that 3D-US measurements were highly reproducible, with an intraclass correlation coefficient of 0.974 (95% CI:0.964-0.981). 3D-US also accurately estimated the real urine volume (intraclass correlation coefficient 0.924; 95% CI:0.888-0.949), but tended to underestimate by about 6.8%. The between-technique Pearson correlation coefficient was 0.94 (P<0.0001) and the regression analysis gave an R-2 of 0.914. Urine retention volumes exceeded 500 ml in 55 women, and 1 l in 7. We concluded that 3D-US was reliably able to measure postpartum urine retention volume.