Perioperative blood loss in robotic urologic surgery: a retrospective evaluation of estimation methods

Abstract

Urology has been a leading field in the adoption of robotic surgery, which offers technical advantages and low complication rates, including notably reduced intraoperative blood loss. In this study, we aimed to examine the relationship between formula-based estimated blood loss and visually estimated intraoperative blood loss in robotic urologic procedures. In this retrospective study, 111 robot-assisted urologic surgery were included. The agreement between the visually estimated intraoperative blood loss and the estimated values calculated using different formulas (Hb dilution method formula, Hb mass method, Gross Formula, L & oacute;pez-Picado Formula). To determine how consistent each estimation was with the visually estimated intraoperative blood loss and with each other, Bland-Altman analysis, Concordance Correlation Coefficient (CCC) and Intraclass Correlation Coefficient (ICC) were applied. Intraoperative blood loss (visual estimation) indicated a mean blood loss of 220.72 +/- 212.61 mL, whereas formula-based calculations consistently yielded higher estimates: L & oacute;pez-Picado, 721.64 +/- 532 mL; Hb mass method, 667.79 +/- 429 mL; Gross formula, 726.97 +/- 540 mL; and Hb dilution method, 737.99 +/- 545 mL. The analyses revealed that all formulas differed statistically significantly from the visually estimated intraoperative blood loss. Evaluation of agreement and consistency demonstrated that the formulas showed poor agreement both with estimated blood loss and with one another. The strongest concordance was observed between L & oacute;pez-Picado and Gross formula. There was a large discrepancy between visually estimated intraoperative blood loss and formula-based estimations. While formula-based methods show strong internal consistency, they differ substantially from the subjective estimates commonly used.