Quality Assurance in Ultrasonographic Measurements of the MyomatousUterus

Abstract

It is important to evaluate quality control in health care. The aim of the current study was to evaluate different sources of variation in estimating the preoperative weight of the myomatous uterus by using repeatability and reproducibility (R&R) method. Material and Methods: Estimation of the total weight of the uterus by a formula combining the formulas for the prolate ellipsoid and cylinder was preoperatively determined using a transvaginal ultrasound probe. Three physicians repeated each round of measurements three times, producing in total 108 findings (12 subjects x 3 investigators x 3 rounds) in 12 women with symptomatic leiomyomas scheduled to undergo hysterectomy. Variation was divided into different components: physicians, patients, and repeated measurements. Variation due to differences across repeated measurements (repeatability), across physicians (reproducibility), and across patients (variability) was then estimated. The estimates of uterine weights were compared to the true weight of the hysterectomy specimen. Results: The more experienced the physician was in taking the ultrasound measurements the less deviation was observed between her own three measurements. Repeatability was 28 %, reproducibility 0% and patient-to-patient variation 72%. There was no significant difference between the accuracy of the measurements of the three physicians. Conclusions: The experience of the physician had an effect on repeatability but not on reproducibility in estimating uterine weight by ultrasound. Our results indicate the importance of conscientiousness in taking measurements of the uterus and cervix.