Reliability of intraobserver and interobserver sonographic endometrial stripe thickness measurements

Abstract

Objective: To determine the intraobserver and interobserver reliability of endometrial stripe thickness measurements in women undergoing controlled ovarian hyperstimulation.

Design: Prospective blinded study.

Setting: Tertiary care, university hospital.

Patient(s): Sixty-three patients undergoing controlled ovarian hyperstimulation and being monitored with transvaginal ultrasound were studied.

Intervention(s): None.

Main outcome measure(s): Intraobserver and interobserver variability of endometrial stripe thickness measurements between 2 blinded observers, with 2 observations made by each observer.

Result(s): A statistically significant correlation was detected between the 2 measurements of each observer. The mean (+/-SD) interobserver difference between the average of the 2 measurements performed by both observers was 1.02 +/- 0.82 mm. A statistically significant correlation between the measurements of the 2 observers was detected. For both observers A and B, who used < or =6 mm as an abnormal endometrial thickness, an excellent level of agreement was detected between the 2 measurements made on each patient. When comparing the average values obtained by the 2 observers for each of the patients, an excellent level of agreement was detected.

Conclusion(s): There is an excellent correlation between intraobserver and interobserver measurements of endometrial stripe thickness.