Intra-observer and inter-observer variability in scoring laparoscopic diagnosis of pelvic adhesions

The objective of this study was to investigate intra-observer as well as inter-observer variability in the assessment of laparoscopic scoring of adhesive disease. Patients with suspected pelvic adhesions underwent a laparoscopic examination and surgical correction, using a standardized adhesion scoring system for severity as well as for extent of the adhesive process. An active agent or placebo was placed in the peritoneal cavity to discourage reformation of adhesions on a double-blind basis. A second laparoscopy was performed in 6-10 weeks and patients were re-scored. The videotapes were viewed and re-scored by the operator and four other gynaecological surgeons on two occasions, 3 months apart. Both adhesion severity and extent scores were lower following laparoscopic intervention. Reliability coefficients as defined by intra-class correlations were large indicating good reliability (0.778 and 0.758 for severity and extent respectively). Intra-observer (replication) variability was 12% of the total versus 19% for inter-observer variability of adhesion severity. These values were 15.4 and 22.3% of the total respectively, for adhesion extent. Individual observers tended to up-grade severity and minimize extent of adhesions when comparing the second with the first videotape views. The surgeon recorded greater adhesion score differences between the two procedures than the other observers. Though two physicians consistently had higher scores than two others, the intra-observer and interobserver variabilities were acceptable. Consultant review of high quality videotaped laparoscopic procedures is a valid method of independent assessment of operative findings.