Quality assurance of second opinion pathology in gynecologic oncology

Objective: To determine the effect of routine second review of pathologic material that was sent to Ohio State University before initiation of therapy. Methods: All the gynecologic-oncologic histopathology review diagnoses made during a 1-year period were compared with original pathologic diagnoses. When there was a discrepant diagnosis with the second interpretation, the case was reviewed by at least two pathologists. Discrepancies were coded as no diagnostic disagreement, no diagnostic disagreement but pertinent information not included, diagnostic disagreement without clinical consequences, diagnostic disagreement with minor clinical significance, or diagnostic disagreement with major clinical significance. Proportions and confidence intervals were calculated. Results: Pathology reports from 295 referred patients were reviewed. Two hundred forty-five (83.1%) showed no discrepancy. Discrepancies were found in 50 cases (16.9%). There was significant information missing in four cases (1.4%), diagnostic disagreement with no clinical significance in 22 cases (7.5%), and diagnostic disagreement with minor clinical significance in 10 cases (3.4%). In 14 cases (4.7%, 95% confidence interval 2.28, 7.12) the changes in diagnoses had major therapeutic or prognostic implications that included changes from malignant or low malignant potential to benign (seven cases), malignant to low malignant potential (three cases), change in tumor type (two cases), and assessment of invasion (two cases). The cost of reviewing 295 specimens was approximately $39,235. The cost of identifying each major discrepancy was about $2802. Conclusion: Routine pathology review of gynecologic-oncologic cases before definite treatment revealed notable discrepancies in diagnoses. In 4.7% of cases, the change in diagnosis had a major effect on proper treatment planning or a significant prognostic implication.