UCSF

Objective

To determine whether center to center discrepancies in the ability to locate sperm in infertile testes with abnormal histology stems in part from inconsistencies in pathologists' readings of testis biopsies.

Design

Prospective cohort study.

Setting

Academic male infertility practice.

Patient(s)

Consecutive series of azoospermic men referred with testis biopsy slides between 1998 and 2003.

Intervention(s)

Testis biopsy histologies on azoospermic patients referred for infertility care were re-reviewed by a single pathologist blinded to the original reading. Subsequent infertility care was guided by the findings from the second histologic reading.

Main outcome measure(s)

Agreement between the outside and in-house review of testis biopsy readings was assessed with the kappa statistic. Pregnancy outcomes that resulted from clinical decisions informed by the second histologic readings were also assessed.

Result(s)

Among 113 histologic specimens, re-review was complicated by fixation artifacts in 18 cases (16%) and insufficient biopsy sample size in 13 cases (12%). The kappa score for interobserver agreement in readings was 0.43 (95% CI 0.32-0.054). Mixed histology patterns in particular were underappreciated by outside pathologists (13% of cases on original reading, 36% of cases on review). In 27% of all cases, the differences in biopsy readings had a significant impact on clinical management.

Conclusion(s)

A correlation between independent testis histology readings in azoospermic men demonstrates frequent inconsistencies. These differences contribute to inaccurate phenotyping of male infertility and can significantly impact the direction of infertility care. These findings highlight the need for a standardized approach to testis histologic review.