19: Rates of inappropriate oophorectomy at the time of benign hysterectomy


      Emerging evidence suggests that premenopausal oophorectomy is associated with worsened long-term health outcomes. As of 2008, the revised ACOG guidelines state, “strong consideration should be made for retaining normal ovaries in premenopausal women.” Given recent findings that nearly one in five hysterectomies in the US may be unnecessary and due to the potential adverse health effects of oophorectomy, we sought to determine the rate of potentially unnecessary oophorectomies being performed in premenopausal women during hysterectomy for benign indications.

      Materials and Methods

      We reviewed all non-radical inpatient hysterectomies performed in California from 2005-2011 using the Office of Statewide Health Planning (OSHPD) patient discharge database (PDD). The PDD includes all nonfederal hospital discharges. In addition to a primary diagnosis and procedure code each discharge includes up to 19 secondary procedure codes and 24 secondary diagnosis codes. Any diagnosis associated with cancer was excluded. Three independent investigators reviewed ICD-9 codes and created a master list of recognized indications for oophorectomy. We defined oophorectomies as “appropriate” if a supporting ICD-9 code (ovarian cyst, BRCA+ carrier status, endometriosis, etc.) was linked and “inappropriate” if no such codes were linked. “Premenopausal” was defined as age less than 50. STATA software was used for analysis and p<0.05 was considered statistically significant.


      A total of 259,294 inpatient, non-radical hysterectomies were performed in California between 2005 and 2011 for benign indications. Of these, 96,976 (37.4%) were performed with concomitant removal of all ovaries (bilateral or removal of the remaining ovary). The majority (53.2%) of oophorectomies were performed in premenopausal women. Based on lack of supportive diagnoses, 32,860 (36.7%) premenopausal oophorectomies were classified as “inappropriate.” Over the 7-year time period, the total number of inpatient premenopausal hysterectomies with oophorectomy decreased (10,166/year in 2004 to 4,672/year 2011); however, the percentage of “inappropriate” oophorectomy remained stable (36-38%). Logistic regression analysis identified Hispanic and Black race as the only demographic factors associated with an increased rate of “inappropriate” oophorectomy at the time of hysterectomy (p<.001), but hospital characteristics did not account for any observed differences. Endometriosis was the most common indication for “appropriate” oophorectomy.