Management of Krukenberg tumors: an 11-year experience and review of the literature.

Purposes: The Montefiore Medical Center (MMC) experience with Krukenberg tumors (gastrointestinal cancer metastatic to the ovary) was reviewed 1) in order to determine whether it could be used to make clinical management recommendations, eg, differences in treatment based on menopausal status, secondary debulking, prophylactic oophorectomy; 2) to compare the MMC experience to the reported literature, since one of our patients survived 1912 months beyond the generally accepted mean of this rare tumor.
Methods: 1) Prospective study from January 1985-April 1996 of 5 patients followed at MMC with a diagnosis of gastrointestinal (GI) cancer metastatic to the ovary; 2) retrospective chart review of all female patients admitted to MMC with a diagnosis of GI cancer during the same time period; 3) computerized literature review from 1966 to 1996.
Results: Five patients were followed prospectively. Retrospective chart review identified 1,016 female patients admitted with gastric (231) or colon (785) cancer, 2 additional cases of Krukenberg tumor were identified. Of the 7 patients with Krukenberg tumors the primary tumor was colon in 5 and gastric in 2. The average age at presentation was 49.3 years (range 35-80); 5 were premenopausal, 2 were postmenopausal, and 2 (28.6%) were postpartum. The average survival of these 7 patients was 12.3 months (range 4 days to 26 months). The 5 patients who had ovarian metastases removed at first laparotomy survived an average of 10.8 months. The 1 patient who had a secondary debulking survived longest (26 months). The 1 patient who had no surgery lived 6 months. These numbers are not statistically significant, but do suggest that further study is warranted. Including the prospectively studied patients, 1,021 patients were evaluated. Seven of 1,021 (0.7%) had Krukenberg tumors. Two of 11 (18.2%) premenopausal gastric cancer patients had a Krukenberg tumor, whereas none of the 222 postmenopausal gastric cancer patients did. Similarly, 3 of 41 (7.3%) premenopausal colon cancer patients had a Krukenberg tumor compared to 2 of 747 (0.3%) postmenopausal colon cancer patients. This difference in menopausal status is significant by Fisher's Exact test (P <.001). Nineteen patients underwent oophorectomy at the time of primary colon cancer surgery: 11 had normal or atrophic ovaries, 2 thecomas, 4 simple cysts, and 2 Krukenberg tumors. Three of 19 (15.8%) were premenopausal.Conclusions: Review of the literature has identified a number of diagnostic and management issues that appear to impact on survival. These include timing of definitive diagnosis of Krukenberg tumors, ie, before, after, or at the same time as diagnosis of the GI primary tumor; menopausal status; concurrent pregnancy; role of debulking; prophylactic oophorectomy. The prognosis worsens when the primary tumor is identified after the metastasis to the ovary is discovered. Krukenberg tumor is more common in premenopausal women than in postmenopausal women. The number of patients with Krukenberg tumors discovered postpartum in our study was significantly increased compared to the other series in the literature. The number of patients who received prophylactic oophorectomy or secondary debulking in our study was inadequate to draw conclusions regarding a benefit of these procedures. However, a benefit may be suggested for debulking, as survival appears to be increased. More importantly, there may be a role for prophylactic oophorectomies in both pre- and postmenopausal patients, as this would eliminate the need for a repeat laparotomy. A randomized trial is needed to evaluate the role of prophylactic oophorectomy and debulking. Since the clinical and pathologic details in the literature vary widely, it is extremely difficult to compare studies, particularly the treatment and survival of patients with Krukenberg tumors. Reports should include age, site of GI primary, time from diagnosis of primary to ovarian metastasis, and overall survival as well as survival from the time of diagnosis and treatment of the Krukenberg. We wish to alert the clinician that persistent GI symptoms always warrant investigation. Pelvic inflammatory disease, pregnancy, and postpartum endometritis may mask the GI symptoms. Delays in diagnosis should be avoided. During surgery, the gynecologic surgeon must do a complete upper abdominal exploration, and the general surgeon must do a complete pelvic evaluation. Since Krukenberg tumors are rare, a national registry should be started to gather information on these patients; this might lead to better diagnosis and treatment.