Objective
The aim of this study was to evaluate the effect of lymph node dissection in patients with atypical endometrial hyperplasia.
Study Design
Patients undergoing surgical management of atypical endometrial hyperplasia during the study period were retrospectively identified. Clinical and pathologic information was analyzed.
Results
Eighty-eight patients comprised our cohort. Median age was 54 years (range, 37–85 years). Sixty-seven patients had lymph node dissection at the time of surgery for atypical endometrial hyperplasia, whereas 21 did not. Twenty-five of 88 (28.4%) had endometrial carcinoma on final uterine pathology. Stages were as follows: 4 IA, 15 IB, 3 IC, 2 IIB, and 1 IIIC. Surgical outcomes were not statistically significant between staged and unstaged groups. Information from lymph node dissection influenced management decisions in 7 of the 25 (28%) cancer patients.
Conclusion
Lymph node dissection did not adversely affect surgical outcomes in patients with atypical endometrial hyperplasia. Because many of these patients have concurrent endometrial cancer, we recommend consideration of lymph node dissection in atypical endometrial hyperplasia patients undergoing definitive surgical treatment.
Atypical endometrial hyperplasia (AEH) is a putative precursor to type I endometrial carcinoma. AEH arises by the same hormonal mechanism as type I endometrial carcinoma and has been shown to coexist with carcinoma. AEH and carcinoma have similar morphologies, and when they appear concurrently in pathologic specimens, they are frequently in topographic proximity. AEH has been associated with a high risk, up to 29%, of progression to invasive endometrial cancer when left untreated.
Retrospective studies have suggested that the prevalence of concomitant endometrial cancer within hysterectomy specimens in women undergoing surgical treatment for AEH may be as high as 50%. In a recent, multicenter prospective cohort study (Gynecologic Oncology Group [GOG] no. 167), Trimble et al reported that 42.6% (123/289) of women undergoing hysterectomy for definitive management of AEH had endometrial carcinoma in their hysterectomy specimens, regardless of the initial method of diagnosis. Although the majority of cancers were grade 1, 30.9% (38/123 specimens) were myoinvasive and 10.6% (13/123) involved the outer half of the myometrium. The results of this study also highlighted the difficulties inherent in diagnosing AEH even when a panel of expert gynecologic pathologists reviewed the prehysterectomy specimens.
Surgical staging is critical in determining both prognosis and treatment for women with endometrial cancer. It also allows differentiation between those patients who will benefit from adjuvant treatment and those who may be spared the toxicities of further therapy. In a series of patients undergoing definitive treatment for known grade 1 endometrial carcinoma, Ben-Shachar et al reported that data from surgical staging significantly impacted postoperative treatment decisions in 29% of patients.
In recognition of the high risk of endometrial carcinoma in women with AEH, as well as the poor reliability of intraoperative frozen section pathology for detection of invasive disease, we reconsidered our surgical approach to AEH after the initial results of GOG no. 167 were released in 2004. Patients with AEH were counseled about their risk of concurrent endometrial cancer and recommended, in most cases, to undergo surgical staging, including hysterectomy, bilateral salpingo-oophorectomy (BSO), and pelvic and paraaortic lymph node dissection (LND). Our objective is to report the results of this surgical approach to AEH.
Materials and Methods
After institutional review board approval, a chart review was performed of AEH patients treated at 2 institutions from February 2004–July 2008. Outpatient billing databases at the New York University Cancer Institute and the Bellevue Hospital gynecology clinic were searched and subjects identified using International Classification of Diseases revision 9 (ICD-9) codes for endometrial hyperplasia. Subjects included all patients with a preoperative diagnosis of complex or simple atypical endometrial hyperplasia who underwent hysterectomy and removal of the adnexa. Surgery was performed by 1 of 5 faculty gynecologic oncologists. Electronic medical records, pathology reports, hospital charts, and outpatient medical charts were reviewed. Information collected included patient demographics, intraoperative findings, operative time, pathologic findings, surgical complications, and perioperative morbidities. Complications considered directly or potentially related to the surgical procedure performed included infection (wound, urinary tract, pulmonary, or gastrointestinal infection); hematologic condition (anemia requiring blood transfusion, deep vein thrombosis, or pulmonary embolism); bowel obstruction or ileus; lymphedema and lymphocysts; and urinary tract, vein, or nerve injury. Intraoperative frozen section to assess the presence of invasive cancer, and, if present, the depth of myometrial invasion, was performed at the discretion of the attending surgeon. Frozen sections were performed by a staff surgical pathologist. Descriptive statistics were used to summarize the demographic and clinical patient data. Statistical analysis included χ 2 test, Fisher’s exact test, and Student t test as appropriate, and was performed using Microsoft Excel 2004 (Microsoft, Redmond, WA). Significance was set at the standard level of P < .05.
Results
We identified 88 patients who were surgically treated for AEH during the study period. The median age of the cohort was 54 years (range, 37–85 years). The majority (67%) of patients were postmenopausal.
Sixty-seven patients had lymph node sampling at the time of definitive hysterectomy and adenexectomy for AEH, whereas 21 patients underwent hysterectomy with removal of the adnexa only. Table 1 shows the demographic and clinical characteristics of the study groups. All characteristics were similar, and differences between groups were not statistically significant.
| Characteristic | Hysterectomy/BSO (n = 21) | Hysterectomy/BSO with LND (n = 67) |
|---|---|---|
| Age, y | 57 (38–75) | 53 (37–85) |
| BMI, kg/m 2 | 31 (21–45) | 28 (19–54) |
| Invasive endometrial cancer | 7 (33.3) | 18 (26.8) |
| Menopausal status | ||
| Pre | 5 (24) | 24 (36) |
| Post | 16 (76) | 43 (64) |
| Ethnicity | ||
| White | 12 (57) | 45 (67) |
| Hispanic | 4 (19) | 9 (13) |
| African American | 3 (14) | 5 (7) |
| Asian | 2 (10) | 8 (12) |
| Medical comorbidities a | ||
| Yes | 11 (52) | 25 (37) |
| No | 10 (4) | 42 (63) |
| Sampling method | ||
| Endometrial biopsy | 7 (33) | 30 (45) |
| D&C | 14 (67) | 37 (55) |
| Surgical approach | ||
| Laparotomy | 13 (62) | 35 (52) |
| Laparoscopy | 8 (38) | 32 (48) |
a Medical comorbidities included: hypertension, coronary artery disease, cerebrovascular accident, diabetes mellitus, and pulmonary vascular disease.
All patients undergoing LND had pelvic nodes removed, whereas only 4 subjects underwent dissection of the paraaortic nodes. The average number of nodes removed was 10 (range, 1–39; standard error of the mean, 1.03). Blood loss and hospital stay were comparable between the staged and unstaged groups ( Table 2 ). Operative time was on average 16 minutes longer in the staged group but did not reach statistical significance ( P = .19). No patient required blood transfusion.
