In endometrial carcinoma, extrauterine advancement and metastases are poor prognostic factors(1–4). Frequent sites of advancement and metastases of endometrial carcinoma are the peritoneum, adnexa, and lymph nodes; thus, histologic examination of these organs is important. However, omentectomy is not included in the recommendations for primary treatment in the National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology-v.1.2005(5). There have been very few reports of the value of omentectomy for the treatment of endometrial carcinoma, particularly endometrioid adenocarcinoma. We performed an omentectomy in addition to routine staging laparotomy as primary surgery for clinical stage I endometrioid adenocarcinoma and prospectively investigated omental metastases. In addition, the relationships between omental metastases and lymph node metastases, peritoneal cytology findings, and adnexal metastases were investigated to clarify the significance of omentectomy.
The clinical benefit of an omentectomy in endometrioid adenocarcinoma is unclear. The objective of this study was to clarify the significance of an omentectomy performed for clinical stage I endometrioid adenocarcinoma. A prospective study was performed on 134 patients with clinical stage I endometrioid adenocarcinoma who underwent omentectomy in addition to a staging laparotomy between 1998 and 2004: simple total hysterectomy, bilateral salpingo-oophorectomy, pelvic and para-aortic lymph node dissection, and peritoneal cytology. The frequency and prognosis of omental metastases and their relationships with extrauterine spread to other sites were investigated. Omental metastasis was noted in four patients (3.0%). As for extrauterine spread, the positivity rate of lymph node metastases was 13/128 (10.2%), peritoneal cytology was 13/133 (9.8%), and adnexal metastases was 10/134 (7.5%). Omental metastases correlated with peritoneal cytology and adnexal metastases (P < 0.05 for both); however, two of the omental metastases–positive patients were peritoneal cytology negative. All omental metastases–positive patients died shortly after surgery, showing that their prognosis was poor. The omental metastases rate for clinical stage I endometrioid adenocarcinoma was lower than the positive rates for extrauterine spread to other sites; thus, the routine application of omentectomy as a part of a staging laparotomy may not be efficacious. However, omental metastases are a significant poor prognostic factor, and intraoperative examination of the omentum by close inspection and palpation as well as pathologic examination, if possible, may be indicated.
Materials and methods
Omentectomy was performed in addition to routine staging laparotomy (simple total hysterectomy, bilateral salpingo-oophorectomy, pelvic and para-aortic lymph node dissection, and peritoneal cytology) in patients who were not suspected as having extrauterine advancement of the lesion preoperatively. The study was conducted from 1998 through 2004 at Jichi Medical University Hospital (Tochigi, Japan), and all patients gave written informed consent. This study was approved by the Institutional Review Board of Jichi Medical University Hospital. Of 178 patients histologically diagnosed with endometrioid adenocarcinoma, 134 patients were at clinical stage I and underwent omentectomy; these patients were prospectively investigated.
Because nonendometrioid adenocarcinoma has a higher frequency of extrauterine than endometrioid adenocarcinoma(6–8), omentectomy was routinely performed for patients with nonendometrioid adenocarcinoma in our hospital. For this reason, nonendometrioid adenocarcinoma patients were excluded from the study.
Infracolic omentectomy was performed on all patients. A minimum of 10 × 5 cm was excised. When metastases were macroscopically suspected, a complete omentectomy was performed. A pathologist prepared samples primarily from the region suspected to contain metastases by inspection and palpation; subsequently, a histologic diagnosis was made. When inspection and palpation did not suggest metastases, samples were prepared from at least one region and histologically examined. Concurrently, the presence or absence of metastases in the excised lymph nodes, adnexa, and peritoneal washings were histologically investigated. The relationships between omental metastases and extrauterine tumor advancement, the histologic grade, and myometrial depth were investigated. The presence of an estrogen receptor was not measured. For statistical analysis, the Chi-square test or Fisher exact probability test was used.
The patients’ backgrounds are shown in Table 1. The median age was 57 years with an age range of 29–81 years. The surgical stage was I in 98 patients, II in 15, III in 17, and IV in 4.
|Age (29–81, median 57)|
Omental metastases were noted in 4 of the 134 patients (3.0%). As for the other stage-determining factors, the positive rate of lymph node metastases was 13 of 128 (10.2%), the positive peritoneal cytology rate was 13 of 133 (9.8%), and the adnexal metastases rate was 10 of 134 (7.5%) (Table 2). Lymph node dissection (mean number of removed lymph nodes: 42; range 20–88) was not performed in six patients because of risk factors or advanced patient age; peritoneal cytology was not done in one patient because of a lost cytology sample.
|Lymph node||13/128 (10.2)|
|Positive peritoneal cytology||13/133 (9.8)|
The incidence of omental metastases was significantly higher in patients positive for adnexal metastases and peritoneal cytology (P < 0.05 for both); however, no significant difference was noted in regard to the depth of myometrial invasion, histologic grade, or lymph node metastases (Table 3).
|Factors||n||Omental metastases (n)||%||P|
|Myometrial invasion ≥1/2 (IC)||37||1||2.7||0.69|
|Myometrial invasion <1/2 (IA, IB)||97||3||3.1|
|Grades 2 and 3||46||3||6.5||0.12|
|Positive lymph node||13||1||7.7||0.27|
|Negative lymph node||115||2||1.7|
|No adnexal involvement||124||2||1.6|
|Positive peritoneal cytology||13||2||15.4||0.047|
|Negative peritoneal cytology||120||2||1.7|
Table 4 shows the clinicopathologic features of four omental metastases–positive patients. In cases 1 and 3, peritoneal cytology was negative, the depth of myometrial invasion was less than 1/2, and lymph node metastases was negative, showing omental metastases in low-risk cases. In cases 1, 2, and 4, intraoperative inspection and palpation strongly suggested metastases; however, the final diagnosis for case 3 was made by histologic examination of the adhesion site between the uterine body and the omentum. All four patients with omental metastases died early (at 8, 5, 5, and 15 months), showing a very poor prognosis for this finding.
|Case||Age (years)||Grade||Peritoneal cytology||Muscle invasion||Lymph node metastases||Adnexal involvement||Omental metastases (cm)||Prognosis (month)|
|4||59||2||Positive||<1/2||Not done||Positive||>1||Death (15)|
The frequency of omental metastases in patients with clinical stage I endometrioid adenocarcinoma was 3.0%. The reported frequency of omental metastases in clinical stage I endometrial cancer ranged from 6.0% to 8.3%(9–11), which is higher than that of our study. However, nonendometrioid adenocarcinomas, such as serous adenocarcinoma, were included in these reports. Serous adenocarcinoma is more likely than endometrioid adenocarcinoma to invade beyond the uterus(6–8), which may explain the increased frequency in their reports. When nonendometrioid adenocarcinomas, such as serous adenocarcinoma, were excluded from their reports, the frequency of omental metastases was 3.1%(9) and 5.3%(10), and these were consistent with our findings. Thus, the frequency of omental metastases in clinical stage I endometrioid adenocarcinoma may be lower than the positive rates of the other extrauterine invasions, such as lymph node metastases, peritoneal cytology, and adnexal metastasis.
It might be possible to diagnose omental metastases by inspection and palpation. We suspected metastases based on inspection and palpation in three of the four cases. Chen(3) reported that metastases were confirmed by histologic examination in 3 of 202 cases (1.5%) diagnosed to be free of metastases by intraoperative inspection and palpation. Gehrig et al.(12) reported that most cases of omental metastases could be detected by inspection because the sensitivity was 0.89 and the specificity was 1.00 for the intraoperative diagnosis of endometrial serous adenocarcinoma via surgical inspection. In contrast, other studies(9,10) found that more than half of the cases of omental metastases could not be diagnosed before histologic examination, showing that it is not possible to diagnose all cases of omental metastases solely by inspection. Although close inspection and palpation during surgery are important for the diagnosis of omental metastases, histologic examination should also be performed when adhesions and/or thickening are noted.
All patients with omental metastases died of the primary disease shortly after surgery; thus, they had a very poor prognosis. Omental metastases may be a poor prognostic factor for endometrioid adenocarcinoma. Potent postoperative adjuvant therapy may be necessary for patients with omental metastases.
This prospective clinicopathologic study clarified that the frequency of omental metastases was lower than that of lymph node metastases and positive peritoneal cytology in clinical stage I endometrioid adenocarcinoma. Thus, performing an omentectomy on all patients as a part of routine laparotomy cannot be recommended. However, close inspection and palpation during surgery are essential because omental metastases are an important poor prognostic factor. When adhesion or thickening is noted, the omentum should be excised and histologically examined.