Introduction

Pre-treatment hemoglobin level is a prognostic factor in a number of epithelial malignancies including carcinoma of the cervix (Logsdon and Eifel 1999; Mundt et al. 1998; Gücer et al. 1998), ovary (Obermair et al. 1998), and endometrium (Tamussino et al. 2001). The relationship between elevated platelet count and the malignant tumors was first reported by Reiss et al. in 1872 (Tranum and Haut 1974). The thrombocytosis was previously found to be related with poor prognostic factors in ovarian carcinoma (Bozkurt et al. 2004; Zeimet et al. 1994), cervical (Hernandez et al. 1992, 2000; Lopes et al. 1994), and endometrial carcinomas (Tamussino et al. 2001; Menczer et al. 1996; Gucer et al. 1998; Scholtz et al. 2000; Ayhan et al. 2006). Tumors can produce or induce cytokines such as interleukins, interferon gamma, and tumor necrosis factor that can induce hemolysis, suppress erythropoiesis, and inhibit the response of erythroid progenitor cells to erytropoietin (Spivak 1994; Dowlati et al. 1997). Absolute or relative erytropoietin deficiency seems to be the final pathway leading to anemia in patients with tumors (Moliterno and Spivak 1996). Erytropoietin may be involved in the development of thrombocytosis associated with tumor anemia and has been shown to promote platelet formation in rats (Berridge et al. 1988). Ovarian cancer and cervical cancer can produce interleukin-6 (van der Zee et al. 1995). However, the findings of the studies on IL-6 levels in patients with endometrial cancer are conflicting.

The aim of this study was to evaluate the relationship between preoperative hemoglobin and platelet count and prognostic factors in patients with endometrial cancer.

Materials and methods

The study population consisted of 61 patients who were primarily treated by total abdominal hysterectomy, salpingo-oophorectomy, bilateral pelvic and para-aortic lymphadenectomy, infracolic omentectomy, and peritoneal cytology for endometrial carcinoma between January 2002 and December 2005 at Ankara Oncology Education and Research Hospital. The study protocol was approved by the ethics committee of Ankara Oncology Education and Research Hospital.

The surgical procedures for endometrial cancers in our institution are defined as the extended surgical staging consisting of a washing peritoneal cytology, total abdominal hysterectomy and bilateral salpingo-oophorectomy, infracolic omentectomy with full pelvic and para-aortic lymphadenectomy. The pelvic and peri-aortic lympadenectomy involved a systemic dissection of all pelvic and peri-aortic nodes, including the peri-aortic and common iliac nodes, from above the inferior mesenteric artery to the deep circumflex vein distally. The obturator fossa is entered removing all obturator nodes above the obturator nerve. The tumors were surgically staged according to the FIGO staging system (News FIGO 1989). Endometrioid adenocarcinomas were graded according to FIGO classification. The histological classification was based on the classification of tumors by World Health Organization Classification (Silverberg et al. 2003). The histologic types of endometrial cancer in the present study included endometrioid carcinoma, undifferentiated carcinoma, clear cell carcinoma, and papillary serous carcinoma. They were categorized as endometrioid and non-endometrioid (clear cell carcinoma, papillary serous carcinoma, and undifferentiated carcinoma) groups. Formalin-fixed hematoxylin and eosin-stained slides of 5 μm were prepared again from the tumor tissue of the same patients and revised by a senior pathologist (GB) to verify the diagnosis. Myometrial invasion was evaluated by assessing the percentage of myometrial thickness of the deepest tumor extension. Presence or absence of vascular invasion, cervical stromal invasion, omental metastasis, lymph node metastasis, and adnexal involvement from endometrial cancer were evaluated. Tumor size was measured by its maximum diameter. To determine the presence or absence of cancer cells, peritoneal cytology was studied.

Vascular invasion, adnexal involvement, cervical involvement, omental metastasis, positive washing cytology, and lymph node metastasis were dichotomized based on the presence or absence of each factor. Tumor size (≤2 vs. >2 cm), tumor grade (grade 1 vs. grade 2 and grade 3), and depth of myometrial invasion (myometrial invasion <1/2 vs. ≥1/2) were evaluated as two groups.

Blood hemoglobin and platelet count of the patients were obtained 3 days before the surgery. Anemia was defined as a hemoglobin level of ≤12 g/dL. Thrombocytosis was defined as a platelet count of ≥400,000 μL−1 (Zeimet et al. 1994).

Statistical analyses

Statistical analyses were performed using “SPSS 10.05 for Windows” computer program. Kruskal–Wallis, Mann–Whitney U, Fishers Exact and Chi-square tests were used for univariate analyses. The multivariate analysis was performed with logistic regression in a forward stepwise fashion. P values less than 0.05 derived from two-tailed tests were considered statistically significant.

Results

Sixty-one women underwent surgery for endometrial carcinoma in 48 months. The age range of the patients was 35–80 years with a mean age of 57.40 ± 8.99 years.

Twenty-six (42.6%) patients had a pre-treatment hemoglobin value of ≤12 g/dL. These 26 patients had significantly higher rates of positive cytology (P = 0.008), advanced FIGO stage (P = 0.03), omental metastases (P = 0.01), cervical involvement (P = 0.02), and thrombocytosis (P = 0.02) (Table 1). According to the results of the multivariate analysis, cervical involvement (P = 0.019, OR = 4.030, 95% CI = 1.255–12.947) and positive peritoneal cytology (P = 0.022, OR = 12.509, 95% CI = 1.43–109.36) were significantly associated with low hemoglobin level.

Table 1 Patients and clinocopathologic profile according to pre-treatment hemoglobin (Hb) of patients with endometrial carcinoma
Full size table

The univariate comparisons of the median preoperative platelet counts are presented in Table 2. No differences were determined between the patients for age, depth of myometrial invasion, peritoneal cytology, and lymphovascular invasion (LVSI). Cervical involvement and lymph node metastasis were associated with significantly higher median preoperative platelet counts when compared with the absence of cervical involvement and lymph node metastasis (P = 0.01 and P = 0.05, respectively). Based on the results of the multivariate analysis, thrombocytosis was not an independent prognostic factor.

Table 2 Univariate analysis; the association of preoperative platelet count with clinicopathologic factors in patients with endometrioid adenocarcinoma
Full size table

Discussion

Majority of endometrial carcinomas are diagnosed, while the disease is still confined to the uterus. A prognostic factor to identify patients at high risk for recurrence in this subgroup could be clinically useful.

Anemia in patients with cancer can be caused by bleeding or paraneoplastic mechanisms. Although most patients with endometrial carcinoma present with postmenopausal vaginal bleeding, this is rarely severe enough to cause anemia. Tumors can produce or induce cytokines such as interleukins, interferon gamma, and tumor necrosis factor that can induce hemolysis, suppress erythropoiesis, and inhibit the response of erythroid progenitor cells to erytropoietin (Spivak 1994; Dowlati et al. 1997). Absolute or relative erytropoietin deficiency seems to be the final pathway leading to anemia in patients with tumors (Moliterno and Spivak 1996). Anemia in these patients seems to be a marker of more aggressive disease (Obermair et al. 1998). In our study, cervical involvement and positive peritoneal cytology was significantly associated with low hemoglobin level. Erytropoietin may be involved in the development of thrombocytosis associated with tumor anemia and has been shown to promote platelet formation in rats (Berridge et al. 1988). Interleukin-6 (IL-6) and granulocyte–macrophage colony stimulating factor (GmCSF) probably stimulate thrombopoiesis and may provoke thrombocytosis (Salgado et al. 1999). Ovarian cancer and cervical cancer can produce IL-6 (van der Zee et al. 1995). However, the findings of the studies on IL-6 levels in patients with endometrial cancer are conflicting. Chopra et al. (1997) found normal IL-6 levels in 59 women with stage I–IV disease, whereas Scambia et al. (1994) found elevated levels in 37% of a series of patients.

The relationship between elevated platelet count and the malignant tumors was previously reported in several studies (Tamussino et al. 2001; Tranum and Haut 1974; Bozkurt et al. 2004; Zeimet et al. 1994; Hernandez et al. 1992; Hernandez et al. 2000; Lopes et al. 1994; Menczer et al. 1996; Gucer et al. 1998; Scholtz et al. 2000). In the present analysis, thrombocytosis was significantly more common in endometrial cancer patients with anemia than in those with a hemoglobin level >12 g/dL. Tamussino et al. (2001) have also described an apparent inverse relationship between hemoglobin and thrombocyte levels in patients with endometrial cancer.

Literature reveals few studies evaluating the association between preoperative platelet counts and endometrial carcinoma (Tamussino et al. 2001; Menczer et al. 1996; Gucer et al. 1998; Scholtz et al. 2000). Menczer et al. (1996) concluded that when 300,000 platelet count was taken as a threshold value, unfavorable grades were significantly more common in the patients with higher platelet counts. On the other hand, Gucer et al. (1998) reported that at a cut-off value of 400,000, thrombocytosis was a poor prognostic factor. In the study by Scholtz et al. (2000), 5-year DFS and overall survival rates were significantly influenced by stage, cervical involvement, and thrombocytosis. Tamussino et al. (2001) confirmed that thrombocytosis was associated with advanced FIGO stage, poor grade, and non-endometrioid histology in the presence of anemia in endometrial cancer patients. According to the results of Ayhan et al. (2006) advanced stage, poorly differentiated tumor grade, the presence of cervical and adnexal involvements are associated with preoperative platelet counts in endometrial cancer patients. In our study, the univariate analysis showed that the presence of cervical involvements and lymphatic metastasis were associated with significantly higher median preoperative platelet counts; however, in the light of the results of the multivariate analysis, thrombocytosis was not a prognostic factors.

In conclusion, low pre-treatment hemoglobin level may reflect poor prognostic factors such as positive cytology and cervical involvement in patients with endometrial cancer that is associated with thrombocytosis.