Between January 2006 and March 2010, 107 patients with EOC underwent surgery at the Departments of Obstetrics and Gynecology of the Provincial Hospital Affiliated to Shandong University (Jinan, China; n=71) and the Haimen City People's Hospital (Haimen, China; n=36). Diagnoses were confirmed by the pathology results. All the patients were classified by FIGO (2000) standard staging (26), and the age of the patients at the first visit was 34–79 years (mean, 55.28±10.34 years). The control group comprised females with benign ovarian tumors [n=32; age, 32–72 years (mean age, 51.38±9.05 years)], including simple ovarian cyst, benign serous or mucinous cystadenoma, and healthy adult females [n=12; age, 38–61 years (49.50±6.53 years)]. There was no significant age difference between the groups.

For enrollment in the study, patients had to meet the following criteria: i) Ovarian cancer without surgical contraindication and EOC confirmed by postoperative pathology; and ii) normal white blood cell count and neutrophil ratios, as shown by routine blood examination on admission. Patients were excluded due to infection, trauma, coronary heart disease, hypertension and connective tissue diseases, as these conditions could result in elevated CRP concentrations. The study protocol was approved by the Ethics Committee of Haimen City People's Hospital and the Provincial Hospital Affiliated to Shandong University, and all participants provided their written informed consent to participate in the study.

On the second day after admission, 6 ml morning fasting blood was collected from each patient and the serum of 3 ml was used for CRP quantification via nephelometry rates (High-Sensitivity CRP kit; Beckman Coulter, Miami, FL, USA). With the development of a high-sensitivity assay method, the associations between inflammation and disease risk allow a wide spectrum of CRP levels, particularly in the range of concentrations previously widely considered ‘normal’, i.e. levels <10 mg/l. Due to the intra-assay variability (1.64–3.34%), CRP serum levels ≤8 mg/l were defined as the normal range in the two hospitals in the present study. The serum of another 3 ml blood was used to determine the content of CA125 via electrochemiluminescence (Roche Diagnostics GmbH, Manneheim, Germany), and the CA125 normal reference value was set to 0–35 U/ml.

A total of 69 patients received surgery without residual tumor or residual lesions <2 cm; 38 patients received subtotal tumor resections with residual lesions ≥2 cm. Sixty-five cases underwent pelvic and para-aortic lymph node dissection: 21 patients had lymph node metastasis and 44 cases were without metastasis. Intraperitoneal combined with intravenous chemotherapy or simple intravenous chemotherapy was started 7 days after surgery once every 4 weeks, and regular follow-ups were performed for 11–74 months with a mean duration of 28.5 months.

SPSS 17.0 statistical software was used for the statistical analyses, and the measurement data are presented as the mean ± standard deviation or median. Comparisons of measurement data were performed using the Student's t-test, rank sum test or analysis of variance, and count data were compared using the χ² test. Single or multiple factors were compared using logistic regression analyses. The correlation of CRP with CA125 was analyzed by a Spearman rank correlation analysis. P<0.05 was considered to indicate a statistically significant difference.

According to the FIGO staging criteria, 29 cases were in stage I (27.10%), 14 cases in stage II (13.08%), 44 cases in stage III (41.12%) and 20 cases in stage IV (18.69%). The analysis of pathological type revealed that 43 cases were serous cystadenocarcinoma (40.19%); 38 cases were mucinous cystadenocarcinoma (35.51%) and 35 cases were other types of epithelial cancers (32.71%), including endometrial cancer, clear cell carcinoma, transitional cell carcinoma and epithelial cancer that could not be classified due to tissue necrosis. Ascites occurred in 42 cases (39.25%), and 65 cases (60.75%) were without ascites. For histological cell grading, 28 cases were in the G1 phase (26.17%), 44 cases in the G2 phase (41.12%) and 35 cases in the G3 phase (32.71%).

In the present study, 62% of the patients (66/107) were positive for CRP (CRP >8 mg/l) in the EOC group, while no positive results were found in the control group. The mean preoperative serum CRP concentration of the patients with EOC was 14.32 mg/l (1.02–163.00 mg/l), which was ∼6.6-fold that of the ovarian benign tumor group [2.175 mg/l (1.00–4.59 mg/l)] (P<0.001), as well as ∼9.7-fold the serum concentration of the healthy adult females [1.47 mg/l (1.00–3.16 mg/l)] (P<0.001). The serum concentration of CA125 in the EOC group was 475.90 U/l (16.26–5,000 U/l), whereas in the benign ovarian tumor group it was 23.85 U/l (7.74–61.31 U/l) and in the healthy adult females it was 8.77 U/l (1.35–18.33 U/l) (P<0.05). The CA125 serum concentration in the EOC group was ∼20-fold that of the benign ovarian tumor group and 54-fold that of the healthy adult females. A rank correlation analysis of preoperative serum CRP and CA125 concentrations in the EOC group revealed that the Spearman correlation coefficient (rs) was 0.603 (P<0.001, two-way), suggesting that CRP was positively correlated with CA125; however, no significant correlation between the CRP and CA125 concentrations existed in the control group (rs=-0.095, P=0.708).

It was next analyzed whether elevated preoperative serum CRP concentrations correlated with the clinicopathological parameters of the patients with EOC. As shown in Table I, the serum concentrations of CRP correlated with histological grading, FIGO staging, ascites, tumor size and lymph node metastasis, whereas only positive serum CRP did not correlate with histological grading. No significant difference was found in the serum CRP concentrations between different ages and different pathological tumor types.

Based on the results of single factor analyses, a positive serum CRP value was correlated with FIGO tumor staging, histological grading, presence of ascites and lymph node metastasis (Table II), while CA125 had no effect on CRP-positive rates. When the factors that correlated with CRP in the single factor analyses as independent variables were used in a multiple factor logistic regression analysis it was found that the FIGO tumor staging, presence of ascites and lymph node metastasis were correlated with CRP-positive rates (Table III).

The overall five-year survival rate of patients with serum CRP ≤8 mg/l was 75.8%, compared with a survival rate of 33.3% for patients with CRP concentrations >8 mg/l (P<0.001) (Fig. 1).