PURPOSE: To investigate whether the total number of removed lymph nodes (LNs) and the number of metastatic LNs would prove to be independent prognostic factors for survival in patients with cervical cancer (CC). METHODS: Data from patients with CC who underwent radical surgery between March 1980 and September 2009 were reviewed. A total of 526 patients were included in the statistical analysis. Full pathologic evaluation was performed. The total number of examined LNs and their histopathological status were analyzed for their prognostic effect on survival by means of multivariable Cox proportional hazard regression models. RESULTS: The median number (interquartile range) of total, pelvic, and para-aortic nodes removed was 37 (29-47), 34 (27-42), and 19 (14-24), respectively. Positive pelvic nodes were found in 102 of 526 (19%) patients. All 8 patients with para-aortic metastases had also pelvic node metastases. At multivariable analysis, vaginal involvement, type of lymphadenectomy and LN status all significantly negatively affected disease-free survival and overall survival, whereas the number of total LNs removed did not affect survival. CONCLUSIONS: LN metastasis and number of LN metastases confer an independent risk for worse survival in patients with CC. Pelvic lymphadenectomy is important for staging and regional disease control when LNs are involved. If a standardized complete lymphadenectomy is performed, the number of LNs is not a significant factor per se.
PURPOSE: To investigate whether the total number of removed lymph nodes (LNs) and the number of metastatic LNs would prove to be independent prognostic factors for survival in patients with cervical cancer (CC). METHODS: Data from patients with CC who underwent radical surgery between March 1980 and September 2009 were reviewed. A total of 526 patients were included in the statistical analysis. Full pathologic evaluation was performed. The total number of examined LNs and their histopathological status were analyzed for their prognostic effect on survival by means of multivariable Cox proportional hazard regression models. RESULTS: The median number (interquartile range) of total, pelvic, and para-aortic nodes removed was 37 (29-47), 34 (27-42), and 19 (14-24), respectively. Positive pelvic nodes were found in 102 of 526 (19%) patients. All 8 patients with para-aortic metastases had also pelvic node metastases. At multivariable analysis, vaginal involvement, type of lymphadenectomy and LN status all significantly negatively affected disease-free survival and overall survival, whereas the number of total LNs removed did not affect survival. CONCLUSIONS: LN metastasis and number of LN metastases confer an independent risk for worse survival in patients with CC. Pelvic lymphadenectomy is important for staging and regional disease control when LNs are involved. If a standardized complete lymphadenectomy is performed, the number of LNs is not a significant factor per se.