Risk Factors for Pelvic Lymph Node Metastasis and Recurrence in Patients Undergoing Radical Hysterectomy for Cervical Cancer

Background: Radical hysterectomy and bilateral pelvic lymphadenectomy are standard treatments for early-stage cervical cancer. Pelvic lymph node metastasis (PLNM) is one of the critical factors affecting the postoperative prognosis of patients. Therefore, the identication of preoperative risk factors for PLNM will minimize its occurrence and improve prognosis. The purpose of this study was to investigate the risk factors for PLNM and its recurrence in patients undergoing radical hysterectomy for cervical cancer. Methods: Medical records of 245 patients who underwent radical hysterectomy and bilateral pelvic lymphadenectomy as primary treatment for the International Federation of Gynaecology and Obstetrics (FIGO) stage IA-IIA cervical cancer between January 2010 and December 2015 at our hospital were reviewed. Age, FIGO stage, preoperative hemoglobin level, depth of stromal invasion, lymphovascular space invasion (LVSI), human papillomavirus(HPV) infection, parametrial inltration, tumor diameter, number of lymphadenectomies, and pathological type were retrospectively analyzed. All patients were followed up for 5–10 years. Results: Among the 245 patients, 185 (75.51%) had no PLNM, whereas 60 (24.49%) had PLNM. Preoperative hemoglobin level, FIGO stage, LVSI, parametrial inltration, and tumor diameter differed signicantly between the two groups (P<0.05). Multivariate analysis revealed preoperative hemoglobin <110 g/L, FIGO stage II, LVSI, parametrial inltration, and tumor diameter ≥ 4 cm as signicant risk factors for PLNM and recurrence of cervical cancer after surgery (P<0.05). PLNM was identied as the independent risk factor for recurrence in patients with cervical cancer after surgery (P<0.05). Conclusions: PLNM is an important prognostic indicator for the clinical treatment of cervical cancer. Patients

The purpose of this retrospective study was to identify the risk factors for PLNM and recurrence, then provide insights into the treatment and prognosis of cervical cancer.

Clinical data
The clinical data of 245 patients with cervical cancer who underwent radical hysterectomy and bilateral pelvic lymphadenectomy at Tianjin Central Hospital of Gynecology and Obstetrics between January 2010 and December 2015 were retrospectively analyzed using the hospital medical records. The inclusion criteria in our study were as follows:1) con rmed diagnosis of cervical cancer by histopathological examination; 2) International Federation of Gynaecology and Obstetrics (FIGO) stage: IA-IIA; and 3) radical hysterectomy and bilateral pelvic lymphadenectomy. The exclusion criteria were as follows: 1) incomplete clinical data; 2) preoperative metastatic cervical cancer; 3) cervical lymphoma, cervical melanoma, and other cervical nonepithelial tumors; and 4) case complicated by malignant tumors in other organ systems.
All patients were followed up for 5-10 years, and based on follow-up data, they were divided into two groups: 1) Recurrence group: patients with recurrence (39, 15.92%) and 2) no recurrence group: patients without recurrence (206, 84.08%).
Patient anonymity was preserved as the data were collected from the hospital's electronic medical records. The research ethics committee of Tianjin Central Hospital of Gynecology and Obstetrics waived the requirement for ethics approval and informed consent because the study used previously stored data.
Statistical analysis SPSS version 21.0 (SPSS Inc, Chicago, IL, USA) was used for statistical analysis. The two-by-two or fourfold contingency table (chi-square) test employing exact probabilities was used. Multivariate analysis was performed using a logistic regression model. All tests were two-sided, and the level of signi cance was set at P < 0.05.

Logistic regression analysis of risk factors for PLNM
Logistic regression analysis was used to identify independent predictors for PLMN, and preoperative hemoglobin < 110 g/L, FIGO stage II, LVSI, deep stromal invasion, parametrial in ltration, and tumor diameter ≥ 4 cm were found to be independent risk factors for postoperative PLNM of cervical cancer ( P < 0.05) ( Table 2).  (Table 3).

Discussion
Radical hysterectomy and bilateral pelvic lymphadenectomy are still the primary clinical treatment methods for patients with early-stage cervical cancer. The 5-year survival rate after surgery is reported to be relatively high in patients with FIGO stage IA-IIA [7].. While for some patients, a radical cure is possible, there is a possibility of relapse within 18-24 months after the initial treatment [7]. PLNM is the main metastatic route of cancer cell proliferation and an essential determinant of prognosis [7,8]. The incidence of PLNM in cervical cancer patients after surgery is reported to be about 30% [9]. In this study, the incidence of PLNM was similar at 24%.
The hemoglobin level reduces commonly in the perioperative period [10]. Moreover, anemia occurs in more than one-third of cancer patients, and severe anemia is a risk factor for death in such patients [11][12][13]. The level of hemoglobin, the primary oxygen carrier, directly affects the oxygen supply and oxygen content of the tumor. Preoperative blood transfusion and other strategies do not improve prognosis in cervical cancer patients, and in patients complicated with anemia, the tumor is highly aggressive, further deteriorating the prognosis [14,15]. In this study, the number of patients with preoperative hemoglobin < 110 g/L was signi cantly higher in the PLNM(+) group than in PLNM(-) group(P < 0.02).And the number of patients with preoperative hemoglobin < 110 g/L was signi cantly higher in the recurrence groups than in the no-recurrence groups(P < 0.03). Moreover, preoperative hemoglobin < 110 g/L was identi ed as an independent risk factor for postoperative PLNM and recurrence of cervical cancer after surgery. This nding was consistent with the results of previous studies.
Tumor staging is a de ning index of tumor growth and the extent of its spread. As the tumor stage increases, the depth and extent of tumor invasion to the surrounding tissue, the aggressiveness and malignancy of the tumor, and the recurrence rate increase signi cantly [3,16]. LNM rates of Ia, Ib, IIa, and IIb stages are reported to be 10.5%,13.1%, 27.1%, and 50.0%, respectively [17], con rming that LNM increases with advanced FIGO stages. In this study, the number of FIGO stage II cervical cancer patients with PLNM was higher than the number of FIGO stage I cervical cancer patients with PLNM (P = 0.005). This nding is consistent with previously reported results.
LVSI, deep stromal invasion, parametrial in ltration, and tumor diameter are closely related to PLNM and the recurrence of early cervical cancer [18,19]. LVSI is pathologically con rmed by the presence of malignant tumor cells between two layers of vascular endothelial tissue and is an important prognostic index of cervical cancer. Vascular in ltration is an independent risk factor for PLNM [20,21]. When cancer cells invade the lymphatic space, they can promote the formation of tumor thrombosis and invade local lymph nodes through the lymphatic vessels, thus inducing parametrial in ltration and PLMN [22,23]. Consistent with previous studies, in our study, the proportions of patients with LVSI (81.67%) and parametrial in ltration (75%) in the PLNM(+) group were signi cantly higher than those of patients in the PLNM(-) group (P < 0.005).
The tumor diameter can re ect the tumor growth time as tumor growth is a continuous invasion and proliferation process. The longer the growth time, the more likely is the lymph node metastasis [6,7]. With an increase in tumor diameter and a prolonged growth period, the depth of stromal invasion tends to increase. The contact area between tumor tissue and lymphatic vessels and the risk of LNM also tend to signi cantly increase [18, 24,25]. A study on the prognosis of 93 patients with early cervical cancer after surgery found that a tumor diameter ≥ 4 cm is a risk factor for PLNM and recurrence of cervical cancer [26]. In this study, the number of patients with tumor diameter ≥ 4 cm and deep stromal invasion differed signi cantly between the two groups (P < 0.005 and P < 0.05), consistent with previous studies.
Besides preoperative hemoglobin < 110 g/L, LVSI, deep stromal invasion, parametrial in ltration, and tumor diameter ≥ 4 cm, it was shown that PLNM is an independent risk factor for the recurrence of cervical cancer. LNM occurs in late-stage cancer. Postoperative invasion, metastasis, and recurrence are prone to occur in patients with PLNM. Thus, the postoperative survival rate tends to decrease [27,28]. Pelvic lymph node dissection can effectively remove metastatic lymph nodes, reduce the tumor load, prevent LNM, and reduce the risk of distant recurrence [29,30]. For patients with positive pelvic lymph nodes, the interval between recurrence is signi cantly shorter than for those with negative lymph nodes, and the risk of recurrence is relatively higher [6]. In this study, the proportion of patients with PLNM in the recurrence group was (79.47%) signi cantly higher than that in the non-recurrence group (P = 0.000). PLNM was therefore identi ed as an independent risk factor for recurrence in patients with cervical cancer after radical hysterectomy and bilateral pelvic lymphadenectomy.
The main strength of this study was the inclusion of patients with PLNM. The prognosis of this group of patients is poor, and adequate preoperative evaluation and postoperative follow-up are needed to improve the prognosis. Moreover, patients with recurrence after surgery were from the same group of patients with PLNM, thereby reducing bias and achieving more accurate results.
This study also has unavoidable limitations due to its retrospective design. First, we could not assess all variables potentially associated with residual lesions in this single-center study. Furthermore, because the study population was from one hospital, the external validity of our results may be low. Further prospective studies with a larger sample size and a broader context are needed.

Conclusions
In this study, we found that hemoglobin < 110 g/L, FIGO stage II, LVSI, deep stromal invasion, parametrial in ltration, and tumor diameter ≥ 4 cm are independent risk factors for postoperative PLNM and cervical cancer recurrence. Furthermore, PLNM is an independent risk factor for the postoperative recurrence of cervical cancer. LNM is an important prognostic indicator for the clinical treatment of cervical cancer. A comprehensive preoperative evaluation is strongly recommended to improve the curative effect and prognosis of cervical cancer and avoid PLNM. For patients with risk factors for PLNM, careful and systematic pelvic lymphadenectomy should be performed. Patients at a high risk of recurrence, especially those with PLNM, must be strictly followed up, and targeted treatment should be provided after surgery.
Further research is needed to determine whether pelvic lymphadenectomy should be attempted only in high-risk individuals. Tianjin Central Hospital of Gynecology and Obstetrics because the committee did not consider approval was necessary for a retrospective chart review. The data were collected through the institution's electronic medical records while pserving patient anonymity. The research ethics committee waived the requirement for informed consent because the study used pviously stored data. Administrative permissions were not required to access and use the medical records described in our study. Study ow chart