The Surgeon’s Proficiency Affected Survival Outcomes of Minimally Invasive Surgery for Early-Stage Cervical Cancer: A Retrospective Study of 851 Patients

Purpose To compare the clinical outcomes of patients with early-stage cervical cancer who underwent minimally invasive surgery (MIS) by surgeons in different phases and evaluate whether the proficiency of surgeons affects the survival outcomes. Materials and Methods A total of 851 patients with early-stage cervical cancer who underwent radical hysterectomy between January 2008 and June 2018 (every year from January to June) at a tertiary hospital were retrospectively analyzed. We categorized patients into four phases according to their sequence (phase one, 1-10 cases; phase two: 11-20 cases; phase three: 21-30 cases; phase four: > 30 cases). Demographics and clinical and pathological data were collected and analyzed. Results There were no statistical differences between the open surgery and MIS groups regarding three- and five-year overall survival (OS) and disease-free survival (DFS). The OS and DFS of patients in the MIS group in phase one were significantly lower than those in later phases and those in the open surgery group after adjustment (OS, P = 0.009; HR, 2.896; 95%CI, 1.303-6.435; DFS, P = 0.009; HR, 2.712; 95%CI, 1.289-5.706). Survival outcomes were not statistically significant when comparing different surgeons. Conclusion The phase one cases of MIS had lower OS and DFS than those in later phases and those in the open surgery group. Thus, we suggest that the proficiency of surgeons is associated with survival outcomes of MIS. Favorable outcomes can be obtained after a certain number of MIS cases.


INTRODUCTION
According to statistics from the International Agency for Research on Cancer, the incidence of cervical cancer was 604,000 in 2020, making it the fourth most common gynecological cancer worldwide (1). Radical hysterectomy (RH) with lymphadenectomy remains one of the preferred treatments for patients with cervical cancer diagnosed at the International Federation of Gynecology and Obstetrics (FIGO) stage IA-IIA (2). In recent decades, minimally invasive surgery (MIS) has become more common, displacing the use of traditional open surgery for early-stage cancer. However, the Laparoscopic Approach to Cervical Cancer (LACC) trial observed that disease-free survival (DFS) at 4.5 years and overall survival (OS) at three years of MIS were significantly lower than those of open surgery among women with stage IA1-IB1 cervical cancer, which remains controversial (3). Potential explanations include increasing tumor spillage due to the application of a uterine manipulator, the effect of CO 2 insufflation, the volume of surgery, and the surgeon's proficiency (4)(5)(6)(7). Several studies have focused on the correlation between surgeon proficiency and clinical outcomes in recent years. Lan Ying Li et al. found that compared to open surgery, more cases were required for surgeons performing minimally invasive RH to reach an acceptable five-year DFS (8). Kim et al. demonstrated that surgeons' proficiency in the MIS group significantly affected progression-free survival (PFS) (9). Liu et al. concluded that the learning curve could be a probable reason for poor outcomes of MIS by analyzing stage IB cervical cancer patients treated with RH by one surgeon for 15 years (10). Nevertheless, existing studies have limitations such as low sample sizes, incomplete follow-up information, and the absence of clinicopathological features of patients. There is still insufficient evidence regarding whether a surgeon's proficiency is associated with survival outcomes in MIS. We aimed to explore the effects of surgeons' proficiency in MIS on short-and longterm clinical outcomes and whether it accounts for the clinical outcomes of MIS.

Study Design and Participants
This study was approved by the Institutional Ethics Committee of West China Second University Hospital, and all participants provided their written informed consent to participate in this study. Cervical cancer patients with consecutive FIGO (2009) stage IA-IIA treated with RH between January 2008 and June 2018 (every year from January to June) were retrospectively analyzed. The inclusion criteria were: (1) patients with FIGO stage IA1 with lymphovascular space invasion (LVSI), IA2, IB, and selected IIA cervical cancer; (2) patients who underwent standard surgical treatment, which was performed by five specific surgeons according to the National Comprehensive Cancer Network guidelines, a modified RH (Type B of the Querleu and Morrow (Q-M) surgical classification system) with pelvic lymphadenectomy (PLND) in stage IA1 with LVSI and stage IA2, and an RH (Type C of the Q-M surgical classification system) with PLND with/without para-aortic lymphadenectomy in stage IB to IIA (2,11,12). (3) Patients with histological subtypes of squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma, regardless of histological grading. The exclusion criteria were as follows: (1) patients with incomplete data or complete loss to follow-up; (2) patients with severe fundamental diseases such as immune deficiency or other malignant tumors; and (3) patients with cervical cancer in pregnancy.
Complete information, including demographics and clinical and pathological information, was extracted from the hospital information system by two investigators (Y.Y. and Y.H.). The demographics extracted included age, menstruation (menopause or not), and body mass index (BMI). The clinical information extracted included diagnosis, FIGO (2009) stage, surgical approach, surgeon name, date of surgery, hospital stay, duration of surgery, estimated blood loss, CO 2 pneumoperitoneum in MIS, number of resected lymph nodes, pre-and post-operative treatment, and imaging data. Pathological information included histologic subtype, grading, LVSI, stromal invasion depth, parametrial involvement, vaginal margin involvement, and lymph node metastasis. Recurrence was confirmed by clinical findings, radiological examinations, and pathology reports. OS and DFS were the primary outcomes of this study. OS was defined as the time interval between the date of surgery and the date of death. DFS was defined as the time interval between the date of surgery and either the date of the first recurrence or death.

Statistical Analysis
The data were analyzed using the Statistical Package for the Social Sciences software version 22.0 (IBM Corp, Armonk, NY, USA). Missing values were statistically imputed by multiple imputations using logistic regression and predictive mean matching (13). The enumeration data were analyzed using the chi-square test. The measurement data were analyzed via t-test and Mann-Whitney U test between two groups, while an analysis of variance and Kruskal-Wallis H test were used to compare multiple groups. Survival curves were generated using the Kaplan-Meier method and analyzed using the log-rank test and multivariable Cox proportional hazards regression models. All the P-values reported are two-sided. A P-value of < 0.05 was considered statistically significant.

Clinical Outcomes in Different Surgical Approaches
The median follow-up duration was 88. 6 Figure 1]. In addition, we analyzed the survival outcomes of the open surgery and MIS groups for every surgeon and found that the survival outcomes were still not statistically significant (all p-values > 0.05).

Survival Outcomes in Different Phases and Surgeons
For every surgeon, we categorized patients who underwent surgery into four phases according to their sequence (phase one, 1-10 cases; phase two, 11-20 cases; phase three, 21-30 cases; phase four, more than 30 cases). Considering open surgery as an advanced technique, we did not categorize patients in this group into different phases. When stratified by surgical phases, the OS and DFS of the MIS group in phase one were significantly lower than those in later phases and in the open surgery group after adjusting for age, BMI, FIGO stage, histologic subtype, and grading. The three-year OS was 91.8% (45/49), and the fiveyear OS was 87.8% (43/49) in phase one of the MIS group (P = 0.009; HR, 2.896; 95%CI, 1.303-6.435). The three-year DFS was 91.8% (45/49) and the five-year DFS was 85.7% (42/49) in phase one of the MIS group (P = 0.009; HR, 2.712; 95%CI, 1.289-5.706; Figure 2). We obtained similar results when analyzing the surgeons separately ( Table 2). Statistical differences were observed in the median number of resected lymph nodes (p < 0.001). There were no significant differences among the four phases with respect to perioperative characteristics, including hospital stay, duration of surgery, volume of blood loss, and volume of CO 2 pneumoperitoneum after adjustment ( Table 3).
When comparing different surgeons, the survival outcomes were not statistically significant after adjustment. Furthermore, the subgroup analysis in the corresponding phases showed the same result (all P > 0.05; Figure 3).    (2), the standard approach for RH is the open surgery, and the oncologic risks of MIS should be informed carefully to patients given the findings of poorer survival outcomes of MIS recently. The development of MIS was affected by relevant research works. Therefore, it is essential to clear the reasonable application of MIS to achieve maximum benefits for patients. Pedestrian et al. showed that patients with tumor dimension less than two centimeters still suitable for MIS, but more studies are needed to refine on criteria (20).
Several potential reasons may explain the limitations of MIS, such as the use of a uterine manipulator, insufflation gas, intracorporeal colpotomy under CO 2 pneumoperitoneum, and the proficiency of the surgeon (7,(21)(22)(23). Gynecologic   (27). In this regard, our findings were in line with the results of previous studies that the first ten cases were significantly inferior to subsequent cases. Moreover, favorable survival outcomes observed after the first ten cases also demonstrated the effect of surgeon proficiency. There was no statistical significance when comparing the different surgeons, indicating that favorable outcomes can be obtained by practicing the MIS technique after a certain number of cases. Based on our findings, we offer the following suggestions. First, surgeons should undergo standardized training to improve their surgical skills (28). Second, surgeons performing MIS should have some standard qualification. For example, surgeons can only perform laparoscopic surgery after completing a certain number of virtual surgeries through surgery simulators. Third, beginners should be overseen by experienced surgeons, particularly for the first dozen or so cases.
The main strength of this study is the large sample size. In addition, compared with other studies, we analyzed more clinical parameters, making our results more robust and applicable to a larger population. However, our study has several limitations. Due to its retrospective nature, there could be bias in the patient selection. Furthermore, we did not evaluate the effect of different surgical techniques and devices in different phases. More studies on specific surgical approaches and techniques, including conventional multiport laparoscopic and laparo-endoscopic  single-site surgery with/without technological innovations, are needed. Although there was no robotic case in our study, some relevant studies showed that robotic and laparoscopic approaches were similar in perioperative and postoperative outcomes (29), future studies could consider this aspect.
In conclusion, our retrospective study demonstrated that phase one cases of MIS had lower OS and DFS than those in later phases or that underwent open surgery. Thus, we suggest that the proficiency of the operating surgeon is associated with the survival outcomes of MIS. More favorable outcomes can be obtained after a certain number of MIS cases have been performed.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Institutional Ethics Committee of West China Second University Hospital. The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
YY and ZYL contributed to data collection and study design. YY and YH contributed to data analysis. All authors contributed to the article and approved the submitted version.

FUNDING
This study was supported by grants from the Sichuan Youth Foundation of Science of Technology (Grant number: 2015JQ0026).