Does the experience of the first assistant affect organ injuries in laparoscopic hysterectomy for benign diseases?

It is recommended that inexperienced first assistants participate in TLH for training and experience while considering the difficulty of the surgical cases.

We conducted a retrospective cohort study of 1,682 women who underwent TLH for benign indications between January 1, 2014 and December 31, 2020 at Teine Keijinkai Hospital in Sapporo, Japan. All TLH procedures were performed by fully qualified and experienced primary surgeons. All patients were admitted to the hospital one day before surgery, and they received standard prophylactic antibiotics. A blood test was performed on postoperative day (POD) 2 according to our department’s clinical protocol (Table S1). Data on 12 variables were collected (Table S2), and they included demographic data, comorbidities, preoperative use of gonadotropin-releasing hormone (GnRH) agonists, surgical data, surgical techniques in addition to TLH, and diagnosis. We defined an experienced surgeon, including both the operator and the first assistant, as a surgeon certified by the Skill Qualification Committee of the Japan Society of Gynecologic and Obstetric Endoscopy and Minimally Invasive Therapy or a surgeon with equivalent surgical skills. A surgeon with equivalent surgical skills to those of the certified endoscopic surgeon was defined as a surgeon familiar with TLH who had performed at least 75 TLH surgeries in the past, as mentioned above [ 4 , 12 , 13 ]. To acquire this qualification, applicants are required to perform over 100 gynecologic laparoscopic surgeries over a 2-year period after qualification as a specialist recognized by the Japan Society of Obstetrics and Gynecology; they are also required to pass an examination involving a video of the surgery. We believed that this qualification could only be obtained by a surgeon whose skills had improved after appropriate training and that a technically certified surgeon could be considered an experienced surgeon. Because our hospital is a teaching hospital, many young physicians leave within 6 months to a few years. Therefore, by the aforementioned definition, none of the inexperienced surgeons trained at our hospital became experienced surgeons. To identify patients with organ injuries (ureteral, bladder, and intestinal injuries) related to TLH, medical records were evaluated within 90 days of TLH. Evidence of perioperative organ injuries was determined by review of the operative report, imaging, or a urology and surgery consultation note. Data were reported as medians (interquartile ranges [IQRs]) for continuous variables. Categorical variables were depicted as absolute numbers and proportions and reported as numbers (%). We divided patients into two groups according to whether an experienced first assistant was present during surgery. To identify the risk factors associated with organ injuries during the 90 days after TLH, univariate analyses were performed. To compare each variable between the two groups, the chi-squared test or Fisher’s exact test was used for categorical variables, and the Mann–Whitney U test was used for continuous variables because all continuous variables follow nonparametric distributions. Propensity score (PS)-based inverse probability of treatment weighting (IPTW) was fitted to confirm the background differences in the two groups in assessing whether the presence or absence of an experienced first assistant affected the morbidity rate of organ injuries. First, the PS for the group with an experienced first assistant was determined by fitting a binary logistic regression model. Patient demographics, comorbidities, preoperative use of a GnRH agonist, surgical data, surgical techniques in addition to TLH, and diagnoses were entered into the final model. Then, an IPTW approach was used to assign the group with an experienced first assistant a weight of 1/PS, and the group without an experienced first assistant, a weight of 1/(1 − PS). The cases in which the PS values were greater than 0.1 were included in the analysis. In the PS-IPTW model, proportional distribution of baseline covariates was assessed by the standardized mean difference (SMD). The two groups were considered to be well balanced when the SMD was less than 0.1. We tested for the differences in organ injury rate between the two groups using Poisson regression. Two-sided p values < 0.05 were considered to show statistical significance. We used the statistical software R 3.6.1 for the statistical analyses [ 20 ]. This study was approved by the research ethics committee of Teine Keijinkai Hospital (approval no.: 2-020337-00). Informed consent was obtained in the form of opt-out on the website of Teine Keijinkai Hospital.

Of the 1,682 patients, 14 (0.83%) developed organ injuries (ureteral, bladder, and intestinal injuries) during the 90 days after TLH. Five (35.7%) had ureteral injuries, two (14.3%) had bladder injuries, six (42.9%) had intestinal injuries, and one (7.1%) had both bladder and intestinal injuries (Table 1 ). The result of univariate analysis and baseline characteristics of patients before and after inverse probability weighting in this study are presented in Table 2 . In the univariate analysis, in the group with an experienced first assistant, the rates of diabetes mellitus ( p value ≤ 0.001) and prior pelvic surgery ( p value = 0.038) were significantly lower, whereas the rate of preoperative use of a GnRH agonist was significantly higher ( p value ≤ 0.001) than that in the group without an experienced first assistant. Table 1 List of cases according to characteristics of organ injuries, time to recognize injuries, and subsequent injury management Case No Type of organ injury Date of diagnosis Surgical technique With an experienced first assistant Management 1 Ureteral During surgery TLH 2  + LBS 3 Yes Intraoperative abdominal ureterocystoneostomy 2 Ureteral During surgery TLH + LBS Yes Intraoperative abdominal ureteroureterostomy 3 Ureteral During surgery TLH + LLSO 4  + LRC 5  + LRS 6 No Intraoperative abdominal ureteroureterostomy 4 Ureteral POD 1 6 TLH + LLSO + LRS Yes Abdominal ureterocystoneostomy on POD 14 5 Ureteral POD 11 TLH No Ureteral stent placement 6 Bladder During surgery TLH + LBS Yes Abdominal intraoperative cystorrhaphy 7 Bladder During surgery TLH + LRSO 7  + LLS 8 Yes Intraoperative abdominal cystorrhaphy 8 Sigmoid colon During surgery TLH + LBS Yes Intraoperative laparoscopic sutures in the sigmoid colon 9 Sigmoid colon POD 7 TLH + LBSO 9 Yes Laparoscopic high anterior resection and colostomy on POD 7 10 Rectum During surgery TLH + LLSO + LRS Yes Intraoperative laparoscopic sutures in the rectum 11 Rectum During surgery TLH + LLC 10  + LBS Yes Intraoperative laparoscopic sutures in the rectum 12 Rectum During surgery TLH + LLSO + LRS No Intraoperative laparoscopic sutures in the rectum 13 Rectum POD 3 TLH + LBS Yes Laparoscopic high anterior resection and colostomy on POD 3 14 Bladder and rectum During surgery TLH + anterior and posterior colporrhaphy Yes Intraoperative vaginal sutures in the rectum and cystorrhaphy POD postoperative day, TLH total laparoscopic hysterectomy, LBS laparoscopic bilateral salpingectomy, LLSO laparoscopic left salpingo-oophorectomy, LRC laparoscopic right ovarian cystectomy, LRS laparoscopic right salpingectomy, LRSO laparoscopic right salpingo-oophorectomy, LLS laparoscopic left salpingectomy, LBSO laparoscopic bilateral salpingo-oophorectomy, LLC laparoscopic left ovarian cystectomy Table 2 The result of univariate analysis and baseline characteristics of patients before and after inverse probability weighting Valuables n (%) Before inverse probability weighting After inverse probability weighting With an experienced first assistant With an experienced first assistant Yes ( N  = 1119) No ( N  = 563) p value SMD 2 Yes ( N  = 1119) No ( N  = 563) SMD Age (y) 46.0 (43.0–49.0) 46.0 (43.0–49.0) 0.182 0.059 45.0 (42.1–50.0) 45.9 (40.2–53.0) 0.007 BMI 1 (kg/m 2 ) 22.6 (20.6–25.7) 22.4 (20.3–25.5) 0.605 0.011 22.5 (20.1–25.4) 22.5 (18.8–26.9)  < 0.001 Parous women 743 (66.4%) 367 (65.2%) 0.624 0.026 739 (66.8%) 369 (65.7%) 0.004 Comorbidity  Hypertension 75 (6.7%) 42 (7.5%) 0.612 0.030 78 (7.0%) 38 (6.8%) 0.005  Diabetes mellitus 16 (1.4%) 25 (4.4%)  < 0.001 0.179 29 (2.6%) 14 (2.5%) 0.007 Prior pelvic surgery 404 (36.1%) 233 (41.4%) 0.038 0.109 426 (38.0%) 214 (38.1%) 0.002 Prior cesarean section 156 (13.9%) 97 (17.2%) 0.083 0.091 171 (15.3%) 86 (15.3%) 0.002 Preoperative use of a GnRH 3 agonist 596 (53.3%) 248 (44.0%)  < 0.001 0.185 561 (50.1%) 282 (50.2%) 0.002 Emergent surgery 1 (0.1%) 2 (0.4%) 0.261 0.056 2 (0.2%) 1 (0.2%) 0.004 Intraoperative adhesion 102 (9.1%) 44 (8.2%) 0.409 0.047 97 (8.7%) 48 (8.5%) 0.001 Specimen weight, g 290.0 (176.5–462.4) 294.0 (188.0–463.5) 0.324 0.015 285.0 (176.8–464.1) 290.1 (184.1–472.5) 0.002 Additional operative technique  Salpingo-oophorectomy 186 (16.6%) 97 (17.2%) 0.782 0.016 189 (16.9%) 95 (16.9%) 0.002 Diagnosis  Pelvic organ prolapse 9 (0.8%) 4 (0.7%) 1.00 0.011 8 (0.7%) 3 (0.5%) 0.018  Endometriosis 249 (22.3%) 119 (21.1%) 0.618 0.027 245 (21.9%) 123 (21.9%) 0.001 Operation time 131.0 (102.0–168.0) 122.0 (98.0–148.5)  < 0.001 NA NA NA NA Values are presented as n (%) or median (interquartile range). According to the univariate analysis, in the group that underwent surgery with an experienced first assistant, the rates of diabetes mellitus ( p value ≤ 0.001) and prior pelvic surgery ( p value = 0.038) were significantly lower, and the rate of preoperative use of a GnRH agonist was significantly higher ( p value ≤ 0.001) than in the group that underwent surgery without an experienced first assistant. The operating time was not included in the model by propensity score because it was not a pre-treatment variable. After inverse probability weighting, all 12 covariates were well balanced between the two groups (SMD less than 0.1) BMI body mass index, GnRH gonadotropin-releasing hormone, SMD standardized mean difference, NA not assessed List of cases according to characteristics of organ injuries, time to recognize injuries, and subsequent injury management TLH + LLSO 4  + LRC 5 + LRS 6 POD postoperative day, TLH total laparoscopic hysterectomy, LBS laparoscopic bilateral salpingectomy, LLSO laparoscopic left salpingo-oophorectomy, LRC laparoscopic right ovarian cystectomy, LRS laparoscopic right salpingectomy, LRSO laparoscopic right salpingo-oophorectomy, LLS laparoscopic left salpingectomy, LBSO laparoscopic bilateral salpingo-oophorectomy, LLC laparoscopic left ovarian cystectomy The result of univariate analysis and baseline characteristics of patients before and after inverse probability weighting 46.0 (43.0–49.0) 46.0 (43.0–49.0) 45.0 (42.1–50.0) 45.9 (40.2–53.0) 22.6 (20.6–25.7) 22.4 (20.3–25.5) 22.5 (20.1–25.4) 22.5 (18.8–26.9) Preoperative use of a GnRH 3 agonist 290.0 (176.5–462.4) 294.0 (188.0–463.5) 285.0 (176.8–464.1) 290.1 (184.1–472.5) 131.0 (102.0–168.0) 122.0 (98.0–148.5) Values are presented as n (%) or median (interquartile range). According to the univariate analysis, in the group that underwent surgery with an experienced first assistant, the rates of diabetes mellitus ( p value ≤ 0.001) and prior pelvic surgery ( p value = 0.038) were significantly lower, and the rate of preoperative use of a GnRH agonist was significantly higher ( p value ≤ 0.001) than in the group that underwent surgery without an experienced first assistant. The operating time was not included in the model by propensity score because it was not a pre-treatment variable. After inverse probability weighting, all 12 covariates were well balanced between the two groups (SMD less than 0.1) BMI body mass index, GnRH gonadotropin-releasing hormone, SMD standardized mean difference, NA not assessed After PS-IPTW, all 12 covariates were well balanced between the two groups (SMD less than 0.1), and no significant difference was observed in the incidence rate of organ injuries between the two groups ( p  = 0.348). IPTW estimator was 0.52 (95% Confidence interval [− 0.72, 1.75]) (Table 3 ). Table 3 Univariate and Propensity score-based inverse probability of treatment weighting analyses of organ injury rate in patients who underwent laparoscopic hysterectomy for benign diseases with or without an experienced first assistant Outcome With an experienced first assistant Without an experienced first assistant IPTW 1 estimator p value Organ injury rate (%)  Unadjusted 0.98% 0.53% NA 2 0.408  PS 3 adjusted 0.94% 0.57% 0.52 95% CI 4 [− 0.72, 1.75] 0.348 The involvement of an inexperienced first assistant did not affect the morbidity rate of organ injuries ( p  = 0.348) IPTW inverse probability of treatment weighting, NA not assessed, PS propensity score, CI confidence interval Univariate and Propensity score-based inverse probability of treatment weighting analyses of organ injury rate in patients who underwent laparoscopic hysterectomy for benign diseases with or without an experienced first assistant 0.52 95% CI 4 [− 0.72, 1.75] The involvement of an inexperienced first assistant did not affect the morbidity rate of organ injuries ( p  = 0.348) IPTW inverse probability of treatment weighting, NA not assessed, PS propensity score, CI confidence interval

In a population of women who underwent TLH performed by experienced primary surgeons for benign diseases at Teine Keijinkai Hospital, the involvement of an inexperienced first assistant did not affect the morbidity rate of organ injuries. Considering the experience level of the first assistant and the difficulty of surgical cases, the participation of inexperienced first assistants in TLH should, therefore, be encouraged for training and exposure.

In this study, we evaluated the impact of the first assistant’s experience level on the occurrence of perioperative organ injuries. As a result, the involvement of an inexperienced first assistant in TLH for benign diseases was not found to affect the morbidity rate of organ injuries. Regarding different surgical procedures, several studies determined the complications associated with the experience level of the first assistant. In surgeries for benign diseases, Tarik et al. reported that the training level of the first assistant in bariatric procedures did not negatively affect the occurrence of serious perioperative complications [ 21 ]. In surgeries for malignant diseases, comparable results have been reported in laparoscopic sigmoidectomies and laparoscopic hepatectomies [ 14 , 22 ]. During TLH, the first assistant has several important roles. The skills of the first assistant are an essential part of an effective surgical team because first assistants serve as an integral bridge between primary surgeons and patients [ 23 ]. Above all, the surgical camera plays a vital role in the success of any laparoscopic procedure. Techniques that ensure optimal visualization are imperative for first assistants [ 24 ]. Moreover, they must identify the dissection plane, grasp tissue, add adequate counter traction, and provide the appropriate surgical field view for primary surgeons [ 25 ]. First assistants also provide suitable guidance and second opinions intraoperatively if needed [ 26 , 27 ]. A primary surgeon may experience stress when working with an inexperienced assistant [ 14 ]. Cai et al. described potential reasons why surgery with an inexperienced assistant does not affect the morbidity of complications [ 22 ]. First, the lack of experience of the first assistant surgeon could be offset by the rest of the surgical team. If the primary surgeon knows that the first assistant surgeon is less experienced, he or she may consciously verbalize all instructions more clearly to enable better support by the first assistant surgeon. The primary surgeon may also involve more assistants in the surgery for a more even division of tasks. Second, appropriate selection of the surgical assistant based on his or her experience level may occur in consideration of case difficulty and the primary surgeon’s experience. In our study, the group of experienced first assistants had a slightly longer operative time than the group that did not. It may be that experienced first assistants were selected for more difficult cases. Furthermore, Igor et al. stated that primary surgeons may alter their technique based on the experience level of assistant surgeons to ensure surgical safety [ 27 ]. The median BMI of the operated women was approximately 22. Compared to patients with normal BMI, the high BMI patients who underwent TLH reported no particular increase in complications [ 28 ], which would be unlikely to account for the relatively low rate of organ damage in this study. One strength of this study is that it included detailed perioperative information from our surgical database. Moreover, all patients received standardized and homogeneous management at our hospital. On the contrary, this study also has several limitations, and the results should be interpreted with caution. First, this study has inherent bias associated with operative factors that may not be referenced from the medical records (e.g., the degree of surgical difficulty and how to handle the organization by the surgeon). Second, our study reflected the experience of a single hospital, and the findings may not be generalizable to other institutions and settings. Lastly, TLH has low rates of perioperative complications and is considered relatively safe. Therefore, it may be difficult to detect significant differences in the rates of perioperative complications between patients in this cohort.

Throughout the last three decades, laparoscopic surgery has replaced laparotomy for gynecological surgery [ 1 – 3 ]. Hysterectomy is one of the most common gynecologic surgical procedures worldwide, especially for benign conditions [ 4 , 5 ]. Compared with open surgery, laparoscopic surgery for hysterectomy is regarded as a minimally invasive approach on the basis of decreased blood loss, length of hospital stay, postoperative pain, risk of infection, and shorter overall recovery time [ 6 – 8 ]. Because of the obvious advantages of total laparoscopic hysterectomy (TLH), it is imperative that gynecologists receive adequate exposure to this technique to gain experience [ 9 ]. Compared with abdominal and vaginal hysterectomies, TLH requires greater surgical expertise [ 7 ], a longer time to master, and a longer operating time [ 7 , 10 ]. As surgeons perform more laparoscopic procedures, it is prudent to consider how many operations an individual surgeon must perform to be considered competent [ 11 ]. The number of cases needed to achieve an experienced level of performance of TLH has been reported to range from 21 to 75 [ 4 , 12 , 13 ]. However, primary surgeons cannot successfully perform laparoscopic surgery alone and inevitably require a surgical team that includes assistant surgeons [ 14 ]. At our institution, TLH is aided by two assistants. The first assistant holds the laparoscope, while the second assistant moves the uterine manipulator. In a teaching hospital such as ours, trainees including novices in laparoscopic surgery and residents are continuously rotated for surgical training purposes. The experience of the first assistant is essential for trainees of TLH so that they can acquire the skills of a primary surgeon. Assistants understand the details of the operative procedure, learn effective methods to support the operator, and ensure that their experience is useful for their own future performance of the procedure. Though TLH has many advantages mentioned above, when serious complications, especially injuries in organs adjacent to the uterus (ureteral, bladder, and intestinal injuries), occur during the surgery, conversion to laparotomy is frequently needed [ 15 ], which increases the secondary burden and stress on patients [ 16 ]. To find a balance between patient safety and trainee education, the effects on the occurrence of severe complications must be scrutinized. Previous investigations have highlighted the relationship between the experience of primary surgeons and perioperative complications in TLH [ 9 , 17 – 19 ]. However, to our knowledge, little is known about the impact of the first assistant surgeon’s experience in TLH on the development of complications. Therefore, in the current study, we aim to identify the influence of the first assistant surgeon’s experience on the occurrence of perioperative organ injuries (ureteral, bladder, and intestinal injuries) in TLH for benign diseases.

Below is the link to the electronic supplementary material. Supplementary file1 (PDF 83 KB) Supplementary file2 (TXT 328 KB) Supplementary file1 (PDF 83 KB) Supplementary file2 (TXT 328 KB)