Two-Layer Compared With One-Layer Vaginal Cuff Closure at the Time of Total Laparoscopic Hysterectomy to Reduce Complications.

The University of Pittsburgh Institutional Review Board approved this study. We performed a retrospective cohort study of women undergoing 1-LVC or 2-LVC closure during TLH at the University of Pittsburgh Medical Center between 2011–2017. Cases were gathered using department surgical calendars of fellowship-trained minimally invasive gynecologic surgeons (FMIGS). Eleven surgeons participated in the study. 99.9% of the procedures were performed by high volume surgeons defined as >12 hysterectomies per year. Four surgeons, who on average performed >40 hysterectomies per year, accounted for 97.4% of all surgeries. All surgeons received their FMIGS training at the same institution and their surgical technique with the exception of the cuff closure was uniform. Colpotomy was performed with 60 Watts cutting current along a Pelosi uterine manipulator cup. Laparoscopic vaginal cuff closures commenced in a similar fashion in both groups. The vaginal cuff angles were sutured bilaterally, tied extracorporeally and then temporarily suspended through the lateral lower ports (Video 1). This was followed by either a 1-LVC or a 2-LVC closure using a 180-day absorptive 2–0 V-loc barbed suture ( Figure 1 ). 1-LVC closure consisted of continuous full thickness bites from anterior to posterior through the vaginal mucosa and muscularis from right to left and then tracking back towards the middle from the left angle by at least two suture throws. In contrast, a 2-LVC closure entailed first closing only the vaginal mucosa in a running continuous fashion from right to left followed by a second layer utilizing the same suture to close the vaginal muscularis from left to right. Every attempt was made to not pass the needle through the first layer when suturing the overlying tissue, but rather imbricated over it. Patients were instructed to abstain from intercourse for 8–12 weeks postoperatively. Surgeons either exclusively performed a 1-LVC or 2-LVC closure in all their patients and did not individualize selection of cuff closure based on patient risk factors at the time of surgery. Eligibility criteria included ages 18–100, benign primary indications for total laparoscopic hysterectomies including pelvic pain, endometriosis, leiomyomas, abnormal uterine bleeding, dysplasia, hyperplasia or cancer prophylaxis. Subjects undergoing hysterectomies for oncologic indications were excluded. Robotic procedures and vaginal cuff closures that were not performed laparoscopically were omitted. Electronic medical data was used to abstract intraoperative and 180-day postoperative complications after TLH. Operative notes, anesthesia records, emergency room (ER), as well as inpatient and outpatient clinic and phone call documentation were reviewed. Patient demographics, surgical indications and factors influencing vaginal cuff complications including age, postmenopausal status, body mass index, diabetes, tobacco use, sexual activity, and immunosuppressant medications were recorded. The primary outcome was a composite of total postoperative complications, defined as all 30-day postoperative medical and surgical complications as well as vaginal cuff complications up to 180 days. 30-day postoperative complications consisted of visceral injury, blood transfusions, venous thromboembolism, abdominal or skin hematomas, surgical site infections, and medical complications. 180-day vaginal cuff complications measured dehiscence (with or without bowel evisceration), mucosal separation, hematomas, cellulitis, abscess, granulation tissue formation and persistent bleeding defined as evaluation for vaginal bleeding on two or more occasions. Cuff cellulitis was defined as purulent drainage with cuff tenderness, erythema, or swelling. A CT scan was performed to rule in a pelvic abscess if warranted based on exam findings. Vaginal cuff dehiscence referred to the complete separation of the vaginal cuff incision with visible intraperitoneal content. Mucosal separation was defined as a disruption of the vaginal epithelium/ lamina propria but not the underlying vaginal muscularis thus still providing a barrier between the vagina and the peritoneal cavity. All cuff complications were verified by pelvic examination. Postoperative readmissions to the emergency room and hospital were also recorded. Varying rates of postoperative complications after laparoscopic vaginal cuff closure have been reported. 4 , 5 , 7 , 8 , 16 Rates as low as 7.3% for total postoperative and 4.7% for vaginal cuff complication, respectively, have recently been cited in a large randomized control trial. 7 Using these published rates, we estimated that 1049 women in each cohort were required to detect a 50% reduction in both parameters with 80% power and a 2-sided alpha of 0.05. However, given inherent limitations with retrospective cohort studies we expanded the data collection to a total of 6 years for added statistical robustness. Post hoc analysis of the final sample size confirmed that the groups yielded a 97% power to detect a difference between the group proportions of 0.016 with a significance level of 0.05. Both patient populations were analyzed through descriptive statistics. Continuous variables were presented as means with standard deviations and categorical variables were noted as frequencies in percentages. Student t-tests or Wilcoxon rank sum testing were used for continuous variables, and χ 2 or Fisher exact test were used for comparisons of proportions. Univariate analysis was performed on factors associated with postoperative complications and if near significance ( P <0.1) included in a multiple logistic regression model. SAS software version 9.4 (Cary, NC) was used for all analyses. Type I (alpha) error was set at 0.05. Values of P ≤.05 were considered statistically significant.

During the 6-year study period, 2973 women qualified for analysis (1-LVC, n=1760 (59.2%); 2-LVC, n=1213 (40.8%)). Indications for hysterectomy most commonly included one or more of the following diagnoses: abnormal uterine bleeding (31.2%), leiomyomas (28.9%), and pelvic pain with or without endometriosis (44.2%). Another 15.6% of hysterectomies were performed for precancerous conditions, cancer prophylaxis or other conditions. Demographic characteristics varied by BMI, obesity and diabetes status with the 2-LVC group exhibiting lower values ( Table 1 ). Women in this group were also 3.6% less likely to smoke ( P <.01). There were no other statistical differences between the two cohorts. Postoperative complications were lower in the 2-LVC group (3.5% vs. 5.7%; P <.01) ( Table 2 ). The primary difference stemmed from lower vaginal cuff complications within 180 days (0.9% vs. 2.6%; P <.01); no differences in 30-day postoperative medical and surgical complications were observed between the two groups (2.6% vs. 3.1%; P = 0.77. Likewise, postoperative emergency room visits and hospital readmissions (11.1% vs. 12.7%; P =.18) did not differ. There were no dehiscences or mucosal separations noted in the 2-LVC ( Figure 2 ). In contrast, 17 cases were observed in the 1-LVC group (0.0% vs. 1.0%; P <.01). Dehiscence rates in the 1-LVC closure group occurred in equal proportions among surgeons utilizing this closure technique (data not shown). 50% of mucosal separations and all 13 dehiscences in the 1-LVC group required repair in the operating room. All women who experienced a dehiscence or mucosal separation in the 1-LVC group reported sexually activity at the time of their pre-operative consultation. All dehiscences occurred after patients’ first postoperative resumption of insertional sexual activity. 64.7% of patients were of normal weight and nearly half were current smokers ( Appendix 1 , available online at http://links.lww.com/AOG/C329 ). Infections did not play a significant role in our cohort. Granulation tissue along the vaginal cuff was primarily managed in the office with silver nitrate application (1-LVC: 7 out of 8 patients; 2-LVC: 2 out of 3 patients). Cases of persistent bleeding generally did not require treatment and self-resolved (1-LVC: 9 out of 12 cases; 2-LVC: 1 out of 2 cases). Although 56.1% (n=32/57) of cuff complications were diagnosed and managed in the ER or inpatient hospital setting, this comprised only a small fraction (9.0%; n=32/357) of all postoperative representations to the hospital after TLH. Table 3 highlights factors associated with vaginal cuff complications in univariate and multivariate logistic regression modeling. Higher BMI (aOR 0.96, 95% CI 0.92–0.99) and a 2-LVC closure (aOR 0.38, 95% CI 0.19–0.74) represented the only factors protective of postoperative cuff complications.

We demonstrate that a 2-LVC closure leads to significant decreases in total postoperative complications in women undergoing TLH. The driving force behind this difference was a reduction in vaginal cuff complications, specifically dehiscence and mucosal separation rates in the 2-LVC cohort. Hence, implementation of 2-LVC closures during TLH could lower patient morbidity if implemented for all TLH. Although several other publications have examined individual surgical risk factors linked to vaginal cuff complications, these studies are largely heterogeneous in their methodology and focus primarily on mode of hysterectomy. 7 , 12 – 14 , 17 , 18 Unfortunately, closure techniques often differ in suture material (Vicryl, Maxon vs. barbed V-loc suture), suture technique (continuous running, figure of 8 or a combination with reinforcing stiches), or colpotomy technique and energy use (cutting vs. coagulation current). 9 , 11 – 13 , 19 – 22 Surgeon experience has also increasingly been recognized as a critical factor in preventing postoperative complications, including cuff suturing. 16 , 23 Given low vaginal cuff complications, especially dehiscence rates, studies quickly become underpowered to draw comparative conclusions. Our study standardizes these variables by examining only TLH performed by high volume, identically trained surgeons using indistinguishable colpotomy techniques and vaginal cuff closure suture material. Closure with a 1-LVC vs. 2-LVC closure was the only differing factor. Our study also assesses the benefits of a unique imbricating 2-LVC closure with one continuous barbed suture. One of the greatest oppositions to laparoscopic vaginal cuff closures has been the long-standing belief that vaginal cuff closures during minimally invasive procedures are associated with higher dehiscence rates compared to vaginal procedures. 9 , 12 , 14 Large, yet mostly retrospective studies report incidence rates of 0.15% (0.00–0.32%) for total vaginal hysterectomy (TVH) vs. 0.87% (0.64–1.59%) for TLH. One of these studies showed that in 12,000 women undergoing hysterectomies, vaginal cuff closure technique rather than the mode of hysterectomy dictated these outcomes since vaginal cuff suturing after TLH carried the same dehiscence rates as TVH 12 . However, this position has recently been challenged in a well-designed follow up prospective randomized controlled trial of 1400 women conducted by the same researchers. 7 Similar to our results, Uccella et al. (2018) demonstrated lower rates of total postoperative and cuff complications in the laparoscopic vaginal cuff closure compared to TVH. The authors concluded that this marked difference is likely a result of technological advances, surgeon skills and past misleading conclusions based on retrospective study designs that inadequately captured vaginal cuff dehiscences given lack of standardized postoperative assessment and delayed occurrence. In our study, dehiscences generally occurred at 6–12 weeks, but one case was observed as far out as 5 months postoperatively, thus highlighting that long-term follow up is often needed to capture these adverse events. Prospective studies will be required to confirm these findings and corroborate our results. However, it is encouraging that our unique 2-layer imbricating laparoscopic vaginal cuff closure resulted in no cuff dehiscences and overall lower postoperative cuff complications compared to the traditional 1-layer closure, which has primarily been studied in the literature. As more general Ob/Gyn surgeons become comfortable with laparoscopic suturing, our 2-layer closure technique may further aid lower volume surgeons to minimize postoperative cuff complications. Limitations associated with our research are those linked to above mentioned retrospective cohort studies including missing or inaccurate data and discrepancies in data collection. This was encountered specifically during documentation of tobacco use or postmenopausal status where approximately 1% of patients had missing values. Sexual activity rates were not consistently recorded and hence could not be included as a variable. Patients were instructed to abstain from intercourse for 8–12 weeks based on individual surgeon preference, which could represent a confounding factor. However, both preferences were represented in each group. Additionally, cuff dehiscences in the 1-LVC closure group also occurred at and beyond the 12-week mark and always occurred at first intercourse, hence suggesting that sexual activity rather than the timing likely played a more pivotal role. While demographics varied between the groups with tobacco use and BMI lower in the 2-LVC cohort, these two factors were controlled for in a uni- and multivariate analysis, and often have opposing effects on cuff complications with tobacco negatively impacting wound healing whereas a higher estrogenic state with increasing BMI is protective. 15 Our analysis supports the observation that obesity reduced postoperative cuff complications. Although operating time differences for cuff closures were not specifically recorded, in our hands a 2-LVC closure only adds approximately three minutes to our 1-LVC closure duration. It is unlikely that this would lead to clinically significant increases in postoperative complications. Our data was almost exclusively restricted to four high-volume, fellowship-trained minimally invasive gynecologic surgeons at one institution, which can limit external validity. However, the homogeneity of surgeon skills and utilization of identical surgical technique including the same uterine manipulator and electrosurgical energy setting, allows for a more accurate outcome determination of studied variables. Although our energy settings vary slightly from the traditional watts used in most studies, we find that a slightly increased power allows for more efficient cutting of the tissue and hence reduced thermal spread. Systematic reviews of different uterine manipulators have not been shown to favor one with respect to safety outcomes. 24 Furthermore, our 1-LVC closure complication rates are similar to previous randomized controlled trials and hence these two variables are less likely to have impacted the results. 7 Another strength of our study includes the large sample size, which is necessary to study these relatively uncommon events. Furthermore, our 35-hospital health care system allows for the capture of most postoperative complications as patients are instructed to present to our nearest network hospital for evaluation if transport to the hospital where the original surgery was conducted is not possible.

Hysterectomies are the second most common gynecologic procedure performed in the United States with nearly 400,000 cases a year. 1 , 2 Nearly one in three women undergo a hysterectomy by age 65 6 . Over the last decade, total laparoscopic hysterectomy (TLH) has emerged as the leading minimally invasive approach with nearly half of all cases now being performed with this modality. 1 – 5 Although vaginal cuff dehiscences are uncommon (0.8–5.8%), they continue to represent one of the most morbid complications after hysterectomies. 8 – 10 Risk factors for cuff dehiscence can largely be divided into factors that affect wound healing and surgical technique. Patient characteristics including postmenopausal age, immunosuppressive state, diabetes, active tobacco use and precocious intercourse postoperatively are among the most common elements credited to negatively affect cuff healing. 9 , 11 – 14 In contrast, obesity and older premenopausal age are protective patient factors. 11 , 13 , 15 Dehiscences from surgical factors are two-fold; those associated with technique (mode of hysterectomy, suture material, energy use during colpotomy, tissue bites, closure approach) and those attributed to postoperative complications (surgical site infection, hematomas). 9 , 12 , 14 Optimizing and standardizing the vaginal closure during TLH may be a key element to lowering postoperative complications including vaginal cuff dehiscence. The purpose of this study was to explore whether a two-layer laparoscopic vaginal cuff (2-LVC) closure compared to a one-layer (1-LVC) closure in women undergoing TLH would reduce total postoperative complications. We hypothesized that 2-LVC closures would significantly lower total postoperative complications primarily by reducing vaginal cuff complications.