Robot assisted laparoscopy vs laparoscopic transabdominal vesicovaginal fistula repair: a prospective comparison

Article information

Abstract

INTRODUCTION

Vesicovaginal fistula (VVF) has always remained and will remain a challenge to a surgeon, considering its scope of challenges extending well beyond the surgical domain, as its symptoms create social concerns for the patient. With time and technological advances, various newer techniques of refinement have come into the practice of VVF repairs. The most recent of which is the robotic technology. The wellknown advantages of the robot in surgery, such as excellent ergonomics, superior three-dimensional vision and accurate translation of hand movements into instrument motion, perfectly position its utility for VVF repairs, where the inherent difficulties of the procedure remain the challenging task of providing adequate exposure of the local site and the laborious maneuvering during pelvic dissection and suturing [1]. Minimally invasive approaches to VVF repair, as opposed to the traditional open approaches, are being increasingly utilized for the obvious advantage of reducing patient morbidity. But the earlier attempted laparoscopic repairs were technically demanding owing to the steep learning curve associated with intracorporeal suturing and difficulty in performing dissections between the bladder and vagina [1]. The introduction of robotic systems overcame these drawbacks of laparoscopic surgery, making it an ideal prospect to be utilized in pelvic surgery. Since the first demonstration of robotic repair in a middle-aged woman post-vaginal hysterectomy VVF in 2005, little progress has been made in terms of the available published and reliable literature on robotic repairs of VVF [1]. The majority of literature on the subject describing outcomes is seen to be limited to either brief case series or isolated case reports [2-7]. As a result, it goes without saying that the study population in the available literature has always been small, making it difficult to translate them into any recommendations that would guide real-world practice. The lower number of patients studied could be attributed to the relative rarity of VVF, which ranges between 0.3% and 2% across various etiologies, as well as the limited availability of robotic technology in many centers around the world [8,9].

The utility of robotic surgery has its own set of challenges that demand elucidation. A few things to enumerate would be the heightened expenses, greater demand for operating room resources such as space and the availability of skilled technical staff, in addition to the issues related to its accessibility in resource-poor countries. The cost of a Da Vinci robotic machine ranges between 1 and 2.3 million dollars depending on the version and its configurations, with an additional 180,000 dollars annually for maintenance [10]. In comparison, the conventional laparoscopic system has decent availability and fairly mitigates these cost and spacerelated issues, but carries the downsides of difficult intracorporeal suturing and its associated learning curve [10]. With its dependence on extensive and precise dissection with intracorporeal suturing, transabdominal VVF repair is one of the laparoscopic procedures in a urologist’s armamentarium most likely to benefit from robotic assistance. We aim to compare the efficacy of robotic and laparoscopic transabdominal VVF repairs through a prospective interventional comparative study comparing success rates, perioperative and post-operative parameters of clinical significance.

MATERIALS AND METHODS

A prospective interventional comparative study was performed on 40 eligible consenting patients of VVF with the primary intent to treat, as per protocol inclusion criteria (all patients between 18–70 years who presented with VVF not suitable for transvaginal repair). Patients that underwent VVF repair for complex VVF (fistula size >4 cm, multiple fistulae, fistulas with extensive tissue loss, malignant VVF and radiation induced VVF) and VVF associated with other genitourinary fistulae were excluded from the study. The study was conducted over a total period of 18 months, from 1 October 2022 to 31 March 2024. Patient allocation was done alternatively into the 2 groups, Arm A (robot assisted laparoscopy arm) and Arm B (laparoscopy arm). Preoperative demographic and diagnostic details were recorded at admission and then patients in both arms underwent the modified ‘o’ Conor transabdominal VVF repair with the usual pelvic port configuration and established standard described steps of the procedure (Fig. 1). The procedure was started with cystoscopy and bilateral ureteric catheter placements and a guidewire placement through the VVF site. Then, general peritoneal evaluation for any adhesions and adhesiolysis of the same was done (Fig. 1A). Localization of the tentative site of fistula was performed by manipulation of the guidewire placed through the fistula site (Fig. 1B). The above enables us to lessen our bivalving incision to a limited conservative cystotomy (Fig. 1C). Fistula site was then delineated, dissected away from the bladder and closed eventually with barbed absorbable sutures (Fig. 1D-F). Cystotomy closure was then performed with omental interposition and drain placement (Fig. 1G, H). The procedure concluded with the observation of a negative leak test with 200 mL bladder saline inflation (Fig. 1H). Intraoperative data including fistula dimensions, operative time, estimated blood loss, any concomitant procedures, need for conversion to open surgery and intraoperative complications were recorded during the procedure, while post-operative data such as the duration of abdominal drain, length of hospital stay (LOS) and post-operative complications as per Clavien-Dindo classification were also collected at the post-operative recovery phase. Patients were then followed up on the 1st (D₃₀) and 3rd (D₉₀) month after surgery to assess for surgical success, which was defined as complete resolution of symptoms with resumption of normal voids. The flow of study is summarized in Fig. 2. The present study protocol was reviewed and approved by the Institutional Review Board of Vardhman Mahavir Medical College and Safdarjung Hospital under S. No IEC/VMMC/SJH/Thesis/9/2022/CC-30. Informed consent was obtained by all subjects when they were enrolled.

Fig. 1.

Image depiction of the steps of transabdominal vesicovaginal fistula repair. (A) Adhesiolysis – separation of omental adhesions from the pelvic region to approach the fistula site. (B) Identification – demarcating the probable site of the vesicovaginal fistula by manipulation of the sponge placed in the vagina. This step aids in planning the cystotomy. (C) Limited cystotomy – a vertical incision given at the probable site of fistula. (D) Delineation – assessing the extent and size of the fistula and its relation to the bilateral ureteric orifices through the cystotomy incision. (E) Dissection – fistula dissection with separation of bladder and vaginal flaps. (F) Fistula closure – vesicovaginal fistula closure done with V-Loc 3’0 sutures. (G) Bladder closure – cystotomy closure done with V-Loc 2’0 sutures. (H) Leak test - done with 200 cc inflation of bladder with saline followed by omental interposition.

Fig. 2.

Flow of study description. VVF, vesicovaginal fistula.

RESULTS

A comparative demographic and pre-treatment patient profile has been depicted in Table 1 while Table 2 provides the comparative data on the post-treatment parameters. Demographic and relevant pre-treatment data were comparable between both arms with the lone exception of greater number of comorbid patients in the Laparoscopic arm (P=0.028). It is noteworthy to mention in regard to post-treatment data, statistically significant superior improvements were seen in the robotic arm in terms of faster operative duration (149.7±24.48 min vs. 228.7±28.63 min, P<0.001), lesser estimated blood loss (23.75±14.59 mL vs. 68.25±32.41 mL, P<0.001) and earlier abdominal drain removal times (3.05±0.83 days vs. 4.15±1.31 days, P=0.003). No major complications were observed with the study population. 2 patients had failure of procedure in the laparoscopic arm while all patients that underwent robotic repair had procedural success.

Table 1.

Patient demographics & baseline characteristics

Table 2.

Peri-operative details comparison

DISCUSSION

In the present study of 40 patients underwent transabdominal minimally invasive VVF repair through both laparoscopic and robotic platforms. A comparative analysis between both groups based on their perioperative and postoperative parameters was attempted. We portray our comparative analysis to be unique as it would be the sole study on the subject to attempt comparing the two minimally invasive modalities as most available literature are limited to largely to case reports or case series involving either robotic or laparoscopic operating platforms.

Success rates observed in our study were 100 and 90% in the robotic and laparoscopic arms respectively, with 2 failures in the latter arm. Technical difficulty owing to larger fistula size (3 cm fistulas) and dense local adhesions encountered during the procedure were the attributing factors for the two accounted failures. Subsequently, these patients were evaluated again during the 3rd month post primary procedure and underwent subsequent successful transabdominal robotic repairs. These patients remained recurrence free at 6 months post-operatively. Our results were comparable to the global trends that were observed in the worldwide literature published on both modalities [2,4,6,11-17].

In terms of mean operative time, the robotic arm displayed a statistically significant swifter procedure time when compared to the laparoscopic arm (149.7±24.48 vs. 228.7±28.63, P<0.001). Owing to a lack of direct comparative studies on the subject, we observed the operative times published in short case series reports for comparison. Our times were comparable and mirrored the results of most published literature on robotic repairs [3,7,11,18]. But on the contrary, published reports on laparoscopic VVF repair depicted largely different operative times with a range between 280 minutes [19] to 141 minutes [17]. A notable exception was the study by Mallikarjuna et al. [20], which projected an extraordinarily swift operative time of 54 minutes. The above findings just reflect the factors that influence results in laparoscopy, such as learning curves, surgeon experience, challenging intracorporeal suturing, case profile and so on. While, on the other hand, the uniformity in operative times with robotic repair in published studies clearly reflect the advantages conferred by the robotic systems such as superior magnification, extended degree of instrument movement and excellent ergonomics, all translating into better facilitation of procedural learning and hence the uniform stable outcome parameters [3,4,7,11,18].

In terms of the average estimated blood loss in our study, visibly lesser losses were observed in the robotic arm, with their difference reaching statistical significance (23.75±14.59 mL vs. 68.25±32.41 mL, P<0.001). Comparisons with available relatively large case series articles on both robotic and laparoscopic VVF repairs revealed findings in line with our observations [5,6,11,16,17,18,21].

When assessing the average duration to post-operative abdominal drain removal, the robotic arm fared significantly better in comparison to its counterpart (3.05±0.83 days vs. 4.15±1.31 days, P=0.003). On comparing similar studies of robotic VVF repairs [5,9,11,12], our results were comparable to the general trend observed while the drain removal durations were much earlier in the available articles pertaining to laparoscopic VVF repairs [14,21,22]. This deviation in our study could be explained by the factor of evolving surgical experience of our operative surgeons leading to them exercising extra caution and thereby keeping the drains for an extended duration. We are positive that with evolving experience over time, the difference observed with the drain durations would diminish. The mean LOS observed in both the groups were 7.30±3.33 days (robotic) and 8.70±2.75 days (laparoscopic) respectively with no statistical difference. When comparing identical robotic studies [2,6,12,18], our results appeared comparable to the general global trend while on the contrary the length of stay was shorter in articles pertaining to laparoscopy than the findings observed in our study [14,17,19,23]. Surprisingly, published data by Miklos et al. [16], Giannakopoulos et al. [21] and Mallikarjuna et al. [20] described even earlier discharge times of less than 3 days on an average. Our prolonged discharge times may be attributed to our institutional protocol of discharging patients post removal of ureteric catheters, surgeon discretion and the patient profile. Being an apex government center for urology, we received patients from very humble backgrounds, with most of them being devoid of easy access to a tertiary care center from their place of residence. Hence, our extended length of stay in both arms does not exactly portray the post-operative patient recovery profile but rather reflects on the interlude of multiple factors that influence the hospital stay of patients like varied institutional protocols of post-operative management and discharge policies, surgeon discretions on ureteric catheter and drain removal, complexity of fistula repair and patient recovery profile.

No major complications were observed in both arms of the study with concomitant procedure of Suprapubic catheter placement intraoperatively done in one and three patients of the robotic and laparoscopy arms of the study respectively, based on surgeon discretion. A comparative compilation of all available published literature on robotic and laparoscopic VVF repairs is portrayed in Table 3 and 4.

Table 3.

Summary of published studies on RAL VVF repair

Table 4.

Summary of published studies on LAP VVF repair

Some limitations exist in our study that demand addressal. Although carrying a unique comparative design between two minimally invasive modalities, the sample size of our study was considerably lower. The factors that played a role would be the limited study duration of 18 months and a relatively diminishing incidence of VVF patients in the outpatient department with improving health care facilities and patient awareness. Secondly, our study focussed mainly on the detailed analysis of the anatomical success rates and perioperative outcomes. We unfortunately failed to address the functional outcomes following successful VVF repair, such as the quality-of-life improvements and observations on bladder and sexual function post successful surgical repair. Another drawback worth mentioning was the limited follow-up duration because of our constricted study duration. Even though a 3-month follow-up duration was enough to assess operative success and early perioperative period, a longer follow-up could have given insights on long term complications and functional issues that occurred post VVF repair. Multiple surgeons of varying surgical experience operated on the patients. Although this could have affected our success rates and other perioperative outcomes, our study in true sense represents a better real-world scenario. We aim to address all our drawbacks in the future with a prospective randomized study design incorporating functional outcome analysis with a longer follow-up.

In conclusions, robot-assisted VVF repairs seem to be a viable better option when compared to laparoscopic repairs, especially with the reduced patient morbidity, comparable operative success, superior perioperative outcomes and better surgical ergonomics. Large, randomized studies with greater patient population and inclusion of even complex VVFs would be the way forward to validate our findings and improve our understanding on both operative platforms.

Article information Continued