Robotic versus laparoscopic revisional bariatric surgeries: a systematic review and meta-analysis

Abstract

Purpose

In recent years, the need for revisional bariatric surgery (RBS) procedures has experienced a noteworthy surge to confront complexities and weight recidivism. Despite being a subject of controversy for many, the utilization of the Da Vinci robotic platform (Intuitive Surgical, Inc.) may present benefits in RBS. This study aimed to evaluate the outcomes of robotic RBS in comparison to Laparoscopic RBS.

Methods

A meticulous and thorough analysis was ensured through a comprehensive exploration of the literature, which included PubMed, Medline, Scopus, and Cochrane. This exploration was conducted in adherence to the directives outlined in the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. The Newcastle-Ottawa scale was used for quality assessment.

Results

A total of 11 studies were included in this meta-analysis, comprising 55,889 in the laparoscopic group and 5,809 in the robotic group. No significant differences were observed in the leak, bleeding, operative time, or length of stay across both groups. However, the robotic group showed higher rates of conversion to open surgery (odds ratio [OR], 0.65; 95% confidence interval [CI], 0.53–0.79; p < 0.0001; I² = 0%), reoperation (OR, 0.70; 95% CI, 0.57–0.87; p = 0.0009; I² = 6%), and readmission (higher rate of readmission in the robotic group; OR, 0.76; 95% CI, 0.62–0.92; p = 0.005; I² = 30%)

Conclusion

Robotic-assisted bariatric surgery has no significant advantage over conventional laparoscopic surgery. Further research is warranted to explore and evaluate surgeons’ methodology and proficiency differences.

INTRODUCTION

Bariatric surgery is the most efficacious approach for achieving sustainable weight reduction and mitigating weight-associated complications [1]. The prevailing modality for the majority of these procedures is through laparoscopic technique, which offers a multitude of benefits. These include prompt convalescence, mitigation of postoperative pain, and amelioration of wound-related complications compared to open surgery. In modern times, bariatric surgery has attained a high level of safety, reflected in a mortality rate of three per 1,000 patients [2].

The requirement for revisional bariatric surgery (RBS) after initial bariatric surgery has significantly increased over the past few decades, and this trend is expected to continue. The RBS procedures are intricate in nature and are linked to elevated rates of morbidity and mortality in contrast to primary surgeries [3].

The utilization of the robotic platform presents numerous advantages, particularly in intricate procedures, such as advanced ergonomics, three-dimensional stereotactic vision, and seven degrees of instrument flexibility. This degree of freedom enables surgeons to conduct dissections in narrow spaces, facilitating precise intracorporeal suturing [4]. Over the past decade, the use of robotic platforms for both primary and RBS has been on the rise. Notably, the first instance of a robotic approach for bariatric surgery was demonstrated by Himpens in 1998, who successfully performed a robotic AGB [5].

The promise of the robot in RBS seems significant as this technology has already exhibited perioperative outcomes comparable to traditional laparoscopy [5]. Our objective in this meta-analysis is to juxtapose the perioperative outcomes of robotic and conventional laparoscopic techniques solely in RBS.

METHODS

Study design

The present investigation was conducted with the utmost fidelity to a previously established methodology unanimously agreed upon by all contributing authors. This included strict adherence to the directives outlined in the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines.

Literature search strategy

A comprehensive literature exploration was undertaken on September 20, 2023 to ensure a meticulous and thorough analysis. The literature review was performed using PubMed, Medline, Scopus (Elsevier), and Cochrane Central Register. We used the following terms (“Robotic” OR “robotic surgery” OR “robot OR” “Davinci robot”) AND (“Laparoscopy” OR “laparoscopic bariatric” OR “conventional bariatric”) AND (“revisional surgery” OR “revisional bariatric surgery” OR “secondary bariatric surgery” OR “Conversion”).

Data extraction

Data were extracted for each included study, pertaining to demographics such as sample size for each group, age, sex, preoperative body mass index, and comorbidities, as well as perioperative outcomes, including mean operative time, mean hospital stay, and incidence of intraoperative and postoperative complications. This process was carried out by two investigators (KA and HEB).

Statistical analysis

For categorical outcomes, we evaluated the odds ratio (OR) and the 95% confidence interval (CI) utilizing the random-effects model (Mantel-Haenszel statistical method). Continuous outcomes were determined through the weighted mean difference and its 95% CI with random-effects (inverse variance statistical method) models. The Cochrane Collaboration RevMan version 5.3 was employed for data analysis.

Eligibility criteria

Studies were deemed eligible for inclusion if they involved a comparative analysis between patients undergoing robotic and laparoscopic revisional bariatric procedures. We excluded studies with insufficient or duplicated data, a missing control arm, or a sample size of less than 10 patients. Only articles in English language were included.

Quality and publication bias evaluation

The Newcastle-Ottawa quality assessment scale (NOS) 7 in Table 1 [3,6-15] was used as an evaluation tool to assess non-randomized controlled trials (non-RCTs). This scale ranges from 0 to 9 stars. Studies scoring five stars or higher were considered to have adequate methodological quality and were included. Non-RCTs were found in the literature for inclusion. Two investigators (KA and HEB) independently rated the studies, and a final decision was reached by consensus. The risk of publication bias was evaluated through visual inspection of funnel plots.

Table 1.

Quality assessment

Study	Exposed cohort	Non-exposed cohort	Ascertainment of exposure	Outcome of interest	Comparability	Assessment of outcome	Length of follow‐up	Adequacy of follow‐up	Total score
Clapp et al. [12] (2019)	*	*	*	*	**	*			7
Beckmann et al. [10] (2020)	*	*	*	*	*	*	*	*	8
Acevedo et al. [11] (2019)	*	*	*	*	**	*			7
Ugliono et al. [3] (2022)	*	*	*	*	**	*			7
Gray et al. [13] (2018)	*	*	*	*	*	*	*	*	8
King et al. [9] (2020)	*	*	*	*	**	*	*		8
El Chaar et al. [7] (2020)	*	*	*	*	**	*			7
Nasser et al. [8] (2019)	*	*	*	*	*	*	*	*	8
Buchs et al. [6] (2013)	*	*	*	*	*	*	*	*	8
Moon et al. [14] (2020)	*	*	*	*	**	*			7
Seton et al. [15] (2022)	*	*	*	*	*	*	*	*	8

RESULTS

Overview of article selection

A total of 11 studies were included in this meta-analysis, involving 55,889 individuals in the laparoscopic group and 5,809 in the robotic group (Fig. 1). The majority of patients were females, with 47,626 (85.2%) in the laparoscopic group and 4,335 (74.6%) in the robotic group. These studies were published between 2013 and 2022 and originated in the United States, Germany, Italy, and Switzerland. Table 2 [3,6-15] depicts individual study characteristics.

Fig. 1.

PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram.

Table 2.

Summary of the studies selected for systematic review and meta-analysis

Study	Type of study	Journal	Country	Date Published	Sample size			No. of female patients			Mean age (yr)a)			Body mass index (kg/m2)a)
					Laparoscopic	Robotic		Laparoscopic	Robotic		Laparoscopic	Robotic		Laparoscopic	Robotic
Clapp et al. [12]	Retrospective	Surgery for Obesity and Related Disease	USA	March 17, 2019	35,988	1,929		30,701	1,686		48.5 ± 11.3	48.5 ± 11.3		41.5 ± 7.2	41.9 ± 7.1
Beckmann et al. [10]	Retrospective	Karger Open Access	Germany	April 3, 2020	18	41		11	30		48.8±9.2	46.5±10.3		47.1±6.2	42.9±8.2
Acevedo et al. [11]	Retrospective case control study	Surgical Endoscopy	USA	June 17, 2019	1,144	1,144		1,040	1,040		47.0 ± 10.6	46.9 ± 10.7		41.3 ± 8.6	41.6 ± 8.7
Ugliono et al. [3]	Retrospective	Updates in Surgery	Italy	November 23, 2022	44	150		33	128		48.0 ± 10.5	49.0 ± 10.1		35.6 ± 7.2	38.0 ± 8.5
King et al. [9]	Retrospective	Obesity Surgery	USA	October 19, 2020	115	52		102	41		47.9 ± 10.2	50.4 ± 11.5		37.0 ± 9.3	39.5 ± 10.2
Gray et al. [13]	Retrospective	Obesity Surgery	USA	February 7, 2018	66	18		NA	NA		NA	NA		NA	NA
Nasser et al. [8]	Retrospective	Surgery for Obesity and Related Diseases	USA	December 3, 2019	15,935	1,077		13,435	159		48.1 ± 10.0	48.3 ± 11		42.5 ± 7.6	42.0 ± 8.2
El Chaar et al. [7]	Retrospective	Journal of Robotic Surgery	USA	May 30, 2020	220	220		186	185		50.7 ± 11.3	49.5 ± 11.3		42.6 ± 8.0	42.2 ± 7.6
Buchs et al. [6]	Retrospective	The International Journal of Medical Robotics and Computer	Switzer­land	October 24, 2013	21	11		17	10		44.0 ± 10.4	45 ± 9.5		41 ± 6	40 ± 8
Moon et al. [14]	Retrospective	Surgery for Obesity and Related Diseases	USA	March 21, 2020	64	30		56	26		38.8 ± 7.9	45.4 ± 13.0		50.3 ± 10.1	44.6 ± 7.4
Seton et al. [15]	Retrospective	Obesity Surgery	USA	October 20, 2022	2,274	1,137		2,045	1,030		46.15 ± 10.31	46.17 ± 10.84		39.11 ± 7.61	39.31 ± 7.73

NA, not applicable.

^a)Mean ± standard deviation.

Mean operative time

Nine studies mentioned the mean operative times, ranging from 103.7 to 270 minutes in the laparoscopic group and 101 to 352 minutes in the robotic group. However, no statistically significance difference was observed, with a mean difference of –29.77 (95% CI, –61.74 to 2.19; p = 0.07; I² = 100%) (Fig. 2A).

Fig. 2.

Forest plot of the mean operative time (A) and the mean length of hospital stay (B). SD, standard deviation; IV, inverse variance; CI, confidence interval; df, degree of freedom.

Length of hospital stay

Among the 11 studies included, eight provided data on the length of hospital stay. Similar durations were observed in both groups, ranging from 1.7 to 8 days in the laparoscopic group compared to 1.674 to 6 days in the robotic group. The absolute mean difference was 0.37 (95% CI, –0.23 to 0.97; p = 0.22; I² = 67%), and no statistically significant was observed (Fig. 2B).

Conversion to open surgery

Six studies reported incidences of conversion to open surgery, with total events of 1,250 in the laparoscopic group and 118 in the robotic group. This resulted in an OR of 0.65 (95% CI, 0.53–0.79; p < 0.0001; I² = 0%). The analysis revealed that conversion to open surgery is significantly more likely to occur among patients undergoing robotic revisional weight loss surgery (Fig. 3A).

Fig. 3.

Forest plot of the odds ratio of conversion to open surgery (A), leakage (B), bleeding (C), surgical site infection (D), reoperation rate (E), readmission rate (F), mortality rate (G), and total complications (H). SD, standard deviation; IV, inverse variance; CI, confidence interval; df, degree of freedom.

Complications

Leakage was detected in seven studies, with 432 incidences occurring in the laparoscopic group compared to 54 incidences in the robotic group. No significant difference was detected in the development of the leak, indicated by an OR of 0.94 (95% CI, 0.44–1.99; p = 0.87; I² = 39%) (Fig. 3B).

Bleeding was reported in five studies, and it occurred in 808 incidences in the laparoscopic group and 61 incidences in the robotic group. However, no significant difference was observed (OR, 2.48; 95% CI, 0.72–8.55; p = 0.15; I² = 94%) (Fig. 3C).

Surgical site infection was reported in five studies, with 1,396 incidences in the laparoscopic group and 104 in the robotic group, resulting in an OR of 0.94 (95% CI, 0.77–1.16; p = 0.59; I² = 0%), showing no significant difference between both groups (Fig. 3D).

Reoperation rate was assessed in seven articles. Patients in the laparoscopic group had a statistically significant lower reoperation rate (OR, 0.70; 95% CI, 0.57–0.87; p = 0.0009; I² = 6%) (Fig. 3E).

Seven studies reported the readmission rate, encompassing a total population of 23,466 individuals, with 19,818 in the laparoscopic group and 3,678 in the robotic group. The analysis revealed a lower rate of readmission in the laparoscopic group, demonstrating an OR of 0.76 (95% CI, 0.62–0.92; p = 0.005; I² = 30%) (Fig. 3F).

Mortality rate was reported in six studies. No significant difference between both groups was detected with an OR of 2.82 (95% CI, 0.62–12.80; p = 0.18; I² = 76%) (Fig. 3G).

Overall complications were reported in all the studies, accounting for 4,552 incidences in the laparoscopic group compared to 445 incidences in the robotic group. Notably, no significant difference was observed (OR, 1.02; 95% CI, 0.79–1.31; p = 0.90; I² = 58%) (Fig. 3G).

DISCUSSION

The increasing utilization of robotic platform for bariatric procedures executed annually underscores the significance of comprehending its role in revisional surgery. These subsequent operations involve a more intricate technicality, leading to a heightened risk of complications in comparison to primary bariatric surgery [6]. Laparoscopic surgery swiftly emerged as the primary approach during the era of minimally invasive surgery and is currently deemed the gold standard for all bariatric surgeries. Nevertheless, laparoscopic surgery is not immune to limitations, which include a loss of degrees of freedom in a confined space, restricted rotational motion, reduced depth perception, a two-dimensional video system, spatial disorientation, decreased haptic feedback, significant ergonomic challenges, and interference of the natural coupling of the surgeon’s eyes and hands by an independently moving video camera. Owing to these limitations, there has been a surge in the adoption of new forms of minimally invasive surgery, such as robotic surgery [7].

Our systematic review and meta-analysis revealed a higher incidence of conversion to open surgery within the robotic group compared to the laparoscopic group, a result consistent with findings by Nasser et al. [8] that reported a 0.5% rate for robotic-assisted surgery. This contrasts with the study conducted by King et al. [9], where no conversion to an open approach was encountered. This discrepancy could be partly explained by the increase in expertise and surgeon comfort. The accomplishment of a learning trajectory is an absolute prerequisite, even for seasoned laparoscopic surgeons, prior to embarking on technically intricate robotic procedures [16], especially considering that revisional surgeries are considered highly complex compared to primary bariatric surgeries, owing to changes in anatomy and the presence of adhesions from prior surgeries [4]. On the other hand, conducting these surgeries in high-volume robotic centers may decrease the risk of conversion.

In contrast to Edwin et al. [11], we found no difference between the two interventions in terms of length of hospital stay. In his study, Edwin et al. demonstrated a longer length of stay in the robotic group (2.01 ± 3.2 days vs. 1.8 ± 1.9 days, p < 0.0001), potentially contributable to the surgery’s complexity. Revisional surgeries include extensive dissection from previous surgeries in addition to multiple bowel resections. Although Klock et al. [17] did not report a statistically significant difference in hospital length of stay, they found the robotic approach to be linked with a 31% increase in hospital cost, resulting in an average of 3,700 US dollars greater cost per case. This finding is consistent with previously reported data comparing robotic to laparoscopic assisted. Despite this, arguments in favor of cost-effectiveness still lean toward the laparoscopic approach.

Our meta-analysis did not reveal any significant difference between robotic and laparoscopic revisional surgeries in terms of mean difference (–29.77; 95% CI, –61.74 to 2.19; p = 0.07, I² = 100%). This lack of distinction could be attributed to the varying experience levels of surgeons in the robotic group, as not all practitioners are adept at handling technically challenging revisional surgery on the robot platform. Surgeons who start with robotics often handle simpler cases to enhance their learning curve, a notion supported by Buchs et al. [6], who observed that once surgeons surpass the learning curve for revisional surgery, the time taken to perform the procedure becomes comparable between the laparoscopic and robotic-assisted methods. However, this observation may not apply universally to primary bariatric procedures. For example, when performing straight bariatric surgery, such as sleeve gastrectomy, the robotic platform does not show superiority over laparoscopy in terms of efficiency. The time taken for both approaches, including docking and undocking time, remains comparable. Moreover, laparoscopy typically offers an easier learning curve compared to robotics in these cases.

Although the utilization of a robot-assisted approach may not have a direct positive impact on patients and may result in additional costs, there are a multitude of immeasurable yet valuable advantages for surgeons. Antanavicius and Sucandy’s research [18] revealed that the Da Vinci robotic system (Intuitive Surgical, Inc.) presents significant benefits, particularly in the creation of complex anastomosis. The main advantages include enhanced operational ease, visualization, accuracy, and range of motion, especially during critical stages of the operation. The robotic system enables bariatric surgeons to create anastomosis in a similar manner to the open technique.

While informative, our meta-analysis does have certain limitations. The most significant of these limitations relates to the heterogeneity among the included studies. Additionally, the sample sizes were unbalanced, with some studies having considerably higher sample sizes than others.

In conclusion, our systematic literature review and meta-analysis revealed no significant advantage of using robotic-assisted RBS. These findings are consistent with previous research studies exploring the effectiveness of robotic techniques for bariatric surgery. Nevertheless, it is crucial to conduct further comprehensive research; examining the diversity in surgeon methodology and surgical proficiency is necessary to ascertain the variables contributing to the rise in these distinctions.