Robotic operations in urgent general surgery: a systematic review

Two of the authors (AR and JL) independently searched PubMed (1980-present) and the Cochrane Library (1980-present). The systematic review was performed according to the preferred reporting items for systematic reviews and meta-analysis (PRISMA) guidelines [7].

A two-stage analysis was conducted. First, a preliminary search and screening of 500 results was performed on October 26th and 27th, 2020, to elaborate the principles of the further analysis. The results of this first search indicated that the number of publications of interest was very limited. Therefore, we decided to include case reports and case series in the systematic analysis. Furthermore, this first search showed a massive heterogeneity of data. To give the reader a full overview of the topic, we decided not to attempt a meta-analysis since a meta-analysis would have limited the useable studies to only a few. A strict systematic review or a meta-analysis would necessarily have meant that most urgent surgical diseases could not have been analyzed. The primary goal of this work, however, was to cover robotic-assisted operations in urgent surgery as broadly as possible, which was only possible through a partially descriptive evaluation.

The second definitive search was run on December 31st, 2021.

The following search terms were used in combination with “robotic” and “robotic surgery”: adhesiolysis, appendectomy, appendicitis, bowel obstruction, cholecystectomy, cholecystitis, diverticular disease, diverticulitis, hernia, incarceration, perforation, peritonitis, ulcer, urgency and emergency (Appendix 1).

The search terms were partially truncated to include as many grammatical variables as possible.

All studies, case reports and case series describing urgent general and visceral surgical procedures in connection with robotic surgery were included. Manuscripts were included in which typical general or visceral surgical operations were described; inclusion was not decided by which specialty the operation performed. If study populations were published multiple times, the more recent publication was included.

Publications that did not report the original data were excluded, as were those with overlapping study populations (see above). Urological, gynecological or vascular surgery emergency interventions were excluded. Reports on the thoracoscopic robotic treatment of diaphragmatic hernias were also excluded. If the full text was not available, the study was also excluded.

If the specified information was extractable, complications were classified according to the Clavien–Dindo classification, whenever applicable [8]. In all other cases and biliary complications in cholecystectomies, the complications were mentioned separately.

All included manuscripts were examined with a focus on the following factors: (a) primary objective of the study and technical aspects, (b) complications (see above) and whether these complications could be related to the use of the robot, (c) pros and cons of the use of the robotic operation, (d) financial aspects, (e) factors related to the acute/urgent situation and (f) further outcomes of interest.

Seven papers reported on urgent appendectomies using robotic surgery in a total of 196 patients (Table 1). The vast majority of patients (n = 185) were analyzed in two studies, both using RCM. Three more publications were case reports or case series. As a result, only 11 patients were urgently appendectomized using an SR. No complications or conversions were reported.

The three case reports/case series specified closure of the appendix stump (sutures), and the other studies did not provide any information on this. There were no reports of robotic single-incision appendectomies.

Thirty publications reported on urgent robotic cholecystectomies, none of which were randomized (Table 2). Three studies were prospectively controlled, and 14 were retrospectively controlled. Of the uncontrolled studies, three were prospective, and six were retrospective. Four case reports or case series have been summarized in the table for better clarity [13, 30‐32].

A total of 804 urgent cholecystectomies were performed with robotic assistance, 101 of them with an RCM. The remaining 703 patients underwent surgical intervention with an SR. The Da Vinci^® system was used in 546 (77.7%) patients; the robot type was not reported for 130 patients. One study explicitly analyzed urgent operations as the main interest [23].

Eleven publications reported cholecystectomies using the single-incision robotic technique (SIRC).

The reports of complications differed greatly, as did the follow-up periods (0–6 months). Kubat et al. showed 12% complications in urgent cholecystectomies, including one (1.5%) biliary tract injury [23]. The highest reported complication rate was 20% (including seromas), and 9 studies stated that no complications occurred. The incidence of Clavien–Dindo grade ≥ 3 complications varied between 1.7 and 4.2%. Biliary tract problems (injury, leakage) were demonstrated in three studies, each with one patient. Hernias represented a particular complication of single-site operations; these were described in three studies, with an incidence of 0.9–6.5%. Conversion to a laparoscopic procedure or to open surgery was reported in 7 studies, and two studies indicated that conversion was necessary due to inflammation [33, 34]. In one study, conversions were an exclusion criterion [35]. Six authors reported a prolonged operation time and one reported a decreased operation time in the robotic group compared to laparoscopic cholecystectomies. One study reported higher costs for robotically operated patients [36].

Urgent gastrointestinal operations were investigated in 12 studies (Table 3). There were three studies in this group of topics that specifically investigated urgent operations: Anderson et al., Beltzer et al. and, most recently, Robinson et al. [24, 37, 38]. The publication by Ohmura et al. on the use of an RCM evaluated for appendicitis and cholecystitis also reports on the operation of 16 perforations in the upper gastrointestinal tract [14]. Three very similar case reports of Kudsi et al. were combined into a case series [9‐11].

It remains unclear how many of the 60 robotically operated patients reported by Beltzer et al. were treated in an emergency situation. For this review, we accepted two of the 60 patients with diverticulitis type IIc of classification of diverticular disease (CDD) since they undoubtedly belonged to the group of urgent operations [39]. The study did not distinguish the complications between urgent and elective operations. The overall complication rate was 30%, with percentages of 8.3% for Clavien–Dindo grade 3b complications and 6.7% for anastomotic leakage. One fatality (Clavien–Dindo grade 5), and the need for conversion in another case were described [37]. Anderson et al. listed a complication rate of 20%, though with no further specifications (n = 1) [24]. Robinsons’ primary focus was on a typical urgent operation: perforated gastrojejunal ulcers. He reported noninferiority to the laparoscopic approach but dramatically higher costs for robotic operations. None of the case reports/case series reported any complications.

SRs are commonly used in hernia surgery; however, the urgency of these operations arises from the incarceration of the hernia, which is often associated with intestinal obstruction. We identified eight reports of urgent robotic operations for hernias (Table 4). Four studies retrospectively analyzed urgent hernia operations, two of which were controlled studies [25, 40, 41]. In addition, we identified a database analysis that included urgent robotic hiatal hernia operations [27]. A rare indication for urgent robotic surgery was the operation of posttraumatic splenic bleeding reported by Giulianotti et al. [15]. No reports on the use of an SR in adhesive intestinal obstruction were found.

For urgent robotic hernia operations, the published complication rates were very heterogeneous; the database analysis of Hosein et al. reported one death (0.1%) and an overall complication rate of 2% [27]. Muysoms et al. reported 3.5% minor and 0.2% major complications [41]. Kudsi et al. published a complication rate of 36.6% (23% minor and 13% major complications); however, this rate was significantly lower than that in the open surgery subgroup of Clavien–Dindo grade ≥ 3 complications [26]. A further interesting outcome was the significantly shorter hospital stay but higher costs and the recurrence rate of 3.2% reported by Myosoms et al. [41]. All studies emphasized that the interventions were technically feasible using an SR.

To date, there is only limited research on urgent robotic operations in general surgery; therefore, the data available are still very limited. Against this background, we consider a review such as ours, which systematically collects and examines the existing data and presents it descriptively, to be as comprehensive as possible and to be valuable and necessary.

Our review gives clear indications that robotic surgery has not yet arrived in urgent general surgery on a larger scale. A randomized controlled trial has not yet been performed, which is not totally unexpected given that randomized trials dealing with emergency and urgent surgery are generally very rare. However, even the nonrandomized studies offer specific data only to a limited extent: only six studies named urgent operations as a main interest or important variable [23‐27, 38]. This makes a systematic assessment of the evidence complicated and a meta-analysis practically impossible. It can be stated that there is only a low level of evidence regarding robotic surgery in the context of urgent operations. Nevertheless, we were able to identify and summarize a notable number of publications covering a wide range of diseases.

Case reports and series are sources with limited evidence and are not suitable for a meta-analysis. In the absence of sources with better evidence, these are nevertheless presented in our study, as these reports fulfill the function of documenting the technical feasibility of certain interventions in the sense of a proof of concept.

With the recently published position paper of the WSES, the topic of urgent robotic operations was highlighted for the first time. De’Angelis et al. also conducted an extensive literature search. The number of papers screened was comparable to our study, but de’Angelis et al. used only ten manuscripts for their analysis: five retrospective cohort studies, and five case reports/case series [4]. We decided against reducing the number of publications used through a stricter assessment since we wanted to provide a maximum amount of information about robotic emergency operations. Nevertheless, our presentation goes beyond a purely narrative review, as, under verifiable conditions, we offer the first complete overview of the published data. However, the conclusions the authors of the WSES statement published are very similar to our findings, though we covered a wider field of operations [4]. The combination of de’Angelis et al. and our work creates, for the first time, a deeper impression of the significance of robotic operations in urgent surgery.

The main indications for robotic emergency interventions were gallbladder, hernias, and gastrointestinal surgery, as well as appendectomies. Some common aspects can be identified. First, there are supposedly higher costs of robotic interventions. Our review shows that robotic operations are significantly more expensive than laparoscopic surgeries [36, 38, 41, 42]. To date, there are no data that demonstrate the amortization of the extra costs by a reduction in the length of stay or complications that stem from the use of an SR. Moreover, the abovementioned studies that showed a cost increase per procedure did not take the considerable acquisition costs of the systems into account. The option of refunding by additional charges for the use of an SR is not possible in all health systems. Second, an often cited counterargument against robotic emergency interventions is the increased time requirements due to longer preparation or operation times. Here, the data were ambiguous and ranged from lengthening to shortening of time intervals with a tendency toward increased operative time. Minimally invasive surgery is primarily intended for stable patients, which of course also applies to robotic operations. Thus, even the moderate increases in operating time described above do not seem to be a contraindication for robotic operations.

RCMs can be classified as robotic surgery only to a limited extent. However, they offer some of the alleged advantages, in particular a stable image and ergonomic advantages. However, there are also might be disadvantages, e.g., complications due to technical malfunctions.

According to this review, this type of surgery seems to be not very widespread, especially in urgent surgery, as only two publications have described the use of these systems [14, 21].

No clear statements about the advantages and disadvantages of RCMs can be drawn from the data presented here, but it can be stated that this type of surgery is undoubtedly more similar to laparoscopic than to robotic surgery. However, it seems important that no RCM-associated complications are reported.

Appendectomy is a very frequent operation, although there have been very few studies that deal with robotic operations for acute appendicitis. This is of particular interest since appendicitis was one of the diseases that significantly led to the development of modern laparoscopy, initially starting with the confirmation of the diagnosis and the first laparoscopic appendectomy by Semm in 1983 [43]. Today, laparoscopic appendectomy is the therapy of choice for appendicitis [2].

There may be various reasons for the lack of studies on SR for appendicitis: The presumed high costs of robotic operations have been discussed since the start of robotic surgery [30]. Most likely, a “minor” procedure such as an appendectomy with correspondingly low remuneration is estimated as economically unreasonable and will therefore not become part of clinical routine for robotic surgery. It is also conceivable that the advantages of robotic operations for an often rather simple operation without complex preparation are not regarded as sufficient to implement this technique. An urgent appendectomy with the Da Vinci^® system was reported in only two cases, while this robot system has been generally used most often. The feasibility of appendectomies with the Da Vinci^® system was shown by several studies reporting on appendectomies in the context of other operations [44, 45].

An option to make a robotic appendectomy economically worth considering is to overthink the closure of the appendix stump: the use of the stapler or clip applicator for the Da Vinci^® system causes relevant costs. This could be circumvented by closing the stump with a Roeder loop or suture. However, a current meta-analysis showed that stump closure with staples is superior in terms of postoperative complications [46].

In the overall view, appendectomy via SR is technically possible, but the additional expense compared to laparoscopic appendectomy most likely cannot be justified.

Regarding this indication, the number of published studies is better, although far from satisfactory. However, against the background of more available studies, a case report analysis was less necessary. In addition to the fact that one of these case reports was among the first publications on robotic surgery in general surgery (Hanisch et al.), these reports did not provide any crucial information [30]. The statements regarding the advantages and disadvantages of the robotic approach in urgent cholecystectomies were very heterogeneous without a clear trend.

Remarkably, many studies compared single-incision robotic cholecystectomy (SIRC) with nonrobotic surgical procedures. Therefore, we assume a reporting bias: multi-incision robotic cholecystectomy (MIRC) was primarily examined in studies that were published before 2010, and acute cholecystitis was often an exclusion criterion [19]. However, the scientific perception of robotic cholecystectomy may have changed in two aspects: MIRC appears to be such a standard procedure that it is less examined in studies. At the same time, and based on their increased robotic experience, more researchers have opted to include acute cholecystectomy in their studies. In our estimation, these interventions are currently often performed as SIRC since this technique is assumed to be more innovative. Furthermore, there is another inaccuracy in the analysis: a number of studies reported robotic surgery for acute cholecystitis, with acute inflammation diagnosed intraoperatively. Therefore, this might not be labeled “urgent surgery” correctly in the narrower sense. The subject of this analysis is not the question of single- vs. multi-incision operations. The currently most up-to-date Practice Guideline on Safe Cholecystectomy votes for MIRC, in particular due to an increased rate of biliary tract injuries in the single-incision group [47]. Regarding the literature analyzed in our study, there was one biliary tract injury during an operation for acute cholecystitis using the SIRC technique. The data on complications and conversions were heterogeneous, as were the statements on the possible advantages and disadvantages of robotic operations. Biliary tract problems were reported in three patients, which resulted in a biliary complication rate of 0.4% among all patients operated on with an SR, which was within the expected range. Except for one study, these complications were not differentiated between urgent and elective operations. Notably, Kulaylat et al. reported an increase in hospital costs of 38% for the robotic procedure compared to laparoscopic cholecystectomy [36]. In summary, at the moment, neither the advantages nor the risks of robotic operations in urgent, acute cholecystectomies can be adequately assessed. However, no clear contraindications for the use of robots in this situation were found in this review.

The advantage of robotic surgery in acute conditions of the gastrointestinal tract, especially perforations, seems obvious: since a robot provides a significant advantage for suturing and tying knots, perforations and ruptures can be closed excellently. Furthermore, robotic operations for benign and malignant diseases of the colon or the upper gastrointestinal tract are clinical routine, with an increasing proportion being performed in the elective setting [48]. Our review shows that for urgent operations, there are hardly any data available. However, our own experience and the case reports/series listed demonstrate that these interventions were definitely possible in an urgent setting.

Beltzer et al. summarized that there was no advantage in using the robot for surgical procedures in diverticular disease [37]. Anderson et al. also did not state any explicit advantages and reported a longer operation time [24]. There are currently no data that propagate the use of SR in this indication group for a better outcome in a population operated urgently.

Robinson et al. presented a study primarily focused on urgent robotic surgery for perforated gastrojejunal perforations [38]. The authors were able to show that robotic surgery was not inferior to laparoscopic surgery. In addition, they reported two interesting aspects that otherwise hardly received attention: (1) the immediate preparation time in the OR was even shorter than that with laparoscopic operations; and (2) 54% of the operations were performed on weekends or during the evening, night or early morning. This invalidates some of the arguments that have been put forward against robotic emergency operations.

As mentioned above, in regard to surgery of the gastrointestinal tract, the cost disadvantages of SRs matter again: stapler and sealing devices in addition to draping, scissors, forceps, etc., are significant cost factors. Schiergens et al. showed that the use of an SR for an elective sigma resection increases the cost of surgical supplies by more than a factor of 4 compared to open surgery and by more than a factor of 2 compared to laparoscopic surgery. These numbers were very comparable to the cost increase reported by Robison et al. for urgent gastrojejunal ulcers [38, 49].

SRs are used regularly in minimally invasive hernia surgery [3]. This may contribute to the fact that there are quite a few studies on urgent robotic hernia surgery. Even if the evidence cannot be described as satisfactory, the overall picture is similar to those of the other indication groups: robotic interventions are feasible in incarcerated hernias and in urgent situations. The complication and conversion rates were low and comparable to those of nonrobotic, minimally invasive procedures. To date, however, no clear advantages of the robotic technique have been demonstrated.

The situation is different in hiatal hernias: in most cases, hiatal hernias are technically more demanding than inguinal or ventral hernias. The sutures required to reconstruct the hiatus and to create the fundoplication as well as the preparation make this operation ideal for the use of an SR. Current studies showed fewer complications and a shorter hospital stay for robotic hiatus hernia surgery in the elective setting [50]. Even if no data are available for urgent hiatoplasty, it is quite likely that such effects can be observed here.

The report of a splenectomy for bleeding (along with several reports of elective splenectomies) indicated that this operation is also possible with an SR. If the patient’s circulatory system remains stable and a minimally invasive procedure is conceivable, the procedure can also be carried out with an SR if the appropriate expertise is available.