Laparoscopic cholecystectomy (LC) is the gold standard for treating symptomatic gallstones, acute cholecystitis, and acute gallstone pancreatitis. Introduced about 40 years ago, its popularity grew with the support of patients, surgeons, and commercial partners. Robotic surgery, in contrast, was introduced approximately 20 years ago and was initially limited to a few centers that could afford a robotic system. Recently, the landscape has shifted due to decreased costs, the introduction of new platforms that have stimulated market competition, and the growing body of evidence that supports robotic surgery over laparoscopic and open surgeries. Consequently, the use of surgical robots has been expanded to virtually all surgical fields. However, robotic cholecystectomy (RC) is still struggling to gain widespread clinical acceptance, primarily due to the higher costs associated with the use and maintenance of the robotic platform compared to more affordable laparoscopic instruments. These additional expenses are not easily reimbursed by government or private insurances.

Nonetheless, RC offers the advantage of a minimally invasive approach with equipment that is potentially superior to that used in traditional laparoscopic methods. The robotic approach enables more precise dissection, eliminates tremor, allows full endo-wrist movements, and on some platforms, provides haptic feedback. In this publication, we present our perspective on RC versus LC, regardless of costs and financial implications.

The discussion on RC versus LC gained prominence following the publication of a paper by Kalata et al. [1], which analyzed 1,026,088 cholecystectomies performed in the US from January 1, 2010, to December 31, 2019. The study found that RC is associated with a bile duct injury (BDI) risk that is twice that of LC. This finding resonated significantly on social media platforms [2].

There is concern that the message in this publication could be considered misleading due to a suboptimal methodology that may introduce bias into the study.

The first bias is selection bias. The authors only considered patients aged between 66 and 99 years (mean age 72±12 years), which does not represent the typical cholecystectomy population, whose mean age is significantly lower [3], with only about one-third being older than 60 years [4]. Additionally, the authors included patients who were monitored for 23 hours or more post-operation, representing a quarter of patients undergoing minimally invasive cholecystectomy [5]. The patient cohort analyzed was predominantly comprised of urgent or emergency cases (870,405, 84.8%) compared to elective cases (155,683, 15.2%), and for acute cholecystitis (897,116, 87.4%) as opposed to other diagnoses such as biliary pain or biliary pancreatitis (128,972, 12.6%). This cohort is not reflective of the general cholecystectomy population, where elective procedures are usually more common than emergency cases.

The second bias is related to study design. The research employed a retrospective, non-randomized case-control design, rated as 3B level evidence [6]. This method is not ideal for comparing two surgical techniques; a randomized controlled trial (RCT) is considered the gold-standard approach. Although LC was adopted in clinical practice without a proper RCT, at that time, withholding a minimally invasive option from randomized patients was deemed unethical due to LC’s proven benefits. In contrast, a RCT comparing RC and LC is entirely feasible, ensuring that all patients benefit from minimally invasive and effective treatment.

The third bias is comparison bias, where the two groups compared were not equivalent. The respective sample sizes were unbalanced, with 25,084 in the RC group and 1,001,004 in the LC group, a difference of 40-fold. Patients in the RC group had significantly higher Elixhauser comorbidity indexes and higher rates of obesity, hypertension, hypothyroidism, anaemia, kidney failure, depression, complicated diabetes, liver disease, weight loss, and solid tumors with or without metastases and were more frequently admitted as elective cases compared to LC patients (33.7% vs. 14.7%). RC was less frequently performed for acute cholecystitis compared to LC patients (78.1% vs. 87.7%). Importantly, a portion of patients undergoing RC were actually admitted for acute cholecystitis and treated electively or as delayed emergencies after nonsurgical treatment of the gallbladder inflammation, possibly due to the unavailability of the robot in the emergency theater. Unfortunately, the publication does not reveal how many patients in the two groups were initially treated conservatively with the goal of performing a minimally invasive operation at a later stage. This consideration is crucial when assessing if the preference for a robotic operation influenced the decision for an initial nonsurgical approach, thus delaying definitive cholecystitis treatment and allowing the pericholecystic inflammation to become more complex. It is well recognized that the timing of LC for acute cholecystitis depends on multiple factors, including the organization and availability of an operating theater, the surgical robot, and the presence of a specifically skilled surgeon. Given these factors, the robotic approach may have been preferred in more challenging cases. Unfortunately, the publication lacked a difficulty grading system for LC, such as that proposed by Sugrue et al. [7]. Instead of using conventional sensitivity multi-variant analysis, propensity score matching (PSM) would have been much more effective at reducing selection and comparison biases while maintaining an acceptable sample size for both groups. A more recent study including a PSM comparative analysis of robotic, laparoscopic, and open emergency cholecystectomies on 233,945 cases showed that the outcomes of RC were at least as good as those of LC in terms of BDI [8]. A recent meta-analysis of 13 comparative studies with 22,440 patients indicated that although RC is associated with a longer operative time compared to LC, it showed no difference in intraoperative complications or BDI [9]. This study analyzed a more realistic sample population, where most patients were females (65.2%) and the average age was 48.5 years.

From a conceptual viewpoint, comparing an established technique such as LC with a relatively new and non-yet-standardized technique like RC makes little sense. The study by Kalata et al. [1] may be affected by this temporal bias, and these conceptual errors are evident in a publication by Aguayo et al. [10] as well. In this publication, 3,193,697 patients from the 2008 to 2017 National Inpatients Sample database who underwent cholecystectomy were analyzed. The study reports that in 2008, only 2 out of 10,000 cholecystectomies were performed using a surgical robot, and this figure increased to 3 in 100 by 2017. Thus, a total of 1.3% of patients included in this study underwent an RC. These statistics suggest that the study spanned a period during which robotic surgery, especially RC, was in its developmental stages. This would affect the results generated, as those initial RCs were likely not uniformly performed and were executed by inexperienced robotic surgeons. The results of this cohort of patients, involved in a learning curve, cannot be reliably compared with those from the well-established and extensively standardized LC cohort. Interestingly, the publication notes that patients who underwent RC had a significantly higher Elixhauser index of comorbidities. This raises the ethical dilemma of why more comorbid patients were subjected to the less standardized and relatively new surgical technique rather than being offered the well-established laparoscopic approach.

When LC was introduced in clinical practice in the late ‘80s, several publications labeled it as a “scandal” due to its disruption of traditional open cholecystectomy techniques, contributing to a 2-3-fold increase in the BDI risk [11]. This risk was attributed more to the inherent characteristics of the LC technique rather than to the experience or skills of the practitioners [12]. Since the introduction of LC, it has been shown that the risk of BDI during LC decreases with increased procedure volume, dropping to 0.19% after 150 LCs/year [13]. It is likely that a similar trend will be observed with RC.

The current literature on the subject of comparing RC to LC generally agrees that RC is at least equivalent to LC in terms of clinical outcomes. It is recognized that RC has perceived drawbacks, such as longer operative times and higher costs. However, a recent publication from Austria demonstrated no differences in operative time or financial costs when comparing LC to RC [14]. RC may actually offer advantages over LC. In agreement with the retrospective study by Rifai et al. [15], a PSM analysis of RC versus LC revealed that despite higher costs, RC is associated with shorter hospital stays and a reduced readmission rate compared to LC [16]. From our personal experience, the robotic platform provides greater visibility, more precise handling of the gallbladder and associated structures, thus facilitating easier dissection than with a laparoscopic approach. This could explain why, in expert hands, the outcomes of RC can surpass those of LC. With any surgical technique used to remove the gallbladder, the golden rules of a safe cholecystectomy must be adhered to at all times. Preparing Calot’s triangle and obtaining the “Critical View of Safety” are essential in every case. Both can arguably be better achieved with the robotic than with the laparoscopic approach due to the larger range of movements allowed by the robotic platform.

Undoubtedly, robotic surgery requires training and practice, and RC offers an excellent opportunity for training in robotic surgery [17]. A recent UK multicentre study analyzed a retrospective sample of 600 RCs performed within the UK Robotic Hepatopancreaticobiliary training program and demonstrated that RC is at least as safe and effective as LC and should be considered an index training procedure [18].

Finally, financial considerations and optimal resource usage should drive or at least influence our decisions. In other words, the ability to perform a procedure does not necessarily mean it should be performed in all cases. A 16-year-old paper from Switzerland based on a small sample size—only 50 cases of RC compared with 50 matched cases of LC—showed that overall hospital costs were higher with RC, with a raw difference of about $1,600 per case [19]. Likewise, the comparative study by Aguayo et al. [10] showed a cost difference of about $2,300. However, as previously mentioned, this study also presents significant biases that limit its reliability. The paper by Gantschnigg et al. [14] reported a much less significant difference between RC and LC of about 350 euros. The analysis of costs of RC versus LC is beyond the aim of this narrative review. The present considerations merely indicate that although RC may currently be regarded as more expensive than LC, the difference is not substantial, and it is likely to decrease over time.

The various papers discussed in this manuscript are summarized in a synoptic table (Table 1). While this table does not claim to be exhaustive of the published literature on the comparison between LC and RC, a complete systematic review is beyond the scope of this paper. The inconsistent literature on this topic has led to diverse opinions on the appropriateness of using technologically advanced (and expensive) tools like surgical robots for treating perceived trivial ailments such as benign diseases of the gallbladder, whether elective or emergency. Clearly, LC remains the gold standard, and most cholecystectomy cases may not require the advanced manipulation capabilities of robotic platforms. However, surgeons should maintain an open mind and consider what is in the best interest of patients. The available literature should be critically evaluated, taking into account possible biases and preconceptions of various studies while expecting a degree of variability depending on study designs and the authors’ opinions. Our paper does not have sufficient evidence to support RC over LC, or vice versa, but it aims to raise awareness about possible biases and the unreliability of some published studies.

Table 1 . Synopsis of the papers analyzed in the present study

Author (year)	Study design	Key finding	Limitation	LoE
Kalata et al. (2023) [1]	Comparative, retrospective, non-randomized RC: 25,084 cases LC: 1,001,004 cases	RC associated with increased risk of BDI No difference in overall 30-day morbidity	Retrospective Included only patients aged 66–99 years Included only patients observed for 23+ hours after operation: 85% emergency operation, 87% acute cholecystitis Unbalanced samples of RC vs LC Higher comorbidity index in RC patients No PSM	3b
Campbell et al. (2023) [8]	Comparative (PSM 1:1), retrospective, non-randomized RC: 10,019 cases LC: 221,239 cases OC: 2,687 cases	Reduced conversion rate with RC vs. LC Longer operation time with RC vs. LC Reduced morbidity and mortality with RC/LC vs. OC No different risk of BDI with RC, LC and OC	Retrospective Included only emergency cholecystectomies Unbalanced samples of RC vs. LC vs. OC Risk of miscoding	2b
Delgado et al. (2024) [9]	Meta-analysis, 13 comparative studies Only RCT and PSM studies included RC: 10,758 cases LC: 11,682 cases	Longer operative time with RC vs. LC No different risk of BDI and overall morbidity between RC and LC	High heterogeneity Also retrospective study included	2a
Aguayo et al. (2020) [10]	Comparative, retrospective, non-randomized RC: 41,518 cases LC: 3152179 cases	RC associated with higher morbidity No different mortality rate RC associated with higher costs	Retrospective Unbalanced samples of RC vs LC Higher comorbidity index in RC patients No PSM	3b
Gantschnigg et al. (2023) [14]	Comparative, retrospective on prospectively collected data, non-randomized (non-systematic randomization by patient choice) RC: 110 cases LC: 110 cases	RC associated with higher costs No different mortality, morbidity, operation time, postoperative stay Higher conversion rate in LC vs. RC	Retrospective Small sample size Not clear patient selection Emergency cases excluded Converted cases excluded from the cost analysis	4
Rifai et al. (2023) [15]	Comparative, retrospective, non-randomized RC: 165 cases LC: 105 cases	No different operative time between RC and LC Longer length of stay with LC Higher conversion rate with LC	Retrospective Only one surgeon performed RC, while two surgeons performed LC Only emergency cholecystectomy considered Possibly more difficult cases underwent LC	4
Kane et al. (2020) [16]	Comparative (PSM 1:10), retrospective, non-randomized RC: 106 cases LC: 3,149 cases (1,060 after PSM)	RC associated with longer operative time and higher costs RC associated with shorter length of stay and lower readmission rate	Retrospective Morbidity rate not analyzed Preoperative diagnosis and presentation not considered	2b
Breitenstein et al. (2008) [19]	Comparative (PSM 1:1), retrospective, non-randomized RC: 50 cases LC: 50 cases	RC associated with higher costs No different conversion rate, mortality, morbidity, operation time, postoperative stay	Retrospective Small sample size	4

Studies are listed in the order they are mentioned in the text. Only recent studies were considered.

RC, robotic cholecystectomy; LC, laparoscopic cholecystectomy; LoE, level of evidence; PSM, propensity score matching; BDI, bile duct injury.

Current evidence suggests that RC is safe and effective and may offer some advantages over LC, but it also requires a learning curve, like any new technique or procedure. Further research, particularly through RCTs, is necessary to draw definitive conclusions on the safety and efficacy of RC compared to LC. RC is a breakthrough in minimally invasive surgery, providing an opportunity to familiarize with the robotic device and enhance surgeons' skills in preparation for more complex robotic operations. The robotic approach can be advantageous in selected cholecystectomy cases where finer dissection is required, such as in acute cholecystitis (see Supplementary Video) or when exploring the common bile duct is necessary. With ongoing reductions in the costs of robotic platforms and disposable equipment, RC could become the gold standard for benign gallbladder disorders, in both urgent and elective scenarios, as was the case with the introduction of LC. Dismissing it on the basis of low-level evidence is reminiscent of the cold reception Erich Mühe received after introducing his pioneering LC approximately 40 years ago [20].

Supplementary data related to this article can be found at https://doi.org/10.14701/ahbps.24-192.

None.

No potential conflict of interest relevant to this article was reported.

Conceptualization: GDT, DC. Data curation: PPB, GBS, AC, FP, GP. Methodology: GDT, GBS, DC. Visualization: GDT, GBS, DC. Writing - original draft: GDT, GBS, DC. Writing - review & editing: All authors.