Minimal-invasive surgery has become the treatment of choice in many abdominal procedures, including hiatal hernia repair since it is associated with less postoperative pain, easier mobilization of the patient, and shorter hospital lengths of stay. Laparoscopic hiatal hernia repair is a time-consuming and technically demanding procedure, commonly accepted to have two major disadvantages; the lack of depth perception because of the two-dimensional imaging, and limited maneuverability. On the other hand, the robotic-assisted approach offers better ergonomic conditions, a three-dimensional view, better access to the mediastinum, and a wide range of motions, wrist-like movements of the instruments, significantly useful for suturing or mesh placement, but it is also associated with a higher cost. The aim of this systematic review and meta-analysis is to find out whether robotic-assisted paraesophageal hernia repair is superior to conventional laparoscopic in terms of morbidity and operation time, and therefore worth the cost. The PICO criteria were used in order to conduct the scientific question: Is the robotic approach (Intervention) superior to laparoscopic (Comparison) in male and female patients with hiatal hernia, who are older than 18 years (Population), in terms of postoperative complications, length of stay (LOS), operative time and recurrence rates (Outcomes)?

Eligibility criteria

Inclusion criteria

Only comparative studies of robotic and laparoscopic paraesophageal hernia repair in adults were included in the study.

Exclusion criteria

Case reports, small case series, articles not in the English language, as well as articles concerning patients younger than 18 years old were excluded from the study. In addition, articles about other types of diaphragmatic hernia repair like Morgagni, Bochdalek, iatrogenic, and postesophageal, are excluded from this study.

Search strategy

From December 1, 2022 to December 31, 2022, a comprehensive literature search of MEDLINE, Scopus, and the Cochrane Library was conducted, using the keywords robotic AND laparoscopic AND hernia AND (hiatal OR paraesophageal OR diaphragmatic). The evaluation of the studies was performed by two reviewers who worked independently. Cohen’s kappa was calculated at 98.6%, indicating an almost perfect level of agreement [1]. The search strategy algorithm and the data extraction are illustrated in Fig. 1. Most of the studies included in this systematic review are single-center retrospective studies or observational cohorts selected from databases where the data were registered prospectively. No randomized control studies are reported in the literature. For this reason, the risk of bias in literature, unfortunately, exists, especially regarding indication and selection bias. The design of the study was performed according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 guidelines. No automation tools were used. The study focuses on the hiatal hernia repair, not the fundoplication technique.

Statistical analysis

IBM SPSS version 29.0 (IBM Corp.) was used for the meta-analysis and the results are presented in Forest plots. Egger test was used to estimate publication bias, which is illustrated in funnel plots. Any missing results were excluded from the analysis. The ReviewManager (RevMan) calculator version 5.2, a useful tool provided by the Cochrane Library, was used to estimate missing standard deviations in cases of continuous variables. The random effects model was used for continuous or binary variables and the effect was considered statistically significant when the p-value was below 0.05, with a 95% confidence interval (CI). Cohen’s d, the standardized mean difference was used for the effect size of continuous variables. Sidik-Jonkman estimator with Knapp-Hartung adjustment was applied [2].

The overall postoperative complications, the mean operation time, the hospital LOS, the estimated blood loss, and the recurrence rates, are the outcomes intended to be studied in this review. The data collected from the studies included in the review are presented in Tables 1, 2, and 3.

Only six studies provided enough evidence about the overall postoperative complication rates, as illustrated in Table 1.

Ward et al. [3] published the study with the most participants, from the National Inpatient Sample database during the time period 2010–2015, and reported that the complications in the robotic group were significantly higher odds ratio (OR) (1.17; 95% CI, 1.07–1.27), and specifically respiratory failure (OR, 1.68; 95% CI, 1.37–2.05) and esophageal perforation (OR, 2.19; 95% CI, 1.42–3.93), even in high volume centers. On the other hand, Soliman et al. [4] reported that older age and laparoscopic approach were associated with more postoperative complications, but it was not a randomized study. Meta-analysis for categorical data was performed, according to the meta-analysis, there was no statistically important difference considering the overall postoperative complications rates between laparoscopic and robotic hiatal hernia repair. A random effects model was applied; OR, 0.56 (95% CI, 0.28–1.11), without any statistical significance. A p-value for Egger was 0.24, indicating statistically significant publication bias, with considerable heterogeneity. This is explained by the fact that four of the studies included in the meta-analysis are retrospective single-center studies with a smaller number of participants and the rest of the studies are database studies with a much bigger number of participants. Sensitivity analysis could not be performed because multiple studies have similar weights. The results of the meta-analysis are interpreted in Fig. 2A and the funnel plot in Fig. 3A.

Some of the most frequent complications are dysphagia, pleural effusion [5], pneumonia, venous thromboembolism, cardiac failure [3], atrial fibrillation [6], atelectasis, delayed gastric emptying, wound infections [7], thoracic or abdominal infections [6], and mediastinitis.

Regarding intraoperative complications, the most common are bleeding [5], perforation [8], and pneumothorax [7]. The important information missing from the majority of the studies is the type of fundoplication used and whether a mesh was placed or not.

Another area of interest is whether robotic-assisted hiatal hernia repair is related to shorter hospital LOS. Nine out of 10 studies provided data about the LOS, as shown in Table 2.

There was no statistically significant publication bias (p for Egger test = 0.073; 95% CI, –0.913 to 0.550), as shown in the funnel plot (Fig. 3B). According to the meta-analysis, as illustrated in Fig. 2B, there is no statistical significance considering the LOS between the robotic and the laparoscopic approach (t = –1.54, degree of freedom [df] = 8; 95% CI, –0.53 to 0.11; p = 0.16). However, there is a significant heterogeneity (I², 97%), especially between single-center studies with a smaller number of participants and database studies, where a bigger number of participants is included. Sensitivity analysis could not be performed because multiple studies had similar weights.

Although the meta-analysis proved equivalent results regarding the LOS, a considerable number of studies, as shown in Table 2, reported shorter hospitalization in the robotic group. A different adaptation of ERAS (Enhanced Recovery After Surgery) protocols might be the explanation for this fact [9].

Concerning the mean operation time, only six studies provided information about this outcome, as indicated in Table 3.

There is no significantly important difference concerning the operation time between robot-assisted and laparoscopic hiatal hernia repair (t = 1.41, df = 5; 95% CI, –0.15 to 0.52; p = 0.22) (Fig. 2C). Regarding publication bias, p for Egger test = 0.921, so there is no significant important publication bias, as illustrated in the funnel plot (Fig. 3C). However, it should be noticed that the operation time might decrease over time, as for example mentioned by Benedix et al. [5], as the surgeons become more experienced. This notice might consist of a learning curve bias, especially for small-size studies. The surgeons who performed the operations reported by Soliman et al. [4] had little or no experience with robotic surgery. In addition, mesh placement, which might affect the operation time, is mentioned in only two studies [5,8]. Another important factor is that the article reported by Gerull et al. [10] is a retrospective database study, whereas all the other articles included are single-center studies. This fact might have affected the heterogeneity which was significant, up to 82.7%, as well as the funnel plot (Fig. 3C). In order to deal with the heterogeneity, sensitivity analysis was performed and the study published by Gerull et al. [10] was excluded, as the study with the highest weight. Meta-analysis was performed again showing less heterogeneity (I², 58%), but still no statistically important difference (t = 0.3, df = 5; 95% CI, –0.02 to 0.62; p = 0.06) (Fig. 4A, B).

In general, the use of mesh is a very important element for hiatal hernia reconstruction, but unfortunately, few studies provide evidence about this factor. Soliman et al. [4] reported that mesh was used in only four of a total of 293 patients. O’Connor et al. [9] reported that mesh placement was performed more frequently in the laparoscopic group with a p < 0.001, whereas Tjeerdsma et al. [8] found no difference in the use of mesh between the two groups, although in a smaller sample. More studies regarding the use of mesh and its correlation with the operative time, the complications, and the recurrence rates are needed.

Only three studies provided evidence in regard to estimated blood loss. Gehrig et al. [7] stated that there is no statistically important difference between the laparoscopic and robotic groups.

Not enough data were provided concerning the recurrence rates in order to perform the meta-analysis. Only O’Connor et al. [9] stated that the robotic group had a lower radiologic recurrence rate (13.3% compared to 32.8% in the laparoscopic group with a p-value of 0.008; OR, 0.31 [95% CI, 0.17–0.57]) in a 1-year follow-up (Table 2).

Unfortunately, all of the studies included in the meta-analysis are retrospective and none of them is randomized. Due to the considerable heterogeneity, the results of the meta-analysis are not reliable. Further studies and especially randomized control studies are needed in order to reach reliable conclusions.

Robotic-assisted surgery is becoming more and more popular among surgeons since it overcomes the technical difficulties of conventional laparoscopy. Its safety and efficacy have been proven even for oncologic procedures, while its cost has been justified for procedures with limited anatomic space, such as radical prostatectomy, low anterior resection, and bariatric procedures. Especially for the paraesophageal hernia repair, which is a technically challenging procedure, demanding high mediastinal dissection, complete removal of the hernia sac, a low-tension hiatal reconstruction with sutures, with or without mesh placement, robotic technology seems a very useful tool to overcome these difficulties, restricting the need to conversion to open surgery [11,12]. However, according to the presented meta-analysis, the short-term outcomes with regard to operation time, LOS, and postoperative complications are equivalent, whereas there is not enough data in the literature concerning the long-term outcomes and specifically the recurrence rates.

Robotic approach might be very useful for the repair of recurrent or complex hiatal hernias [9], which are accompanied by higher morbidity and mortality as well as less satisfactory symptomatic outcome [13]. Seetharamaiah et al. [14] reported a series of 19 robotic repairs of giant paraesophageal hernias, with only two surgery-related complications, one conversion to open approach and no recurrence in a mean follow-up period of 15.6 months. Taking into consideration that the majority of complications are pulmonary events and pneumonia, the robotic approach enables better visualization and ergonomics during the dissection of the hernia sac from the delicate pleura [9]. In addition, it provides high-quality hiatal reconstruction and suturing even in the reoperative field, which is characterized by changed anatomical planes and demands extensive adhesiolysis. As a result, the surgeons, being aware of the capabilities of the robot, feel more confident, a fact which explains the lower conversion to open procedure rates for redo hiatal hernias [13]. The conversion rate for redo hiatal hernias with the laparoscopic approach might be as high as 11% [15]. Gerull et al. [10] also reported a higher percentage of redo hernias in the robotic group, as well as a lower percentage of esophageal lengthening procedure, such as Collis gastroplasty and wedge fundectomy. Mertens et al. [6] presented a large series of both primary and redo robotic-assisted hiatal hernia repair with major complication rate up to 5.2% and 2.6% respectively, while the incidence of complications of any severity were 17.1% and 10.6%, indicating that the low number of complications in the redo group was not significantly different from the primary procedure group. The presence of strong adhesions and the strangulation of a significant portion of abdominal viscera consist the two main reasons for conversion in the robotic group [6]. Sowards et al. [16] mentioned longer operative times, increased hospital LOS, and the use of mesh in the recurrent group compared to the primary group, with no significant difference in intraoperative complications, estimated blood loss, or postoperative dysphagia, while no conversion was noticed. On the other hand, a single-center retrospective study by Tolboom et al. [13] reported a significant reduction in conversion rates and in hospitalization time in the robotic-assisted redo group, whereas the most common complication was the perforation of the esophagus or the stomach. Nevertheless, previous laparoscopic antireflux surgery does not suggest an indication for open approach in case of recurrence; if a robotic platform is available in combination with an experienced surgical team, a robotic approach is recommended.

Experience from robotic redo hiatal hernia repair after primary robotic procedure is provided by Arcerito et al. [17], who mentioned the increased possibility of conversion to open approach due to the severe scar tissue lying between hiatal crura and fundoplication, which developed more likely from the placement of an absorbable mesh. However, no mesh-related complications are mentioned during a two-and-a-half-year time period [17], thanks to the absorbable property, even in longer follow-up periods [18].

Regarding the learning curve, it is believed that the robotic approach has a shorter learning curve compared to laparoscopy for advanced surgical procedures, while a significant case volume and dedicated operation room staff can significantly reduce the operation time [19]. Galvani et al. [20], in a large cohort of 61 robotic procedures all performed by one surgeon, claimed a learning curve of 36 cases, comparable to conventional laparoscopy. Sarkaria et al. [12] in a series of 24 patients, noticed that the operation time was decreased by 98 minutes between the second and the first half of the series. Washington et al. [21] also reported shorter operative time between the early and the late robotic experience, 184 and 142 minutes respectively, as well as a significant decrease in conversion rates, after only one year of thirteen robotic procedures.

Another topic of interest is the application of the robotic platform in the emergency setting in cases of strangulation of hernia contents. According to a study published by Hosein et al. [22], minimally invasive approaches predominate even in the urgent setting, with better postoperative outcomes, with the open approach being selected only for extremely ill patients. The robotic approach was superior to the open one for mild to moderate ill patients, but not superior to the laparoscopic approach. In another study by Vasudevan et al. [19], 40% of the robotic procedures were performed on patients presenting with acute symptoms and neither higher operative time nor conversions were observed in comparison to elective cases. Arcerito et al. [17] also suggested the robotic approach for the treatment of acutely presented hiatal hernias even on admission day. Equivalent outcomes were noticed even for complete upside-down stomach hiatal hernias [23]. Because of the higher cost, the robotic approach is reserved for giant or revisional hiatal hernias in some centers [22].

Regarding the recurrence rates, there is not enough evidence in the literature concerning the long-term outcomes. Mertens et al. [6] reported two cases of early symptomatic recurrence requiring redo surgery during the 30-day postoperative period. Draaisma et al. [24] reported a low midterm recurrence rate after 1 year of follow-up. Brenkman et al. [25], in a cohort of 40 patients and during an 11-month follow-up period, reported only one symptomatic recurrence (2.5%). He attributed this result to the application of the robotic platform and the Toupet-fundoplication, which attached to the crus, provides further support to the hiatal repair, while no mesh was placed. Galvani et al. [20] in a cohort of 61 patients, where mesh was used in all cases, and during a median follow-up period of 24 months, reported a radiologic recurrence rate up to 42%, pointing out that the majority of the patients were asymptomatic and indicating than the durability of the repair decreases over time. Arcerito et al. [17] also provided promising evidence regarding the long-term recurrence rate during two-and-a-half years of follow-up. In one of the largest studies with paraesophageal hiatal hernia repair by Gerull et al. [11], a radiographic recurrence rate of only 9% during 5 years of observation is reported, in combination with clinically significant quality of life benefits. The presentation of recurrence may not solely depend on the surgical technique but also on the presence of ultrastructural abnormalities in the muscular tissue of the crura in patients with hiatal hernia. These patients often exhibit a high incidence of severe muscular lesions, which are not observed in individuals with a normal gastroesophageal junction [26]. This evidence supports the application of mesh in order to strengthen the hiatal repair and lower the recurrence rates.

Robotic procedures in general are associated with higher costs. Interestingly, Gerull et al. [11] reported that the operation equipment costs were similar between robotic and traditional laparoscopic paraesophageal hernia repair, with a mean difference of only $89. Lekarczyk et al. [27] found similar hospital profits for the robotic group despite higher supply costs and charges. However, Kulshrestha et al. [28], in a study with more participants, reported that both open and robotic-assisted procedures had significantly higher median index hospitalization costs compared to laparoscopic ones. The most common reasons that increased the cost were upper endoscopy and reoperation, followed by emergent priority, increased comorbidity index, and LOS. Hosein et al. [22] also demonstrated that the laparoscopic approach was the least expensive. However, the increased experience of the surgeons with the robotic platform, which will lead to fewer complications, LOS, and postoperative exams, as well as the fact that the cost of the robotic equipment will go down over time, might diminish this difference in the future.

The robotic platform has been used successfully for the repair of other diaphragmatic hernias, such as Morgagni [29,30], Bochdalek [31], and even postesophagectomy hiatal hernias [32], indicating its potential in technically demanding procedures.

Over the last decades, esophageal surgery has evolved from open approaches including both laparotomy and thoracotomy to minimally invasive procedures. The robotic-assisted hiatal hernia repair is superior to the traditional open approach in terms of overall complication rate, mortality [20], postoperative pain, and hospital LOS. However, its superiority over conservative laparoscopy has not been proven yet.

The present meta-analysis did not demonstrate any advantage of the robotic-assisted paraesophageal hernia repair over the conventional laparoscopic approach, which remains the most cost-effective approach. Robotic paraesophageal hernia repair is safe and feasible, but still not superior to laparoscopy, unless in cases with recurrent complex hernias. It is obvious that more comparative and mainly randomized control studies with subgroup analysis need to be performed in order to reach more accurate conclusions and find out which patients would benefit most. A long-term follow-up of the patients is essential because there is a lack of evidence in the literature concerning the recurrence rates of each approach.

Ethical statements

This systematic review and meta-analysis were conducted in accordance with ethical principles and guidelines. As this study is based on literature review, institutional ethics committee approval was not required.

Authors’ contributions

Conceptualization, Investigation, Methodology, Software: SE

Data curation, Validation: GI

Formal analysis: KA

Supervision, Resources: MK

Project administration: KA

Visualization: PP

Writing–original draft: SE

Writing–review & editing: PP

All authors read and approved the final manuscript.

Conflict of interest

All authors have no conflicts of interest to declare.

Funding/support

None.

Data availability

The data presented in this study are available on request from the corresponding author.

Data for postoperative complications

Data for the length of stay, the estimated blood loss, and the recurrence rates

Data for operation time