Background

Primary ventral hernia repair (PVHR) is a very common general surgery procedure, with 270′000 performed annually in the United States [1]. Standard open PVHR approach with pro-peritoneal or retro-muscular mesh placement is associated with rate of surgical complications up to 25% led by surgical site infections (SSI), and 6 months recurrence rate up to 13% according to large scale, multicenter or nationwide studies [2], [3], [4]. Yet often benign, open PVHR complications lead to significant re-admissions and follow-up costs [5]. Alternatively, intra-peritoneal onlay mesh (IPOM) laparoscopic approach for PVHR reduces SSI rate [6] and has gained popularity for incisional ventral hernia repair, especially in overweight patients [7]. For routine PVHR and small sized hernias, however, laparoscopic PVHR with intraperitoneal mesh placement is controverted. Intraperitoneal meshes are known to cause adhesions and potentially severe complications, port-site hernia may occur, primary defect closure can be challenging and is not systematically performed despite its proven effect in reducing seroma and hernia-site events [8], [9], [10]. In addition, poorer results can be observed after laparoscopic PVHR compared to open repair in terms of quality of life, long-term pain and functional disabilities, which may be explained by use of tackers to hold the mesh in place [11]. Laparoscopic transabdominal pre-peritoneal (TAPP) PVHR technique has been described and may address complications related to intra-abdominal mesh placement [12]. It combines the advantages of both open (no use of tackers, no intra-peritoneal material) and laparoscopic (minimally invasive access reducing SSI risk) approaches, but remains technically demanding and time consuming even in expert hands, especially if defect closure is performed.

The daVinci Xi robotic system (Intuitive Surgical, Sunnyvale, CA, USA) helps to overcome technical limitations such as difficulty to perform upside-down suture or precise dissection in limited workspace. Moreover, this device allows using only 8 mm trocars; reducing concerns about port site hernia, which occurs on 12 mm ports in about 2.5% laparoscopic PVHR [8]. Data from recent retrospective studies of TAPP robotic PVHR established the feasibility and safety of robotic approach for ventral hernia surgery [13], [14]. The authors underlined its potential advantages over standard laparoscopic approach in terms of recurrence rate, surgical site complications and ability to perform primary defect closure [15], [16].

To date, no study investigated in a prospective randomized design the performance of robotic PVHR compared to open PVHR.

The present study protocol aims to determine if robotic PVHR reduces surgical site complications when compared to standard open technique, and if its cost-benefit balance is favorable.

Intervention and control

Randomization, allocation concealment and blinding

Fixed block randomization of blocks of 6 participants, on a 1:1 basis, will be generated using the RedCap software system. Block randomization is necessary for logistic reasons (management of operating room schedule, as the use of the robot needs to be spread throughout the year). As this study is single blinded, and as the patient doesn’t have access to the operating room schedule and decisions made for previously included patients, there is no need for a varying block size randomization method.

Patients will be blinded from the chosen technique pre-operatively until post-operative day two. To ensure blindness, opaque wound dressing will be applied both on surgical wound(s) and on spot(s) where wounds would have taken place with the other procedure (open PVHR or robotic PVHR). Blindness will be maintained until wound dressing changing or removal at post-operative day two.

Definition of endpoints and outcome measures

The primary outcome will be the incidence of surgical site complication within 30 days after surgery. Surgical site complication is defined according to Clavien and Dindo (31) as any deviation from the normal postoperative course, and classified as follow:

• •  Grade 1: minor risk events not requiring therapy (with exceptions of analgesic, antipyretic, antiemetic, and antidiarrheal drugs or drugs required for lower urinary tract infection). For example, simple hematoma or abscess needing only nursing wound care.
• •  Grade 2: potentially life-threatening complications with the need of intervention or a hospital stay longer than twice the median hospitalization for the same procedure in a similar patient (1 day without comorbidity, 3 days with multiple or major comorbidities). Grade 2 is divided into 2 subgroups based on the invasiveness of the therapy selected to treat the complication; grade 2a require medications only, and grade 2b require invasive procedure.
• •  Grade 3: complications leading to lasting disability or organ resection. For example intestinal resection for perforation or major tissue loss due to severe wound infection.
• •  Grade 4: death of a patient due to a complication.

Primary outcome assessment will be based on physical examination of surgical wounds and former hernia site. Imaging techniques (ultrasound or computed tomography) may be required to confirm diagnosis, if clinically relevant (no imaging exam will be performed for the sole purpose of the study).

Follow-up period will be of 30 days after surgery and the study will end when the last included patient will complete this 30 days follow-up.

Secondary endpoints

Secondary outcomes will be:

• •  detailed surgical site complications
• •  general complications (not directly relate to surgical site, Clavien-Dindo’s scoring system)
• •  pain (Visual Analogue Scale (VAS), at every post-operative visit)
• •  painkillers consumption (recorded from patient’s medical record for in-hospital stay and, for out-hospital period evaluated with consumption recall at each visit)
• •  esthetic satisfaction (qualitative appreciation by the patient, at wound dressing removal and at last visit)
• •  quality of life (EuraHS-QoL, at lwound dressing removal, at threads removal and at last visit)
• •  in-hospital costs (hospitalization costs including medications, care and labs, cost of the procedure, diagnosis related group class, total amount billed to the insurance)
• •  out-hospital costs (derived using REKOLE method)
• •  early recurrence at 1 month (physical examination ± imaging)

Safety outcomes will be assessed in patients undergoing robotic PVHR :

• •  rate of device related adverse events
• •  rate and nature of robotic system malfunctions

Data collection and management

A member of the study team will identify any eligible patient undergoing primary umbilical or epigastric hernia repair during pre-operative appointment. A member of the study team will give the patient the study information leaflet and discuss it. The patient will have an opportunity to read through the leaflet and ask any questions relating to the study. If the patient wishes to participate, a written agreement will be signed. Patients who do not wish to participate may also be offered robotic PVHR (as well as open PVHR or IPOM), as this technique is now part of the surgical armamentarium of our Division to treat ventral hernias. Included patients will be randomized using the RedCap system. Data of every included patient will be filled in an anonymized case-report form using the RedCap system.

Missing data will thus be managed depending on the cause of missingness. If missingness may directly or indirectly be related to the treatment, it may cause a bias even in intention-to-treat analysis. In such situation, the following measures will be taken: reporting of the missing data in the final report, missing data in intention to treat analysis will be analysed according to the worst case scenario method (failure)., Pre-operative drop-outs will be replaced by new patient inclusion and will thus not lead to attrition bias. Post-operative drop-outs are unlikely to happen as post-operative study evaluations and schedule do not differ much from standard post-operative care. Eventual post-operative patients’ drop-outs, however, are taken into account in the power calculation.

All study data will be archived for a minimum of 10 years after study termination or premature termination of the clinical trial.

Statistical analysis

All study practices and statistical methods are based on the International Conference on Harmonization (IHC) document “Statistical Principles for Clinical Trials.”

Baseline characteristics and safety outputs will be summarized overall and by intervention group.

In summary tables variables will be presented as follows. Continuous variables, the minimum and maximum values, the arithmetic mean and standard deviation will be presented to the same number of decimal places as the original data. Qualitative variables, absolute frequencies and percentages will be used. The denominator for each percentage will be the number of subjects within the population treatment group unless otherwise specified.

Chi2 test (or exact Fisher test when expected effectives are lower than 5) will be run on binary and categorical outcomes such as primary outcome (rate of surgical site complication). Student’s t test will be run on continuous variables if they are normally distributed. With 80 patients per group, according to the central limit theorem, t test is applicable even if the data are not normally distributed.

Analysis will be carried in intention to treat.

Subgroup analyses are planned for the following subgroups: weight groups: obese (BMI ≥ 30 kg/m²) / non-obese (BMI < 30 kg/m²), defect size groups: ≤2 cm/>2 cm.

All hypothesis testing will be carried out at the 5% (2-sided) significance level. P-values will be rounded to three decimals. P-values less than 0.001 will be reported as <0.001 in tables.

No interim analysis is planned for this study considering low risk of adverse events, short planned duration of the sudy and low risk of surgical failure as the type of repair is the same in both groups (preperitoneal repair with mesh), the only difference being the surgical access, and both surgical accesses are validated and commonly used in abdominal surgery.

Discussion

In the present trial, we expect to demonstrate that robotic PVHR leads to lower incidence of surgical site complication rate than the same procedure performed through standard open approach (open PVHR) while being an acceptable solution from economic, operative time and functional standpoints.

The study seeks primarily to determine the incidence of surgical site complication after robotic PVHR versus open PVHR. Secondary objectives are to assess general complications rate, early recurrence rate, pain, esthetic results and costs.

As a comparator against robotic PVHR, we chose open PVHR with mesh reinforcement, as this constitutes the gold standard procedure, IPOM being usually reserved for obese patients with small defects only. Moreover, functional results between laparoscopic IPOM and robotic PVHR are not likely to be the same as mesh reinforcement is not performed in the same anatomic plane, and defect closure is not performed in IPOM.

Conceptually, both approaches (open PVHR and robotic PVHR) lead to the same hernia repair involving closure of musculo-fascial defect with non-absorbable mesh reinforcement in the preperitoneal plane.

Essential difference comes from the minimally invasive approach to perform the procedure in the robotic group and the trans-abdominal approach. Some risk historically described as related to the laparoscopic approach (intra-abdominal organ lesions and adhesions creation), have to be balanced as they were described in the early days of laparoscopy and on small cohorts, and as a substantial number of open PVHR end up with peritoneal effraction too during hernia reduction. As such, a recent US based national quality database of more than 5000 patients over a 4 years course showed that rate of inadvertent enterotomies during ventral hernia repairs is not significantly different between open (2%) and minimally invasive approaches (1.8%, including robotic and standard laparoscopic cases) (29). The subject of adhesion formation is purely conceptual: any peritoneal penetration is likely to produce adhesions, but open ventral hernia repair induce most of the time an opening of the peritoneal sac, and in that way is not less traumatic for the peritoneum than a laparoscopic approach. More adhesions are seen in the IPOM technique than in open repair but this is not applicable to robotic PVHR technique as IPOM implicates an intra-abdominal synthetic material whereas robotic PVHR places the mesh extraperitoneally. No data, until now, allows expecting more adhesions after laparoscopic (or robotic) TAPP ventral hernia repair than after open PVHR.

Adverse events related to the robotic system itself may occur, but in our experience they are frequently related to mis-knowledge and mis-use of the system. In a recent review (30), it was clearly demonstrated that instrument malfunction occur rarely (13 malfunctions over 10 000 cases), and that they result in complication only in case of inappropriate management of the situation by the operating team. Thus, over more than 150 robotic cases performed last year in the visceral surgery department of the HUG, no complication directly related to system malfunction was observed. In addition, strong experience in using the robotic system and a dedicated surgical team allow dealing with eventual device malfunctions without causing any harm to the patient. To reduce these risks the surgical team, which will perform robotic procedures will be constituted of surgeons trained for this kind of procedure and experienced in using the daVinci robotic system. On the other way, to limit bias related to surgical team experience the following measures are deployed:

• •  at least one surgeon familiar with general surgery but not with the robotic surgery nor robotic PVHR procedure, will be teached to do the procedure under supervision in the robotic group.
• •  although open umbilical hernia repair is a very common procedure learned very early during surgical residency (often the first learned abdominal procedure), all procedures of the open group will be performed or at least supervised by an experienced surgeon.

Using this design, we are as close as possible from the “real life” conditions of surgical practice in a teaching hospital, with practice from trained surgeons as well as from trainees under supervision.

Trial status

Recruiting.

Disclosures

Dr. Douissard received a research grant from the Intuitive Foundation to support this study (http://www.intuitive-foundation.org/clinical-research-grants/). Outside the scope of this work, Dr. Douissard received personal fees from Verb Surgical Inc. and non-financial support from Intuitive Surgical Inc. Dr. Monika Hagen has no conflicts of interest or financial ties to disclose pertinent to this work. Outside the scope of this work Dr. Monika Hagen received personal fees from Intuitive Surgical Inc. and Ethicon Endosurgery and non-financial support from Intuitive Surgical Inc. and Ethicon Endosurgery. Dr. Meyer, Dr. Dupuis, Dr. Peloso, Mrs. Mareschal and Prof. Toso have no conflicts of interest or financial ties to disclose pertinent to this work.

Funding

This study is supported by a research grant from the Intuitive Foundation. More information can be found at the following adress: http://www.intuitive-foundation.org/clinical-research-grants/