Follow-up of complex hernia repair with intraoperative fascial traction

Patients from five different high volume certified hernia centers in Germany treated between August 2019 and September 2023 were included. Between August 2019 and April 2021, the fasciotens^® Abdomen device was used for IFT, thereafter the fasciotens^® Hernia system was applied. All patients were treated for midline incisional hernia repair including IFT which was previously described [16, 17]. For the study presented, only cases in which anterior fascial closure at low tension was not possible without using IFT were included for data analysis.

Postoperative routine follow-up was realised within 30 days after surgery. Afterward, patients are encouraged to undergo yearly follow-ups. For this follow-up, study patients were individually invited by each certified hernia center for an additional outpatient consultation.

All patients received a CT scan or MRI for preoperative planning. In 87% of cases, the patients were treated with Botulinum Toxin A (BTA) preoperatively. The injected units and protocols differed by each hernia center. Intraoperatively, an attempt was made to leave the hernia sac intact (if possible) or the peritoneal flap technique according to Malik et al. was performed [19]. Restrained adhesiolysis and retrorectus dissection as described by Rives and Stoppa was performed. In some cases, an additional TAR was carried out. Main causes were a former stoma side, an additional lateral defect or unavailability of a peritoneal flap for posterior sheath closure (especially after incisional hernia following laporostomy). The posterior rectus sheaths were either closed directly or by using peritoneal flap. All patients but one received mesh augmentation in the sublay (retrorectus) position using different types of meshes as described in Fig. 3. IFT was used if a low-tension fascial closure was not achievable for the anterior rectus sheaths. All centers are using a specially designed medical device from fasciotens (Fasciotens Gmbh, Germany) for controlled IFT, which is a combination of a reusable arm attached to the OR table and a single-use traction unit. The cost for the single-use part is comparable to a small to medium sized biological mesh. An increasing number of German private and public health insurance companies are already covering the patient-related cost. To carry out IFT, 12 polyfilamentel surgical sutures (Vicryl^TMPlus, USP 2, Ethicon^®, USA) (6 U-sutures distributed equally on each side) are sutured about 1 cm lateral to the margin of the anterior rectus sheaths on each side with a stitch length of 2–3 cm. The sutures are then aligned crosswise and clamped in the suture retention frame of the IFT device [20]. We used a traction force between approx. 14–20 kg in this cohort (Fig. 1). Traction was carried out for 30 min. Generally, the handling of the device is quite intuitive and the surgeons in our group felt comfortable to use the technique after two proctored cases on average. If a low-tension closure of the anterior sheaths was not feasible after IFT, additional bridging using either part of the hernia sac (peritoneal flap technique) or a further mesh was performed.

During outpatient consultation written consent for data collection and processing as part of the study was given. Prospective data on 30 days postoperative outcome were retrospectively collected individually by each certified hernia center. The minimum follow-up period was 3 months for long-term clinical outcomes. During outpatient consultation, each patient underwent a physical examination of the abdomen by one of the specialized abdominal wall reconstruction (AWR) surgeons in each hospital. Additionally, a standardized dynamic abdominal wall ultrasound (DAWUS) was performed. We established this examination according to the dynamic inguinal ultrasound (DIUS) of groin hernias [21]. The following parameters were assessed using the DAWUS protocol:

In case of the possibility of hernia recurrence patients were referred for a CT scan. Each patient was assessed for patient-reported outcomes (PRO) using the HerQles hernia-related quality of life (QOL) survey [4]. Data analysis was carried out using Minitab^® Statistical Software Version 21.3 (Minitab, LLC; State College, Pennsylvania, USA). Differences between groups were tested using Chi-square or Fisher’s exact test. The level of significance was determined with α = 0.05.

Hernia recurrence as the primary clinical outcome was assessed by physical examination and DAWUS. Patient-related outcomes (e.g., BMI, age), surgery-related outcomes, and 30-day postoperative findings were collected (e.g., length of stay, surgical site occurrence) retrospectively. For standardization, we used the classification for postoperative complications which was proposed by Haskins et al. [22]. Using a standardized protocol for DAWUS abdominal wall properties were collected. The protocol can be found in the supplementary data. Data sets were stored by each hernia center in an individual Excel sheet.

BTA for chemical relaxation of the lateral abdominal wall has become quite popular in recent years. Several studies have shown very positive results, although there are no randomized controlled trials available yet [26‐28]. According to recent studies, reduction of hernia width of about 5 cm and myofascial advancement of approx. 4 cm is achievable per side [27, 29]. In our cohort, 87% of all cases were pretreated with BTA. Four of five hernia centers participating in this study use BTA routinely for midline hernias with a transverse diameter above 8–10 cm. In our study, intraoperative findings on myofascial elongation in each case were not recorded. Therefore, the influence of BTA cannot be differentiated. Nevertheless, all participating hernia centers did not utilize IFT if low-tension closure of the anterior rectus sheaths were deemed possible by the responsible surgeon. Interestingly, BTA had no significant influence on the closure rate of the anterior rectus sheaths.

Bridging of the anterior rectus sheath should absolutely be avoided as it significantly increases the risk of recurrence [30]. We had a primary closure rate for the anterior rectus sheaths of 94% in our cohort which is in line with results from high-volume single center data (92%) and slightly lower than findings from a systematic review (98%) and newly published data from Ghent (98.2%) regarding TAR [31‐33]. However, apart from data from Novitsky, which were included in the systematic review, and Zolin, the defect width was considerably smaller in the other studies. Recently, a cadaveric model has shown that IFT provides comparable medialisation of the anterior rectus sheath compared to established component separation techniques [34]. If low-tension closure of the anterior rectus sheaths could not be achieved after IFT, the peritoneal flap technique or mesh for bridging was used [19]. Interestingly, performing a TAR has not had a significant impact on the anterior sheath closure rate in our data, which is in line with findings that the TAR is less effective than anterior CS in terms of anterior sheaths approximation [35, 36]. In conclusion, IFT shows similar fascial closure rates compared to component separation techniques but avoids extensive preparation.

Concerning the posterior rectus sheaths, cadaveric studies have shown superiority of the TAR over the anterior component separation concerning medialization [35, 36]. However, there is an ongoing debate about whether direct closure of the posterior sheaths is also necessary. To our knowledge, there are no studies showing an advantage of direct closure of the posterior sheath over bridging. Additionally, an older biomechanical analysis found that the posterior rectus sheaths are significantly less resistant to bursting than the anterior rectus sheaths [37]. In our opinion, utilising retrorectus mesh augmentation using a synthetic mesh always needs to be combined with a closed layer between the abdominal contents and the mesh to avoid adhesions and potential bowel erosion. If direct closure of the posterior sheaths was not possible, we performed the peritoneal flap technique (some call it closure, some call it bridging), which (as a standalone technique) has shown favourable outcomes regarding SSO and recurrence rates [19, 38].

The use of IFT in abdominal wall reconstruction of midline incisional hernias is normally combined with a retrorectus dissection as described by Rives-Stoppa [39, 40]. The resulting space is used for sublay mesh augmentation. Therefore, the median mesh width appears relatively low with 20 cm in our cohort. Since the first description of posterior component separation (PCS) by Carbonell in 2008, a large mesh overlay has been advocated for the repair of complex abdominal wall hernias. Interestingly, Carbonell only mentioned the possibility of using a larger mesh if a PCS is carried out without providing evidence that this is truly necessary [41]. To our knowledge, there is no study to date that shows a reduction in the recurrence rate with the insertion of a larger mesh in open ventral hernia repair [42]. Like other authors, we believe that restoring and closing the myofascial ring of the abdominal wall is a key step in ventral hernia repair, perhaps more than having a large mesh overlap [38]. However, performing PCS and especially TAR results in an extended plane up to the spine which allows the placement of a large mesh. This is undoubtedly necessary if an additional lateral defect is present. In our clinical experience, once this space has been created, a wider mesh is usually inserted regardless of the circumstances. Also in our cohort, when an additional TAR was performed, a significantly wider mesh was used (p = 0.012; α = 0.05). Therefore, the defect-to-mesh ratio was higher if an additional TAR was performed (1.97 vs. 2.51 respectively). In addition, some authors of this study performed a TAR in selected cases and inserted a larger mesh to avoid possible hernia formation when a former stoma site was present. Interesting results regarding mesh strength and dislocation were reported by the group of Kallinowski from Heidelberg. They developed a model called dynamic intermittent strain (DIS) to simulate coughing by applying short impacts of 250 mmHg on a piece of the mesh using a pig tissue model [43]. Following the DIS concept, to bridge a gap of 5 cm in diameter, an overlap of 5 cm in all directions is sufficient if a DIS class A mesh is used. If the peritoneum/posterior sheath is closed, no mesh dislocation is seen. The overlap could even be reduced to 2.5 cm [44]. The Dynamesh Cicat^® mesh (FEG Textiltechnik, Aachen, Germany) which was used in 92.9% of all cases in our cohort was defined as a DIS class A mesh. We therefore consider that an excessive overlap is not mandatory and a TAR is not needed if the lateral abdominal wall is left intact and the midline restored. In our study, the recurrence rate was not different in the TAR group (1/28) with a larger overlap and the IFT group (1/72) with Rives Stoppa only.

For data collection, we conducted a retrospective analysis of the 30-day postoperative outcome. We found an overall SSO rate of 33% and an overall SSI rate of 9%. The majority of these SSOs were seromas (54.5%). Remarkably, both the SSO rate and the seroma rate, was significantly higher if an additional TAR was performed (26.4% vs. 50% and 12.5% vs. 32.1%; p = 0.024 and p = 0.008; α = 0.05). 16 patients needed an intervention either as aspiration and drain of seroma/hematoma or as wound revision. And again, the rate of intervention was higher if an additional TAR was performed (25% vs. 12.5% respectively). In general, the SSO rate of 33% appears to be higher than previously published data for IFT [16, 17]. However, more recent data have shown similar results [34]. One explanation is the relatively obese patient cohort with a mean BMI of 31.3 kg/m². A retrospective review of the ACHQC has shown that the risk for complications significantly increases when the BMI exceeds 30 kg/m² [45]. Furthermore, 54.5% of all SSOs in our patient cohort were seromas. Since all participating hospitals are specialised hernia centers, patients are closely followed up by out-patient appointments within 30 days after surgery including physical examinations and routine ultrasound. Hence, even small and clinically silent seromas were meticulously recorded. Some authors suggest that seromas that do not cause any symptoms or lead to further complications should possibly not be considered complications or SSOs at all [46, 47]. In addition, seroma formation can be found in up to 100% of cases after laparoscopic IPOM for incisional hernia repair using ultrasound examination [48]. Morales-Conde has even suggested a classification solely for seromas and categorising them into non clinical seroma, incident or complication [49]. Looking at our data, the overall rate of SSO would decrease to 20% if only the seromas requiring intervention were counted as SSO. In addition, there was not standard for the treatment of seromas and the thresholds for seroma aspiration were different in the participating hernia centres.

Regarding open TAR, a recent systematic review and meta-analysis have found an overall complication rate of 33.34%, an intervention rate of 9.82%, and a SSI rate of 9.13% in twenty-two studies [50]. Oprea et al. found an overall wound morbidity rate of 28.63% and a SSI rate of 10.6% in a systematic review of anterior and posterior component separation [51]. A systematic review for giant incisional hernias (as defined in a hernia width above 10 cm) found a SSO rate of 21.4% for open anterior CS and 23.7% for TAR respectively [5]. Apart from the latter, the SSO and SSI rates are consistent with our results for IFT. Additionally, intervention rates of 9.82% and 8.36 have been reported for TAR, which is slightly lower than the overall intervention rate in our cohort [50, 51]. However, it is noteworthy that all systematic reviews had heterogeneity regarding the definition of SSOs and intervention in the reviewed studies which could lead to different rates. Interestingly, like the SSO rate, the SSI rate was higher if an additional TAR was performed in our cohort but not statistically significant (10.7% vs. 8.3% respectively). In our cohort, when IFT was combined with TAR, the risk for wound morbidity increased. This is in line with results from a propensity-scored matched study by Marturano comparing BTA and CS showing no difference in fascial closure rates but higher incidence of SSO when performing a CS [52].

Different tools and measurements are used to detect hernias or assess possible recurrence after hernia repair. Clinical examination is the easiest and probably oldest method to diagnose a hernia/recurrent hernia. Regarding recurrences, it has a reported sensitivity of 75% and specificity of 90% [7]. The most common diagnostic tools are probably CT scans or MRIs. Interestingly, even though most authors and clinicians are using CT scans for preoperative planning in ventral/incisional hernia repair the sensitivity and specificity are on the same level as clinical examination (79% and 94%, respectively) [53]. In the authors’ practice ultrasound has a great value and is used daily. It was shown that dynamic abdominal wall ultrasound has a sensitivity of 98% and specificity of 88% for incisional hernias and is therefore not inferior to a CT scan [7]. Since the EHS either recommends ultrasound or a CT scan for follow-up, we decided to combine physical examination and dynamic abdominal wall ultrasound (DAWUS) using a specially designed protocol (supplementary data) [54]. Hence, we avoided exposure to radiation. A recurrence was detected in two patients. They received a CT scan afterwards, which confirmed the recurrence. Both patients were treated surgically.

Utilising physical examination and a standardized ultrasound protocol, we found a recurrence rate of 2% at a mean follow-up time of 19.6 ± 10.7 months. Novitsky found a recurrence rate of 3.7% after a mean follow-up of 31.5 months in his single center TAR cohort, which is in line with a systematic review by Wegdam et al. from 2019 and Oprea et al. from 2023 (4% after 2 years and 6.11% after minimum 1-year follow-up respectively) [31, 51, 55]. However, the group of Rosen recently found a composite recurrence rate of 26% for TAR after a median follow-up of 2 years in their patient cohort [33]. It is noteworthy that in our cohort a recurrence occurred in two cases, whereas one patient was initially treated with an additional TAR and the other had IFT and retrorectus dissection.

Surgeons often focus on clinically measurable outcomes like SSO and recurrence rates. For patients, an even greater significance is quality of life (QOL), including mental health and other areas of well-being [56‐58]. Several attempts have been made to measure QOL, especially concerning the function of the abdominal wall in hernia patients with the Carolina Comfort Scale (CCS) and HerQles being the most sophisticated and established [59, 60]. This is the first study to investigate QOL in ventral hernia repair using IFT. We used the HerQles questionnaire as CCS focuses more on mesh sensation. In our cohort, we found a mean HerQles summary score of 68.5 ranging from 0 (worst) to 100 (best). This is compatible with a large survey of 1817 patients with ventral hernia repair from the Abdominal Core Health Quality Collaborative (ACHQC) that showed a mean HerQles score of 74.9 1-year postoperatively. However, it should be noted, that all types of surgical approaches were included (open, laparoscopically, robotically) in the ACHQC analysis. Regarding TAR, other retrospective analyses have shown higher HerQles summary scores (82 and 91.8 respectively) [61, 62]. However, these studies presented single-surgeon resp. single-center data with a short follow-up of 6 months and one year respectively. A limitation of our study is the lack of a preoperative assessment of QOL which unfortunately does not allow any conclusions to be drawn about postoperative improvement.

All patients in our cohort answered the HerQles questionnaire only postoperatively during follow-up but not before surgery. It is therefore not possible to compare QOL pre-and post-operatively. Furthermore, our standardised dynamic ultrasound, although seeming feasible for follow-up, needs further validation by comparing the results with CT scan or MRI. Additionally, since hernia recurrences can occur even after many years, a longer follow-up period for patients treated with IFT is needed. Finally, comparative studies with other techniques like component separation are needed to determine which technique is most suitable for the surgeon’s intended goal and for each patient individually.