Introduction

The use of prosthetic mesh to augment suture repair of large paraoesophageal hernias is widespread but controversial. Several meta-analyses have demonstrated that crural reinforcement with non-absorbable mesh increases the durability of repair of a large hiatus hernia [1,2,3,4,5]. More recent meta-analyses that include randomised controlled trials only have shown no durability benefit when TiMesh is used [6,7,8]. Surgeon surveys in the USA and Europe indicate preparedness by surgeons to utilise both absorbable and non-absorbable mesh to augment hiatal repair [9,10,11]. The risk of mesh-related complications provides an additional tempering influence on the universal uptake of mesh augmentation. Mesh-specific complications appear infrequent but may be catastrophic [12,13,14]. Our aim was to identify the risk of mesh-specific complications from a large series of consecutive patients undergoing hiatal hernia repair augmented with a lightweight polypropylene mesh (TiMesh) (PFM Medical, Koln, Germany) over a 12-year period. We hypothesise that mesh augmentation of sutured hiatal repair is possible to be performed with acceptable short-term mesh-specific complications.

Patients and Methods

A single unit case series review of patients undergoing hiatal hernia repair was undertaken. For the purposes of our ongoing audit, patients were divided into (1) elective paraoesophageal hernia repair, (2) emergency paraoesophageal hernia repair, (3) elective fundoplication, and (4) revision hiatal repair. All patients undergoing prosthesis-reinforced hiatal repair with TiMesh between February 2005 and October 2017 were identified. Pre-operative, intra-operative, and post-operative data were collected for all patients undergoing hiatal repair. Patients were routinely followed up at 1 month and as required thereafter. Delayed complication rates and outcomes of interest were identified on a review of case notes. The primary outcome of interest was the rate of mesh-specific complications, namely mesh fixation–related injuries, mesh erosion, oesophageal or hiatal strictures and stenosis, and mesh explantation. Routine endoscopic and radiological follow up was not performed in this cohort. Patients were seen in our clinic 1 month post-operatively and then as clinically required. No long-term symptomatic or quality-of-life data is presented. Categorical variables were presented as counts and percentages, and continuous variables were expressed as a median and range. This study was approved by the Northern Sydney Local Health District Human Research Ethics Committee (2022/ETH02380).

Operative Technique

Our operative technique has previously been reported [15, 16]. All operations were performed by the senior authors or trainees under their supervision. The hernia sac was dissected from its mediastinal attachments using ultrasonic shears (ultrasonic coagulating shears, Ethicon Endosurgery Inc., Cincinnati, OH, USA). A posterior sutured hiatoplasty was performed with interrupted polyester sutures (0 Ethibond Ethicon, Johnson & Johnson, Cincinnati, OH, USA) placed 1 cm apart. In cases where the hiatal aperture was particularly large, the hiatal pillars were attenuated, or hiatal closure was performed under tension, and a posterior onlay mesh reinforcement was performed. Onlay mesh reinforcement was performed with TiMesh (PFM Medical, Köln, Germany) and fixed with helical screws (ProTack 5 mm; Covidien, Mansfield, MA, USA). The mesh was cut to size. Typically, the configuration is a rectangular piece measuring 3 × 4 cm with a shallow concavity superiorly to confirm with the hiatal aperture. The mesh was fixed to the hiatal repair, ensuring that the mesh did not come into contact with the posterior aspect of the distal oesophagus (Fig. 1). The mesh did not extend anteriorly beyond the posterior cruroplasty. Fixation screws were, therefore, not placed anterior to the hiatal aperture. A posterior partial (Toupet) fundoplication was then performed routinely.

Fig. 1
figure 1

TiMesh reinforcement of hiatal suture repair

Full size image

Results

A total of 393 patients underwent TiMesh augmented hiatal repair between February 2005 and October 2017. Patient characteristics are summarised in Table 1. The median age for patients undergoing paraoesophageal hernia was 71 years of age (range 31–93), with a preponderance of women. Patients undergoing fundoplication for gastro-oesophageal reflux were younger, with a median age of 53 (range 16–83). The predominant indication for the use of TiMesh was for paraoesophageal hernia repair. Mesh was used less frequently in patients undergoing hiatal repair and fundoplication for gastro-oesophageal reflux.

Table 1 Summary of patient characteristics
Full size table

There were no mesh-specific complications in the perioperative period, and particularly, no instances of fixation-associated injury. In this series, seven patients underwent up front open repair, six who had previously undergone open hiatal repair and one patient with an acute gastric obstruction. Four patients underwent laparoscopic converted to open repair: two who had undergone previous open hiatal repair, one converted for splenic capsular bleed, and another obese patient due to difficult posterior hiatal access. Mesh was explanted in one patient (1/393, 0.25%) who underwent emergency paraoesophageal hernia repair complicated by sepsis following an oesophageal leak. Mesh erosion was incidentally found in two patients (2/393, 0.51%) at 3 years (41 months) and 9 years (115 months) following surgery, respectively. Both patients underwent endoscopy at other institutions, one for investigation of iron deficiency anaemia and the other for post-fundoplication symptoms. Neither patient required mesh explantation, and both remain asymptomatic. No cases of oesophageal or hiatal strictures were identified. Perioperative complications are presented in Table 1. There were two perioperative deaths. An 88-year-old woman undergoing elective repair with Ti-Mesh suffered an oesophageal injury at the cardioesophageal junction due to traction with a nylon sling. This was recognised and repaired at the time of surgery. The patient underwent a thoracotomy for an empyema presumably related to an ongoing leak from the repair. The patient refused further active treatment and succumbed. An 82-year-old woman underwent repair with Ti-Mesh for acute gastric volvulus. The patient returned to the operating theatre 12 days postoperatively for gastroscopy to investigate postoperative dysphagia. The patient failed to regain consciousness and ultimately succumbed to a basal ganglia infarct.

Discussion

The increasingly widespread use of prosthetic mesh to reinforce hiatal repair has been demonstrated in recent surveys of American and European gastrointestinal surgeons [7,8,9]. These studies have also highlighted the large diversity in practice and approach to using prostheses at the hiatus, where the indication for use, mesh type, configuration, and placement technique vary. The advantage of prosthetic mesh in preventing hiatal hernia recurrence is not conclusively established. Five meta-analyses examining mesh-reinforced hiatal repair versus suture-only repair demonstrated a lower recurrence rate in patients undergoing a non-resorbable mesh-reinforced repair at short- to medium-term follow up [1,2,3,4,5]. However, more recent meta-analyses involving randomised controlled trials have not shown this benefit [6,7,8]. A recent systematic review and meta-analysis of randomised controlled trials demonstrated no benefit of mesh augmentation with respect to recurrence rates, both in the short and long term. Quite reasonably, the authors conclude that ‘as both techniques delivered comparable clinical outcomes, a suture technique for primary hiatus hernia repair is simpler and should be recommended’ [6]. It is worth noting that the prosthetic materials used in the included studies varied, and some may no longer be in widespread use. This is particularly the case with expanded polytetrafluoroethylene (ePTFE), which was shown to decrease recurrence rates in one randomised trial [17]. There are two prospective randomised trials comparing TiMesh augmentation to primary closure [18, 19], with 5-year follow up recently published for one trial [20]. Neither study reported a benefit in recurrence rate with TiMesh repair compared to primary suture repair.

Reports of complications with various types of mesh and fixation techniques illustrate that mesh repair at the hiatus is not without risk. Cardiac vascular injury may occur at the time of mesh fixation, resulting in bleeding or cardiac tamponade [20,21,22,23]. Mesh erosion may require foregut resection, which may be technically difficult and result in compromised patient quality of life and nutrition [12,13,14, 24, 25]. It must be noted that various mesh types have different physical characteristics that influence tissue incorporation and propensity to erode [26], and not all case reports of complications identify the type of mesh utilised. While mesh-related complications appear to be rare, their true incidence is unknown. Complications such as these probably suffer from natural publication and reporting bias against poor outcomes [27]. Secondly, complication reporting has primarily been the focus of case reports and small case series, which lack denominators through which incidence can be calculated.

In a survey of European surgeons, approximately one-third of the 165 respondents had seen one or more complications related to mesh use in hiatal repair during their careers [11]. From a survey of SAGES members, the authors propose the mesh-related complication rate to be approximately 2 to 4 patients per 1000, reporting an incidence for erosion and strictures of 0.27% and 0.20%, respectively [9]. The authors recognise this should be taken only as an estimate of real-world results as the respondents’ data could not be verified, but it does appear to align with other reports of mesh complications. A systematic review of the literature calculated the incidence of oesophageal erosion and dense fibrosis causing oesophageal stenosis as 0.2% and 0.5%, respectively [2].

Our reported rate of mesh erosion of 0.51% appears to be consistent with previously reported rates of mesh complications. In both cases, the patients were asymptomatic and did not require mesh explantation. We believe our erosion cases reflect inadequate fixation of the mesh to the diaphragm, resulting in incomplete incorporation. Our single case of explantation occurred in a patient who had undergone laparoscopic repair of an acute hernia. A subsequent laparotomy was performed to address an unrecognised oesophageal perforation, and at that time, the mesh was explanted. There were no cardiac or vascular injury related complications relating to fixation screws in this series. Cardiac injuries are likely to be avoided if mesh is placed posteriorly over the hiatal repair, obviating the need to place screws anteriorly.

The most obvious limitation in a study such as this is the lack of long-term symptomatic and objective follow-up with either contrast study or endoscopy. Routine anatomical follow-up with either of these modalities outside the context of a clinical trial on a cohort this large would prove challenging from a logistical viewpoint. With that in mind, we are unable to report rates of asymptomatic hernia recurrence or mesh erosion.

Conclusion

Whether there is a role for non-absorbable mesh in repairing hiatal defects remains to be seen. In the absence of supportive data, the use of suture repair should remain the standard of care for large hiatus hernias. Whether particularly large defects or defects closed under tension stand to benefit from mesh augmentation should be the subject of further studies. We have used TiMesh lightweight polypropylene mesh as our standard hiatal augmentation for selected cases since 2002. While there are various configurations and materials used in mesh-augmented repair, we have limited our use of mesh over the study period to a posterior onlay repair using TiMesh. From our large series, albeit without routine endoscopic and radiological follow-up, we demonstrate, we believe, acceptably low rates of mesh-related complication. Two cases of asymptomatic mesh erosion were incidentally identified in the course of 393 TiMesh repairs, and mesh was explanted in one patient. For the purposes of patient consent and ongoing discussion, we report the risk of mesh erosion and mesh explantation to be 0.51% and 0.25%, respectively.