Gastrogastric herniation: an unusual complication following greater curve plication for the treatment of morbid obesity

Abstract

Greater curve plication is an emerging procedure for the treatment of morbid obesity. A median weight loss of up to 61% at one year has been reported in initial reports. Thus far, operative morbidity is low and there is no reported mortality. We present a case of gastric herniation after greater curve plication. Severe nausea and vomiting occurred in our patient with an excessively tight greater curve plication. Two gastric hernias developed through the plication suture. Surgical reduction of these hernias and revision of the original procedure was required. We recommend that greater curve plication is performed over a bougie and that two rows of closely spaced interrupted sutures are used to secure the plication.

The surgical treatment of morbid obesity, as part of a multidisciplinary approach, is the only current therapy that results in long-term weight loss.^1–3 However, substantial improvements in obesity related co-morbidities are seen with small amounts of sustained weight loss. Most existing bariatric procedures produce a median five-year excess weight loss (EWL) of at least 50%.^1,3 Current bariatric procedures induce weight loss by intake restriction (adjustable gastric banding, sleeve gastrectomy), malabsorption (biliopancreatic diversion) or elements of both with the Roux-en-Y gastric bypass (RYGB).

Vertical gastric plication is an emerging bariatric procedure. Laparoscopic invagination of the greater curve of the stomach markedly reduces gastric capacity, causing weight loss.⁴ The purported benefits of this procedure are the absence of gastric transection, the absence of a permanent prosthesis, the potential reversibility and the relative economy compared with existing alternatives.^5,6 Both anterior gastric and greater curve plication (GCP) have been reported in human and animal studies.^7,8 However, greater volume reduction and weight loss is seen with GCP.^5–7

Initial reported series of GCP are promising.^5–8 The technique is straightforward and there is minimal morbidity and no reported mortality.^5–7 Patients experience appetite reduction, food intake restriction and early satiety, and weight loss is relatively rapid.^5–7,9 Talebpour and Amoli describe a reduction in the functional volume of the stomach to 50ml after plication with painful gastric distension discouraging further oral intake after this point.⁶ The median one-year EWL lies between 53.4% and 61%.^5–7,9

We have selectively used GCP in the management of morbid obesity and report a novel complication arising after this procedure.

Case history

A 58-year-old woman was seen with failed weight loss after gastric banding. Her pre-band body mass index was 38.5kg/m² and at consultation it was 36kg/m². Obesity related co-morbidities included diabetes and ischaemic heart disease. Our standard practice is to offer these patients conversion to RYGB but she declined a stapled procedure after a discussion about the risk profile. In lieu of other options, the patient was informed about GCP and its role in weight loss. The potential of combining GCP with gastric banding^9,10 was also discussed. Possible risks and the emerging role of gastric plication were highlighted. The patient elected to proceed.

At operation, the peri-band gastroplasty was undone and the greater curve mobilised before being plicated from below the existing band to 6cm from the pylorus with a single row of interrupted non-absorbable monofilament sutures (2/0 Novafil™, Covidien, Dublin, Ireland). No bougie was used.

Post-operatively, the patient experienced severe nausea and vomiting and was treated with a regular gastrokinetic agent (metoclopramide; Valeant, Mississauga, Ontario, Canada) and a selective serotonin receptor antagonist (on-dansetron; GlaxoSmithKline, London, UK). The glucocorticoid dexamethasone (Phoenix Pharmaceuticals, Burlingame, CA, US) was used as an antiemetic and to minimise oedema in the intraluminal fold of stomach.

In contrast to our previous experience with plication, these symptoms persisted for over 48 hours and so upper gastrointestinal endoscopy was arranged. At endoscopy, grade 1 oesophagitis was noted. The plication was tight (and tightest in the region of the incisura) but it could be traversed with the endoscope. Despite a feeding tube being placed, the nausea and vomiting persisted for a further 48 hours so a decision was made to reoperate.

At relaparoscopy, in addition to a tight plication, two gastrogastric hernias were seen protruding through the imbrication stitch (Fig 1). This stitch was removed, showing the gastric serosa at the hernias to be viable but oedematous and engorged. This normalised shortly after stitch removal and it was deemed safe to perform a reimbrication. An 11mm gastroscope (Olympus, Center Valley, PA, US) was used as a bougie to minimise the risk of another gastric obstruction. The patient’s symptoms settled following the second procedure and she was discharged after a further 72 hours tolerating a liquid diet.

Figure 1.

Gastric herniation through plication suture

Discussion

Promising results have been reported with GCP.^5–7 As with all new technology, appropriate governance ensures safe and ethical introduction and allows audit of follow-up. Unexpected issues and complications must be analysed.

We have selectively used GCP in patients too frail to tolerate complications from a stapled procedure. We are particular in stressing the emerging nature of this operation and emphasise the limited supporting data. We also limit this procedure to older patients in whom durability is perhaps not such an issue.

We have found that after GCP most patients experience temporary nausea and vomiting. This is thought to arise from the presence of the intragastric fold and may be similar to the symptoms seen in patients treated with an intragastric balloon (Inamed, Santa Barbara, CA, US).¹¹ This can usually be managed with standard antiemetic therapy and does not normally persist.⁷

Our case highlights a number of important issues, primarily the consequence of an overly tight plication. Although not completely occluded, the in-folded stomach was large enough to cause a functional gastric outlet obstruction. Enough of a problem in itself, the ensuing vomiting created high pressures in the gastric tube. This was exacerbated by gas insufflation during our post-operative endoscopy. These high pressures contributed to the formation of gastric hernias (Fig 2). Our second technical fault was the placement of too widely separated plication sutures. This allowed a potential space through which pressurised stomach could herniate.

Figure 2.

Raised intraluminal pressure leading to gastric herniation (A: normal stomach; B: stomach after plication, persistent vomiting and endoscopy leading to raised intraluminal pressures; C: gastric herniation between plication sutures subsequent to raised intraluminal pressures)

In this patient the gastric hernias were identified serendipitously, only because of a need to loosen the gastric plication. Although these hernias may not have developed without the persistent vomiting, it is interesting (and alarming) that the herniation seemed to be asymptomatic. The serosa of the herniated stomach was dusky until ‘reduced’ by removing the plication stitch. If untreated, the venous congestion would have undoubtedly progressed to full thickness ischaemia and perforation.

Conclusions

From our experience and the lessons in this case we make the following recommendations for GCP. The gastric plication should be performed over a bougie. We now use the same 36Fr bougie we use for sleeve resection. Furthermore, two rows of sutures placed 1–2cm apart would likely prevent the occurrence of transplication gastric herniation. This is consistent with the findings of Menchaca et al, who also proposed that two layers of sutures provide the most durable serosa-to-serosa adhesions.⁸