Interventional procedure overview of laparoscopic insertion of a magnetic ring for gastro-oesophageal reflux disease

Overall morbidity and complication

Overall postoperative morbidity ranged from 0% to 3% of patients who had laparoscopic insertion of a magnetic ring and from 0% to 7% of patients who had fundoplication in the systematic review and meta-analysis of 7 studies (Aiolfi 2018).

The intraoperative complication rate was 2% in the laparoscopic insertion of a magnetic ring group and 1% in the fundoplication group, and the procedure-related complication rate was about 2% in each group in the non-randomised comparative study of 631 patients (Bonavina 2020).

Major complications were reported in 2 patients in the case series of 553 patients. These complications included CO₂ retention needing reintubation (n=1) and mediastinal abscess needing drainage and intravenous antibiotic (n=1; Ayazi 2020a).

Minor complications were described in 9% (49/553) of patients in the case series of 553 patients. These complications included poor postoperative pain control (n=4), significant nausea during immediate postoperative period (n=5), hypoxia needing supplemental oxygenation (n=7), lethargy (n=3), abdominal pain needing additional evaluation (n=5), bothersome nausea or vomiting needing emergency department visit (n=11), dysphagia needing hospital admission (n=7), abdominal wall haematoma at gastric pacer insertion site (n=1), deep vein thrombosis (n=1), urinary retention (n=1), cardiac arrhythmia (n=1), dyspnoea (n=2), and aspiration pneumonia (n=1; Ayazi 2020a).

Overall complications were reported in 3% (n=2) of patients without hiatal hernia, 11% (n=23) of patients with small hiatal hernias, 14% (n=8) of patients with large hiatal hernias, and 27% (n=6) of patients with paraesophageal hernias (p=0.015) in the non-randomised comparative study of 350 patients. The overall complication rate was 11% (n=39), consisting of 2 major complications and 37 minor complications. Major complications happened in 2 patients with large hiatal hernias, and minor complications occurred in 2 patients without hiatal hernia, 23 patients with small hiatal hernias, 6 patients with large hiatal hernias, and 6 patients with paraesophageal hernias (Ayazi 2020b).

Dysphagia (needing treatment)

The pooled incidence of postoperative dysphagia was 29% in 1 year after laparoscopic insertion of a magnetic ring (95% CI 13% to 46%; Cochrane Q P<0.00001, I²=96%; 6 studies) in the systematic review and meta-analysis of 15 studies. Only 1 study reported that 4% to 6% of patients complained of dysphagia at 1 year after laparoscopic insertion of a magnetic ring. The pooled rate of endoscopic dilation needed was 7.4% (95% CI 2.9% to 13.7%; Cochrane Q P=0.0005, I²=80%; 5 studies) in 1 year. Comparison between laparoscopic insertion of a magnetic ring and LNF showed that there was no statistically significant difference in postoperative dysphagia (RR 1.21, 95% CI 0.92 to 1.60, p=0.18; Cochrane Q P=0.89, I²=0%; 2 studies; Zhuang 2021).

Analysis of 13 single-arm cohort studies showed that postoperative dilation was reported in 8% of patients (164/2,112) in the systematic review and meta-analysis of 19 studies. When comparing laparoscopic insertion of a magnetic ring with fundoplication, there was no statistically significant difference in postoperative dysphagia (pooled OR 0.94, 95% CI 0.57 to 1.55, p=0.822; I²=20.4%; 4 studies; Guidozzi 2019).

Dysphagia needing endoscopic dilation was described in 9% of patients who had laparoscopic insertion of a magnetic ring and 7% of patients who had fundoplication (OR=1.56, 95% CI 0.61 to 3.95, p=0.119; I²=35%; 5 studies) in the systematic review and meta-analysis of 7 studies (Aiolfi 2018).

Dysphagia was reported in 32% (15/47) of patients at 6 months after laparoscopic insertion of a magnetic ring in the RCT of 134 patients having laparoscopic insertion of a magnetic ring or twice-daily PPIs. Intervention was needed in 7 patients (oral corticosteroids, n=3; endoscopic dilation, n=3; laparoscopic hiatal hernia repair, n=1; Bell 2019). At study completion (12 months), the rate of dysphagia was 40% (n=19) in the primary laparoscopic insertion of a magnetic ring group and 33% (n=10) in the crossover group (Bell 2020).

Postoperative endoscopic dilation was needed in 169 patients (99 patients needed 1, and 70 needed more than 1) in the case series of 553 patients. The indications for endoscopic dilation were dysphagia (n=129), chest pain (n=14) and both dysphagia and chest pain (n=26; Ayazi 2020a).

Postoperative dysphagia was statistically significantly higher in the severe GORD group than the non-severe GORD group (25% compared with 14%, p=0.0124) in the non-randomised comparative study of 336 patients. Endoscopic dilation was needed in 3% and 2%, respectively (p=0.6562). At baseline, dysphagia was also statistically significantly higher in the severe group (12%) than the non-severe group (6%, p=0.0460; Ferrari 2021).

Postoperative dysphagia was reported in 16% of patients without hiatal hernia, 16% of patients with small hiatal hernias, 6% of patients with large hiatal hernias, and 6% of patients with paraesophageal hernias (p=0.08) at a mean follow up of 13.6 months in the non-randomised comparative study of 350 patients. Endoscopic dilation was needed in 20%, 26%, 24% and 5%, respectively (n=0.12; Ayazi 2020b).

Dysphagia needing device removal was reported in 292 patients at a mean of 10.9 months after implantation in a review of the MAUDE database and the Ethicon's complaint database of 27,779 patients implanted with LINX devices (DeMarchi 2021).

Device erosion

Analysis of 13 single-arm cohort studies revealed that the overall rate of oesophageal erosion was less than 1% (31/11,530) in the systematic review and meta-analysis of 19 studies (Guidozzi 2019).

Erosion needing device removal was reported in 27 patients at a mean of 25 months after LINX device implantation in the review of the MAUDE database and the Ethicon's complaint database. The cumulative risk of erosion at 7 years was 0.28% (95% CI 0.17% to 0.46%; DeMarchi 2021).

Device removal

Device removal was reported in 15 patients (5 studies) at 5-year follow up in the systematic review and meta-analysis of 15 studies (Zhuang 2021). Fourteen patients had their implants removed because of persistent dysphasia, chest pain or unresolved GORD symptoms, and 1 patient was because of implant erosion.

Analysis of 13 single-arm cohort studies showed that device removal or reoperation was reported in 3% of patients (69/2,098) in the systematic review and meta-analysis of 19 studies. When comparing laparoscopic insertion of a magnetic ring with fundoplication, there was no statistically significant difference in postoperative reoperation (pooled OR 1.23, 95% CI 0.26 to 5.8; p=0.797; I²=48.5%; 4 studies; Guidozzi 2019).

Device removal was reported in 1.5%, 2.0% and 2.4% of patients at 1, 2 and 3 years after laparoscopic insertion of a magnetic ring in the non-randomised comparative study of 631 patients with GORD (Bonavina 2020).

Device removal was reported in 7% (37/553) of patients in the case series of 553 patients (Ayazi 2020a). The reasons for removals included troublesome dysphagia or chest pain not responding to dilation (n=20), recurrence of hernia and migration of the device (n=3), recurrence of hernia and mediastinal abscess (n=1), worsening typical reflux symptoms (n=4), worsening atypical reflux symptoms (n=3), possible titanium allergy (n=1), unexplained leukocytosis (n=1), need for subsequent operation (n=2), and removal indicated by device malfunction (n=2).

Device removal was done in 3 patients between 6 to 12 years after laparoscopic insertion of a magnetic ring in the case series of 124 patients. The reasons for device removal were heartburn/continued reflux symptoms (n=1), dysphagia (n=1) and need of magnetic resonance study/planned magnetic resonance imaging (n=1; Ferrari 2020).

Device removal was reported in 8 patients with severe GORD and 24 patients with non-severe GORD in the non-randomised comparative study of 336 patients (Ferrari 2021).

Device removal was reported in 2% (609/27,779) of patients at a mean of 14.6 months after implantation in the review of the MAUDE database and the Ethicon's complaint database. The overall 7-year cumulative risk of explant was 5% (95% CI 4.31% to 5.36%; DeMarchi 2021). The reasons for device removal included dysphagia (n=292), persistent or recurrent GORD (n=125), erosion (n=27), abdominal pain or pain (n=46), discontinuous device (n=17), need for MRI (N=11), vomiting (n=16), gastroparesis (n=4), device migration (n=3) and other unknown reasons (n=68).

Device removal was reported in 6% of patients without hiatal hernia, 6% of patients with small hiatal hernias, 2% of patients with large hiatal hernias, and 0% of patients with paraesophageal hernias (p=0.28) at a mean follow up of 13.6 months in the non-randomised comparative study of 350 patients. The overall rate of device removal was 5% at a mean follow up of 13.6 months, and all the removals were for persistent dysphagia or oesophageal spasm unresponsive to endoscopic dilation (Ayazi 2020b).

Reoperation

Reoperation was needed in 13 patients who had laparoscopic insertion of a magnetic ring (12 device removals and 1 crural release) and 11 patients who had fundoplication (5 herniation of the fundic wrap, 3 persistent GORD, 2 retroesophageal abscess and 1 crural release) in the systematic review and meta-analysis of 7 studies. The estimated pooled OR of reoperation was 0.54 (95% CI 0.22 to 1.34, p=0.183, I²=0.0%; 3 studies) between groups (Aiolfi 2018).

The surgical intervention rate at 1, 2 and 3 years after treatment was 1.6%, 1.2% and 0.6% of patients who had laparoscopic insertion of a magnetic ring and 1.9%, 0% and 0% of patients who had fundoplication in the non-randomised comparative study of 631 patients. The intervention for the laparoscopic insertion of a magnetic ring group was the removal of device for dysphagia (45%), ongoing GORD (18%), vomiting/regurgitation (18%), gastric pain (9.5%) and need for MRI (9.5%). The intervention for the fundoplication group was revision of a Nissen wrap because of ongoing GORD, reherniation, and a sigmoid resection secondary to diverticulitis (Bonavina 2020).

Reoperation was needed in 7 of the 24 patients who were found to have hiatal hernia recurrence on endoscopy at a mean follow up of 13.6 months in the non-randomised comparative study of 350 patients. Of these 24 patients, 20 patients had small hernias, 1 patient had large hernias, and 3 patients had paraesophageal hernias. The incidence of recurrent hiatal hernia increased in direct correlation with the preoperative hiatal hernia size (small hiatal hernia, 10%; large hiatal hernia, 17%; paraesophageal hernia, 20%; p=0.032). When comparing with patients with a full dissection, patients with a minimal dissection had a statistically significantly higher hiatal hernia recurrence rate (minimal dissection, 21%; full dissection, 8%; p=0.033), and were more likely to need reoperation (minimal dissection, 11%; full dissection, 2%; p=0.0133; Ayazi 2020b).

Readmission

Readmission was reported in 31 patientswithin 90 days after the procedurein the case series of 553 patients. Of these patients, 23 patients were readmitted to hospital within 30 days and 8 patients were readmitted between 30 to 90 days (Ayazi 2020a).

Readmission was described in 19 patients within 90 days after the procedure, including 14 patients with small hiatal hernias, 2 patients with large hiatal hernias, and 3 patients with paraesophageal hernias (p=0.049) in the non-randomised comparative study of 350 patients (Ayazi 2020b).

Dysphagia needing hospital admission was reported in 7 patients in the case series of 553 patients (Ayazi 2020a).

Anecdotal and theoretical adverse events

In addition to safety outcomes reported in the literature, professional experts are asked about anecdotal adverse events (events that they have heard about) and about theoretical adverse events (events that they think might possibly occur, even if they have never happened).

For this procedure, professional experts did not list any additional anecdotal or theoretical adverse events.