Management Experiences of Post-Gastrectomy Severe Alkaline Reflux Esophagitis

Abstract

Alkaline reflux esophagitis is a recognized complication of procedures that compromise the lower esophageal sphincter (LES), including gastrectomy. Incidence of reflux is dependent on the reconstructive procedure, with Roux-en-Y (RY) esophagojejunostomy commonly accepted as the optimal method. The authors report their experience of 5 patients who underwent remedial intervention for severe alkaline reflux esophagitis following gastric cancer surgery, over a 6-year period (2014-2020). Primary diagnoses encompassed 4 gastric adenocarcinomas and 1 gastric neuroendocrine tumor. Four patients previously underwent total gastrectomy and 1 subtotal gastrectomy with RY reconstruction. Onset of postoperative reflux symptoms ranged from 2 weeks to 3 years. Failing medical management, all patients underwent jejunojejunal anastomosis and Roux limb length revision with surgical jejunostomy. At follow-up, 4 out of 5 patients had some degree of symptom resolution and one with unresolved symptoms. The authors report our experience of managing this complication following gastrectomy with jejunojejunal anastomosis and Roux limb length revision.

Introduction

Alkaline reflux esophagitis (ARO) is a recognized complication of procedures that compromise the lower esophageal sphincter (LES), including gastrectomy. Incidence of reflux is dependent on the reconstructive technique, with Roux-en-Y (RY) esophagojejunostomy commonly accepted as the optimal method. ¹ Roux-en-Y reconstruction significantly decreases the incidence of this complication; however, when it does occur, it is often challenging to manage. Alkaline reflux esophagitis results from mucosal injury secondary to jejunal contents passing into the esophagus.

Initial management of ARO is conservative, entailing anti-reflux medication and lifestyle advice. Failing this, surgical intervention is required. Roux-en-Y reconstruction has been shown to reduce debilitating symptoms of ARO and improving patient quality of life. It also reduces the risk of long-term complications of ARO including Barrett’s esophagus and potential for malignant transformation. ² The authors report their experience of 5 patients who underwent remedial intervention for severe alkaline reflux esophagitis following gastric surgery.

Approach at Our Center

Operations Performed

These 5 patients were managed in a national tertiary center over a 6-year period (2014-2020). Ethical approval was granted by the St. James’s Hospital Dublin ethics committee for purpose of audit. This center performs high volume upper gastrointestinal resectional surgery, undertaking a mean number of 54 esophagectomies and 29 gastrectomies annually during the study period described. Overall, 3/5 had their initial resectional surgery at this unit, with the remaining 2/5 undergoing surgeries elsewhere. From the ones performed in this unit, the overall prevalence of revisional surgery for (i) gastrectomies in total was 3/204 (1.27%), (ii) total gastrectomy (TG) was 2/124 (1.6%), and (iii) subtotal gastrectomy (STG) was 1/49 (2%).

Management

The management of reflux symptoms consists of – sucralfate, dietary counseling and in cases of failure, significant impact on health-related quality of life (HRQL), nutrition and esophagitis, and evidence of short or sub-optimal Roux limb length the center would offer a revisional surgery.

Roux Limb Lengthening

In regards to the revisional surgery, in all cases there is revision to the RY anastomosis in order to lengthen the limb, with average lengthening of limb being 40.5 cm, but ranging from 35-65 cm from original anastomosis site. There were no significant intraoperative or postoperative complications, and there were no intraoperative or perioperative deaths.

Clinical Summary

During a 5-year period 5 patients who had originally had TG/STG surgery for oncological and non-oncological purposes and had subsequently go on to develop signs and symptoms of ARO were examined. These patients then underwent revision exploratory operations of their reconstruction (Table 1). The average age of the patient to date was 73 (range 58-91). The average age of the patient at time of their original gastrectomy was 65. There were 4 males (4/5 (80%)) and 1 female (1/5 (20%)). In total, 4/5 (80%) of the patients had gastric adenocarcinomas, and one (20%) had a gastric neuroendocrine tumor. In total, 4/5 (80%) had TG, one (20%) had STG. The index surgeries were undertaken between 2014 and 2020.

Table 1.

Baseline Characteristics of Post-Gastrectomy Patients With Severe Alkaline Reflux Esophagitis.

Patient No.	Age	Age at Time of Original Diagnosis	Sex	Underlying Disease	Original Operation	Interval Between Original Surgery and Revisional Surgery (Months)	Complications of Original Operation	Symptoms/Clinical Presentations	Time Interval Prior From Index to Symptoms	Revisional Surgery Details	Details of the Operation	Outcome of Operation (Resolved/Unresolved)	Oncological Recurrence (Y/N)
1.	75	62	M	Gastric AC	STG with esophagojejunostomy Roux-en-Y with subsequent completion gastrectomy due to pouch dysfunction	63	- SVT	Burning sensation, post-prandial pain, WL/epigastric pain, nausea, vomiting, productive cough, multiple admission for aspiration pneumonia, biliary esophagitis, severe bile reflux, poor intake, recurrent pulmonary inflammation	3 months	NR	No complications recorded	Unresolved—still being follow-up at tertiary center for ongoing dysphagia and has to stick to soft diet. Weight is stable. Most recent CT normal	N
							- Lung consolidation angiogram: 90% stenosis noted—stent inserted
2.	72	67	F	NET	Laparoscopic TG with Roux-en-Y reconstruction	26	NR	Severe bile reflux, biliary vomiting, nausea, anorexia, esophagitis	NR	Roux limb lengthened by 25 cm to total distance of 75 cm	No complications recorded	Resolved	N
3.	91	89	M	Gastric AC	Open TG with Roux-en-Y reconstruction	9	Nil	Choking episodes, nausea, bile acid reflux with retching and nausea, bilious vomiting, anorexia, diarrhea	3 months	Roux limb lengthened by 30 cm (total distance not recorded) and placement of distal feeding jejunostomy tube	Admitted to ICU, hypotension in ICU requiring noradrenaline, ongoing melena	Dumping symptoms developed post-revisional surgery, alkaline reflux symptoms resolved	N
4.	64	63	M	Gastric AC	Open TG with Roux-en-Y reconstruction	11	Nil	Significant heartburn grade B erosive esophagitis on endoscopy at 6 weeks post-op, significant dumping syndrome, heartburn, diarrhea, steatorrhea, cramps, nausea, anorexia, retching, abdominal pain, weakness, cachexia, bile reflux attacks, constipation	2 weeks	Roux limb lengthened by 30 cm to total distance of 70 cm and placement of distal feeding jejunostomy tube	Uncomplicated operative stay	Resolved	N
6.	58	49	M	Gastric AC	Open TG with abdominal Roux-en-Y	43	NR	Heartburn, endoscopy showed severe esophagitis with barium swallow showing persistent reflux Los Angeles grade B, ongoing bilious vomiting and reflux, abdominal pain, flatulence, vomiting, poor QOL, post-prandial abdominal pain	3 years	Roux limb lengthened by 30 cm to total distance of 65 cm	N/a	Resolved with late-recurrence of severe reflux symptoms due to peritoneal disease recurrence and subacute bowel obstruction	Y—with liver metastasis

Abbreviations: No = number of; M = male; F = female; AC = adenocarcinoma; NET = neuroendocrine tumor; STG = subtotal gastrectomy; TG = total gastrectomy; SVT = supraventricular tachycardia; WL = weight loss; UGI = upper gastrointestinal; QOL = quality-of-life; ICU = intensive care unit; ILD = interstitial lung disease; Y = yes; N = no; NR = not recorded.

All of these individuals demonstrated a similar clinical picture, which is displayed in Table 1. Endoscopy was the single most important diagnostic study performed alongside serial outpatient appointments with clinical nutritionists at various centers with descriptive documentation of symptomatology, weight, and quality of life scores. Endoscopic findings were mostly consistent across the board, usually consisting of a reddened, oedamatous oesophagus, often with ulcerations/erosions. Endoscopy was also important in eliminating other entities such as marginal ulceration, and stomal, afferent, and efferent loop obstruction. All 5 patients underwent antacid therapy with PPI or sucralfate or both.

Operative Success and Post-Revisional Reflux Symptoms

Overall, 4/5 patients were significantly improved from ARO point of view and had improved body weight after revisional surgery. In 4/5 cases, there was drastic improvement of their symptoms post-intervention with dramatic weight gain, near-complete symptom relief, and greatly reduced bile reflux with no further pain, alongside self-reported great improvements in quality of life. In one case, it is documented that there was poor initial resolution with continuous follow-up with the upper gastrointestinal team for ongoing dysphagia with chest pain; however, improvement was seen in the coming months with no esophagitis seen on endoscopy and increasing weight gain. Overall, there was one case where there was a good initial response; however, within 3 months there was some symptom deterioration

Nutritional Findings and Body Weights

Overall, body weights and nutritional status figures were available for all patients. Improved nutritional and body weight was noted in 5/5 cases after revisional surgery (Table 2). At the time of the original operation (gastrectomy), the mean weight was 81.7 kg (range: 71.2-92). All patients who required revision surgery had significant WL pre-procedure. At the time of revision, the median weight was 55.25 kg (range: 43-71) and a mean body mass index was 19.5 (range: 17.4-21.7). The median WL prior to revision surgery was 21.6 kg (range: 11.6-41.8 kg) and median percentage WL from time of gastrectomy was 25.5 kg (range: 16.2-41). All weights post-revision are at least 3 months beyond revision. The latest weight available or the new weight to which they plateaued was chosen. Median absolute weight gain after revision surgery was 5.9 kg (range: 3-12.3 kg) and median percentage weight gain was 13.7% (range: 4%–19.5%). Overall, during their revision surgeries 4/5 were given jejunostomy tubes.

Table 2.

Summary of Weight Loss and Weight Regain.

Patient No.	Interval Between Index and Revisional Surgery (Months)	Weight at Gastrectomy (kg)	Weight at Revisional Surgery (kg)	BMI at Revision	WL on Presentation for Revision (kg)	% WL on Presentation for Revision	Weight Post-Revision (kg)	Weight Regain (kg)	% Weight Regain Post-Revisional Surgery	Jejunostomy Feeding Tube at Time of Revision	Height (m)
1	64	75.9	50.9	17.6	25	32.9	54.2	3.3	10.6	N	1.7
2	26	84.8	43.0	17.4	41.8	41.2	48.9	5.9	13.7	N	1.57
3	9	71.2	59.6	20.4	11.6	16.2	70.0	10.4	17.4	Y	1.71
4	11	84.6	63	21.7	21.6	25.5	75.3	12.3	19.5	Y	1.7
5	54	92	71	21.4	21	22.8	74	3 kg	4	N	1.82

Abbreviations: No. = number; J-j = jejunojejunostomy; BMI = body mass index; Kg = kilograms; WL = weight loss; % = percentage; m = meters.

Discussion

Postoperative alkaline reflux esophagitis is a potentially devastating complication after TG. The advent of the RY reconstruction has significantly decreased the incidence of this complication. However, when reflux esophagitis occurs, it is often refractory to medical treatment. Any intervention that interferes with the mechanism of the pyloric sphincter enables increased reflux of duodenal/jejunal contents into the oesophagus and/or the gastric remnant. ³ Management of reflux can be challenging, particularly in the setting of mixed or alkaline reflux. Acid secretions can be readily controlled using proton-pump inhibitors and γ-aminobutyric acid (GABA) agonists, whereas bile salts and pancreatic secretions prove more difficult. ² Failing medical management, patients with refractory symptoms require surgical reconstruction. ²

There are several possible explanations to demonstrate how duodenal fluid refluxes post-gastrectomy: First, the reconstruction method. In Billroth II method, the duodenal/jejunal fluid passes through the anastomosis. This method has a described incidence of ARO of 50%-60% of cases. ⁴ Comparatively, the RY method has a described incidence of 10%-30%. ⁴ Roux-en-Y esophagojejunostomy is widely accepted as the preferred reconstructive procedure and is effective in preventing reflux, maintaining ideal patient weight. ³ Second, the shorter the anastomotic distance, the more likely ARO to occur, and it has been demonstrated with an anastomotic distance of > 40 cm that leads to less ARO. ² Third, in cases of TG requiring lower esophagectomy, the decrease in LES pressure due to its dissection and the thoracic cavity negative pressure on the esophageal jejunal anastomosis may increase the risk of ARO. ³ Finally, occlusive mechanisms such as disease recurrence or postoperative adhesions or strictures.

The major limitation of our study was the small number of patients we were able to analyze, given that we focused on only one tertiary hospital experience of this complication. This did not allow solid conclusions regarding the incidence of reflux esophagitis after gastrectomy with esophagojejunostomy. Another limitation was that the data was highly heterogenous, qualitative, and descriptive in nature, and therefore, no quantitative conclusions could be drawn.

Conclusion

Severe ARO is a recognized complication of gastric procedures compromising the LES. The authors report our experience of managing this complication following gastrectomy with jejunojejunal anastomosis and Roux limb length revision, with a majority of patients having improvement in, if not complete resolution of, reflux symptoms. However, higher power studies are required to contribute to the growing interest in this disease process.