Alkaline Reflux Gastritis

The hypothesis that reflux of upper intestinal content, particularly of bile acids (BA), is responsible for a unique postgastrectomy syndrome, alkaline reflux gastritis, was tested on 28 occasions in 21 postoperative patients (14 symptomatic patients, 7 controls). Parameters evaluated: recumbent (rec.), upright, p.c. intragastric pH, {BA}, net BA reflux per hour, specific BA fractions, fasting and p.c. gastrin, maximal acid output (MAO), gastric emptying of solids by delta-scintigraphy), and the severity of nonstomal histologic gastritis, the "gastritis score," graded 0-15 by an independent senior pathologist. For the entire group, gastritis severity correlated positively with intragastric {BA} and net BA reflux per hour, both in recumbency and p.c. Five symptomatic patients demonstrated rec. and p.c. {BA} and net BA reflux per hour greater than two standard deviations from comparable mean values in control patients. They differed significantly from the remaining symptomatic patients as follows: increased intragastric {BA} and net BA reflux per hour, increased intragastric pH and decreased MAO. They also demonstrated a more severe grade of gastritis. Lithocholic acid was present in their reflux content significantly more often. Bilious vomiting was also more frequent. No other differences could be identified, either objectively or clinically, between the symptomatic groups. Four patients with excessive reflux underwent Roux-en-Y revision and restudy 6-22 months later. BA reflux was completely abolished, histologic gastritis improved, hematocrit rose, MAO increased, and gastric emptying slowed. Burning pain, bilious vomiting, and symptoms of esophageal reflux were eliminated. Vomiting and nausea were improved. Diarrhea was unchanged. The objective criteria outlined can identify symptomatic postgastrectomy patients with a greater than normal reflux and gastritis. Clinical criteria alone cannot. Revisional surgery in these patients eliminates reflux, improves gastritis, and produces symptomatic improvement. The hypothesis under consideration is strengthened but not proven.