[Factors of gastroesophageal acid reflux in severe obesity].

Objective: To study in morbid obesity the relationship between the degree of gastro-oesophageal reflux (GER) and the excess of body weight, or the related factors such as the energy intake or the fat distribution (waist-hip ratio).

Methods: In 20 morbid obese subjects (body weight: 125 +/- 32 kg) consulting in a weight-loss programme, anthropometric measurements, 3-hr oesophageal pHmetry, double isotope labelled meal for studying gastric emptying, study of gastric acid and pepsin secretions using PEG 4,000 as marker, and upper endoscopy were performed.

Results: Nine out of the 20 patients had more than 10 GER per 3-hr period. Seven patients had at least one GER symptom per day. In 6 patients, pH was under 4 for more than 10% of the time. The total number of GER and the number of GER of more than 5 min duration were correlated to the body mass index (P = 0.016 and P < 0.05 respectively). The number of GER was also correlated to the android type of overweight (P < 0.03). These relationships persisted when sex, age, smoking, and obesity complications (such as diabetes) were taken into account. There was a positive correlation between the number of GER and energy and lipid intake (energy intake: 3,119 +/- 1,082 kcal/day; P < 0.003 for both). The degree of GER was positively related to basal acid output (P = 0.049), and to sham feeding-stimulated acid output (P = 0.05); it was negatively related to gastric emptying half time, but was not correlated with basal or stimulated pepsin output. A relationship was found between body mass index (BMI) and gastric emptying half time for solid (P = 0.002) and liquid phases (P = 0.001).

Conclusion: GER seems to be common in long lasting morbid obesity. The number of refluxes increased with waist/hip ratio, BMI and energy or fat intake. GER was also increased by decreased gastric emptying rate, which was in part determined by BMI. The real prevalence of GER in morbid obeses must be determined by a large prospective study.