Double-blind placebo controlled cross-over trial of cisapride in postgastrectomy patients with duodenogastric reflux (DGR)

Abstract

• Double-Blind Placebo Controlled Cross-Over Trial of Cisapride in Postgastrectomy Patients with Duodenogastric Reflux (DGR). M. Vaezi*, R. Sears*, J. Richter #. *U Ala, B'ham;#Cleveland Clin. Excessive DGR is responsible for upper GI symptoms in about 15% of partial gastrectomy patients. Medical treatment has been disappointing, necessi tat ing surgical interventions in many patients. Using ambulatory esophageal bilirubin monitoring (Bilitec 2000), we evaluated the efficacy of cisapride in improving DGR and pat ient-symptoms. METHODS: Ten chronically symptomatic partial gastrectomy VA patients (5 BI, 5 BII) (8M, x age 57) with DGR by Bilitec, without acid reflux, randomized to either placebo or cisapride (20 mg qid) in double-blinded crossover study. Using Bilitec and standard 2200 calorie, liquid diet, DGR evaluated at baseline and after 1 month on each therapy. Daily diary of severity/frequency of upper GI symptoms recorded and mean monthly scores determined. At end of each treatment month , overal l global a s se s smen t made of s y m p t o m improvements. RESULTS: x&+SE. Wilcoxon, Fisher's tests; * p 0.14 Baseline Placebo Cisapride Total 26.2-+3.6 17.5+_2.1 11.2-+3.7 Upright 26.4+4.1 12.8-+3.5 7.1-+2.9 Supine 21.0+3.6 25.7+5.6 11.9-+3.8 Compared to placebo, cisapride significantly decreased DGR by 45% in upright and 54% in supine positions. There was no order effect. Overall symptoms improved in 70% of patients on cisapride compared to 10%'on placebo tp 7 and bili abs. > 0.14. RESULTS: x-+SE;ANOVA;*p 7 7.2-+1.8 6.1-+1.8 6 .0-+1.7 5.7-+1.6 4.6-+1.5 bili>.14 0.7-+0.2 11.1-+5.0 20.1-+8.4 25.9-+5.9* 48.0-+4.8*# B_~dmM) 0.03-+0.0 0.09-+0.03* 0.16-+0.05 0.24-+0.07_.* 0.50+-0,10# % Time pH 0.14 and gastric BA only discriminated complicated Barrett's from other patient groups. Linear regression found significant correlation (R=0.7, p 0.14 (R=0.3, p>.05), while correlation remained strong for % time pH 7 did not differentiate patients. CONCLUSIONS: 1) pH < 4 is a better predictor of GERD severity. 2) Bile reflux shows parallel increase over GERD spectrum, but only discriminates complicated Barrett's patients from other GERD patient groups. 3) Our study supports the 1 ° role of acid in GERD, but suggests a possible synergistic role for bile acids in complicated Barrett's patients.