Prospective Evaluation of Gastric Neurostimulation for Diabetic Gastroparesis in Canada

Abstract

The efficacy of gastric neurostimulation therapy for diabetic gastroparesis (GP) in a 'real-life' Canadian setting has not been assessed. To assess changes in health-related quality of life (QoL), weekly vomiting frequency (WVF), total symptom score (TSS) and health care utilization 12 months before and after gastric neurostimulator implantation in a diabetic GP cohort. Medication-refractory diabetic GP patients (n=7, four female, mean age 42 years) were prospectively recruited from 2008 to 2012. QoL scores were self-administered and obtained at baseline, 24 and 48 weeks postimplantion. WVF and TSS were assessed similarly. Health care usage, measured as hospitalization frequency and medication cost, was obtained six and 12 months before and after implant. Changes from baseline to six and 12 months for all outcomes were compared. The mean ( ± SD) QoL according to EuroQol was significantly better at 24 weeks after the baseline measurement (baseline 29 ± 5, 24 weeks 52 ± 7; P = 0.03). The mean improvement in TSS was significantly better at one year postintervention (baseline score 35 ± 5 versus 12 months 27 ± 3; P = 0.03). Changes in Short-Form 36 Health Survey and WVF were not significant. Days of GP-related hospitalization were highly variable but decreased from a median of 71 days (range 0 to 227 days) to 29 days (range two to 334 days) one year before and after surgery, respectively (P = 0.735). Outpatient medication costs did not decrease to a significant extent. Gastric neurostimulation for diabetic GP appeared to show some beneficial palliative effects overall in the present small open-label series, but the effect is highly variable among patients, and placebo effect cannot be ruled out.