Is symptom relief associated with reduction in gastric retention after gastric electrical stimulation treatment in patients with gastroparesis? A sensitivity analysis with logistic regression models.

BACKGROUND

Enterra gastric electrical stimulation (GES) is an alternative treatment for gastroparesis (GP) when standard medical therapy fails. The aims of this study were to evaluate the association between total symptom score (TSS) and reduction in gastric retention (GR) after GES by GP etiology and to examine the sensitivity of the association to varying cutpoints used to define GR and TSS improvement.

METHODS

Gastric retention assessed with a standardized (99m) Tc radio-labeled egg meal and TSS measured by a five-point Likert scale in 221 GP patients treated with Enterra GES therapy for at least 1 year were analyzed. Bivariate chi-square test and multivariable logistic regression with all possible cutpoints were used to assess the consistency of association and quantitate the relationship across three GP etiologies.

KEY RESULTS

Symptom relief in diabetic GP was more likely attributable to GR reduction as indicated by the consistently significant odds ratios (P < 0.01) across all cutpoints. The association in idiopathic GP was inconclusive because odds ratios were sensitive to cutpoints with P-values ranging from 0.01 to 0.47. No association was found for patients with post surgical gastroparesis (P > 0.1 for all cutpoints). Patient age, gender, baseline TSS and baseline GR had no significant effect at 5% level on clinical improvement regardless of cutpoints for GR.

CONCLUSIONS & INFERENCES: Association between clinical improvements and GR reduction following GES treatment depends on patient etiology and was able to be demonstrated in diabetic GP. The association for idiopathic GP was inconclusive and no such association was found for post surgical GP across all possible cutpoint combinations.