Thoracic Neuromodulation for Diabetic Gastroparesis

The proposed TNM-DGp trial is a pilot sham-controlled RCT to test the preliminary efficacy of ThorS-MagNT to improve DGp symptom severity and related quality of life. The investigators will also explore mechanisms of change and potential moderators of treatment outcome after ThorS-MagNT. The trial will be conducted over the 3-year pilot RO1.

Randomization: Using the magnitude of change in DGp symptom severity in the initial pilot trial, the investigators will randomize 48 patients in a 1:1:1 ratio to either receive sham, 1 Hz, or 10 Hz ThorS-MagNT over a 5-day period. The investigators will use blocked randomization with a block size ranging from 3 to 6 and known only to the biostatistician. The biostatistician will create the randomization sequence for all participants prior to any pre-treatment assessments. Treatment allocation will be made available to the interventionists only after baseline assessments, determination of eligibility, and informed consent.

Participants (N = 48): The investigators will recruit TNM-DGp trial participants at Augusta University. Approximately 2-3 DGp patients are seen per week at our center. To further facilitate recruitment, advertisements will be put into newsletters of both hospitals, local newspaper, and radio. Flyers will be placed in GI, family medicine, internal medicine, and endocrine clinics, and locoregional colleagues will be emailed. Forty-eight adult, DGp outpatients with refractory symptoms of greater than 6 months and moderate-severe disease (total American Neurogastroenterology and Motility Society Gastroparesis Cardinal Symptoms Index (ANMS GCSI) score greater than or equal to 2.0) will be recruited. Participants will be screened by phone to assess eligibility and interest in the study. Those who are eligible and interested in the study will complete informed consent.

Sample size justification: Assuming a responder rate of 30% for the sham arm, we consider a 1:1:1 balanced design for the sham, 1 Hz, and 10 Hz treatment arms, respectively. For the primary clinical outcome, the difference is assessed between sham and the combined treatment arm (1 Hz and 10 Hz), giving a 1:2 allocation. A sample size of 14 per treatment arm is required to observe a responder rate of 65% for the combined treatment arm with 80% power at 5% significance level. This sample size is sufficient to observe a 1-point improvement in the PAGI-QOL score (secondary clinical outcome) with 85% power. Accounting for a 15% dropout in the study, the investigators will need 16 subjects per treatment arm. For the proposed interim analysis at 33% and 66% recruitment, the investigators will have enough sample size (5 and 10 per treatment arm respectively) to achieve 80% power if the responder rate in the treatment arms is 90% and 77% respectively.

Clinical outcome (Aim 1) measures: The primary clinical outcome will be responder rate with responder defined as ≥30% improvement in gastroparesis symptom severity by the Week 4 ANMS total GCSI-DD compared to baseline. The total ANMS GCSI-DD score includes a 7-day average of 5 gastroparesis-specific items. A greater than 30% reduction in total symptom score is defined as a responder by the FDA User Manual for the ANMS GSCI-DD. Secondary clinical outcome will be improvement in quality of life, defined by a 1-point improvement in Patient Assessment of Gastrointestinal Symptoms-Quality of Life (PAGI-QOL). ANMS GCSI-DD will be collected monthly for one year after treatment. Duration of response will be defined as time to loss of response (<30% improvement over baseline total ANMS GCSI-DD score).