Table 5.
Changes in HbA1c level and medication use by changes in gastric emptying diagnosis over 48-weeks of follow-up
| Functional Dyspepsia (FD) at Baseline (N=60) | Gastroparesis (Gp) at Baseline (N=189) | |||||
|---|---|---|---|---|---|---|
| Changes in HbA1c and Medication Use over 48-weeks of follow-up | FD at 48-weeks (N=38) | Gp at 48-weeks (N=22) | P | Gp at 48-weeks (N=110) | FD at 48-weeks (N=79) | P |
| HbA1c (%)‡ | 0.12 (0.83) | 0.60 (1.55) | .40 | 0.29 (1.25) | 0.18 (1.71) | .77 |
| Medication use over 48 weeks (%) | ||||||
| Narcotics use | 13.9% (54.3) | 22.7% (42.9) | .80 | 15.7% (43.6) | 11.0% (35.6) | .37 |
| Proton pump inhibitors | –2.8% (50.6) | –9.1% (52.6) | .74 | 2.8% (48.3) | –8.2% (46.4) | .12 |
| Prokinetics | –2.8% (56.0) | 0% (43.6) | .26 | 6.5% (49.8) | 5.5% (0.50) | .43 |
| Antiemetics | 2.8% (37.7) | 9.1% (29.4) | .62 | 11.1% (43.9) | 5.5% (46.8) | .20 |
| Antidepressants | –22.2% (54.0) | –9.1% (52.6) | .27 | –10.2% (51.0) | 0% (60.1) | .18 |
| Anxiolytics | 11.1% (57.5) | 13.6% (35.1) | .95 | 14.8% (47.0) | 15.1% (43.0) | .48 |
| Pain modulators | 5.6% (53.2) | 9.1% (61.0) | .95 | 9.3% (39.9) | 6.8% (25.4) | .40 |
| Cannabinoids | 0% (33.8) | 4.5% (21.3) | .46 | 2.8% (25.4) | 9.6% (29.6) | .22 |
| Treatment use over 48 weeks (%) | ||||||
| On total parental nutrition (TPN) | –2.8% (16.7) | –4.5% (21.3) | .33 | 0.9% (21.6) | –5.5% (28.3) | .54 |
| Gastric electric stimulation device implantation | 8.3% (43.9) | 4.5% (37.5 | .34 | 13.0% (41.2) | 9.6% (37.9) | .72 |
Data for changes in HbA1c and any medication use presented are mean changes ± standard deviations. Positive change in Hba1c value or medication use at 48-weeks from baseline indicates worsening of the HbA1c and increasing use of medications. Patients included in this analysis had paired (baseline and 48-week) gastric emptying tests and follow-up history case-reports.
Change in HbA1c was analyzed between converter status within 2 subgroups (FD or Gastroparesis (Gp) at baseline) with an ANCOVA, regressing HbA1c change on baseline HbA1c and converter subgroup. P values for medication and treatment use changes over follow-up as percents were derived from a Wald test to assess whether change in use varied by converter status adjusting for the baseline use by each subgroup using ANCOVA with robust variance.
HbA1c required at baseline for all patients and at follow-up for diabetics, and if available, for non-diabetics. For FD converters, N=9 and 11 patients with both values at fup, and for Gp converters, there were N=36 and N=21 patients with paired HbA1c values at follow-up.
There are 2 patients without paired GES and medication use data in the FD converter subgroup (N=36 and 22) and 8 patients without paired data for analysis in the Gp converter subgroup (N=108 and 73).