Background
Gastroparesis is a condition in which the stomach empties too slowly, leading to nausea, vomiting, early fullness, bloating, abdominal discomfort, and poor oral intake. In people with diabetes, gastroparesis can be especially challenging because delayed stomach emptying may make blood sugar levels more unpredictable. Meals may be absorbed later than expected, which can cause both high and low glucose swings and complicate insulin dosing.
When symptoms are severe and do not improve with dietary changes, antiemetic drugs, prokinetic medications, or other standard treatments, some patients are considered for surgical therapy. Options include gastric electrical stimulation, pyloroplasty, and gastric peroral endoscopic myotomy (G-POEM). These procedures are typically aimed at improving symptoms and gastric emptying, but their effect on long-term diabetes control has not been well understood.
Why this study was done
This study asked an important question: in adults with both diabetes and gastroparesis, does surgical treatment help only with stomach symptoms, or does it also improve glucose control and reduce diabetes-related complications?
That question matters because glycemic control is central to diabetes care. Even modest improvements in hemoglobin A1c, the blood test that reflects average glucose over about three months, can lower the risk of eye, kidney, nerve, and cardiovascular complications over time. If a gastroparesis procedure also helps stabilize blood sugar, that benefit would be clinically meaningful when selecting treatment.
Study design
Researchers performed a retrospective multicenter cohort study using a large national electronic health record database that included more than 100 million patients across more than 70 U.S. health care organizations. Adults with both diabetes and gastroparesis diagnosed between 2010 and 2023 were included.
The study compared patients who underwent surgical therapy for gastroparesis with similar patients who did not receive surgery. To make the groups more comparable, the investigators used propensity score matching, a statistical method that pairs patients with similar baseline characteristics. After matching, 2,272 patients remained in each group.
The analysis focused on several outcomes:
1. Mean hemoglobin A1c over follow-up
2. Probability of adequate glycemic control, defined as HbA1c below 7%
3. Probability of severe hyperglycemia, defined as HbA1c 10% or higher
4. Initiation of insulin therapy
5. Diabetes-related complications
6. Health care use, including emergency department visits and hospitalizations
7. Mortality
Who was included
In total, 95,328 patients met eligibility criteria. Of these, 2,272 patients, or 2.4%, underwent a surgical intervention for gastroparesis. The average age was 54.9 years, and most patients were women. The mean follow-up time was 3.5 years.
At baseline, the two groups had similar HbA1c levels, which is important because it suggests they started from a comparable point in terms of diabetes control.
Main findings
Over the follow-up period, average HbA1c improved in the surgical group but worsened in the nonsurgical group. Specifically, mean HbA1c fell by 0.51 percentage points among patients who had surgery and rose by 0.28 percentage points among those who did not.
At 5 years, mean HbA1c was 6.29% in the surgical group and 7.21% in the nonsurgical group. That difference of 0.92 percentage points was statistically significant.
In practical terms, patients who had surgery were more likely to reach acceptable glucose control and less likely to experience severe hyperglycemia. They were also less likely to need new insulin therapy during follow-up. These findings suggest that the benefits of surgery may extend beyond symptom relief and may include improved diabetes management.
Diabetes-related complications and utilization of care
The surgical group also had fewer diabetes-related complications over 5 years. Complications occurred in 1,209 surgical patients (53.2%) compared with 1,438 nonsurgical patients (63.3%). This was a significant reduction.
Health care use was also lower in the surgical group. Emergency department visits occurred in 41.7% of surgical patients versus 54.7% of nonsurgical patients, and hospitalizations occurred in 45.5% versus 59.0%, respectively. Both differences were statistically significant.
These findings may reflect better symptom control, improved nutritional intake, fewer metabolic swings, or a combination of factors. They also suggest that successful treatment of gastroparesis may reduce the burden on patients and the health care system.
Mortality
Five-year mortality was identical in both groups: 232 patients in each cohort, or 10.4%. The study did not find a survival difference between surgical and nonsurgical management over the follow-up period.
This is an important reminder that improved glucose control and fewer hospital visits do not necessarily translate into a measurable mortality benefit within the study timeframe. Mortality in patients with diabetes and gastroparesis may be influenced by many factors, including disease duration, comorbidities, and overall severity of illness.
What the results may mean
These findings support the idea that surgical therapy for gastroparesis can have metabolic benefits in selected patients with diabetes. Possible explanations include improved gastric emptying, more predictable nutrient absorption, reduced vomiting and poor intake, and better ability to match insulin dosing to meals. When food moves through the stomach more consistently, blood sugar patterns may become easier to manage.
The study does not prove that surgery causes better glucose control in every patient, because retrospective studies cannot fully eliminate bias or account for all confounding factors. For example, patients referred for surgery may differ in motivation, access to care, or overall health status. Still, the large sample size and matching methods strengthen the findings.
Clinical implications
For clinicians, the message is that surgical treatment for refractory diabetic gastroparesis should not be evaluated solely as a symptom procedure. It may also affect diabetes outcomes. When discussing treatment options, clinicians may want to consider:
1. Severity and chronicity of gastroparesis symptoms
2. Nutritional status and weight loss
3. Degree of glycemic variability
4. Insulin requirements and risk of hypoglycemia or hyperglycemia
5. Prior response to medications and dietary therapy
6. Patient preferences and procedural risks
This study suggests that in carefully selected patients, procedures such as gastric electrical stimulation, pyloroplasty, or G-POEM may contribute to better long-term diabetes control in addition to improving upper gastrointestinal symptoms.
Limitations
As with all observational research, several limitations should be kept in mind. The study relied on electronic health record data, so coding accuracy and missing information may affect results. The exact severity of gastroparesis symptoms, the details of dietary management, and some diabetes-specific factors may not have been fully captured.
Also, the surgical category combined different procedures, each with its own mechanism and risk profile. The study therefore shows the overall association of surgical therapy with outcomes, but it cannot tell us whether one procedure is superior to another for glycemic control.
Finally, although the groups were matched statistically, unmeasured differences may still exist. Randomized trials would be needed to establish causality more definitively.
Bottom line
In this large national multicenter study, surgical therapy for diabetic gastroparesis was associated with better glycemic control, less need for insulin initiation, fewer diabetes-related complications, and fewer emergency visits and hospitalizations. Mortality was unchanged.
For patients with difficult-to-control diabetic gastroparesis, surgery may offer benefits beyond symptom relief. These metabolic outcomes may deserve consideration when weighing treatment options, especially in patients whose blood sugar remains unstable despite medical therapy.