Abstract
Background
Although gastroparesis is seen in patients with type 2 diabetes mellitus (T2DM), the prevalence of symptoms suggestive of gastroparesis in patients with T2DM is unknown, particularly among African Americans.
Aims
To determine the prevalence of symptoms associated with gastroparesis in a large community-based population of European Americans and African Americans with T2DM.
Methods
Individuals with T2DM in the Diabetes Heart Study were asked to complete the gastroparesis cardinal symptom index (GCSI) and other GI-related questionnaires. GCSI total score ≥ 18 represented moderate or worse symptoms suggestive of gastroparesis.
Results
A total of 1253 participants (700 female, 553 male) completed the GCSI: 750 were European American and 503 African American. GCSI scores ≥ 18 were recorded in 72 participants: 38 (5%) of European Americans and 34 (7%) of African Americans. The average GCSI was 24.1 in European Americans and 24.6 in African Americans, indicating moderate to severe symptoms. Compared to European Americans with GCSI scores ≥ 18, African Americans were younger (59.4 vs. 53.3 years, p = 0.004), had earlier onset of T2DM (46.3 vs. 40.1 years, p = 0.01), higher HbA1c (7.6 vs. 9.1, p = 0.0009), underwent fewer upper endoscopies (55.3% vs. 26.5%, p = 0.02), and had more anxiety and depression (p < 0.001).
Conclusions
Moderate or greater symptoms suggestive of gastroparesis are present in 5–7% of European and African American patients with T2DM in community-based populations. Symptoms suggestive of gastroparesis may be underappreciated in patients with T2DM and account for upper gastrointestinal symptoms, unexplained glycemic control issues, and decreased quality of life.
Keywords: Gastroparesis, Type 2 diabetes mellitus, Gastroparesis cardinal symptom index, Community-based population
Background
Gastroparesis (GP) is defined as delayed gastric emptying in the absence of mechanical obstruction of the stomach or duodenum [1]. Symptoms suggestive of GP include early satiety, prolonged postprandial fullness, bloating, nausea, vomiting, and abdominal pain [1]. Although multiple conditions have been associated with GP, the majority of cases result from type 1 (T1DM) and type 2 diabetes mellitus (T2DM). Gastrointestinal (GI) symptoms are higher in patients with diabetes mellitus compared to controls; nausea and postprandial fullness are the most severe symptoms in T1DM and T2DM [2-6]. The number of GP-related hospitalizations has been increasing in the USA, which warrants the need for greater appreciation of symptom prevalence in community-based populations [7].
T2DM represents 90–95% of the population with diabetes, an estimated 26 million Americans. Patients with T2DM have more GI symptoms compared with non-diabetic individuals [8]. Upper GI symptoms (early satiety, postprandial fullness, bloating, nausea, vomiting) suggestive of GP were reported in 11–18% of patients with diabetes mellitus [2, 9, 10]. However, these studies are limited by small sample sizes, limited data on the ancestry of participants, and/or did not distinguish between individuals with T1DM or T2DM. Because of these limitations, previous studies may not accurately represent the prevalence of symptoms suggestive of GP in community-based populations with T2DM and could not assess differences based on ancestry.
The aim of this study was to determine the prevalence of symptoms suggestive of GP using the gastroparesis cardinal symptoms index (GCSI), a validated 9-question survey of GP symptoms, in a large community-based population of European Americans (EAs) and African Americans (AAs) with T2DM. Age, gender, glucose control, medication, and frequency of GI tests (upper endoscopy or gastric emptying tests) were recorded in patients with moderate or greater symptoms and compared with subjects having mild GCSI symptoms.
Methods
The study protocol was approved by the institutional review board at Wake Forest Baptist Medical Center and all participants provided written informed consent. One thousand four hundred and two individuals with T2DM were enrolled from the Diabetes Heart Study (DHS). T2DM was defined as a physician diagnosis with participant taking oral agents or insulin for treatment, fasting blood sugar ≥ 126 mg/dL, random glucose ≥ 200 mg/dL or HbA1c ≥ 6.5% measurement at the clinic visit. Recruitment came from internal medicine clinics and local advertising, yielding a community-based sample. The DHS measured subclinical CVD with computed tomography-derived measures of coronary artery calcified atherosclerotic plaque and carotid IMT by ultrasound. At the baseline visit, 56% of participants lacked any prior CVD events, whereas 44% reported prevalent CVD. Participants were asked to self-identify their ethnic background as either European American or African American. European Americans were defined as individuals who self-identify as American of European origin/descent, specifically white Americans. Each participant completed the GCSI, a patient-rated symptom assessment tool for determining severity of nine symptoms suggestive of GP. Total GCSI scores range from 0 (no symptoms) to 45 (most symptomatic) and is often used in clinical research studies [11-13].
The GCSI is a part of the patient assessment of upper gastrointestinal disorders-symptom severity index (PAGI-SYM) instrument, developed to assess the severity of GI symptoms associated with GP [14]. The GCSI consists of three subscales: (1) nausea/vomiting, (2) postprandial fullness, (3) early satiety, bloating. The total cumulative score may range from 0–45 and is often used in clinical research studies [11-13]. Validity and responsiveness of the GCSI to changes in clinical status of patients with GP have been documented [14]. In the present study, patients with GCSI scores 18 (of 45 maximum points) were compared with patients with GCSI scores < 18. GCSI scores 18 or greater were considered to be in the moderate to severe symptom range.
Participants also completed the Dyspepsia Symptom Severity Index (DSSI), the DSSI is a 20-item questionnaire designed to quantify the severity of dyspepsia symptoms based on a 5-point Likert scale ranging from 0 (absence of symptoms) to 4 (severe symptoms) [15]. A total scale score is represented by the mean across the questionnaire’s 3 subscales, with higher scores indicating more severe symptoms. The DSSI has been shown to be reliably consistent over time [15].
Depression and anxiety were measured by the CES-D10 and Brief Symptom Inventory-Anxiety Questionnaire (BSIAQ) respectively. The CES-D10 depression scale is a self-report scale designed to measure the current level of depression of an individual [16]. The CES-D10 was originally derived for research purposes in epidemiological studies involving the general population and primary care patients, but has been used extensively in other settings such as in individuals with chronic conditions [17, 18]. The BSIAQ is a questionnaire to assess patient anxiety in the clinical setting [19, 20].
Glucose measurements were obtained from the latest clinical visit for each participant. Medications that could aggravate or treat GP associated symptoms were recorded, including narcotics, antidepressants, proton pump inhibitors (PPIs), insulin, oral hyperglycemic agents, and metoclopramide.
Statistical Analyses
Statistical analysis was completed using SAS software version 9.4 (SAS Institute, Cary, NC, USA). The analyses were limited to participants with T2DM. Summary statistics such as means and standard deviations (SD) were computed for continuous characteristics, and counts and percentages were computed for discrete characteristics. Summary statistics are presented by ancestral population and by symptom severity (GCSI scores 18 vs. < 18). Marginal models incorporating generalized estimating equations (GEE) approaches were used to compare characteristics between racial groups or symptom severity. These models account for familial correlation using a sandwich estimator of the variance under exchangeable correlation. To study the association between patient characteristics and symptom severity, we used the backward elimination to select the associated patient characteristics. The final multivariable marginal model (logit link and binomial distribution) including the selected characteristics—age, sex, race, smoking, depression scale, BMI, PPI use, narcotic use, insulin use, and oral anti-hyperglycemic medicine—were presented.
Results
A total of 1253 DHS participants with T2DM completed the GCSI, DSSI, BSIAQ, and CES-D10 depression scale; 700 were female and 553 male [750 EA (Table 1) and 503 AA (Table 2)]. Seventy-two individuals had GCSI scores ≥ 18, 5% of EAs (n = 38) and 7% of AAs (n = 34). One thousand one hundred eighty-one individuals had GCSI scores ≤ 17, of which 712 were EA and 469 AA.
Table 1.
Comparison of clinical attributes and diagnostic studies in European Americans with GCSI ≤ 17 versus ≥ 18 in the diabetes heart study
| 0–17 (< 95%) (N = 712) |
≥ 18 (≥ 95%) (N = 38) |
Overall (N = 750) |
p value | |
|---|---|---|---|---|
| Age, years | 66.3 (9.7) | 59.4 (10.2) | 66.0 (9.8) | < 0.001 |
| Age at diabetes diagnosis, years | 50.7 (10.3) | 46.3 (11.9) | 50.5 (10.5) | 0.027 |
| Female | 364 (51.1) | 28 (73.7) | 392 (52.3) | 0.013 |
| Education | 0.027 | |||
| < 12(less than high school) | 109 (15.4) | 8 (21.1) | 117 (15.6) | |
| = 12(high school graduate) | 325 (45.8) | 22 (57.9) | 347 (46.4) | |
| > 12(more than high school) | 276 (38.9) | 8 (21.1) | 284 (38.0) | |
| Body mass index, mean (SD) (kg/m2) | 33.0 (6.8) | 34.7 (6.7) | 33.1 (6.8) | 0.137 |
| Glucose (mg/dl) | 146.1 (54.5) | 153.0 (64.8) | 146.4 (55.0) | 0.509 |
| HbA1c | 7.5 (1.4) | 7.6 (1.6) | 7.5 (1.4) | 0.584 |
| Smoking | 0.019 | |||
| Never | 327 (46.1) | 15 (39.5) | 342 (45.8) | |
| Former | 312 (44.0) | 9 (23.7) | 321 (43.0) | |
| Current | 70 (9.9) | 14 (36.8) | 84 (11.2) | |
| Anxiety score (BSIAQ) | 4.2 (3.7) | 7.0 (5.1) | 4.4 (3.8) | < 0.001 |
| Depression scale (CES-D10) | 7.4 (5.0) | 13.3 (5.9) | 7.7 (5.3) | < 0.001 |
| DSSI | 4.8 (6.1) | 30.3 (12.6) | 6.1 (8.6) | < 0.001 |
| GCSI | 2.6 (3.8) | 24.1 (5.2) | 3.7 (6.1) | < 0.001 |
| Gall bladder surgery | 208 (29.2) | 13 (34.2) | 221 (29.5) | 0.495 |
| Gall stones history | 190 (26.7) | 12 (31.6) | 202 (26.9) | 0.508 |
| Upper endoscopy | 248 (34.8) | 21 (55.3) | 269 (35.9) | 0.010 |
| Gastric emptying | 42 (5.9) | 9 (23.7) | 51 (6.8) | < 0.001 |
| Antidepressant | 189 (26.5) | 19 (50) | 208 (27.7) | 0.002 |
| SSRI/SNRI | 156 (21.9) | 19 (50) | 175 (23.3) | < 0.001 |
| Tricyclic antidepressants | 38 (5.3) | 0 (0) | 38 (5.1) | – |
| Benzodiazepine | 77 (10.8) | 11 (29) | 88 (11.7) | 0.001 |
| Metoclopramide | 6 (0.8) | 5 (13.2) | 11 (1.5) | < 0.001 |
| PPI | 207 (29.1) | 22 (57.9) | 229 (30.5) | < 0.001 |
| Ranitidine | 22 (3.1) | 0 (0) | 22 (2.9) | – |
| Narcotics | 107 (15) | 18 (47.4) | 125 (16.7) | < 0.001 |
| Insulin | 241 (33.9) | 19 (50) | 260 (34.7) | 0.046 |
| Oral anti-hyperglycemics | 487 (68.4) | 28 (73.7) | 515 (68.7) | 0.501 |
BSIAQ Anxiety Questionnaire, DSSI Dyspepsia Symptom Severity Index, GCSI gastroparesis cardinal symptom index, SSRI/SNRI Selective Serotonin Reuptake Inhibitors/Serotonin and Norepinephrine Reuptake, Inhibitors, PPI proton pump inhibitor
Table 2.
Comparison of clinical attributes and diagnostic studies in African Americans with GCSI ≤ 17 versus ≥ 18 in the diabetes heart study
| 0–17 (< 95%) (N = 469) |
≥ 18 (≥ 95%) (N = 34) |
Overall (N = 503) |
p value | |
|---|---|---|---|---|
| Age, years | 58.9 (9.7) | 53.3 (7.9) | 58.6 (9.7) | < 0.001 |
| Age at diabetes diagnosis, years | 45.8 (10.1) | 40.1 (8.6) | 45.5 (10.1) | < 0.001 |
| Female | 282 (60.1) | 26 (76.5) | 308 (61.2) | 0.064 |
| Education | 0.806 | |||
| < 12(less than high school) | 57 (12.2) | 3 (8.8) | 60 (11.9) | |
| = 12(high school graduate) | 127 (27.1) | 10 (29.4) | 137 (27.2) | |
| > 12(more than high school) | 285 (60.8) | 21 (61.8) | 306 (60.8) | |
| Body mass index, mean (SD) (kg/m2) | 35.3 (8.8) | 36.1 (6.8) | 35.4 (8.7) | 0.550 |
| Glucose (mg/dl) | 147.5 (59.9) | 188.2 (90.1) | 150.3 (63.1) | 0.008 |
| HbA1c | 8.0 (2.0) | 9.1 (2.2) | 8.1 (2.0) | 0.002 |
| Smoking | 0.026 | |||
| Never | 216 (46.3) | 13 (38.2) | 229 (45.7) | |
| Former | 143 (30.6) | 4 (11.8) | 147 (29.3) | |
| Current | 108 (23.1) | 17 (50.0) | 125 (25.0) | |
| Anxiety score (BSIAQ) | 4.7 (4.4) | 8.9 (5.3) | 5.0 (4.6) | < 0.001 |
| Depression scale (CES-D10) | 7.9 (5.3) | 14.0 (6.5) | 8.3 (5.6) | < 0.001 |
| DSSI | 4.9 (6.5) | 28.8 (13.3) | 6.5 (9.3) | < 0.001 |
| GCSI | 2.9 (4.2) | 24.6 (5.5) | 4.4 (6.9) | < 0.001 |
| Gall bladder surgery | 63 (13.4) | 8 (23.5) | 71 (14.1) | 0.108 |
| Gall stones history | 61 (13.0) | 9 (26.5) | 70 (13.9) | 0.033 |
| Upper endoscopy | 104 (22.2) | 9 (26.5) | 113 (22.5) | 0.563 |
| Gastric emptying | 15 (3.2) | 2 (5.9) | 17 (3.4) | 0.410 |
| Antidepressant | 68 (14.5) | 11 (32.4) | 79 (15.7) | 0.007 |
| SSRI/SNRI | 53 (11.3) | 10 (29.4) | 63 (12.5) | 0.003 |
| Tricyclic antidepressants | 13 (2.8) | 1 (2.9) | 14 (2.8) | 0.953 |
| Benzodiazepine | 25 (5.3) | 6 (17.6) | 31 (6.2) | 0.006 |
| Metoclopramide | 2 (0.4) | 2 (5.9) | 4 (0.8) | 0.008 |
| PPI | 104 (22.2) | 17 (50) | 121 (24.1) | < 0.001 |
| Ranitidine | 14 (3) | 2 (5.9) | 16 (3.2) | 0.362 |
| Narcotics | 76 (16.2) | 10 (29.4) | 86 (17.1) | 0.053 |
| Insulin | 168 (35.8) | 22 (64.7) | 190 (37.8) | 0.001 |
| Oral anti-hyperglycemics | 260 (55.4) | 19 (55.9) | 279 (55.5) | 0.959 |
BSIAQ Anxiety Questionnaire, DSSI Dyspepsia Symptom Severity Index, GCSI gastroparesis cardinal symptom index, SSRI/SNRI Selective Serotonin Reuptake Inhibitors/Serotonin and Norepinephrine Reuptake, Inhibitors, PPI proton pump inhibitor
Table 1 compares EAs with GCSI scores ≥ 18 to EAs with GCSI ≤ 17. EAs with higher GCSI scores were younger (59.4 vs. 66.3 years, p < 0.001), had earlier onset of T2DM (46.3 vs. 50.7 years, p = 0.027) and were more often female (74% vs. 51%, p = 0.013). Mean HbA1c levels were similar in those with GCSI ≥ 18 versus ≤ 17 (7.6 vs. 7.5, respectively). HbA1c was >7.5 in 12 of 38 (31%) EA participants with GCSI ≥ 18 and in 282 of 712 (40%) EA participants with GCSI ≤ 17. Average BSIAQ scores (7) and CES-D10 depression scale scores (13.3) were higher in EAs with GCSI ≥ 18 compared with GCSI ≤ 17 (4.2 and 7.4, respectively; both p < 0.001). EAs with GCSI ≥ 18 had an average DSSI score 30.3 versus 4.8 in the GCSI ≤ 17 group (p < 0.001) and significantly more endoscopies (55.3% vs. 34.8%; p = 0.01) and gastric emptying tests (23.7% vs. 5.9%; p < 0.001) compared with AAs. A greater percentage of EAs with GCSI ≥ 18 prescribed the following medications, compared to individuals in the GCSI ≤ 17 group: antidepressants (50% vs. 21.9%, p = 0.002), benzodiazepines (29% vs. 10.8%, p = 0.001), metoclopramide (13.2% vs. 0.8%, p < 0.001), PPIs (57.9% vs. 29.1%, p < 0.001), narcotics (47.4% vs. 15%, p < 0.001), and insulin (50% vs. 33.9%, p = 0.05). No differences in use of oral hypoglycemic medications were observed between groups.
Table 2 compares AAs with GCSI scores ≥ 18 to AAs with GCSI ≤ 17; those with higher GCSI were younger (53.3 vs. 58.9 years, p < 0.001), had earlier onset of T2DM (40.1 vs. 45.8 years, p < 0.001), more often female (77% vs. 60%, p = 0.064), and had higher HbA1c (9.1 vs. 8.0, p = 0.002). HbA1c > 7.5 was present in 25 of 34 (74%) AA participants with GCSI ≥ 18 and in 212 of 496 (43%) AA participants with GCSI ≤ 17. BSIAQ scores averaged 8.9 (p < 0.001) and the CES-D10 depression scale averaged 14 (p < 0.001) in AAs with GCSI > 18 versus 4.7 and 7.9, respectively. The AAs with GCSI ≥ 18 had an average DSSI score 28.8 versus 4.9 in those with GCSI ≤ 17 (p < 0.001). Among AAs with GCSI ≥ 18, 26.5% underwent endoscopies and 6% had gastric emptying tests, rates not statistically different from those with GCSI ≤ 17. A greater percentage of AAs with GCSI ≥ 18 took the following medications, compared to individuals in the GCSI ≤ 17 group: benzodiazepines (17.6% vs. 5.3%, p = 0.006), metoclopramide (5.9% vs. 0.4%, p = 0.008), PPIs (50% vs. 22.2%, p < 0.001), narcotics (29.4% vs. 16.2%, p = 0.053), and insulin (64.7% vs. 35.8%, p = 0.001). The percentage of AA individuals with GCSI ≤ 17 or ≥ 18 taking oral hyperglycemic medications was similar in both groups (Table 2).
Table 3 shows comparative data in EAs and AAs with GCSI scores ≥ 18. Both groups had similar percentages female (73.7% vs. 76.5%, p = 0.785); BMI (34.7 vs. 36.1, p = 0.378), anxiety scores (7.0 vs. 8.9, p = 0.124); depression scores (13.3 vs. 14.0, p = 0.641), average GCSI scores (24.1 vs. 24.6, p = 0.683), DSSI scores (30.3 vs. 28.8, p = 0.602), and percentage of individuals using antidepressants (50% vs. 32.4%, p = 0.131), benzodiazepines (29% vs. 17.6%, p = 0.263), metoclopramide (13.2% vs. 5.9%, p = 0.310), PPIs (57.9% vs. 50%, p = 0.502), narcotics (47.4% vs. 29.4%, p = 0.121), insulin (50% vs. 64.7%, p = 0.210), and oral anti-hyperglycemic agents (73.7% vs. 55.9%, p = 0.116). Among those with GCSI ≥ 18, AAs were significantly younger (59.4 vs. 53.3 years, p = 0.004); had earlier onset of T2DM (46.3 vs. 40.1 years, p = 0.011); higher HbA1c (7.6 vs. 9.1, p < 0.001); and underwent fewer upper endoscopies and gastric emptying studies, (26.5% vs. 55.3%, p = 0.015) and (6% vs. 24%, p = 0.051), respectively, than EAs.
Table 3.
Comparison of clinical attributes and diagnostic studies in European American versus African Americans with GCSI ≥ 18 in the diabetes heart study
| EA (N = 38) |
AA (N = 34) |
Overall (N = 72) |
p value | |
|---|---|---|---|---|
| Age, years | 59.4 (10.2) | 53.3 (7.9) | 56.5 (9.6) | 0.004 |
| Age at diabetes diagnosis, years | 46.3 (11.9) | 40.1 (8.6) | 43.3 (10.8) | 0.011 |
| Female | 28 (73.7) | 26 (76.5) | 54 (75.0) | 0.785 |
| Education | 0.001 | |||
| < 12(less than high school) | 8 (21.1) | 3 (8.8) | 11 (15.3) | |
| = 12(high school graduate) | 22 (57.9) | 10 (29.4) | 32 (44.4) | |
| > 12(more than high school) | 8 (21.1) | 21 (61.8) | 29 (40.3) | |
| Body mass index, mean (SD) (kg/m2) | 34.7 (6.7) | 36.1 (6.8) | 35.3 (6.7) | 0.378 |
| Glucose (mg/dl) | 153.0 (64.8) | 188.2 (90.1) | 169.9 (79.4) | 0.057 |
| HbA1c | 7.6 (1.6) | 9.1 (2.2) | 8.3 (2.0) | < 0.001 |
| Smoking | 0.486 | |||
| Never | 15 (39.5) | 13 (38.2) | 28 (38.9) | |
| Former | 9 (23.7) | 4 (11.8) | 13 (18.1) | |
| Current | 14 (36.8) | 17 (50.0) | 31 (43.1) | |
| Anxiety score (BSIAQ) | 7.0 (5.1) | 8.9 (5.3) | 7.9 (5.3) | 0.124 |
| Depression scale (CES-D10) | 13.3 (5.9) | 14.0 (6.5) | 13.6 (6.2) | 0.641 |
| DSSI | 30.3 (12.6) | 28.8 (13.3) | 29.6 (12.9) | 0.602 |
| GCSI | 24.1 (5.2) | 24.6 (5.5) | 24.3 (5.3) | 0.683 |
| Gall bladder surgery | 13 (34.2) | 8 (23.5) | 21 (29.2) | 0.321 |
| Gall stones history | 12 (31.6) | 9 (26.5) | 21 (29.2) | 0.634 |
| Upper endoscopy | 21 (55.3) | 9 (26.5) | 30 (41.7) | 0.015 |
| Gastric emptying | 9 (23.7) | 2 (5.9) | 11 (15.3) | 0.051 |
| Antidepressant | 19 (50) | 11 (32.4) | 30 (41.7) | 0.131 |
| SSRI/SNRI | 19 (50) | 10 (29.4) | 29 (40.3) | 0.078 |
| Tricyclic antidepressants | 0 (0) | 1 (2.9) | 1 (1.4) | – |
| Benzodiazepine | 11 (29) | 6 (17.6) | 17 (23.6) | 0.263 |
| Metoclopramide | 5 (13.2) | 2 (5.9) | 7 (9.7) | 0.310 |
| PPI | 22 (57.9) | 17 (50) | 39 (54.2) | 0.502 |
| Ranitidine | 0 (0) | 2 (5.9) | 2 (2.8) | – |
| Narcotics | 18 (47.4) | 10 (29.4) | 28 (38.9) | 0.121 |
| Insulin | 19 (50) | 22 (64.7) | 41 (56.9) | 0.210 |
| Oral anti-hyperglycemics | 28 (73.7) | 19 (55.9) | 47 (65.3) | 0.116 |
BSIAQ Anxiety Questionnaire, DSSI Dyspepsia Symptom Severity Index, GCSI gastroparesis cardinal symptom index, SSRI/SNRI Selective Serotonin Reuptake Inhibitors/Serotonin and Norepinephrine Reuptake, Inhibitors, PPI proton pump inhibitor
Table 4 compares EAs and AAs with GCSI scores ≤ 17. Similar depression scores (7.4 vs 7.9, p = 0.082), DSSI scores (4.8 vs. 4.9, p = 0.6667) average GCSI scores (2.6 vs. 2.9, p = 0.156), and metoclopramide use (0.8% vs. 0.4%, p = 0.338), ranitidine (3.1% vs. 3%, p = 0.519), narcotics (15% vs. 16.2%, p = 0.317), and insulin (33.9% vs. 35.8%, p = 0.485) were observed. Among those with GCSI scores ≤ 17, AAs were younger (66.3 vs. 58.9 years, p < 0.001), had earlier onset of T2DM (50.7 vs. 45.8 years, p < 0.001), higher BMI (33 vs 35.3, p < 0.001), higher HbA1c (7.5 vs. 8.0, p < 0.001), higher anxiety scores (4.2 vs. 4.7, p = 0.052), underwent fewer upper endoscopies and gastric emptying studies, (34.8% vs. 22.2%, p < 0.001) and (6% vs. 3%, p = 0.037), respectively, and fewer took antidepressants (26.5% vs. 14.5%, p < 0.001), tricyclic antidepressants (5.3% vs. 2.8%, p = 0.009), benzodiazepines (10.8% vs. 5.3%, p = 0.001), PPIs (29.1% vs. 22.2%, p = 0.002), and oral anti-hyperglycemic agents (68.4% vs. 55.4%, p < 0.001) compared to EAs.
Table 4.
Comparison of clinical attributes and diagnostic studies in European American versus African Americans with GCSI ≤ 17 in the diabetes heart study
| EA (N = 712) |
AA (N = 469) |
Overall (N = 1181) | p value | |
|---|---|---|---|---|
| Age, years | 66.3 (9.7) | 58.9 (9.7) | 63.4 (10.3) | < 0.001 |
| Age at diabetes diagnosis, years | 50.7 (10.3) | 45.8 (10.1) | 48.7 (10.5) | < 0.001 |
| Female | 364 (51.1) | 282 (60.1) | 646 (54.7) | 0.002 |
| Education | < 0.001 | |||
| < 12(less than high school) | 109 (15.4) | 57 (12.2) | 166 (14.1) | |
| = 12(high school graduate) | 325 (45.8) | 127 (27.1) | 452 (38.3) | |
| > 12(more than high school) | 276 (38.9) | 285 (60.8) | 561 (47.6) | |
| Body mass index, mean (SD) (kg/m2) | 33.0 (6.8) | 35.3 (8.8) | 33.9 (7.8) | < 0.001 |
| Glucose (mg/dl) | 146.1 (54.5) | 147.5 (59.9) | 146.7 (56.7) | 0.704 |
| HbA1c | 7.5 (1.4) | 8.0 (2.0) | 7.7 (1.6) | < 0.001 |
| Smoking | 0.026 | |||
| Never | 327 (46.1) | 216 (46.3) | 543 (46.2) | |
| Former | 312 (44.0) | 143 (30.6) | 455 (38.7) | |
| Current | 70 (9.9) | 108 (23.1) | 178 (15.1) | |
| Anxiety score (BSIAQ) | 4.2 (3.7) | 4.7 (4.4) | 4.4 (4.0) | 0.052 |
| Depression scale (CES-D10) | 7.4 (5.0) | 7.9 (5.3) | 7.6 (5.1) | 0.082 |
| DSSI | 4.8 (6.1) | 4.9 (6.5) | 4.8 (6.2) | 0.666 |
| GCSI | 2.6 (3.8) | 2.9 (4.2) | 2.7 (4.0) | 0.156 |
| Gall bladder surgery | 208 (29.2) | 63 (13.4) | 271 (23.0) | < 0.001 |
| Gall stones history | 190 (26.7) | 61 (13.0) | 251 (21.3) | 0.001 |
| Upper endoscopy | 248 (34.8) | 104 (22.2) | 352 (29.8) | 0.049 |
| Gastric emptying | 42 (5.9) | 15 (3.2) | 57 (4.8) | 0.748 |
| Antidepressant | 189 (26.5) | 68 (14.5) | 257 (21.8) | < 0.001 |
| SSRI/SNRI | 156 (21.9) | 53 (11.3) | 209 (17.7) | < 0.001 |
| Tricyclic antidepressants | 38 (5.3) | 13 (2.8) | 51 (4.3) | 0.009 |
| Benzodiazepine | 77 (10.8) | 25 (5.3) | 102 (8.6) | 0.001 |
| Metoclopramide | 6 (0.8) | 2 (0.4) | 8 (0.7) | 0.338 |
| PPI | 207 (29.1) | 104 (22.2) | 311 (26.3) | 0.002 |
| Ranitidine | 22 (3.1) | 14 (3) | 36 (3.1) | 0.519 |
| Narcotics | 107 (15) | 76 (16.2) | 183 (15.5) | 0.317 |
| Insulin | 241 (33.9) | 168 (35.8) | 409 (34.6) | 0.485 |
| Oral anti-hyperglycemics | 487 (68.4) | 260 (55.4) | 747 (63.3) | < 0.001 |
BSIAQ Anxiety Questionnaire, DSSI Dyspepsia Symptom Severity Index, GCSI gastroparesis cardinal symptom index, SSRI/SNRI Selective Serotonin Reuptake Inhibitors/Serotonin and Norepinephrine Reuptake, Inhibitors, PPI proton pump inhibitor
Table 5 displays the association between risk factors and symptom severity in all participants (GCSI ≤ 17 and GCSI ≥ 18). Age (OR 0.95, p < 0.001), female gender (OR 2.11, p = 0.021), current tobacco use (OR 2.42, p = 0.006), depression (OR 1.14, p < 0.0001), PPI use (OR 3.24, p < 0.001), narcotic use (OR 2.64, p < 0.001), insulin use (OR 2.39, p = 0.003), and oral anti-hyperglycemic medication use (OR 1.90, p = 0.039) were associated with increased symptom severity. Ancestry, history of tobacco use, and BMI were not significantly associated with symptom severity.
Table 5.
Multivariable association between risk factors and symptom severity in all subjects (GCSI ≤ 17 and GCSI ≥ 18) in the diabetes heart study
| Variables | OR (95% CI) | p value |
|---|---|---|
| Age, years | 0.95 (0.92, 0.98) | < 0.001 |
| Female versus Male | 2.11 (1.12, 3.98) | 0.021 |
| AA versus EA | 0.71 (0.40, 1.26) | 0.240 |
| Current smoking versus never smoking | 2.42 (1.28, 4.57) | 0.006 |
| Former smoking versus never smoking | 0.62 (0.30, 1.29) | 0.198 |
| BMI (kg/m2) | 0.97 (0.94, 1.00) | 0.082 |
| Depression scale | 1.14 (1.09, 1.19) | < 0.001 |
| PPI | 3.24 (1.87, 5.61) | < 0.001 |
| Narcotics | 2.64 (1.48, 4.68) | < 0.001 |
| Insulin | 2.39 (1.34, 4.25) | 0.003 |
| Oral anti-hyperglycemic medications | 1.90 (1.03, 3.49) | 0.039 |
AA African American, EA European American, PPI proton pump inhibitor
Discussion
To our knowledge, this is the first study using the GCSI to determine prevalence of GP symptoms in EAs and AAs in a large community-based population with T2DM. The cumulative scores of 24.1 for EAs and 24.6 for AAs yield an average GCSI score of 2.67 and 2.73, respectively. Thus, the average GCSI scores are > 2.5 in our EA and AA cohorts and indicate moderate symptoms. Results from the GCSI revealed moderate or greater symptoms suggestive of GP in 5% of EAs and 7% of AAs with T2DM. Patients with GP had a higher average GCSI score compared with those with normal gastric emptying (average GCSI score 24.6 vs. 20.7) [21]. In DHS participants, the average GCSI score was 24.1 in EAs and 24.6 in AAs. GCSI scores in this range indicate an increase likelihood of gastroparesis. GCSI ≥ 18 in our cohort also identified a group with clinically relevant symptoms since 32 of the 74 subjects (42%) had endoscopy. Only 15% (11 of 72) had gastric emptying studies, suggesting symptoms associated with gastroparesis may be underappreciated [22]. Thus, GCSI ≥ 18 in individuals with T2DM may identify subjects at risk for GP who would benefit from further workup with gastric emptying studies after upper endoscopy is performed to exclude organic causes of these symptoms. Further, our findings suggest that 1.5–2.1 million patients of an estimated 30 million subjects with T2DM would have GCSI scores > 18 and that many may have clinically relevant symptoms and gastroparesis [23].
The DSSI is designed to quantify the spectrum of dyspepsia symptoms, which include reflux-like, ulcer-like, and dysmotility-like symptoms [15]. Dysmotility-like symptoms with normal gastric emptying are similar to symptoms of gastroparesis [24, 25]. Furthermore, the majority of the dysmotility-like symptoms in the DSSI questionnaire overlap with the GCSI symptoms. The present study shows a correlation between the composite DSSI symptoms and GCSI ≥ 18. EAs and AAs with a GCSI ≥ 18 had higher DSSI scores compared to EAs and AAs with a GCSI ≤ 17. Thus, the GCSI ≥ 18 is associated with a higher DSSI score and corroborates further the severity of GI symptoms in the subjects identified with GCSI ≥ 18.
Our results showed AAs underwent fewer diagnostic procedures such as upper endoscopy and gastric emptying studies, suggesting disparities in diagnostic evaluation of symptoms suggestive of GP. A recent study showed AAs had higher GCSI scores and a higher percentage of hospitalizations related to gastroparesis [26]. These studies suggest that a large number of AAs with T2DM may have undiagnosed GP. Nonwhite individuals with diabetes have a higher proportion of GP compared to white individuals (55% vs. 19%) [27]. Patients with gastroparesis and diabetes oftentimes have difficulty controlling glycemia. Individuals with symptoms suggestive of GP have more emergency department (ED) visits in the year preceding the diagnosis of GP with a gastric emptying test [21]. Thus, further diagnostic workup of patients with GCSI ≥ 18 with endoscopy to exclude mucosal diseases and four-hour solid-phase GE study may identify subjects with GP and lead to diet and drug treatments to improve symptoms and glycemia. EAs and AAs with GCSI ≥ 18 had similar percentages of individuals taking PPIs and prokinetic drugs (Table 3). Our study does not identify why AAs had fewer diagnostic GI studies than EAs; however, possible reasons include economic disparities, access to health care issues, cultural biases, or under appreciation of symptoms suggestive of GP.
The average HbA1c in EAs with GCSI ≥ 18 was significantly lower compared with AAs with GCSI ≥ 18 scores (7.6 vs. 9.1, p < 0.01) despite similar usage of oral glucose lowering agents and similar average GCSI scores (24.1 vs. 24.6) (Table 3). AAs with GCSI ≤ 17 had significantly higher HbA1c and less use of oral anti-hyperglycemics compared with EAs, but both groups had similar GCSI scores as shown in Table 4. These findings suggest that glycemic control, at least to this level, was not a factor influencing the severity of symptoms as assessed by GCSI. A recent study in individuals with T1DM and T2DM and GP also showed no relationship between HbA1c and GCSI scores, suggesting overall glycemic control as assessed by HbA1c does not appear to influence GI symptoms [6]. This report is in contrast to other studies that have suggested a correlation between HbA1c and severity of gastroparesis symptoms [28, 29]. Furthermore, a prospective study of T1 and T2DM with defined GP who had intense glycemic therapy with insulin pump and CGM had a significant decrease in GCSI scores after 6 months of treatment [30]. These findings support the importance of identifying T2DM patients with GCSI ≥ 18, as a portion of the patients may have undetected GP.
Risk factors associated with the severity of GCSI scores (GCSI ≥ 18) included older age, female gender, current smoking, depression, and a variety of prescribed drugs: PPI, narcotics, insulin, and oral anti-hyperglycemics (Table 5). Women comprised the majority of subjects in the EA (77%) and AA (74%) groups with GCSI scores ≥ 18. Women also have a higher incidence of idiopathic and diabetic GP compared with men [2, 31-33]. Fluctuations in sex hormones may impair niterergic-mediated gastric motility or delay GI transit time in the luteal phase of menstruation, but the cause of female predominance in GP is likely multifactorial [31, 34]. Psychosocial factors may also play a role as women reportedly seek health care more frequently than men [31]. Other risk factors associated with ≥ 18 GCSI included smoking, PPIs, and narcotics, which may decrease gastric emptying. DHS participants with T2DM volunteered to participate in this prospective study and many had prevalent clinical or subclinical cardiovascular disease. [35, 36] This aspect reflects similar comorbidities as diabetic individuals in the general population [37]. Depression was an additional risk in the group with GCSI scores ≥ 18. The higher depression and anxiety scores in subjects with GCSI ≥ 18 are consistent with other studies that showed these symptoms were associated with increased GP severity [38]. The combination of depression and anxiety and symptoms suggestive of GP likely contributes to decreased quality of life in T2DM patients [39].
The prevalence of symptoms suggestive of GP using GCSI has not been studied in large community-based American populations with T2DM. We found the prevalence of GCSI ≥ 18 was slightly higher in AAs (7%) than EAs (5%) and it has been suggested that AA individuals experienced more severe symptoms associated with GP compared with white individuals [26, 27]. The prevalence of symptoms suggestive of GP in T2DM subjects was 10.8% in Saudi Arabians and 10% in a study from India [40, 41]. Cassilly et al. reported that no single aspect of the GCSI predicted the presence of gastroparesis, but a cumulative score of 18 had a PPV of 51.7% and NPV of 62.7%. However, their population included only 15% of participants with diabetes [42]. Use of GCSI to study the prevalence of symptoms suggestive of GP may be an important tool in identifying T2DM patients who may actually have gastroparesis or other gastric neuromuscular disorders that cause these symptoms. Limitations of the present analysis included the lack of data on socioeconomic status in participants and reasons medications affecting the gastrointestinal tract were prescribed.
In summary, the prevalence of GCSI scores ≥ 18 was 5% and 7% in EAs and AAs, respectively, in a community-based cohort of patients with T2DM. (Average GCSI scores were approximately 24 in both groups.) A portion of these subjects may have gastroparesis or other gastric neuromuscular disorders. AAs underwent fewer GI procedures, suggesting either poorer access to healthcare or under appreciation of symptoms suggestive of GP. AAs with T2DM and GCSI ≥ 18 were younger and had longer duration diabetes compared to their EA counterparts. Individuals with GCSI ≥ 18 had higher prevalence of depression and anxiety, suggesting that symptoms of GP have a strong negative influence on quality of life. Measurement of the GCSI in T2DM population studies may uncover patients with undiagnosed gastroparesis or gastric neuromuscular disorders.
Supplementary Material
Funding
This funding was provided by National Institutes of Health (Grant Nos. NIH R01 HL92301 (DWB); R01 HL67348 (DWB) and R01 DK071891 (BIF)).
Footnotes
Conflict of interest All authors have no conflicts of interest or financial disclosures to disclose.
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s10620-019-05974-z) contains supplementary material, which is available to authorized users.
References
- 1.Koch KL, Calles-Escandan J. Diabetic gastroparesis. Gastroenterol Clin North Am. 2015;44:39–57. [DOI] [PubMed] [Google Scholar]
- 2.Kofod-Andersen K, Tarnow L. Prevalence of gastroparesis-related symptoms in an unselected cohort of patients with type 1 diabetes. J Diabetes Complicat. 2012;26:89–93. [DOI] [PubMed] [Google Scholar]
- 3.Spangeus A, El-Salhy M, Suhr O, Eriksson J, Lithner F. Prevalence of gastrointestinal symptoms in young and middle-aged diabetic patients. Scand J Gastroenterol. 1999;34:1196–1202. [DOI] [PubMed] [Google Scholar]
- 4.Ricci JA, Siddique R, Steward WF, Sandler RS, Sloan S, Farup CE. Upper gastrointestinal symptoms in a U.S. national sample of adults with diabetes. Scand J Gastroenterol. 2000;35:152–159. [DOI] [PubMed] [Google Scholar]
- 5.Dickman R, Kislov J, Boaz M, et al. Prevalence of symptoms suggestive of gastroparesis in a cohort of patients with diabetes mellitus. J Diabetes Complicat. 2013;27:376–379. [DOI] [PubMed] [Google Scholar]
- 6.Koch KL, Hasler WL, Yates KP, et al. Baseline features and differences in 48 week clinical outcomes in patients with gastroparesis and type 1vs type 2 diabetes. Neurogastroenterol Motil. 2016;28:1001–1015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wang YR, Fisher RS, Parkman HP. Gastroparesis-related hospitalizations in the United States: trends, characteristics, and outcomes, 1995–2004. Am J Gastroenterol. 2008;103:313–322. [DOI] [PubMed] [Google Scholar]
- 8.Intagliata N, Koch KL. Gastroparesis in type 2 diabetes mellitus: prevalence, etiology, diagnosis, and treatment. Curr Gastroenterol Rep. 2007;9:270–279. [DOI] [PubMed] [Google Scholar]
- 9.Janatuinen E, Pikkarainen P, Laakso M, Pyörälä K. Gastrointestinal symptoms in middle-aged diabetic patients. Scand J Gastroenterol. 1993;28:427–432. [DOI] [PubMed] [Google Scholar]
- 10.Bytzer P, Talley NJ, Hammer J, Young LJ, Jones MP, Horowitz M. GI symptoms in diabetes mellitus are associated with both poor glycemic control and diabetic complications. Am J Gastroenterol. 2002;97:604–611. [DOI] [PubMed] [Google Scholar]
- 11.Parkman HP, Van Natta ML, Abeil TL, et al. Effect of nortriptyline on symptoms of idiopathic gastroparesis: the NORIG randomized clinical trial. JAMA. 2013;310:2640–2649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Cassilly DW, Wang YR, Friedenberg FK, Nelson DB, Maurer AH, Parkman HP. Symptoms of gastroparesis: use of the gastroparesis cardinal symptom index in symptomatic patients referred for gastric emptying scintigraphy. Digestion. 2008;78:144–151. [DOI] [PubMed] [Google Scholar]
- 13.Friedenberg FK, Palit A, Parkman HP, Hanlon A, Nelson DB. Botulinum toxin A for the treatment of delayed gastric emptying. Am J Gastroenterol. 2008;103:416–423. [DOI] [PubMed] [Google Scholar]
- 14.Revicki DA, Rentz AM, Dubois D, et al. Gastroparesis Cardinal Symptom Index (GCSI): development and validation of a patient reported assessment of severity of gastroparesis symptoms. Qual Life Res. 2004;13:833–844. [DOI] [PubMed] [Google Scholar]
- 15.Leidy NK, Farup C, Rentz AM, Ganoczy D, Koch KL. Patient-based assessment in dyspepsia: development and validation of Dyspepsia Symptom Severity Index (DSSI). Dig Dis Sci. 2000;45:1172–1179. 10.1023/a:1005558204440. [DOI] [PubMed] [Google Scholar]
- 16.Radloff L. The CES-D scale: a self-report depression scale for research in the general population. Psychol Assess. 1977;1:385–401. [Google Scholar]
- 17.Kato N, Kinugawa K, Shiga T, et al. Depressive symptoms are common and associated with adverse clinical outcomes in heart failure with reduced and preserved ejection fraction. J Cardiol. 2012;60:23–30. [DOI] [PubMed] [Google Scholar]
- 18.Giese-Davis J, Collie K, Rancourt KM, Neri E, Kraemer HC, Spiegel D. Decrease in depression symptoms is associated with longer survival in patients with metastatic breast cancer: a secondary analysis. J Clin Oncol. 2011;29:413–420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Crameri A, Schuetz C, Andreae A, et al. The brief symptom inventory and the outcome questionnaire-45 in the assessment of the outcome quality of mental health interventions. Psychiatry J. 2016;2016:7830785. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Derogatis LR, Melisaratos N. The brief symptom inventory: an introductory report. Psychol Med. 1983;13:595–605. [PubMed] [Google Scholar]
- 21.DiBaise JK, Patel N, Noelting J, Dueck AC, Roarke M, Crowell MD. The relationship among gastroparetic symptoms, quality of life, and gastric emptying in patients referred for gastric emptying testing. Neurogastroenterol Motil. 2016;28:234–242. [DOI] [PubMed] [Google Scholar]
- 22.Parkman HP, Hasler WL, Fisher RS, American Gastroenterological Association. American Gastroenterological Association medical position statement: diagnosis and treatment of gastroparesis. Gastroenterology. 2004;127:1589–1591. [DOI] [PubMed] [Google Scholar]
- 23.Centers for Disease Control and Prevention. Diabetes Report Card 2014. Atlanta, GA: Centers for Disease Control and Prevention, US Department of Health and Human Services; 2015. [Google Scholar]
- 24.Stanghellini V, Tack J. Gastroparesis: separate entity or just a part of dyspepsia? Gut. 2014;63:1972–1978. [DOI] [PubMed] [Google Scholar]
- 25.Cherian D, Sachdeva P, Fisher RS, Parkman HP. Abdominal pain is a frequent symptom of gastroparesis. Clin Gastroenterol Hepatol. 2010;8:676–681. [DOI] [PubMed] [Google Scholar]
- 26.Parkman HP, Yamada G, Van Natta ML, et al. Ethnic, racial, and sex differences in etiology, symptoms, treatment, and symptom outcomes of patients with gastroparesis. Clin Gastroenterol Hepatol. 2019;17:1489–1499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Friedenberg FK, Kowalczyk M, Parkman HP. The influence of race on symptom severity and quality of life in gastroparesis. J Clin Gastroenterol. 2013;47:757–761. [DOI] [PubMed] [Google Scholar]
- 28.Homko C, Siraj ES, Parkman HP. The impact of gastroparesis on diabetes control: patient perceptions. J Diabetes Complicat. 2016;30:826–829. [DOI] [PubMed] [Google Scholar]
- 29.Almogbel RA, Alhussan FA, Alnasser SA, Algeffari MA. Prevalence and risk factors of gastroparesis-related symptoms among patients with type 2 diabetes. Int J Health Sci (Qassim). 2016;10:397–404. [PMC free article] [PubMed] [Google Scholar]
- 30.Calles-Escandón J, Koch KL, Hasler WL, et al. GpCRC Glucose sensor-augmented continuous subcutaneous insulin infusion in patients with diabetic gastroparesis: an open-label pilot prospective study. PLoS One. 2018;13:e0194759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Jung HK, Choung RS, Locke GR III, et al. The incidence, prevalence, and outcomes of patients with gastroparesis in Olmsted County, Minnesota, from 1996 to 2006. Gastroenterology. 2009;136:1225–1233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Dickman R, Wainstein J, Glezerman M, Niv Y, Boaz M. Gender aspects suggestive of gastroparesis in patients with diabetes mellitus: a cross-sectional survey. BMC Gastroenterol. 2014;14:34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Samsom M, Vermeijden JR, Smout AJ, et al. Prevalence of delayed gastric emptying in diabetic patients and relationship to dyspeptic symptoms: a prospective study in unselected diabetic patients. Diabetes Care. 2003;26:3116–3122. [DOI] [PubMed] [Google Scholar]
- 34.Gangula PR, Sekhar KR, Mukhopadhyay S. Gender bias in gastroparesis: is nitric oxide the answer? Dig Dis Sci. 2011;56:2520–2527. 10.1007/s10620-011-1735-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Chan GC, Divers J, Russell GB, et al. FGF23 concentration and APOL1 genotype are novel predictors of mortality in African Americans with type 2 diabetes. Diabetes Care. 2018;41:178–186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Raffield LM, Hsu FC, Cox AJ, Carr JJ, Freedman BI, Bowden DW. Predictors of all-cause and cardiovascular disease mortality in type 2 diabetes: diabetes heart study. Diabetol Metab Syndr. 2015;28:58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Celeng C, Maurovich-Horvat P, Ghoshhajra BB, Merkely B, Leiner T, Takx RA. Prognostic value of coronary computed tomography angiography in patients with diabetes: a meta-analysis. Diabets Care. 2016;39:1274–1280. [DOI] [PubMed] [Google Scholar]
- 38.Woodhouse S, Hebbard G, Knowles SR. Psychological controversies in gastroparesis: a systematic review. World J Gastroenterol. 2017;23:1298–1309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Hasler WL, Parkman HP, Wilson LA, et al. Psychological dysfunction is associated with symptom severity but not disease etiology or degree of gastric retention in patients with gastroparesis. Am J Gastroenterol. 2010;105:2357–2367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Almogbel RA, Alhussan FA, Alnasser SA, Algeffari MA. Prevalence and risk factors of gastroparesis-related symptoms among patients with type 2 diabetes. Int J Health Sci (Qassim). 2016;10:397–404. [PMC free article] [PubMed] [Google Scholar]
- 41.Anudeep V, Vinod KV, Pandit N, et al. Prevalence and predictors of delayed gastric emptying among Indian patients with longstanding type 2 diabetes mellitus. Indian J Gastroenterol. 2016;35:385–392. [DOI] [PubMed] [Google Scholar]
- 42.Cassilly DW, Wang YR, Friedenberg FK, Nelson DB, Maurer AH, Parkman HP. Symptoms if gastroparesis: use of the Gastroparesis Cardinal Symptoms Index in symptomatic patients referred for gastric emptying scintigraphy. Digestion. 2008;78:144–151. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.