A wide variety of gastrointestinal symptoms have been reported to occur in adults with type 1 diabetes mellitus (T1DM) (1,2). An autonomic neuropathy (3,4) and acute hyperglycemia (5,6) have been implicated in the pathogenesis. In children with T1DM, gastrointestinal symptoms are frequently seen, although their pathophysiology and impact on diabetes control are poorly defined (7–9). The aim of this study was to define the etiological role of autonomic neuropathy and potential motility impairment in children and adolescents with T1DM and gastrointestinal symptoms.
PATIENTS AND METHODS
The study group contained 33 consecutive children and adolescents (mean [SD] age, 15.3 [4.9] years; 13 males) with T1DM (median duration of disease, 7.7 years; range, 4–17.5) attending the Outpatient Paediatric Gastroenterology Clinic because of chronic dyspepsia (CD; n = 14; mean [SD], age 15.3 [5.8] years; 7 males; severe in 3, moderate in 4, and mild in 7) or chronic constipation (CC; n = 19; mean [SD], age 14.8 [4.6] years; 6 males) in whom cardiovascular assessments were performed (Table 1).
A study subgroup contained 23 patients (mean [SD] age, 15.0 [4.5] years; 10 males) selected from the initial cohort of the 33 patients with T1DM in whom fasting serum motilin concentrations were measured. Fourteen of these patients had CD, whereas 9 were patients with CC selected with simple random sampling from the initial cohort of patients with constipation. Another two study subgroups used for gastrointestinal motility assessments were 14 patients with T1DM and CD and the 19 patients with CC in whom gastric emptying and mouth-to-anus transit time were measured, respectively.
The control group contained 36 healthy individuals (mean age [SD], 14.3 [4.6] years; 18 males) who were siblings of patients with T1DM or offspring of the hospital staff, matched in age and gender as control subjects for the cardiovascular tests. A control subgroup contained 22 (mean [SD] age, 13.8 [5.3] years; 9 males) healthy control subjects, selected with simple random sampling from the initial cohort, in whom serum motilin concentrations were measured. Another two control subgroups used for the motility comparisons contained patients who did not have T1DM but who attended the gastroenterology outpatient clinic during the study period because of CD (n = 18 patients; mean [SD] age, 14.1 [0.8] years; 5 males; severe in 4, moderate in 5, and mild in 9) or CC (n = 30; mean [SD] age, 13.5 [0.6] years; 13 males) in whom gastric emptying or mouth-to-anus transit time were measured, respectively. All patients with CD had negative Helicobacter pylori antibodies (by enzyme-linked immunosorbent assay).
The study was approved by the local Ethics Committee.
The clinical definition of CD included the presence of postprandial fullness or discomfort as the predominant symptom, associated with two or more of the following symptoms: early satiety, bloating or upper abdominal distention, burning, recurrent belching or eructation, nausea, vomiting, epigastric or retrosternal burning, retching, and anorexia. To fit the definition of “chronic dyspepsia,” the symptoms had to have been present for more than 25% of the time for a period of more than 2 months. The severity of the predominant symptom was classified according to the following criteria: mild when it did not interfere with the child's usual activities, moderate when it affected the child's usual activities, and severe when it forced the child to end his activities.
The clinical definition of CC included passing fewer than three stools per week, with stools of hard consistency, with or without soiling, for more than 3 consecutive months.
Weight and height were recorded, and the nutritional status was expressed as body mass index. Gastric emptying t1/2 and mouth-to-anus transit time were measured as indicated below. Blood samples for motilin, blood glucose, and HbA1C concentrations were performed on the same day with the patient in a fasting state. Autonomic function tests were checked on a separate day.
Tests of Autonomic Function
Cardiovascular tests were performed according to a previously described protocol (10). Supine blood pressure was measured after 10 minutes of rest with an electronic sphygmomanometer (mean of three measurements).
- Assessment of the R-R interval variation during deep breathing. The patient was instructed to perform six respiratory cycles for 1 minute. The result was expressed as the mean of the difference between maximum and minimum R-R intervals converted to beats per minute.
- Assessment of the heart-rate response to Valsalva maneuver (Valsalva ratio). The patient was instructed to attempt a forcible expiration against a closed glottis lasting for 15 seconds, after a deep inspiration. The Valsalva ratio was defined as the ratio of the longest R-R interval during the post-Valsalva reflex bradycardia divided by the shortest.
- Assessment of the immediate heart rate response to standing (30/15 heart-rate ratio). The longest R-R interval around beat 30 was divided by the shortest R-R interval around beat 15.
- Determination of the blood pressure response to standing, defined as the difference between the systolic blood pressure lying and standing, during the previous maneuver. The first three tests primarily assess cardiac parasympathetic activity, whereas the last one assesses sympathetic activity.
Gastric Emptying of a Solid Meal and Mouth-to-Anus Transit Time
Gastric emptying (GE) half time (t1/2) of a solid meal was determined in all patients with CD by gastric scintigraphy using 99Tc-phytate (11). The given meal consisted of 150 g of minced meat mixed with egg labeled with 150 to 250 μCI of 99Tc-phytate. Blood glucose concentrations were evaluated on the same day before gastric scintigraphy.
Mouth-to-anus transit time (MATT) was measured using carmine red (12) in patients with T1DM and CC and was compared with that of non-T1DM patients with the same diagnosis.
After the patients and subjects fasted overnight, blood samples were collected in chilled tubes and immediately cooled in an ice bath. Motilin concentrations were measured by radioimmunoassay using the Eurodiagnostica kit.
HbA1C was assessed using a Bayer DCA 2000 analyzer. The assay was based on a latex immunoagglutination inhibition method. The laboratory reference range in normal children is 4.5% to 5.7%. Fasting blood glucose concentrations were measured with a Hitachi analyzer, using the glucose oxidase method.
All comparisons between two groups were made using the unpaired Student t test for quantitative variables or the unpaired Mann Whitney's test where appropriate. Pearson's χ2 test with Yate's correction of continuity was used for comparing categorical variables or Fisher exact test where appropriate. Univariate and multiple linear regression analysis were used to evaluate any possible correlations between motilin concentrations, GE, or MATT, and age, duration of diabetes, body mass index, gender, current or last year's HbA1C concentrations, or blood glucose concentrations. The statistical tests were performed by the SPSS 8.0 for Windows Software (SPSS Inc., Chicago, IL).
Cardiovascular autonomic test results were similar for patients with T1DM with gastrointestinal symptoms and control subjects, thus excluding autonomic neuropathy as an etiologic factor of the gastrointestinal symptoms (Table 2).
GE t1/2 of a solid meal was similar in T1DM and non-T1DM patients with CD: mean (SD) GE t1/2, 114.4 minutes (30.2 minutes) versus 94.3 minutes (32.5 minutes), respectively, P = 0.2. In addition, a marginally significant positive correlation was found in the group of patients with T1DM and CD between GE t1/2 and blood glucose concentrations (R = 0.54; B = 0.14; 95% CI, −0.02 to 0.3; P = 0.08). However, it should be noted that on the day of the assessment of GE, most of the patients with T1DM and CD had mild or moderate hyperglycemia (serum glucose, <170 mg/dL).
MATT was comparable in T1DM and non-T1DM patients with CC: median time 30.3 hours (minimum, 24.2 hours; maximum, 84.1 hours) versus 40.4 hours (minimum, 22.3 hours; maximum, 64.2 hours), respectively (P = 0.3).
Serum fasting motilin levels were significantly lower in patients with T1DM than in healthy control subjects (P < 0.0005;Fig. 1). In a univariate linear regression analysis model with motilin concentrations as the dependent variable and serum glucose concentrations as the independent variable, no statistical significance was found. However, when in the same model the independent variable was serum glucose concentration category (I: glucose concentration <120 mg/dL; II: glucose concentration 120–170 mg/dL; III: glucose concentration >170 mg/dL), a significant negative correlation was found (B, −13.9; SE, 6.1; 95% CI, −26.1 to −1.7; r, −0.258; P = 0.03;Fig. 2). A significant negative correlation was also found between motilin concentration and the duration of diabetes (R, 0.37; B, −3.4; 95% CI, −5.7 to −1.1; P = 0.005). However, when it was corrected for patient age, there was no correlation.
In a multifactorial model with age, duration of diabetes, body mass index, gender, and current or last year's HbA1C concentrations as independent variables, no correlation was found among the above variables and motilin concentrations, gastric emptying, or MATT.
The current study shows that autonomic neuropathy was not an etiologic factor of the gastrointestinal symptoms in children and adolescents with T1DM. Furthermore, gastrointestinal motility was comparable in patients with and without T1DM who had similar gastrointestinal symptoms. Serum motilin concentrations were lower in patients with T1DM than in healthy control subjects, and a significant correlation was found between serum motilin concentrations and glucose categories.
This study has certain limitations. First, the control group for the comparisons of gastrointestinal motility was patients without T1DM who had CD and CC, not healthy individuals. This was done because we considered it inappropriate to perform invasive tests (GE scintigraphy) in healthy children. Moreover, it is well known that children with CD or CC have delayed GE (13) or MATT (14), respectively, compared with healthy control subjects. Therefore, to achieve homogeneity between the study and the control groups, we considered it appropriate to use as control subjects not only age- and gender-but also symptom-matched individuals who would differ from the study groups only by the absence of T1DM. Furthermore, the performance of upper gastrointestinal endoscopy in dyspeptic patients who had negative serology for H. pylori was considered inappropriate because it has been reported that endoscopic findings in this group of patients show no distinct association with symptoms (15). Finally, the motilin assays were performed only on a fasting condition and not postprandially. Although it is well known that motilin is a hormone that is released postprandially, it was decided to assess this hormone with respect only to its fasting levels, when no interfering factors (insulin administration) co-existed.
In contrast to previous studies in adults (1,3,4,16,17), our study shows that autonomic neuropathy was not a problem in children and adolescents with T1DM and chronic dyspepsia or chronic constipation. This finding is in agreement with that reported by Aman et al. (18), whereas Massin et al. (19) reported that cardiovascular abnormalities were common in children with T1DM older than 11 years of age and were significantly correlated with long-term metabolic control (mean HbA1C for 4 years).
This study showed that in the absence of autonomic neuropathy, no differences in GE and MATT were found between the T1DM and non-T1DM patients with CD and CC, respectively. In adult patients with T1DM, the effect of autonomic neuropathy on GE is well known (1–6,20), whereas the effect on colonic transit is a matter of controversy (17,21,22). Wegener et al. (17) showed that GE, but not mouth-to-cecum or MATT, was impaired by autonomic neuropathy. In contrast, Folwaczny et al. (22) showed that motility impairment of both the stomach and the colon occurred in middle and long-standing diabetes. The frequency of total bowel dysmotility in children with diabetes with or without autonomic neuropathy is not known. A recent study in pediatric patients with T1DM who did not have autonomic neuropathy showed that GE assessed with ultrasonography was delayed in a high proportion of patients with T1DM compared with healthy control subjects (23). Furthermore, Cucchiara et al. (23) reported a significant correlation between the degree of GE delay and the levels of HbA1C. In our study, we failed to show significant correlations between GE and HbA1C levels, which is in agreement with the findings reported by Merio et al. (24), who showed that GE delay was correlated only to the degree of cardiovascular autonomic neuropathy and not to HbA1C levels. In addition, in a recent study in adult patients with T1DM and gastroparesis, the 12-month treatment with cisapride was reported to reverse dyspeptic symptoms and to normalize GE, but not to improve HbA1C (25).
We showed a significant negative correlation between serum glucose concentration category and motilin concentration. Indeed, the transition from glucose concentration category I to II was associated with a reduction in serum motilin level of 13.9 mg/dL. The suppression of serum motilin level by acute hyperglycemia has been shown previously in adults (26). According to the authors, serum motilin concentrations were suppressed only by severe (serum glucose, 250 and 175 mg/dL) hyperglycemia (26). The reduction in secretion of gastrointestinal regulatory peptides is just one of the mechanisms through which acute hyperglycemia affects gastric motility (26). Other mechanisms include suppression of antral contractions (27), inhibition of vagal nerve activity (28), and greater compliance of the proximal stomach to balloon distention (29). Severe hyperglycemia also may affect colonic transit (30). Inhibition of long and short neural reflexes that modulate colonic motility is an implicated mechanism (30). The absence of delays in gastric emptying or in colonic transit in our study possibly was attributable to the lack of severe hyperglycemia in our patients.
In conclusion, autonomic neuropathy is not an etiologic factor in the gastrointestinal symptoms in children and adolescents with diabetes. Mild and moderate hyperglycemia is not able to induce gastrointestinal motility impairment. The role of motilin in patients with T1DM needs additional investigation.
1. Soler NG. Diabetic gastroparesis without autonomic neuropathy. Diabetes Care 1980; 3:200–1.
2. Feldman M, Schiller ER. Disorders of gastrointestinal motility associated with diabetes mellitus
. Ann Intern Med 1983; 8:378–84.
3. Campbell IW, Heading RC, Tothill P, et al. Gastric emptying
in diabetic autonomic neuropathy Gut 1977; 18:462–7.
4. Werth B, Meyer-Wyss B, Spinas GA, et al. Non-invasive assessment of gastrointestinal motility disorders in diabetic patients with and without cardiovascular signs of autonomic neuropathy. Gut 1992; 33:1199–203.
5. Fraser RJ, Horowitz M, Maddox AF, et al. Hyperglycaemia slows gastric emptying
in type I (insulin dependent) diabetes mellitus
. Diabetologia 1990; 33:675–80.
6. Bjornsson ES, Urbanavicius V, Eliasson B, et al. Effects of hyperglycemia on inderdigestive gastrointestinal motility in humans. Scand J Gastroenterol 1994; 29:1096–104.
7. Vogiatzi M, Gunn SK, Sherman LD, et al. Gastrointestinal symptoms
and diabetes mellitus
and adolescents. Clinical Pediatrics 1996; 35:343–6.
8. Burghen GA, Murrell LR, Whitington GL, et al. Acid peptic disease in children
with type I diabetes mellitus
. Am J Dis Child 1992; 146:718–22.
9. Vazeou A, Papadopoulou A, Booth IW, et al. The prevalence of gastrointestinal symptoms
with type 1 diabetes. Diabetes Care 2001; 24:962–4.
10. Ewing DJ, Clarke BF. Diagnosis and management of diabetic autonomic neuropathy. Br Med J 1982; 285:916–8.
11. Malmud LS, Fisher RS, Knight LC, et al. Scintigraphic evaluation of gastric emptying
. Semin Nucl Med 1982; 12:116–25.
12. Dimson SB. Carmine as index of transit time in children
with simple constipation. Arch Dis Child 1970; 45:232–5.
13. Cucchiara S, Minella R, Iorio R, et al. Real-time ultrasound reveals gastric motor abnormalities in children
investigated for dyspeptic symptoms. J Pediatr Gastroenterol Nutr 1995; 21:446–53.
14. Corazziari E, Cucchiara S, Staiano M, et al. Gastrointestinal transit time, frequency of defecation, and anorectal manometry in healthy and constipated children
. J Pediatr 1985; 106:379–82.
15. Johnsen R, Bernersen B, Straum B, et al. Prevalence of endoscopic and histological findings in subjects with and without dyspepsia. BMJ 1991; 302:749–52.
16. Spangeus A, El-Salhy M, Suhr O, et al. Prevalence of gastrointestinal symptoms
in young and mid aged diabetic patients. Scand J Gastroenterol 1999; 34:1196–202.
17. Wegener M, Borsch G, Schaffstein J, et al. Gastrointestinal transit disorders in patients with insulin treated diabetes mellitus
. Dig Dis Sci 1990; 8:23–36.
18. Aman J, Eriksson E, Lideen J. Autonomic nerve function in children
and adolescents with insulin-dependent diabetes mellitus
. Clin Physiol 1991; 11:537–43.
19. Massin MM, Derkenne B, Tallsund M, et al. Cardiac autonomic dysfunction in diabetic children
. Diabetes Care 1999; 22:1845–50.
20. De Block CE, De Leeuw IH, Pelckmans PA, et al. Delayed gastric emptying
and gastric autoimmunity in type I diabetes. Diabetes Care 2002; 25:912–7.
21. Battle WM, Snape WJ, Alavi A, et al. Colonic dysfunction in diabetes mellitus
. Gastroenterology 1980; 79:1217–21.
22. Folwaczny C, Hundegger K, Volger C, et al. Measurement of transit disorders in different gastrointestinal segments of patients with diabetes mellitus
in relation to duration and severity of the disease by use of the metal-detector test. Z Gastroenterol 1995; 33:517–26.
23. Cucchiara S, Franzese A, Salvia G, et al. Gastric emptying
delay and gastric electrical derangement in IDDM. Diabetes Care 1998; 21:438–43.
24. Merio R, Festa A, Bergmann H, et al. Slow gastric emptying
in type 1 diabetes: relation to autonomic neuropathy and peripheral neuropathy, blood glucose and glycemic control. Diabetes Care 1997; 20:419–23.
25. Braden B, Enghofer M, Schaub M, et al. Long-term cisapride treatment improves diabetic gastroparesis but not glycaemic control. Aliment Pharmacol Ther 2002; 16( 7):1341–6.
26. Barnett JL, Owyang C. Serum glucose concentration as modulator of interdigestive gastric motility. Gastroenterology 1988; 95:262.
27. Ishiguchi T, Tada H, Nakagawa K, et al. Hyperglycemia impairs antro-pyloric coordination and delayed gastric emptying
in conscious rats. Auton Neurosci 2002; 95:112–20.
28. Samson M, Akkermans LMA, Jebbink RJA, et al. Gastrointestinal motor mechanisms in hyperglycaemia induced delayed gastric emptying
in type 1 diabetes mellitus
. Gut 1997; 40:641–6.
29. Rayner CK, Verhagen MA, Hebbard GS, et al. Proximal gastric compliance and perception of distention in type 1 diabetes mellitus
: effects of hyperglycaemia. Am J Gastroenterol 2000; 95:1175–83.
30. Sims MA, Hasler WL, Chey WD, et al. Hyperglycemia inhibits mechanoreceptor-mediated gastrocolonic responses and colonic peristaltic reflexes in healthy humans. Gastroenterology 1995; 108:350–9.