Outpatient studies of gastrointestinal symptoms in diabetes mellitus
The early literature emphasised the high prevalence of gastrointestinal symptoms in patients with diabetes complicated by neuropathy. More than six decades ago, Rundles reported that ‘constipation, chronic diarrhoea, anorexia and nausea often accompany the development of diabetic neuropathy’. He studied 125 patients with peripheral neuropathy selected from more than 3000 patients who were diagnosed with diabetes over a 7 year period. No information was given concerning age, gender or duration of disease. More than 60% of the patients reported gastrointestinal symptoms; 42% had constipation, this being the most frequent symptom, and 22% had chronic diarrhoea. However, it was also suggested, although not specifically quantified, that ‘among an average group of diabetics receiving modern treatment, gastrointestinal disturbances’ were ‘probably no more frequent than among a similar group of non-diabetics’. In a follow-up study among 30 additional diabetic patients with neuropathy and gastrointestinal symptoms, abdominal pain was the most frequent symptom (in 70% of patients), followed by constipation, diarrhoea, vomiting and faecal incontinence.
Subsequently, a number of studies have evaluated gastrointestinal symptoms among outpatients with both type 2 (non-insulin) and type 1 (insulin-dependent) diabetes. In a sample of 136 outpatients attending a diabetes clinic, Feldman and Schiller reported that 76% had one or more gastrointestinal symptoms which were, in most patients, chronic or frequently recurrent; nausea and vomiting occurred in 29%, dysphagia in 27%, abdominal pain in 34%, constipation in 60%, diarrhoea in 22% and faecal incontinence in 20% of the patients.
However, no control group was evaluated and the interview methodology applied was not well standardised, neither was the type of diabetes documented.
Clouse and Lustman interviewed 114 outpatients with type 1 and type 2 diabetes; 68% reported at least one gastrointestinal symptom. Nausea was experienced by 21% of patients, abdominal pain by 32%, constipation by 12%, diarrhoea by 21% and bloating by 20%. However, no control group was evaluated.
Ko et al. interviewed 149 patients with type 2 diabetes, using standard questions from a gastrointestinal symptom questionnaire, and 65 control subjects.
They also found a high prevalence of gastrointestinal symptoms in Chinese outpatients with diabetes. Epigastric fullness was experienced by 17% of patients, abdominal pain by 16%, diarrhoea by 35% and constipation by 28% of patients; all of these symptoms were significantly more frequent than in the control group.
In contrast, Maxton and Whorwell interviewed 200 patients with type 1 and type 2 diabetes attending a diabetic clinic, of whom 59 had signs of autonomic neuropathy, and 200 age- and sex-matched control subjects. They found that constipation was more common in patients with autonomic neuropathy (22% of patients) compared with patients without neuropathy (9%) and controls (7 – 14%). Diarrhoea was found in only 5% of patients with neuropathy and in 11% of patients without, and this was not significantly different from controls (3 – 6%). The prevalence of abdominal pain was also similar in patients with (19%) and without (21%) autonomic neuropathy and controls (20%).
Similarly, in 285 consecutive outpatients with type 1 and type 2 diabetes from a diabetic clinic in England, Dandona et al. found a prevalence of 8% for diarrhoea and 5% for constipation, which was not significantly different from the prevalence in a control group of outpatients from other medical clinics.
While the group of patients with diabetes who received biguanides had a higher prevalence of diarrhoea [20%], the prevalence of diarrhoea in patients who were on insulin or other oral hypoglycaemics was low (6%).
Other studies have evaluated gastrointestinal symptoms in outpatients who had type 1 diabetes. Keshavarzian and Iber assessed gastrointestinal symptoms in 75 consecutive male patients with type 1 diabetes who had been on insulin for at least 5 years. Only 19% of the patients reported gastrointestinal symptoms, the most frequent being diarrhoea and constipation, with a prevalence of 5% each. Similarly, Maser et al. evaluated gastrointestinal symptoms in a group of 168 patients with type 1 diabetes with a mean disease duration of 20.5 years; signs of autonomic neuropathy were present in 63 patients (37%). The prevalence of gastrointestinal symptoms was found to be low, with vomiting being the most frequent with a prevalence of 7%. Constipation was reported by only 3% of patients and none had diarrhoea. Enck and associates evaluated 190 consecutive patients with type 1 and type 2 diabetes recruited from a diabetes research centre, and 180 age- and sex-matched controls.
Symptoms arising from the upper gut were reported by 70% of patients with insulin-dependent diabetes and 44% of patients with non-insulin-dependent diabetes; 31% type 1 and 43% type 2, patients respectively, had symptoms from the lower gastrointestinal tract. However, the prevalence of gastrointestinal symptoms in diabetic patients did not differ from the prevalence in the control subjects.
In another survey using a validated questionnaire, Bytzer et al. studied 892 randomly selected patients from a diabetes support group and 209 outpatients.
To obtain information on recent glycaemic control, the authors measured glycated haemoglobin. Glycaemic control was predictive of upper, but not lower, gastrointestinal symptoms. Patients with diabetic complications had a higher frequency of most symptom groups and a higher symptom complexity.
Thus, although a number of outpatient studies have suggested that gastrointestinal symptoms are frequent, these results have not been confirmed by all investigators. Depending on the population studied, the prevalence of symptoms has varied considerably in patients with both type 1 and type 2 diabetes mellitus.