The pathogenesis of diarrhea in patients with diabetes is poorly understood and probably multifactorial. It may be caused by disturbances directly related to diabetes (primary causes) or to late complications (secondary causes). Among primary causes, visceral neuropathy is a key factor but other factors probably contribute as well. Functional changes such as accelerated transit time and decreased intestinal tone might be associated with enhanced cholinergic and decreased P-adrenergic receptor activities (65). Neuroendocrine peptide dysfuntions might also be involved. El-Salhy and Spangeus (66) have shown that in diabetic mice antral VIP and galanin levels are increased, whereas colonic PYY concentrations are decreased. These particular anomalies in enteric peptide profile would favor the development of diarrhea, whereas other anomalies in peptide levels could favor constipation.
Low et al. (67) examined the correlation between various GI symptoms attributable to autonomic neuropathy (diarrhea among them) and objective autonomic impairment measured by laboratory tests in patients with type 1 diabetes mellitus. Their findings suggest that both symptoms and autonomic dysfunction are common in diabetes.
However, they could not always establish a causal relation between autonomic dysfunction and the clinical manifestations. Rosa-e-Silva et al. (68) observed rapid transit rates in the distal small bowel of patients with type 1 diabetes mellitus with strong correlation with the presence of orthostatic hypotension, suggesting that autonomic neuropathy is the common link between both findings. In patients with a postganglionic sympathetic lesion, Camilleri et al. (69) found manometric evidence of intestinal dys-motility in the form of incoordinated bursts of nonpropagated phase III-like activity, although not all of these patients had diarrhea. Among other primary causes of diabetic diarrhea, microangiopathy and functional mucosal abnormalities might also play a role.
Secondary causes of diarrhea in diabetics potentially include bacterial overgrowth abnormalities in bile acid enterohepatic circulation, and exocrine pancreatic insufficiency. The first two might be facilitated by intestinal transit abnormalities that are themselves secondary to intestinal dysmotility. Exocrine pancreatic insufficiency occurs more frequently in diabetics than controls (70), particularly in patients with associated overweight. Exceptionally, pancreatic exocrine insufficiency becomes severe enough to contribute to malabsorption. Abnormal stools and intolerance of dietary fat are symptoms that might suggest this complication in diabetics. The diagnosis might be established simply and noninvasively by measurement of elastase 1 in the stools (71).
Finally, drugs must be considered as a potential pathogenetic factor. For instance, metformine treatment of diabetes is associated with frequent GI side-effects including diarrhea. The latter is not corrected by concomitant odansetron treatment, suggesting that 5HT3 receptors are not implicated (72).
Diabetic diarrhea was first recognized in 1936 by Bargen et al. (73) (Table 2). The diarrhea is watery, often severe, and preceded by abdominal cramps and it occurs particularly at night and may be accompanied by fecal incontinence. Symptoms are intermittent and might last from a few hours to several weeks. During remissions, patients may shift and complain of constipation, resembling the bowel movement alternance characteristic of irritable bowel syndrome. Mild steatorrhea, although not common is compatible with the diabetic diarrhea syndrome, whereas weight loss is unusual. Other signs of autonomic neuropathy may be present concomitantly.
The diagnosis of chronic diabetic diarrhea is essentially one of exclusion. A major problem with this approach remains that some of the main conditions in the differential diagnosis (pancreatic insufficiency, bacterial overgrowth, celiac sprue) can themselves be part of the diabetic diarrhea syndrome. A careful history should be taken to exclude osmotic diarrhea from excessive ingestion of nonabsorbable hexitols (e.g., sorbitol). A 48-72-hour stool collection for weight and fat measurement used to be standard approach, but it is cumbersome and nowadays tends to be bypassed in favor of other tests, as listed in Table 2.
When steatorrhea is found, pancreatic insufficiency should be excluded by performing a pancreatic function test or, if these are not available, by a trial with oral pancreatic enzymes. The degree of steatorrhea relative to diarrhea might sometimes be a useful clue in distinguishing pancreatic insufficiency from that caused by other GI diseases (74). Bacterial overgrowth might be demonstrated by breath tests. Although a positive
Evaluation of Chronic Diarrhea in a Patient With Diabetes
Stools: weight, fat, occult blood, examination for ova, parasites, and culture Colonoscopy (rectal biopsy)
Radiographic studies: plain film of the abdomen; small bowel barium studies; abdominal
CT scan Small bowel biopsy
Small bowel aspirate for giardia and bacteria
Breath tests for malabsorption and bacterial overgrowth
Serum vitamin B12 and folate
Pancreatic function tests (elastase 1 in stools, other)
Therapeutic trials with antibiotics, gluten-free diet, pancreatic enzyme supplements response to antibiotics is suggestive, it is not reliable evidence because diabetic diarrhea often shows spontaneous remissions.
Celiac disease might be associated with diabetes and accompanied by features of more severe malabsorption (considerable steatorrhea, hypoalbuminemia, anemia, abnormal Schilling and xylose test, low serum folate) than is characteristic of diabetic diarrhea. Serological tests for celiac disease are advisable. Small bowel biopsies might be helpful but are not specific for gluten enteropathy because blunting of villi might occur in bacterial overgrowth and other conditions. A favorable clinical and histologi-cal response to gluten-free diet is helpful in confirming the diagnosis.
Sometimes, diabetics may also experience abdominal pain because of thoracolumbar diabetic radiculopathy with no evidence of GI pathological findings. A careful history of the typical burning, sharp dermatomal pain, the chronic course, and the electromyo-gram and thermoregulatory sweat test findings would point to the correct diagnosis of this syndrome.
From a purely clinical standpoint, it is important to recognize the occurrence of pseu-dodiarrhea (also denominated low volume diarrhea), which in fact represents constipation with impaction of hard stools in the lower colon and rectum. These patients often complain to clinicians of "diarrhea" because they have frequent bowel movements expelling small quantities of liquid residue, which leaks around the impacted solid faeces. There is often associated tenesmus and incontinence. Pseudodiarrhea is in fact constipation, and should be treated as such.
A number of treatments have been proposed: antibiotics for bacterial overgrowth, resin cholestyramine, and pancreatic enzyme supplements. Also, antidiarrheal agents such as loperamide, codeine, and anticholinergics have been tried (75). These antidiar-rheals may reduce stool frequency, but often do not diminish stool volume. Treatment of severe diabetic diarrhea with octreotide has been reported as quite successful. Octreotide inhibits peptide secretion including serotonin, gastrin, and motilin, and at the same time directly suppresses GI motility (76) and improves fluid and electrolyte absorption. Octreotide reduces diabetic diarrhea and its consequences (77,78). Octreotide sc injections may be started at 50 ^g twice daily and increased to 100 ^g thrice daily as needed. If effective, consideration must be given to using monthly im injections of a long-acting octreotide preparation to make it more comfortable to the patients. Attention must be given to reduce insulin dosage needs on account of inhibition of release of glucagon, growth hormone, and other peptides by octreotide. Potential for hypoglycemic episodes exists.
Was this article helpful?