Diabetes Mellitus #

Disturbances in gastrointestinal motility with associated symptoms such as vomiting, diarrhoea and constipation have long been recognized as complications of diabetes mellitus. Rundles (1945) described 4 diabetic patients with diarrhoea or alternating diarrhoea and constipation who also had pronounced anorexia and nausea. Radiological examination in all revealed abnormal gastric retention.

Kassander (1958) described 6 cases of poorly controlled diabetes mellitus with upper gastrointestinal manifestations, a condition he called "gastroparesis diabeticorum". In all cases the stomach showed sluggish peristalsis with retention of barium. There was no evidence of an organic lesion causing obstruction and all cases had a patulous pylorus through which the gastric contents could be expressed by hand. It was concluded that these cases represented a disturbance of propulsion, and more specifically of expulsion, of gastric contents. The gastric retention was considered to be similar to that occurring after truncal vagotomy and to be due to diabetic neuropathy involving the vagus nerves. The gastric manifestations could also occur in asymptomatic diabetic patients.

Marshak and Maklansky (l964) mentioned 4 similar cases and reiterated that diabetic gastropathy consisted of gastric retention in the absence of an obstructing lesion at the pylorus. In some cases there was a marked decrease in peristalsis while in others peristalsis was normal.

Zitomer et al. (l968) found that gastric neuropathy typically occurred in patients with a long history of inadequately controlled, moderately severe diabetes mellitus. The symptoms were generally mild and nonspecific, vomiting of undigested food from the previous day being one of the main manifestations. Radiologically the most constant and striking finding, occurring in 34 of their 35 cases, was sluggish, ineffectual and irregularly occurring peristalsis; on occasion peristalsis was absent altogether. There appeared to be decreased or absent propulsive motor activity, associated with abnormal gastric retention of food residues and barium, resembling a bezoar. The malfunction of gastric emptying and consequent unpredictable food absorption probably contributed to poor diabetic control. Despite the considerable delay in gastric emptying, barium could invariably be expressed by palpation through the pylorus, which remained patulous. If the patient was placed on his right side, passive gravitational emptying of liquid contents occurred, while solids tended to be retained. Overall the incidence of clinically significant gastric neuropathy was extremely low in diabetes mellitus, a figure of less than one per cent being mentioned.

Goyal and Spiro (l971) agreed that clinical manifestations of diabetic gastropathy occurred in a very small proportion of cases. While the gastric retention resulting from the reduction of peristaltic contractions resembled the hypomotility of truncal vagotomy, suggesting an autonomic neuropathy, other factors might play a role. It was known, for instance, that hyperglycaemia as well as glucagon inhibited gastric motility.

Scarpello et al. (l976) measured the rate of emptying of isotopically labelled solid meals, using ^99mTc-labelled sulphur colloid as a marker and a gamma camera. The gastric emptying rate in 29 insulin-dependent well-controlled diabetics was compared with that in l8 normal controls. No statistically significant difference in gastric emptying rate between controls and diabetics with or without autonomic neuropathy was found. Only 3 diabetics had greatly delayed gastric emptying, but in one of these the test had previously given a normal result.

Campbell et al. (l977) used a double isotope scintiscanning technique in order to differentiate between solid and liquid emptying in 12 patients with diabetes mellitus (6 with and 6 without objective evidence of autonomic neuropathy), and in 20 non-diabetic controls. Gastric stasis was demonstrated in 3 patients with autonomic neuropathy. In patients without stasis the normal differentiation between solid and liquid emptying was impaired, suggesting an abnormality of "antral" peristalsis.

Gramm et al. (l978) reviewed the radiological findings in 43 patients with diabetic gastric neuropathy. (It was said that the condition was seen more often on account of the increased longevity of diabetic patients). In the majority the stomach had an elongated, sausage-shaped configuration; there was sluggish, irregular gastric peristalsis with retention of solid food and barium in the absence of an organic obstruction at the pylorus.

Feldman et al. (l979) found that many patients with longstanding, insulin-dependent diabetes mellitus had reduced acid secretory responses to sham feeding (suggesting vagal neuropathy) and normal responses to infused food. Both in diabetics with nausea and vomiting and in patients without upper gastrointestinal symptoms, abnormal gastric retention of a nonabsorbable meal marker occurred. The findings supported earlier observations that gastric secretory and motor functions could be impaired in asymptomatic diabetics.

Malagelada et al. (l980) studied gastric motility by means of pressure transducers in the fornix, "antrum" and duodenum in fasting diabetic patients with symptomatic gastroparesis, as well as in asymptomatic diabetics and normal controls. Healthy subjects and asymptomatic diabetes mellitus patients averaged 2 phase III migrating motor complexes per 180 minutes, while most diabetics with gastroparesis had no such complexes. Whether analogous "antral" muscle dysfunction occurred postprandially in diabetic patients with gastroparesis was uncertain. However, as the emptying of solids might be delayed, it seemed likely that "antral" motor activity could be abnormal as it was the main mechanism responsible for the emptying of solids (Chap. 18). According to Feldman and Schiller (l983) the pathogenesis of diabetic gastroparesis remained uncertain, but the evidence supported the view that a vagal autonomic neuropathy was responsible. Feldman et al. (l984), using a radioloic method for assessing gastric emptying of radio-opaque, indigestible solid particles with simultaneous radionuclide scintigraphy of an ¹¹¹In-labelled standard meal, showed that patients with insulin-dependent diabetes mellitus had an abnormally slow gastric emptying rate for indigestible solids; emptying of digestible solids and liquids was close to normal.

Camilleri and Malagelada (l984) studied the fasting and fed manometric profiles of the stomach and proximal small intestine in 14 patients with the clinical diagnosis of diabetic gastroparesis. In 11 of the patients reduction in "antral" pressure activity with absent interdigestive migrating motor complexes was seen. Three patients exhibited a "peculiar continuous 3 minute antral contractile activity". In the small intestine abnormal manometric patterns were observed in 12 patients, in 9 of whom non-propagated, long bursts of powerful contractions occurred. The findings indicated that both in the stomach and small bowel the motility disorder was not invariably of a paretic type. In the stomach a reduction in frequency of amplitude of "antral" motor activity was the rule, but in a minority of patients prolonged periods of contraction of low amplitude (less than 20 mm Hg), occurred. This motor pattern indicated an abolition of the physiological cyclic activity and was distinctly abnormal; the consistently low amplitude might be insufficient to triturate solid food and empty the stomach. Clinically these patients had the same symptoms as those with "antral" hypomotility, for whom the term gastroparesis was perhaps more appropriate.

In symptomatic diabetic gastroparesis patients Achem-Karam et al. (l985) demonstrated abnormal "antral" and duodenal activity, characterized by absence of phase III of the interdigestive migrating motor complex; in 3 out of 6 patients the complex was also absent in the remainder of the proximal small bowel. The findings confirmed some of those previously described by Malagelada et al. (l980) and Camilleri and Malagelada (l984).

Using a dual radionuclide technique, Wright et al. (l985) investigated gastric emptying of solids and liquids in diabetic gastroparesis patients and normal controls. In both groups the emptying rate of liquids was normal; emptying of solids was markedly delayed in the diabetic group.

With a similar technique Horowitz et al. (l987) determined the effects of the prokinetic drug cisapride in 20 insulin-dependent diabetic patients who had delayed gastric emptying of the solid or liquid component of a meal, or both. It was found that cisapride increased the emptying rate of both solids and liquids.

Yoshida et al. (l988) found no morphologic abnormalities of the abdominal vagus nerves, the gastric musculature and myenteric plexuses in patients with diabetic gastroparesis.

In a review Varis (l989) defined diabetic gastroparesis as a delay of gastric emptying without gastric outlet obstruction in patients with long-standing insulin requiring diabetes mellitus. Characteristically the emptying of solids was delayed; in most cases the emptying rate of liquids was within the normal range, although it might be decreased in some patients. All normal motility patterns, e.g. adaptive relaxation and the maintenance of tone in the proximal stomach, "antropyloric" peristaltic movements and phase III of the interdigestive migrating motor complex might be lacking or abnormal. The "antral" motility disorder, leading to retention of solid foods, was considered to be the most characteristic feature of the condition.

As endoscopy may fail to demonstrate motor disorders of the stomach (Varis l989), it does not play an important role in the diagnosis. However, endoscopy does show gastric mucosal pathology such as acute erosions and gastritis, occurring in the majority of patients with diabetic ketoacidosis (Carandang et al. l969), and it may help to exclude mechanical pyloric obstruction (Camilleri and Malagelada l984).

From the above it is clear that gastric motility disturbances may occur both in symptomatic and asymptomatic diabetes mellitus patients. In a small percentage of cases these disorders may progress to diabetic gastroparesis, a condition usually occurring in patients with a long history of inadequately controlled, insulin-dependent diabetes mellitus.

We have had occasion to do upper gastrointestinal radiological examinations in numerous diabetic patients for a variety of reasons, over a period of years. Motility disturbances were found in a small number of patients, both in insulin-dependent and non-insulin dependent cases. The following are some of the cases with radiologically recognisable abnormalities.

Case Reports #

Case 37.1. D.R., 40 year old male had a long history of ethanol abuse and insulin- dependent diabetes mellitus which had been treated inadequately. For several weeks there had been dyspepsia, loss of appetite and nausea. Radiologically primary and secondary oesophageal peristaltic waves were normal. After an overnight fast the stomach contained food residues; the pyloric aperture was patent, measuring 9.0. mm in diameter (Fig. 37.1A). Gastric peristaltic waves were decreased both in frequency (i.e. less than 3 per minute) and intensity (i.e. failing to "bi-sect" the organ). The pyloric sphincteric cylinder was in a state of partial contraction most of the time (Fig. 37.1B). Cyclical contraction and relaxation of the cylinder, normally occurring at a frequency of 3 per minute (Chaps. 13, 15), was lacking. No obstructing lesion was seen at the pylorus and the duodenum appeared normal. Sonographically the gall bladder, liver, spleen and kidneys were normal; there was no ascites. The patient refused endoscopic examination.

A

B

Fig. 37.1. A,B. Case D.R. A Food residues in stomach. Pyloric aperture patent. Gastric peristaltic activity diminished. B Pyloric sphincteric cylinder (arrows) contracted most of the time.

Fig. 37.1. C-F. Case D.R. After treatment and clinical improvement. Lessened contraction of sphincteric cylinder. Some movement evident but normal cyclical activity absent. Note pancreatic calcification

Six months later, after proper diabetic control, he had improved clinically. On this occasion the stomach contained less residual food. Shallow gastric peristaltic waves were present, while the pyloric sphincteric cylinder remained partially contracted throughout the examination (Fig. 37.1C). Although this was less marked than on the previous occasion, normal cyclical activity remained absent and no maximal or complete contractions were seen. Peristaltic activity in the duodenum appeared to be decreased. Active peristaltic contractions with a fast passage of barium was noted in the jejunum. The films also showed scattered areas of calcification in the pancreas and the case was diagnosed as diabetic gastroparesis and chronic alcoholic pancreatitis. Case 37.2. T.M., 64 year old female with longstanding insulin-dependent diabetes mellitus and psychotic symptoms, was admitted with epigastric pain and episodes of vomiting. Radiological examination showed a decrease in frequency and intensity of gastric peristaltic waves; there was a lack of cyclical contraction and relaxation of the pyloric sphincteric cylinder, which remained in a state of partial contraction throughout the examination; this was associated with a patulous pyloric orifice measuring 1.2 cm in diameter. Sonographically the gall bladder, liver, spleen, kidneys and aorta were normal. The pancreas could not be visualized owing to obesity.

Attempts at controlling the diabetes in a peripheral hospital were only partially successful. She was readmitted 14 months later in a disorientated state with uncontrolled diabetes mellitus, diffuse abdominal pain and diarrhoea. Abdominal sonography was normal. Upper gastrointestinal radiological examination after an overnight fast showed numerous food residues in the stomach. Gastric peristalsis was decreased as before. The pyloric sphincteric cylinder remained partially contracted with absent cyclical activity, and contained food residues and unidentified tablets which she had found in the ward and swallowed the previous day (Fig. 37.2). (Each tablet measured 1.1 cm in diameter on the films; owing to magnification the true diameter would be in the region of 8.0 to 9.0 mm). The pyloric aperture remained patulous and no obstructing lesion was seen. The condition was diagnosed as diabetic gastroparesis.

Fig. 37.2. Case T.M. Absent cyclical activity of pyloric sphincteric cylinder. Pyloric aperture patulous. Retention of food residues and 8.0mm diameter tablets

Case 37.3. F.J., 64 year old female with longstanding, non-insulin dependent diabetes mellitus had received oral anti-diabetic therapy for 8 years. She was admitted for dyspepsia, epigastric pain and diabetic retinopathy. Upper gastrointestinal radiological examination showed a small benign-looking gastric polyp immediately below the gastro- oesophageal junction. Gastric peristaltic waves were shallow and appeared sluggish. The pyloric sphincteric cylinder was partially contracted throughout the examination; while minor degress of movement of the walls were discernable, there was a total lack of cyclical contraction and relaxation, the appearance remaining more or less unchanged (Fig. 37.3). The pyloric aperture remained patent, having a diameter of approximately 1.0 cm; continuous emptying of fluid barium occurred through the patulous pylorus. At gastroscopy a small, benign polyp was removed. The remainder of the stomach, the pylorus and duodenum showed no lesion. The motility disturbance of the pyloric sphincteric cylinder with the patulous pyloric aperture was thought to be compatible with diabetic gastroparesis.

Fig. 37.3. A-D Case F.J. Pyloric sphincteric cylinder permanently contracted. Cyclical activity absent. Pyloric aperture patulous with diameter of 1.0 cm

Case 37.4. M.J., 63 year old female was admitted for ischaemic heart disease, loss of weight and anorexia. She was a known case of non-insulin dependent diabetes mellitus and had received oral therapy for the previous 10 years. There was severe target organ involvement with diabetic retinopathy. Upper gastro-intestinal radiological examination showed a constant contraction of the pyloric sphincteric cylinder which remained unchanged throughout the examination (Fig. 37.4); this was thought to be compatible with early diabetic gastroparesis.

Fig. 37.4. Case M.J. Permanent contraction of pyloric sphincteric cylinder (arrows)

Discussion #

The above patients all had symptoms and signs warranting the clinical diagnosis of diabetic gastroparesis. Radiological studies in all showed absent cyclical activity of the pyloric sphincteric cylinder. It has been shown that normal motility of the sphincteric cylinder consists of alternating cycles of contraction and relaxation, occurring at a rte of approximately 3 per minute (Chap. 13); in diabetic gastroparesis these cycles are absent, the cylinder remaining in a state of permanent, partial contraction. As a maximal or complete contraction of the cylinder does not occur, there is a failure of muscular closure of the pyloric aperture (Chap. 13), which remains patent.

The patulous pylorus allows continuous emptying of fluid barium (Case 37.3). Emptying of solid food residues is retarded (Cases 37.1 and 37.2) due to failure of normal cyclical contraction of the sphincteric cylinder, i.e. failure of the normal mechanism of propulsion of solids. The retention of solid tablets with a diameter of 8.0 mm (Case 37.2) indicates failure of trituration of these tablets.

These conclusions tally with some of those of previous authors, and especially with those of Malagelada et al. (l980) and Camilleri and Malagelada (l984). However, we believe that the "antral" motor dysfunction mentioned by them as well as by Achem-Karam etal (l985) and others, can be placed on a firm anatomical footing if reference is made to the findings of Cunningham (l906), Forssell (l913) and Torgersen (l942) (Chap. 3).

The sustained contraction of the pyloric sphincteric cylinder seen radiologically in diabetic gastroparesis will also explain the "peculiar continuous 3 minute antral contractile activity" found by Camilleri and Malagelada (l984) during manometric studies in some of their patients. One agrees with these authors that the disorder is not invariably of a paretic type; in fact, there appears to be hypomotility due to a spastic or hypertonic condition of the gastric smooth musculature, and in particular of the pyloric sphincteric cylinder.

References #

1. Achem-Karam SR, Funakoshi A, Vinik AI, et al. Plasma motilin concentration and interdigestive motor complex in diabetic gastroparesis: effect of metoclopramide. Gastroenterology l985, 88, 492-499.
2. Camilleri M, Malagelada JR. Gastric motility in disease. In: Gastric and Gastroduodenal Motility, ed Akkermans LMA, Johnson AG, Read NW. Praeger Publ, New York l984, pp 213-214.
3. Camilleri M, Malagelada JR. Abnormal intestinal motility in diabetics with the gastroparesis syndrome. Eur J Clin Invest l984, 14, 420-427.
4. Campbell IW, Heading RC, Tothill P, et al. Gastric emptyin in diabetic autonomic neuropathy. Gut l977, l8, 462-467.
5. Carandang NV, Schuman BM, Whitehouse FW. The gastric mucosa of patients in diabetic ketoacidosis: a gastro-camera study. Gastroint Endosc l969, 16, 156-158.
6. Cunningham DJ. The varying form of the stomach in man and the anthropoid ape. Trans Roy Soc Edin l906, 45, 9-47.
7. Feldman M, Corbett DB, Ramsey EJ, et al. Abnormal gastric function in long- standing, insulin-dependent diabetic patients. Gastroenterology l979, 77, 12-17.
8. Feldman M, Schiller LR. Disorders of gastrointestinal motility associated with diabetes mellitus. Ann Int Med l983, 98, 378-384.
9. Feldman M, Smith HJ, Simon TR. Gastric emptying of solid radio-opaque markers: studies in healthy subjects and diabetic patients. Gastroenterology l984, 87, 895-902.
10. Forssell G, Über die Beziehung der Röntgenbilder des menschlichen Magens zu seinem anatomischen Bau. Fortschr Geb Röntgenstr 1913, Suppl 30, 1-265.
11. Goyal RK, Spiro HM. Gastrointestinal manifestations of diabetes mellitus. Med Clin North Amer 1971, 55, 1031-1044.
12. Gramm HF, Reuter K, Costello P. Radiologic manifestations of diabetic gastric neuropathy and its differential diagnosis. Gastrointes Radiol l978, 3, 151-155.
13. Horowitz M, Maddox A, Harding PE, et al. Effect of cisapride on gastric and oesophageal emptying in insulin-dependent diabetes mellitus. Gastroenterology l987, 92, 1899-1907.
14. Kassander P. Asymptomatic gastric retention in diabetics (gastroparesis diabeticorum). Ann Int Med l958, 48, 797-812.
15. Malagelada JR, Rees WDW, Mazzotta LJ, et al. Gastric motor abnormalities in diabetic and postvagotomy gastroparesis: effect of metoclopramide and bethanecol. Gastroenterology l980, 78, 286-293.
16. Marshak RH, Maklansky D. Diabetic gastropathy. Amer J Dig Dis l964, 9, 366-370.
17. Rundles RW. Diabetic neuropathy: general review with report of 125 cases. Medicine l945, 24, 111-160.
18. Scarpello JHB, Barber DC, Hague RV, et al. Gastric emptying of solid meals in diabetics. Brit Med J l976, 2, 671-673.
19. Torgersen J. The muscular build and movements of the stomach and duodenal bulb. Acta Rad l942, Suppl 45, 1-191.
20. Varis K. Diabetic gastroparesis (a review). Scand J Gastroenterol l989, 24, 897-903.
21. Wright RA, Clemente R, Wathen R. Diabetic gastroparesis: an abnormality of gastri emptying of solids. Amer J Med Sci l985, 289, 240-242.
22. Yoshida MM, Schuffler MD, Sumi SM. There are no morphologic abnormalities of the gastric wall or abdominal vagus in patients with diabetic gastroparesis. Gastroenterology l988, 94, 907-910.
23. Zitomer BR, Gramm HF, Kozak GP. Gastric neuropathy in diabetes mellitus: clinical and radiologic observations. Metabolism l968, l7, 199-211.