Gastrointestinal Motility Among Diabetes Patients

Background GI symptoms are extremely common in patients with DM, often with severe consequences for daily activities, ability to work and quality of life. Moreover, GI-dysfunction causes unpredictable absorption of food and oral medication thus making blood-glucose difficult to control and oral medication less effective. Despite of this fact, the pathophysiology of diabetic bowel dysfunction is only scantily described. Rational treatment of GI motility disorders requires detailed knowledge of the underlying mechanisms and the importance of whole-gut evaluation of patients with motility disorders is increasingly recognized.

Contractions of the GI tract are primary stimulated by cholinergic parasympathetic nerves (from the vagus nerve and the sacral nerves) or cholinergic nerves in the enteric nerve system located between the muscle layers in the bowel wall. It is likely that cholinergic denervation is a major factor behind the development of GI dysfunction in DM.

Through collaboration between basic physiologist and clinical researchers we are in the unique international position of having two highly advanced methods for description of GI innervation and motility:

1. [11C]donepezil (Donepezil connected with a carbon isotope) PET tracer to quantify the density of acetylcholine esterase in abdominal organs, including the intestinal wall. The method constitutes the first-ever validated scan-method for in vivo measuring cholinergic denervation of the GI tract.
2. 3D-Transit for minimal invasive and ambulant describing of regional transit times and contractions pattern of the bowel.

Through studies based on the two methods we aim to test the following hypotheses:

1. The [11C]donepezil PET signal of the intestinal wall is weaker than normal in patients with DM and GI symptoms.
2. Compared to healthy controls, patients with diabetes has prolonged transit time through the stomach, the small intestine and the colorectum.
3. Prolonged transit times through the stomach, small intestine and colorectum of diabetic patients are, at least in part, reversible and approach normal during acetylcholine esterase inhibitor treatment.
4. Gastrointestinal dysfunction in DM is "pan-enteric" and assessment of one region is insufficient as clinical evaluation.

Diabetes and dysmotility of the gastrointestinal tract Autonomic neuropathy can affect all parts of the GI tract, which makes specific diagnosis and correct therapy difficult. Diabetes is strongly associated with nausea, vomiting, diarrhea, abdominal pain, constipation and fecal incontinence. The best-investigated aspect of diabetic dysmotility is gastroparesis. It has been reported that one third of patients with DM having delayed emptying of the stomach. However, gastroparesis is usually detected after at least 10 years of diabetes.

The pathophysiology of diabetic gastroparesis is still incompletely described but includes lack of co-coordinated contractions of the stomach, deficient relaxation of the pylorus and fundus of the stomach relaxation. Furthermore both peripheral and central sensory function is abnormal. Small intestine dysfunction in DM is very scantily investigated. Thus, it is still unknown whether patients with diabetic neuropathy primary have rapid or slow passage time through the small intestine. This is important because treatment of the two conditions is opposite.

Autonomic and enteric neuropathy in diabetes The neural control of GI motility is complex. The most essential components are the enteric nervous system located in the bowel wall and parasympathetic nerve fibers from vagus nerve or the sacral nerves. In both systems acetylcholine is the most important stimulating neurotransmitter. It is well known that DM may cause autonomic neuropathy affecting the visceral organs.

Earlier studies of patients with DM have found demyelination, axonal damage and reduced number of motor fibers in vagal nerves. However, the correlation between GI symptoms and other evidences of autonomic neuropathy is poor. Other studies report degeneration of nerve fibers within the enteric nervous system including the Cajal cells (the pacemaker cell of the bowel). Today, it is assumed that diabetic GI dysfunction is caused by a combination of autonomic neuropathy, enteric neuropathy, dysfunction of the Cajal cells, reduced contractility of the intestinal smooth muscle cells and abnormally high blood glucose. The clinical effect of each component is unknown.

Methodological limitations of previous studies of GI dysfunction in diabetes Previous studies, mentioned above, are all observational in design and the changes in nerve or muscle cell function were not correlated to the specific changes in bowel function. Therefore, it has not been possible evaluate the functional effects of neuropathy of specific groups of neurons or lack of specific neurotransmitters. This important aspect will be addressed in the present study.

Most neurons controlling GI motility are located in myenteric plexus deeply in the bowel wall, not available for ordinary biopsies. Thus, most studies of diabetic GI neuropathy have been based on animal models and results cannot directly be applied to humans. Studies of GI contractions and motility patterns under near normal conditions have been difficult to perform because subjects had to be confined to hospital or have catheters inserted. Especially, the small intestine is difficult to access.

[11C]donepezil PET/CT Donepezil binds with high affinity to acetylcholine esterase in the cholinergic synapses.

Parkinson is a disease characterized by noticeable loss of parasympathetic nerve fibers. Like with DM, Parkinson patients often suffer from severe gastroparesis and constipation. We have recently validated [11C]donepezil PET/CT as a measure of the density of acetylcholine esterase in peripheral organs. A high [11C]donepezil PET signal is seen in heart, liver, and pancreas while and medium signal is seen in in the small intestine, and a weak signal in the colon. The intensity of the signal is correlated to areas with known high density of vagal innervation area. In a study of 12 patients with Parkinson´s disease and 12 healthy the patients had a dramatic loss of signal in the small intestine but also a measurable loss in the pancreas. Thus, [11C]donepezil PET constitutes the first validated scan-method for demonstrating parasympathetic denervation in vivo.

Subjects and methods We aim to include 20 healthy subjects and 25 patients with DM and severe GI symptoms (nausea, vomiting, abdominal discomfort, diarrhea or constipation). Standard assessment will be performed prior to inclusion as part of daily clinical practice. Persons with a history of disease within the central nervous system or the gastrointestinal tract and those with a history of major abdominal surgery are excluded.

1. All Subjects will be examined with [11C]donepezil PET/CT to compare the total cholinergic innervation of the gut in patients with DM and healthy.
2. All subjects will be examined with 3D-Transit to compare gastric emptying time, small intestinal transit time, and colorectal transit time in patients with DM and healthy controls.
3. Patients with DM will be examined with 3D-Transit time a second time after administration of the acetylcholine esterase inhibitor Mestinon (60 mg x 4) to determine whether abnormal transit times and patterns are reversible and thus caused by cholinergic denervation.
4. A minority of patients with severe intractable constipation will on clinical indications have a Malone appendicostomy. This procedure ensures efficient emptying of the colon and rectum. These patients will undergo a third examination with 3D-Transit. Comparing this investigation to their first will provide pilot data on to what degree gastroparesis is secondary to severe constipation Health Ethics Committee for Region Middle, Denmark, has approved the project. The protocol has been submitted to "The Danish Health and Medicines Authority".

PET/CT with [11C]donepezil The total radiation dose for the PET/CT is about 7-8 millisievert (mSv), which corresponds to about 2.5 years of background irradiation, or half that of an ordinary diagnostic CT-scan. This radiation dose is linked to a theoretical increased risk for cancer estimated to about 0.05%.

3D-Transit Every capsule (8x21 mm) Capsules emits a electromagnetic field, which is converted to the coordinates (x;y;z;Φ;θ). The coordinates x;y;z defines the distance with few millimeters precision. Φ;θ defines the angle of the magnet in relation to the sensors. The description of location and rotation of the capsule is dynamic and very precise and permits detailed description of the GI-contraction patterns and regional passage time. Artifacts from breathing and physical activity are recorded from a belt around thorax and an accelerometer in the detector. The 3D-Transit method is without discomfort and has only a negligible risk of capsule retention. The investigation is ambulatory and under circumstances close to the subject's everyday life. This is a considerable advantage compared to traditional investigation methods as scintigraphy and manometry. Validity, safety and reproducibility data from 3D-Transit have been published from our group.

The study consists of four different parts. The DM patients examined in 1-3 and few DM patients are able to participate in part 4. The healthy subjects are only examined in the first two parts (1 and 2).

1. 11C Donepezil PET/CT scan
2. 3D-Transit, capsule 1
3. 3D-Transit capsule 2 while taking acetylcholine esterase inhibitor (Pyridostigmine 60 mg four times with 4 hours between each administration.)
4. 3D-Transit capsule 3 after Malone antegrade continence enema

Each part of the study is completed before the next step begins. The study participants are fasting before each study element. DM patients are taking the morning insulin and closely monitored to reach blood glucoses between 4-10 mmol/L

The practical feasibility of the study The Department of Endocrinology at Aarhus University Hospital has specific interest in diabetic neuropathy and patients will be classified according to international standards.

The Department of Hepatology and Gastroenterology, Aarhus University Hospital has a special interest in neurogastroenterology and has established a professorship in this field. The department has close collaboration with the manufacturer of 3D-Transit (Motilis Medica SA, Lausanne, Switzerland) and the research group in London. Through this ongoing collaboration we have been among the pioneers in the development of magnet tracking for description of GI motility.

The PET Center at Aarhus University Hospital is an internationally recognized research unit in this field. The relevant resources are available, including PET/CT scanners, radiochemical facilities for tracer production, software, and computer scientist assistance for analyzing PET-data.

Future perspectives

1. Decreased cholinergic PET signal in the bowel wall of patients suffering from diabetes and severe GI symptoms, in combination with mechanistic data identifying cholinergic neuropathy as a main cause of abnormal transit patterns will have significant international interest and improve future attempts to develop targeted future treatment.
2. Identification of specific phenotypes of diabetic bowel dysfunction will form the basis of larger descriptive studies to allow future stratified treatment.