Gastric Autonomic Neuropathy in Diabetes

Abstract

Autonomic dysfunction of the stomach typically manifests as delayed gastric emptying or gastroparesis and is seen in individuals with both type 1 and 2 diabetes. However, impaired gastric motility is only modestly associated with the presence of upper gastrointestinal symptoms, and the diagnosis of gastroparesis essentially requires a formal measurement of gastric emptying, ideally employing a sensitive and precise technique such as scintigraphy. There is a bidirectional relationship between gastric emptying and glycemia: insulin-induced hypoglycemia accelerates, while acute elevations in blood glucose may delay gastric emptying. On the other hand, relatively more rapid emptying is associated with a higher initial rise in postprandial glucose. The management of gastroparesis requires an individualized approach, integrating dietary modifications, nutritional supplementation, pharmacological therapies, and, in severe cases, advanced interventions including gastrojejunostomy and gastric electrical stimulation. This review provides an overview of the pathophysiology and diagnosis of autonomic neuropathy of the diabetic stomach and discusses current clinical management strategies.

1. Introduction

Diabetic autonomic neuropathy (DAN), defined by the Toronto Expert Panel on Autonomic Neuropathy as “a disorder of the autonomic nervous system in the setting of diabetes, or metabolic derangements of prediabetes, after the exclusion of other causes” [1], is a common and important, but often overlooked, complication of diabetes, associated with substantial negative health and socioeconomic consequences [2]. The prevalence of DAN may be as high as 40% [1], however, this varies widely between studies, reflecting inconsistencies in diagnostic criteria, methodologies, and study populations [1,2,3,4,5]. DAN, unsurprisingly, has a myriad of manifestations, as autonomic nerves innervate multiple organ systems, including the cardiovascular, gastrointestinal (GI), and genitourinary systems [2,6], of which cardiac autonomic neuropathy (CAN) is the best studied and most lethal complication, resulting from damage to autonomic nerve fibers supplying the heart and the blood vessels [7].

While DAN may impact any location in the GI system from the esophagus to the anus, autonomic dysfunction of the stomach is the most studied. Gastric autonomic neuropathy (GAN), manifesting as abnormally delayed gastric emptying, represents an important cause of morbidity, and is also of major relevance to postprandial glucose homeostasis [7]. Gastrointestinal symptoms such as bloating, early satiety, postprandial fullness, nausea, abdominal pain, constipation, and diarrhea occur frequently in all forms of diabetes and negatively impact quality of life and well-being [8,9]. The association of delayed gastric emptying with GI symptoms is, however, poor, contrary to initial expectations.

This narrative review provides a comprehensive overview of autonomic neuropathy of the stomach in diabetes. We address the underlying pathophysiology, clinical manifestations, and diagnostic methodologies, including potential techniques for assessing gastric emptying. We also address nutritional and pharmacological strategies for the treatment of diabetic gastroparesis to facilitate personalized and evidence-based approaches to management and how newer therapies, particularly GLP-1 receptor agonists, impact gastric emptying.

2. Neuroanatomy of the Gastrointestinal Tract

2.1. The Enteric Nervous System

The enteric nervous system (ENS), the intrinsic nervous system of the GI tract, which consists of a mesh-like system of interlinking neurons, operates in close conjunction with the autonomic nervous system (ANS) (the ‘extrinsic nervous system’) to regulate GI motility and hormone secretion. The ENS typically co-ordinates the digestion, secretion, and motility of the GI system to optimize nutrient absorption, and consists of both the myenteric (Auerbach’s) and submucosal (Meissner’s) plexus and the interstitial cells of Cajal (ICCs), positioned between the two muscular layers of the GI tract, which act as an intrinsic pacemaker, along with fibroblast-like cells expressing platelet derived growth factor-alpha (PDGFRα), which promote muscle contraction. The ICCs control slow-wave propagation and are found throughout the GI tract, mediating the ‘crosstalk’ between the ANS and smooth muscles. The myenteric plexus modulates the activity of the longitudinal and circular muscle layers of the GI tract and is directly responsible for their activation.

2.2. The Autonomic Nervous System and the Gastrointestinal Tract

The efferent activity of the ANS in the GI tract is primarily regulated by autonomic reflexes, where sensory information is transmitted to homeostatic control centers in the hypothalamus and brainstem, typically via the vagus nerve. Efferent responses from the extrinsic nervous system are mediated via two anatomically and functionally divisions, the parasympathetic and sympathetic systems [10], which are antagonistic to each other and typically tonically active, so that some input to targeted tissues is always provided. This tonic activity may be up- or down-regulated depending on the efferent input and concurrent alteration of both pathways, so that the up-regulation of one and down-regulation of another leads to rapid and precise alterations in the function of end-organ tissue [10]. The parasympathetic system is typically excitatory and exerts its effect via the vagus nerve in the upper GI tract. Following vagal nerve activation, the parasympathetic fibers are activated, resulting in the release of acetylcholine and neurokinins, which, in turn, activate the myenteric plexus or release nitric oxide (NO) which inhibits the plexus. The myenteric plexus synapses subsequently on the ICCs within muscle bundles and communicates with smooth muscles via gap junctions.

The sympathetic activity of the GI tract is, in contrast, inhibitory. Sympathetic fibers originate from the T8-L2 levels from the spinal cord, then synapse at the pre-vertebral ganglia, and continue onwards to synapse on the myenteric and submucosal plexuses to modulate smooth muscle tone, as well as endocrine and secretory cells.

3. Physiology of Gastric Emptying

Normal gastric emptying is a complex process involving the proximal and distal parts of the stomach, the pyloric sphincter, and the proximal small intestine, mediated by the interaction between the enteric and extrinsic nervous systems, neurohumoral pathways, and small intestinal feedback mechanisms [11,12]. The key mediators involved include excitatory neurotransmitters such as acetylcholine and substance P, which stimulate smooth muscle contraction, as well as inhibitory factors, including vasoactive intestinal peptide (VIP), nitric oxide (NO), and carbon monoxide, which facilitate muscle relaxation and regulate gastric motility [13].

There are two distinct patterns of upper GI motility, as follows: (a) the fasted pattern, which is characterized by the so-called interdigestive migrating motor complex (MMC), and (b) the fed pattern, which occurs in the postprandial state. During the interdigestive phase, a cyclical pattern of contractions occurs in the antrum (distal stomach) and small bowel, taking 90–120 min to reach the distal ileum. The MMC is involved in propelling undigested food and bacteria from the stomach and small intestine to the colon. There are four phases of the MMC in health, as follows: phase I is a quiescent period lasting 40–50 min, phase II is characterized by irregular, intermittent, low-amplitude contractions and lasts 20–40 min, phase III features rhythmic contractions with a maximum amplitude, lasting about 3 minutes in the stomach and 10–12 min in the small intestine, and is followed by phase IV with irregular contractions, returning to phase I. Indigestible solid foods > ~1 mm in size are emptied from the stomach during the late phase II and phase III of the MMC [14].

Following meal ingestion, the fasted pattern is replaced by the fed pattern. This motor pattern is associated with complex and synchronized activity in the proximal and distal stomach regions, the pylorus, and the small intestine. The proximal stomach relaxes to accommodate the meal, usually with minimal intragastric pressure [15]. Following this, low-amplitude tonic contractions facilitate the movement of gastric contents to the distal stomach, where they are mixed with gastric secretions and ground into particles, typically smaller than 1–2 mm in diameter [16]. Phasic and tonic contractions at the pylorus are essential for regulating gastric emptying, facilitating the delivery of chyme to the duodenum, predominantly in a pulsatile rather than continuous manner, following the opening of the pylorus [17].

In the small intestine, the presence of nutrients triggers the release of a number of gastrointestinal hormones—including glucagon-like peptide-1 (GLP-1), cholecystokinin (CCK), and peptide YY (PYY)—which modulate gastric emptying through feedback inhibition. These hormones, along with others including motilin, ghrelin, somatostatin, leptin, gastric inhibitory polypeptide (GIP), pancreatic polypeptide, oxyntomodulin, and amylin, form a complex regulatory loop that coordinates both gastric motility and pancreatic secretion [11,18].

Motilin, secreted from the duodenojejunal mucosa, plays a role in the regulation of phase III of the migrating motor complex, with exogenous administration shown to enhance gastric antral contractions and fundic tone during fasting—supporting its prokinetic role [19]. Ghrelin, released from the oxyntic mucosa of the stomach, also exhibits gastrokinetic properties, particularly in rodent models, although its effects in humans are less conclusive [19,20]. Dysregulation of these hormonal pathways may contribute significantly to the pathogenesis of diabetic gastrointestinal autonomic neuropathy by disrupting normal digestive function and motility [21].

Both the rate and pattern of gastric emptying are influenced by the meal composition (solid, semi-solid, or liquid), osmolarity, caloric density, and particle size. Liquids are preferentially emptied into the small intestine before solids [22]. Unlike liquids, digestible solids typically exhibit an initial lag phase before emptying commences [23]. Non-nutrient liquids empty with an overall exponential emptying pattern, while liquids with a higher nutrient and caloric content exhibit a more linear emptying rate [24]. The presence of nutrients in the proximal small intestine inhibits gastric emptying. The magnitude of this inhibitory feedback is dependent on the type of nutrient, as well as both the length and region of small intestine exposed [25].

In healthy individuals, the overall rate of the gastric emptying of nutrients usually ranges from 1 to 4 kcal/min [26]. While substantial inter-individual variability in gastric emptying is observed among healthy individuals, this variability is even greater in those with diabetes [27]. This is primarily attributable to the high prevalence of delayed gastric emptying in people with diabetes, although a subset exhibit rapid gastric emptying.

Nearly half of individuals with long-standing type 1 and 2 diabetes and poor glycemic control (HbA1c > 8.5%) experience delayed gastric emptying of solids and/or nutrient liquids, whereas some 5% have more rapid emptying [28,29,30]. There is limited information about gastric emptying in individuals with well-controlled diabetes. A cross-sectional study demonstrated that gastric emptying is accelerated in people with well-controlled type 2 diabetes compared to age- and BMI-matched individuals without diabetes [31]. Another study indicated that gastric emptying is more rapid in individuals with both poorly and well-controlled diabetes compared to age-matched individuals without diabetes [32].

It is not widely appreciated that people with obesity without diabetes may exhibit relatively more rapid gastric emptying, and some racial groups may be predisposed to diabetes. Preliminary studies in various ethnic groups, including Han Chinese people, Mexican Americans, and American Indians, are indicative of a more rapid gastric emptying rate compared to Caucasian people, which could contribute to a higher risk of developing type 2 diabetes [33].

3.1. Gastric Emptying and Postprandial Glycemia

There is a bidirectional relationship between gastric emptying and postprandial glycemia (Figure 1). The rate of gastric emptying is a key determinant of postprandial glycemia (especially in the first 60 min post-meal), such that both healthy people and individuals with type 2 diabetes with relatively more rapid emptying exhibit higher initial postprandial glycemic excursions. In healthy individuals with normal glucose tolerance, relatively more rapid gastric emptying is related to a greater glycemic response at 30 min following a 75 g oral glucose tolerance test, but not at 60 min, and this relationship is inverse at 120 min. In contrast, individuals with impaired glucose tolerance exhibit more rapid gastric emptying, which correlates with a higher glycemic response at both 30 and 60 min, but not at 120 min. In individuals with type 2 diabetes, more rapid gastric emptying is associated with a greater glycemic response at both 60 and 120 min [34,35]. Accordingly, the relationship between glycemia and the rate of gastric emptying, which exhibits a rightward shift, serves as a predictive measure of glucose tolerance.

However, in individuals with type 1 diabetes, delayed gastric emptying is evident with both early and long-term hyperglycemia, suggesting a sustained relationship between delayed gastric emptying and glycemic dysregulation [37]. This contrasts with observations in type 2 diabetes, where more rapid gastric emptying correlates with greater postprandial glycemic excursions [38]. These contrasting patterns suggest that although gastric emptying disturbances occur in both types of diabetes, the underlying pathophysiological mechanisms and their influence on glucose regulation differ substantially between type 1 and type 2 diabetes. We have reported that poor glycemic control, as assessed by a glycated hemoglobin [HbA1c], is associated with a prolonged lag phase (the time period before the emptying of solids commences) in people with type 1 diabetes using scintigraphy [39].

Given the relationship between gastric emptying and postprandial glycemia, dietary interventions that slow gastric emptying may be a potential treatment in type 2 diabetes. One strategy to reduce postprandial glycemia is the consumption of a small amount of protein or fat before a meal (dietary ‘preload’). The presence of these nutrients in the small intestine triggers the release of peptides, such as GLP-1, GIP, and cholecystokinin, leading to slower gastric emptying, enhanced insulin secretion, and a consequent reduction in postprandial glycemia [40]. Several studies have evaluated the effect of preloads of protein or amino acids before carbohydrate meals in individuals with type 2 diabetes. For example, consuming 55 g of whey protein 30 min before a potato meal slowed gastric emptying, reduced postprandial glycemia, and stimulated the secretion of incretin hormones [41]. In another study, the ingestion of 30 mL of olive oil 30 min before a carbohydrate meal was shown to slow gastric emptying, improve postprandial glycemia, and increase incretin hormone secretion [40].

Pharmacological agents, which modulate gastric emptying, also influence postprandial glycemia in people with type 2 diabetes. In ‘proof of principle’ studies, prokinetic agents, such as erythromycin, resulted in higher postprandial blood glucose by accelerating gastric emptying, while opioid analogs, such as morphine, slowed gastric emptying to reduce postprandial glycemia in individuals with type 2 diabetes [42].

Hypoglycemia is one of the most important complications of insulin therapy in people with type 1 and 2 diabetes. Abnormally delayed gastric emptying in insulin-treated individuals has been associated with a decreased insulin requirement during the first two hours after a meal, potentially increasing the risk of hypoglycemia [43]. This concept is consistent with the outcome of a community-based study which identified delayed gastric emptying as a risk factor for hypoglycemia in insulin-treated individuals, particularly among those who experienced recurrent episodes of hypoglycemia soon after meals. The gastric emptying of a scrambled egg meal was notably slower in this group compared to people with insulin-treated diabetes without recurrent hypoglycemia. In approximately 30% of people, the delay in gastric emptying was marked, despite the absence of GI symptoms such as nausea or bloating [44].

Conversely, acute variation in blood glucose impacts the rate of gastric emptying, such that it is accelerated substantially during hypoglycemia and delayed during hyperglycemia [45]. The significant impact of glycemia on the rate of gastric emptying has potentially important implications for the efficacy of orally administered medications, including oral hypoglycemic agents [46]. The absorption of sulphonylureas was reported to be decreased after an acute increase in plasma glucose concentration, presumably leading to the slowing of gastric emptying [47].

3.2. Gastric Emptying in Type 1 and Type 2 Diabetes

Gastroparesis is defined as “delayed gastric emptying of nutrients in the absence of a mechanical obstruction of the stomach” [48]. Gastroparesis is present in 30–50% of individuals with long-standing, poorly controlled type 1 and 2 diabetes [27], and is associated with malnutrition, weight loss, poor dehydration, impaired glycemic control, and variable drug absorption, as well as an impaired quality of life [48]. A population-based study from Olmsted County, Minnesota, found that the cumulative proportion of people developing gastroparesis over a 10-year period was 5.2% in type 1 diabetes and 1% in type 2 diabetes compared with 0.2% in non-diabetic controls [49].

However, there is only a modest association between GI symptoms and gastric emptying. Gastroparesis has historically been assumed to be closely associated with GI symptoms, including nausea, postprandial fullness, and bloating [50], but a recent meta-analysis found only postprandial fullness to be associated with delayed gastric emptying in diabetes, albeit weakly [51]. In other studies, two-thirds of people with delayed gastric emptying did not have bloating, nausea, or postprandial fullness [52,53]. People with type 1 diabetes tend to report symptoms suggestive of delayed gastric emptying more than those with type 2 diabetes, especially those with long-standing diabetes [54].

3.3. Pathogenesis of Gastroparesis

The pathogenesis of gastroparesis, whilst incompletely understood, is recognized to be multifactorial, and autonomic impairment is a major contributor (Figure 2) [55,56].

The major parasympathetic autonomic input for the ENS is the vagus nerve, and a loss of parasympathetic input may lead to impaired pyloric sphincter relaxation, thereby slowing gastric emptying [57]—a phenomenon which is evident following surgical vagotomy [58]. The assessment of gastric biopsies from patients with diabetic gastroparesis revealed a substantial reduction in the number of nerve cell bodies when compared to non-diabetic controls. Studies on gastric biopsies of people with refractory diabetic gastroparesis have reported mild lymphocytic infiltration in the myenteric plexus in about 50%, as well as reductions in nerve cell bodies [59].

In diabetic mice, ICC depletion is associated with delayed gastric emptying, reflecting disrupted electrical pacing and reduced motor neurotransmission [60]. Furthermore, human studies indicate that the depletion of ICCs occurs in up to 50% of people with severe gastroparesis [61]. An additional cause of diabetic enteric neuropathy may be extracellular matrix (ECM) remodeling. Diabetes can alter ECM composition in the gastrointestinal tract, potentially disrupting the structural integrity of the enteric nervous system. These changes may impair cell adhesion, tissue architecture, and neural signal transmission, thereby contributing to neuropathic complications [62].

Diabetes also alters the levels and activity of key neurotransmitters, including nitric oxide (NO), vasoactive intestinal peptide (VIP), and serotonin. These changes contribute to the dysregulation of GI motility and secretion [63]. NO, a key mediator of gastric motility, is generated from neuronal nitric oxide synthase (nNOS) neurons [64]. It has been suggested that the binding of advanced glycation end products (AGEs) to myenteric neurons, a characteristic pathophysiological feature of diabetes [65], may reduce nNOS expression in diabetic rats [66].

In addition, somatostatin, an inhibitory neurotransmitter, is a potent suppressor of various gastrointestinal (GI) functions, including peristalsis [66,67]. It primarily inhibits the production and secretion of several hormones and peptides, such as glucagon, insulin, and growth hormone. Somatostatin also suppresses the release of acetylcholine, a major excitatory neurotransmitter in the enteric nervous system, thereby potentially contributing to reduced gastrointestinal motility [68]. However, in enteric neuropathy, this regulatory balance is disrupted, leading to either excessive or insufficient acetylcholine release and hormonal dysregulation, thereby impairing GI motility and function [69].

Additionally, diabetes may disrupt ion channel activity in enteric neurons, particularly involving calcium and potassium pathways. Such a disturbance can decrease neuronal excitability and neurotransmitter release in the ENS and contribute to GI motility disorder [70,71].

Long-standing hyperglycemia and mitochondrial dysfunction in diabetes may exacerbate neural damage by increasing the production of reactive oxygen species (ROS), which, when combined with impaired antioxidant defenses, can further exacerbate neuronal injury [72]. Impaired autophagy in diabetic enteric neurons may also compromise cellular repair and maintenance mechanisms, increasing neuronal vulnerability and contributing to gastrointestinal complications [72,73].

A role for autoantibodies—specifically ganglionic acetylcholine receptor (gAChR) autoantibodies, which are associated with the development of autoimmune autonomic neuropathy, a condition in which GI dysfunction is common—has been suggested [74]. In addition, increased levels of pro-inflammatory cytokines can induce cellular stress and apoptosis in neurons, thereby exacerbating neuropathic conditions [75]. Inflammation associated with compromised mucosal integrity may also increase gut permeability and contribute to symptom severity. It is hypothesized that, in diabetes, alterations in the excitatory/inhibitory neurotransmitter balance—mediated by pro-inflammatory factors and neuropeptides—play a key role in ENS dysfunction and neuronal loss.

In diabetes, insulin-like growth factor-1 (IGF-1) levels are frequently reduced and may play a role in the development of peripheral neuropathy [76]. Alterations in IGF-1 pathways may contribute to the development of gastroparesis by damaging myenteric cholinergic neurons and ICCs [77]. Furthermore, in diabetes, normal protective mechanisms are disturbed, including heme oxygenase, an antioxidant that regulates oxidative stress by promoting ICCs and nitric oxide synthase [78]. However, in diabetes, its production is disturbed, potentially leading to ICC loss and, subsequently, gastroparesis.

Hormone imbalance may impair the coordination between inhibitory and excitatory neuromuscular transmission in the GI tract, resulting in impaired peristalsis and delayed gastric emptying [79]. Although glucagon does not have a direct effect on gastric emptying, its role in glucose regulation may indirectly influence gastric emptying in diabetes [80]. Overall, dysregulation of these hormonal pathways may play a major role in the development of diabetic gastrointestinal autonomic neuropathy and be responsible for disturbing digestive function and motility [21].

3.4. Diagnosis of Gastroparesis

As described, neither the presence nor severity of GI symptoms reliably predicts gastroparesis. Moreover, such symptoms are reported by individuals with other GI disorders. For example, cardinal symptoms of diabetic gastroparesis, such as early satiety, fullness, vomiting, and weight loss, are also reported in those with gastric outlet obstruction syndrome. Gastroparesis can also be associated with dyspepsia, indigestion, and gastro-esophageal reflux disease (GERD). Furthermore, people with eating disorders such as anorexia nervosa and bulimia may also exhibit similar symptoms [78]. In individuals with potential gastroparesis, gastric outlet obstruction must always be excluded, usually by endoscopy. Medications that can influence gastric emptying (such as anticholinergics, opioids, metoclopramide, and erythromycin) should be documented and their relevance should be considered prior to the diagnosis of gastroparesis [54]. Ideally, blood glucose should be normalized to <10 mmol/L prior to the assessment of gastric emptying.

3.4.1. Diagnostic Methods for Gastroparesis

The diagnosis of gastroparesis is dependent on the objective measurement of gastric emptying [81], and the gold standard test is scintigraphy, which has the capacity to precisely measure the emptying of both liquid and solid components of meals [82]. The limitations of scintigraphy include unavoidable radiation exposure, relatively high costs, the need for specialized equipment, and the requirement for staff trained in nuclear medicine [83]. Single-photon emission computed tomography (SPECT) can indirectly measure gastric tone, however, its use is limited by challenges such as ionizing radiation exposure, high expenses, and restricted availability [84]. Alternative measurements that do not involve radiation exposure include ¹³C-based breath tests and ultrasonography [85]. Other techniques that have been explored for the assessment of gastric emptying include the wireless motility capsule, MRI, and SPECT imaging. Of these, the wireless motility capsule is no longer available, but the others remain primarily research tools [86].

3.4.2. Surrogate Measures of Gastrointestinal Autonomic Neuropathy

Cardiovascular autonomic reflex tests (CARTSs), the “gold standard” test for assessing cardiovascular autonomic function, are a battery of simple, non-invasive tests measuring heart rate and blood pressure in response to a change in posture, and with the Valsalva maneuver and a sustained hand grip, have been traditionally used as surrogate measurements of GI autonomic function [87]. However, the relationship between GI motility, including gastric emptying and cardiac autonomic parameters (such as heart rate and cardiac vagal tone), is not strong [88,89]. Biomarkers such as serum pancreatic polypeptide are an indirect measure of vagal influence on the GI tract [90,91], but their utility remains to be established [92]. Tests of sudomotor function such as the Sudoscan^TM, a non-invasive measure of sweat gland function, which is mediated by the autonomic nervous system, could potentially serve as surrogate measurements for GAN [93], although we are unaware of any studies directly comparing these modalities.

3.4.3. Questionnaires for Assessing Gastrointestinal Symptoms and Autonomic Neuropathy

The use of validated questionnaires has benefits for evaluating GI symptoms such as abdominal pain and bloating, heartburn, dyspepsia, nausea and vomiting, and diarrhea over the self-reporting of symptoms. The Patient Assessment of Upper GI Symptom Severity Index (PAGI-SYM) questionnaire is validated and commonly used for the assessment of GI symptoms. It comprises twenty items and the following six subscales: heartburn/regurgitation (seven items), nausea/vomiting (three items), postprandial fullness/early satiety (four items), bloating (two items), upper abdominal pain (two items), and lower abdominal pain (two items). Each symptom is rated on a scale from zero (no symptoms) to five (very severe), reflecting symptom severity over the past two weeks [94]. A limitation of the PAGI-SYM is that it focuses on upper GI symptoms (e.g., nausea, bloating, and early satiety) and, accordingly, poorly reflects symptoms in other parts of the GI tract (e.g., lower GI symptoms like diarrhea or constipation) [94]. While there is no specific validated questionnaire to assess GAN, the Gastroparesis Cardinal Symptom Index (GCSI), derived from a subset of questions of the PAGI-SYM, has been used to assess the severity of gastroparesis [95].

The Diabetes Bowel Symptom Questionnaire (DBSQ) comprises ten items assessing upper and lower GI symptoms using a six-point Likert scale, ranging from “never” to “several times during the week or every day”. It evaluates gastroesophageal reflux (heartburn/regurgitation), gastroparesis (postprandial fullness, nausea/vomiting, bloating, and abdominal pain), irritable bowel syndrome (pain linked to defecation and stool consistency), and lower GI symptoms (diarrhea, constipation, and anal incontinence) [96].

The COMPASS-31 questionnaire comprises 31 questions with six domains, including orthostatic intolerance, vasomotor, secretomotor, GI, bladder, and pupillomotor function. While it provides a comprehensive tool for the assessment of autonomic dysfunction, it is generally considered non-specific for the presence of GI autonomic neuropathy [97].

4. Treatment of Gastroparesis

The optimal management of diabetic gastroparesis often necessitates a multidisciplinary approach involving input from a diabetologist, nutritionist, gastroenterologist, and specifically trained nursing staff. Broadly, the treatment of diabetic gastroparesis can be considered under the following headings: (i) altering food/fiber composition, particle size, or the frequency of meals, (ii) interventions to accelerate gastric emptying, including in severe or refractory gastroparesis, (iii) the relevance of glycemic control, and (iv) other strategies [98].

4.1. Altering Food/Fiber Composition, Particle Size, and Frequency of Meals

Nutritional interventions are often recommended as the first-line management for diabetic gastroparesis, although this approach lacks a strong evidence base in the form of rigorously conducted, large, randomized control trials. According to Abrahamsson et al. (2007) [99], the optimal diet for gastroparesis should be low in fat and fiber and contain small particles, as well as small, frequent meals. However, more recently, Suresh et al. [100] suggested that low-viscosity, soluble fibers may reduce postprandial glycemia and improve gastroparesis symptoms after four weeks in people with both type 1 and type 2 diabetes. A study comparing the effects of different (high/low fat and liquid/solid) meals on individuals with diabetes and gastroparesis demonstrated that high-fat solid meals increased symptom severity, whereas low-fat liquid meals caused the fewest gastroparesis symptoms, as measured by the PAGI-SYM questionnaire [101]. In a small study (n = 7) involving people with type 1 diabetes with gastroparesis, a shorter initial gastric lag phase was reported after the consumption of a small-particle-size diet, as would be predicted [102].

The consumption of small, frequent meals is recommended to reduce the impact of gastroparesis on nutritional uptake [103]. In some individuals who cannot tolerate solid foods, diet alteration, dietary supplementation with liquid meals, or oral nutrition supplements with high calories are prescribed. As discussed, liquids are emptied by the stomach more rapidly in comparison to solids, so this approach may reduce the impact of gastroparesis [104,105].

4.2. Interventions to Accelerate Gastric Emptying in People with Gastroparesis

Accelerating gastric emptying in people with gastroparesis has been studied with both non-pharmacological and pharmacological approaches. It should be noted, however, that the impact of the acceleration of gastric emptying on symptoms has, in general, yielded only a weakly positive relationship.

4.2.1. Non-Pharmacological Interventions

Isoflavones, phenolic compounds with either estrogen-agonist or estrogen-antagonist properties [106], may accelerate gastric emptying in people with gastroparesis [107]. A cross-over study involving ten individuals with delayed gastric emptying found that isoflavones accelerated gastric emptying, as measured by the [¹³C] octanoic acid breath test, over eight weeks. It is thought that isoflavones enhance gastric emptying by affecting prostaglandin synthetase. The consumption of soy germ pasta daily for 8 weeks has been reported to improve gastroparesis symptoms among individuals with 2 diabetes [108]. There is also some evidence that the essential oil of Pistacia atlantica gum may accelerate gastric emptying by increasing peristaltic contraction in the fundus and antrum, and may improve symptoms among individuals with type 1 and type 2 diabetes [109].

4.2.2. Pharmacological Treatments

A number of so-called ‘prokinetic agents’, designed to accelerate gastric emptying, have been studied and are commonly used in clinical practice. A liquid form of metoclopramide, a dopamine receptor antagonist/serotonin 5-HT₄ receptor agonist and a weak inhibitor of 5-HT₃ receptors, can be prescribed at 5–10 mg up to three times per day, 15 min before meal, as a safer and more effective form of the drug [104]. Nevertheless, because of potential adverse effects on the cardiovascular and central nervous systems (especially tardive dyskinesis), its consumption should usually be restricted to less than 12 weeks [110]. The results of a double-blind study reported that metoclopramide at a dosage of 10 mg four times a day for three weeks could reduce gastric symptoms and accelerate gastric emptying in 40 people with type 2 diabetes and gastroparesis [111]. In the trial by Ricci et al. [95], the parenteral administration of 10 mg metoclopramide was shown to accelerate gastric emptying, while the oral administration of 10 mg metoclopramide before meals and at bedtime resulted in a 53% reduction in gastric symptoms.

Domperidone is another dopamine receptor antagonist with a lower prevalence of adverse effects in comparison to metoclopramide. It is usually prescribed at a dose of 10–20 mg three times a day to target nausea and vomiting [80]. Prescribing 10 mg of domperidone three times a day to 24 people with diabetic or idiopathic gastroparesis improved ‘gastric’ symptoms after six weeks of treatment [112]. It should be noted, however, that domperidone is not approved for use by the FDA in the USA due to concerns of cardiac arrhythmias.

The antibiotic Erythromycin is a motilin agonist and an alternative treatment for gastroparesis. It acts by binding to motilin receptors and can initiate phase III activity of the gastric MMC. It is effective in accelerating gastric emptying and addressing motility symptoms such as vomiting, early satiety, and fullness. Erythromycin, like other motilin agonists, improves gastric symptoms when given orally, however, the down-regulation of motilin receptors occurs with long-term therapy, leading to tachyphylaxis [113]. A study comparing the effects of erythromycin (250 mg three times/day) and metoclopramide (10 mg three times/day) in individuals with both type 1 and 2 diabetes with gastroparesis reported meaningful improvements in gastric emptying and symptoms in both groups, however, erythromycin appeared to be more effective [114].

Ghrelin stimulates GI motility via both vagal and non-vagal mediated mechanisms. In a cross-over study, the infusion of ghrelin (5 pmol/kg/min) over two hours accelerated gastric emptying in people with type 1 and 2 diabetes with gastroparesis [115]. However, due to its short half-life, potential treatment with endogenous ghrelin is not possible. Hence, synthetic ghrelin agonists, including TZP-101, TZP-102, and relamorelin (RM-131), with longer half-lives, have been developed for the treatment of diabetic gastroparesis. Relamorelin, a novel and selective ghrelin receptor agonist, was shown to accelerate gastric emptying in individuals with both type 1 [116] and 2 diabetes [117,118,119].

Other motilin agonists such as camicinal [120,121] and mitemcinal [122] have also been shown to accelerate gastric emptying in diabetic gastroparesis.

In addition to the prokinetics used to accelerate gastric motility, antiemetics, including phenothiazines and antihistamine agents, may be prescribed for relieving nausea and vomiting [80]. Small studies in healthy individuals have indicated that serotonin (5-HT₃) receptor blockers, including ondansetron and granisetron, may decrease nausea [118,123], but there is a lack of evidence to support their efficacy in individuals with gastroparesis. The 5-HT₄ receptor agonist cisapride also showed promise, with initial studies showing the proliferation of antral contractions and accelerated gastric emptying, but it is no longer FDA-approved due to the risk of serious electrocardiogram changes, such QT prolongation and cardiac arrhythmias.

4.3. Severe and Refractory Gastroparesis

Enteral feeding may be needed for patients with severe gastroparesis who cannot meet their dietary requirements through oral intake [45], which can also provide hydration and medication. Post-pyloric tube feeding is an alternative method for the transmission of nutritional formula directly to the small intestine. Short-term feeding may be achieved with nasoduodenal or naso-jejunal intubation, but for a longer duration, gastrojejunostomy or jejunostomy tubes should be placed [104]. Tube feeding is an effective option for optimizing glycemic control in individuals with diabetes who suffer from frequent vomiting and malnutrition, particularly in response to meals [124]. However, there is a risk of pulmonary aspiration in those who have impaired lower esophageal sphincter function [124]. Parental nutrition is rarely required for patients with gastroparesis and is used when enteral feeding cannot be tolerated [125]. Enteral nutrition is preferrable to parenteral nutrition, particularly as the latter can increase the risk of infections, liver disease, thrombosis, and metabolic bone disease. Since gastroparesis may lead to poor oral intake and malnutrition, close supervision by an experienced dietician is pivotal to the management of nutritional status and hydration [126].

4.4. Gastric Electrical Stimulation

In people with refractory gastroparesis, gastric electrical stimulation (GES) may be an option to improve symptoms. The Enterra GES system (high frequency/low energy) is FDA-approved and is implanted surgically via laparotomy, with recipients needing to stay in hospital for approximately two days post-implantation. GES may also impact nutritional status and quality of life. However, evidence related to its efficacy is inconsistent. The Enterra GES system may be more helpful in people with diabetic gastroparesis in comparison to individuals who have idiopathic or postsurgical gastroparesis.

5. The Impact of GLP-1 Receptor Agonists on Gastric Emptying

GLP-1 receptor agonists (GLP-1 RAs), based on the gut-derived incretin hormone, GLP-1, are widely used today in the management of type 2 diabetes and obesity [127].

The half-life of native GLP-1 is ~five minutes; however, GLP-1 RAs differ in their chemical structure and pharmacokinetic profiles, resulting in a prolonged duration of action and therapeutic efficacy. Based on their duration of action, these agents are categorized as either short-acting (such as exenatide twice daily and lixisenatide) or long-acting (including liraglutide, once-weekly exenatide, dulaglutide, albiglutide, and semaglutide) [128]. We have shown that endogenous GLP-1 has the potential to slow gastric emptying using the specific GLP-1 antagonist Exendin 9-39. Accordingly, gastric emptying is accelerated when exendin 9-39 is administered and delayed by the exogenous administration of GLP-1, resulting in both an initial rise and fall in postprandial glucose, respectively [129].

Due to their pharmacokinetics, short-acting agents are usually administered before meals and are useful in attenuating postprandial glucose spikes [130]. Studies using scintigraphy and breath tests have demonstrated that short-acting GLP-1 RAs lower postprandial glucose primarily through delayed gastric emptying, with more pronounced effects in individuals with relatively faster baseline gastric emptying [131,132,133]. In healthy individuals, the continuous infusion of GLP-1 (~24 h) reduces gastric emptying compared to placebo; however, this effect is less marked than that observed with acute or intermittent administration. This indicates that the initial potent effect on slowing gastric emptying can lessen with sustained exposure to GLP-1 [134].

It is generally believed that long-acting GLP-1 RAs initially slow gastric emptying, but that effect diminishes over time due to tachyphylaxis, typically occurring after several weeks or months of continuous exposure. As a result, their glucose-lowering action may be explained primarily through their actions on insulin secretion and glucagon suppression, particularly improving fasting plasma glucose [135]. However, it should be appreciated that the number of studies on the effects of long-acting GLP-1 RAs on gastric emptying is small, and these are indicative of a sustained effect on slow emptying, which may potentially diminish over time. For example, in healthy individuals with a BMI of >30, subcutaneous liraglutide (3 mg) was shown to delay the gastric emptying of solids at 5 weeks; however, this effect was attenuated by week 16, while still significant [136]. A secondary analysis showed significant interindividual variability in both the magnitude and timing of this response [137]. We have recently shown that Exenatide (once weekly) also delayed the gastric emptying of solids when assessed by scintigraphy, although the magnitude of this delay was less than that seen with short-acting agents [133].

A delay in gastric emptying is one of the main mechanisms accounting for postprandial glucose lowering by GLP-1 RAs [138]. The magnitude of the slowing of gastric emptying depends on the baseline rate of emptying, with a more pronounced slowing in individuals with initially relatively rapid gastric emptying [139]. In contrast, in critically ill patients—where gastric emptying is often already markedly delayed—exogenous GLP-1 does not exert additional slowing effects [139].

Both upper and lower GI symptoms are commonly seen with GLP-1 RAs and are often the main reason for discontinuation. Upper GI events are thought to be primarily the result of direct action on the central nervous system by GLP-1 receptors (area postrema in the brain stem) [140]. In the lower GI tract, GLP-1 may increase motility and induce diarrhea through the intramural autonomic plexus [141,142].

However, there is a growing concern that the empirical usage of GLP-1 RAs may unmask or worsen pre-existing gastroparesis. An increasing number of case reports on guidance for GLP-1 RA gastroparesis have been published. In response, guidelines from medical societies such as The American Society of Anesthesiologists have been published, recommending withholding a long-acting agent for at least one week prior to the procedure/surgery in people using long-acting GLP-1 RAs (once-weekly agents). These, however, lack an evidence base. For example, we have previously shown that exenatide QW’s ability to delay gastric emptying is longer than a week. There is the possibility that gastric emptying may be profoundly delayed, impacting glycemia to the point of hypoglycemia, especially when insulin is concomitantly administered with these agents. This would reflect a mismatch between insulin delivery and carbohydrate absorption due to prolonged gastric retention by GLP-1 RAs [133].

A case could be made for the routine assessment of GI autonomic nervous function in people with diabetes, including the use of validated questionnaires for GI symptoms and gastric emptying studies. At the very least, clinicians should be wary of prescribing a combination of GLP-1 RA and insulin in those with an impaired awareness of hypoglycemia or with a known history or risk of developing gastroparesis.

6. Conclusions

Gastric autonomic neuropathy is an important and common complication of diabetes, which may lead to substantial morbidity and impacts on glycemic control. Delayed gastric emptying occurs frequently in diabetes, and severe gastroparesis can be debilitating. There have been recent, important insights into the pathophysiological mechanisms underlying gastroparesis. The current management options for gastroparesis, however, remain sub-optimal, and the beneficial impact of accelerating gastric emptying on symptoms is only modest.