Idiopathic gastroparesis

Epidemiology

To the best of our knowledge, only three large population-based studies have been performed on the epidemiology of IG. One of these studies was from the Rochester Epidemiology Project, a database of medical records from 3604 patients from Olmsted County, Minnesota from 1996 to 2006, and 83 of these patients met criteria for definite gastroparesis . Definite gastroparesis was defined as having delayed gastric emptying by standard scintigraphy and typical symptoms for more than 3 months. IG was seen in 49.4% of these cases (n=41), 25.3% were diabetic, 23% medication-related, 10.8% from connective tissue disease, 7.2% postsurgical, 2.4% malignancy-related. The age-adjusted incidence per 100,000 person-years of definite gastroparesis for the years 1996–2006 was 9.8 for women (95% confidence interval [CI], 7.5–12.1) and 2.4 for men (95% CI, 1.2–3.8). The age-adjusted prevalence of definite gastroparesis per 100,000 persons was 37.8 in women (95% CI, 23.3–52.4) and 9.6 in men (95% CI, 1.8–17.4). Among the 41 IG patients, nearly 22% (n=9) were associated with a preceding acute viral illness.

The second large population-based study on IG was on 243 patients enrolled in the GpCRC where clinical features including sex, body mass, symptom onset, and symptom severity were described . Data from this study showed that the mean age of patients with IG was 41 years, 88% of patients were female, 90% Caucasian, 20% overweight (BMI, 25–30 kg/m ² ), 26% obese (BMI, >30 kg/m ² ), 19% had an infectious prodrome, and 50% had an acute onset of symptoms. Severe gastric emptying delay was noted in 28% of patients (defined as >35% retention at 4 hours where <10% retention is normal), and these patients had more severe nausea, vomiting, and anorexia compared to patients with mild (10–25% retained after 4 hours) or moderate (25–35% isotope retained at 4 hours) gastric emptying. The study concluded that IG is a heterogeneous syndrome that primarily affects females, and for reasons not understood, also affects obese patients.

The third large population-based study performed on IG was by Parkman et al. on 718 patients enrolled in the GpCRC from 2007 to 2017 . This study demonstrated that the etiology, symptom severity, and treatments for gastroparesis vary according to race, ethnicity, and gender. From an ethnic standpoint, whites were more likely to have IG compared to Hispanics and blacks (78% in whites, 11% in Hispanics, 9% in blacks; P= 0.60). Females were more likely to have IG compared to males (69% vs 46%; P <.001). Even though females had more severe gastrointestinal symptoms including bloating and postprandial fullness, they also had better clinical improvement over time compared to males.

Similar patient characteristics to the three aforementioned population-based studies were seen in a tertiary referral case series of 146 patients with IG with most of these patients showing a strong female predominance (82%) and with a mean age of onset of 45 years . The etiology of gastroparesis for these 146 patients were 36% idiopathic, 29% diabetic, 13% postgastric surgery, 7.5% Parkinson’s, 4.8% collagen vascular disease, 4.1% intestinal pseudoobstruction, 6% miscellaneous. Subgroups within the IG group further noted that 23% of these patients had a pre-viral prodrome, 62% were female, and 48% had predominantly abdominal pain. Other studies corroborate similar patient characteristics to the three previously mentioned population-based studies in IG: predominantly female, mean-age of onset in the early 40’s, obesity or overweight predominance, and a preceding infectious prodrome linked to a good prognosis .

Obesity is a consistent feature identified as a clinical characteristic in IG worth noting. A study published by Homko et al., evaluated factors influencing body weight in 29 patients with IG compared to 39 healthy controls by assessing their resting energy expenditure, body composition, dietary intake, symptoms, and physical activity . Their results demonstrated that a subgroup of patients with IG gained weight because of increased caloric intake, consumption of energy-deficient foods (diet deficient in calories, vitamins, and minerals), and reduced energy expenditure. In a study of 305 patients enrolled in the GpCRC with IG (n=204) and DG (n=101), T2DM gastroparetic patients were significantly more overweight or obese compared to IG: 46% IG versus 70% DG (50.9% T1DM and 91.3% of T2DM) . Also, this study showed that patients with IG were more likely to be deficient in vitamin B ₆ , vitamin K, and iron compared to DG patients. Ogorek et al. also found he diet of IG patients to have marked deficiencies in many essential vitamins and minerals consistent with the past observations of energy-deficient foods with overall increased caloric content .

Symptoms

Cardinal symptoms of gastroparesis includes nausea, vomiting, early satiety, postprandial fullness, abdominal pain, and bloating . These symptoms result from several proposed mechanisms including visceral hypersensitivity, decreased gastric tone, hypomotility of the gastric antrum, and impaired antroduodenal coordination . The symptom severity of patients with gastroparesis can be reliably assessed using the Gastroparesis Cardinal Symptom Index (GCSI) . The GCSI questionnaire is a subset of the longer Patient Assessment of Upper Gastrointestinal Symptoms (PAGI-SYM) questionnaire validated to assess the symptom severity in patients with upper gastrointestinal disorders (gastroesophageal reflux, dyspepsia, gastroparesis) . The GCSI measure of gastroparesis-related symptom severity was developed by Revicki et al. after 169 gastroparesis patients were recruited from seven clinical centers across the US to complete this survey at a baseline visit and again in 8 weeks. The GCSI is comprised of nine symptoms in three sub-scales: post-prandial fullness/early satiety, nausea/vomiting and bloating. It should be noted that abdominal pain is not included in this scoring system. The GCSI has been validated as a reliable tool to assess the symptom severity and symptom profile in patients with both IG and DG .

A number of similarities and differences exist in both the symptomatology and clinical presentation of patients with DG compared to IG. Common symptoms between both groups of patients include nausea, vomiting, early satiety, and postprandial fullness . In a large multicenter study of 416 patients enrolled in the NIH GpCRC (61% IG [n=254], 33% DG [n=137; 78 with type 1, 59 with type 2]), the most common symptom prompting evaluation was abdominal pain for IG and vomiting for DG . Additionally, patients with IG were found to have more severe postprandial fullness and early satiety, whereas patients with DG were found to have more severe retching, nausea, and vomiting. Objectively, patients with IG tend to have a less severe delay in gastric emptying but have more severe abdominal pain, whereas patients with DG tend to have a more severe delay in gastric emptying but with more nausea and vomiting. When it came to hospitalizations, patients with type 1 DM reported more hospitalizations compared to IG, mostly for vomiting and dehydration (5.1±6.4 per year for T1DM versus 1.6±3.0 per year for IG; mean±SD). Similarly, a study from the Beth Israel Deaconess Medical Center and the Joslin Diabetes Center reported that patients with DG had more hospitalizations than patients with DM with gastrointestinal symptoms who had normal gastric emptying (i.e. 25.5 versus 5.1 per 1000 patient days) . From an epidemiological standpoint, females predominate in both DG and IG but this predominance is more pronounced in IG . Pathophysiological differences also exist between DG and IG in that patients with DG may have more vagal nerve dysfunction compared to patients with IG . From a cellular standpoint, patients with IG appear to have a more decreased expression of nitric oxide synthase compared to DG . Similarities in cellular abnormalities between both DG and IG are that both have been shown to have depleted interstitial cells of Cajal, loss of nerve fibers, and inflammatory infiltrates .

Several studies show a poor correlation between symptom severity and degree of delayed gastric emptying . Overall, the gastrointestinal symptoms that best seem to be associated with a delay in gastric emptying include early satiety, nausea, vomiting, and postprandial fullness . Research from the group at Texas Tech University in El Paso recently investigated the long-term relationship between gastric emptying changes and gastrointestinal symptoms in patients with IG. We assessed 23 patients with IG (87% females, mean age 48.4) who were treated with antiemetics and prokinetics and followed over 1 year. A gastric emptying study (GES) was performed at baseline and again at 1 year follow-up. Remarkably, up to 59% of IG patients normalized their fourth hour gastric retention at 1 year (<10% retention) with overall mean improvement of 17% in gastric emptying (mean fourth hour retention improved from 26.7% at baseline to 9.6% at follow-up). Additionally, their PAGI-SYM symptom scores also improved from a score of 44.61 at baseline to 35.6 a year later, but clearly, symptoms remained. Hence, in these patients with IG, there was some disparity between the good gastric emptying response with treatment and time, and symptom status which was still slowly resolving at follow-up. Therefore, longer term follow-up with larger patient numbers might help us determine if normalization of both gastric emptying (GE) and GI symptoms might occur. Certainly by initiating prokinetic therapy as was done in this report, as well as allowing the role of time to pass, gastric neuromuscular function may be receiving a “jump-start” in healing the insult induced by a presumed preceding offending agent (i.e. infection, toxin).

For patients with IG, several case series and clinical studies have emphasized that abdominal pain can be a common presenting symptom , whereas nausea and vomiting are most severe in DG . For example, in the McCallum series, “very prominent” abdominal pain was noted in 48% of IG patients (52 of 146 patients had IG) . In a large study of 346 gastroparetic patients (212 IG, 134 DG), abdominal pain was reported in 91% of IG and those with severe/very severe pain also had associated somatization symptoms and concomitant use of opiates . In another large study of 416 patients enrolled in the GpCRC (254 IG, 137 DG), abdominal pain was the most prominent symptom prompting medical evaluation in 76% of IG patients . The reason why IG patients experience more significant abdominal pain compared to other subtypes of gastroparesis is not well understood. Since poor correlation exists between the severity of abdominal pain with the severity of gastric emptying delay, one can hypothesize that – in addition to gastric motor dysfunction – several other factors may be playing a role.

Several possible factors may be contributing to abdominal pain predominance in patients with IG. Psychiatric dysfunction such as severe depression (up to 23%) and anxiety (36%) are common clinical findings in IG and is one of the factors proposed to contribute to the severity of gastrointestinal symptoms, including abdominal pain . In a GpCRC study involving 393 patients (256 IG, 137 DG), moderate upper abdominal pain was reported in 66% of IG, and increased pain was correlated with worsened depression, anxiety, quality of life, and increased narcotic use . Psychosocial factors can further influence the clinical trajectory of IG and result in increased hospitalizations during exacerbations and possibly even lead to increased surgical interventions and/or alternative nutrition . Additionally, surgical adhesions can further decrease quality of life and contribute to chronic gastrointestinal symptoms such as abdominal pain, bloating, and nausea . Oftentimes narcotics are required for pain control which can lead to even worsening symptoms since opioids worsen gastric emptying, thereby further exacerbating gastrointestinal manifestations, and thus creating a vicious cycle. In the tertiary referral case series of 146 patients with IG, 48% had predominantly abdominal pain, and their psychological status correlated with a predictive response to prokinetic therapy. For example, 62% of women with IG reported a prior history of physical or sexual abuse, and these patients were less likely to respond to prokinetics and they also had a significant association with depression, somatization, abdominal pain, and lifetime surgeries. These factors can be associated with visceral hypersensitivity which might also help explain the higher prevalence of abdominal pain in IG compared to other types of gastroparesis . Hence psychosocial dysfunction, surgical procedures, adhesions, narcotics, and/or a history of physical and/or sexual abuse are all factors that have been proposed to explain abdominal pain in IG.

Abdominal pain can also be a predominant symptom in other gastrointestinal and systemic disorders such as cyclic vomiting syndrome, dyspepsia, biliary colic, irritable bowel syndrome (IBS), and fibromyalgia which can result in significant overlap of symptoms and also confuse the diagnosis of IG. Several overlapping clinical features exists between IG and the aforementioned gastrointestinal disorders including female predominance, psychological disturbance, idiopathic nature of illness, and prior history of sexual and/or physical abuse . In a large multicenter study from the GpCRC, 86% of patients with IG met criteria for functional dyspepsia . As many as 70% of patients with FD also have anxiety and seek more frequent psychiatric consultations . A greater number of FD patients also seek more medical consultations for abdominal pain and other somatic complaints . Just like IG, abdominal pain is often times the predominant symptom in FD with up to 40% of this patient population reported to have abnormal GE . In IBS, emotional disorders such as anxiety, depression, somatization, stress, hypochondriasis are prevalent . Similarly to IG, abdominal pain and higher levels of psychological distress are also seen in IBS patients which greatly affect their quality of life and lead to a higher trend in health-care resource utilization . Fibromyalgia is a well-recognized explanation for diffuse pain ranging from joints and muscles to the abdomen and may be a concomitant diagnosis in some IG patients.

Above examples raise the question whether psychological experiences might alter the way the brain processes and interprets pain. Geeraerts et al. showed that experimentally-induced anxiety in healthy subjects reduced their pain threshold and resulted in significant reduction in gastric compliance . In 139 patients with FD (102 women), the severity of their anxiety negatively correlated with abdominal discomfort and pain thresholds and gastric compliance during balloon distension . Therefore, anxiety can lead to gastric sensory and motor dysfunction and can correlate with the severity of abdominal pain in patients with IG, IBS, and FD. This might serve as rationale to further study the relationship between the pain response and areas in the brain that play a key role in emotional responses and chronic pain modulation such as the amygdala and hippocampus. Functional magnetic resonance imaging of the brain might therefore help us map emotional motor system structures associated with abdominal pain in IG patients. It is therefore important to recognize and treat psychiatric factors in patients with IG and other functional gastrointestinal disorders as emotional disturbance can contribute to the patient’s symptom severity and disease progression if emotional and well-being issues are not recognized.

Another possible factor that can contribute to enhanced abdominal pain might be from musculoskeletal abdominal wall pain. Forceful vomiting alone can cause pain from the abdominal wall related to rectus muscle strain, cramping, or spasm. A positive Carnett’s sign on physical exam would further support this theory: a clinical exam finding characterized by worsening epigastric abdominal pain elicited by tightening of the abdominal wall muscles when the patient sits up or with leg raising maneuvers. A positive test indicates an increased likelihood of abdominal wall musculoskeletal pain and not abdominal cavity pain.

In summary, many factors explaining the predominance of abdominal pain in patients with IG is multifactorial including a possible co-existing secondary gastrointestinal disorder (i.e. IG with IBS or fibromyalgia) which may further complicate the clinical presentation. Other factors include psychosocial factors, history of physical and/or sexual abuse, chronic narcotics, adhesions from prior surgical interventions, and musculoskeletal causes. Further research and studies are needed to better understand the pathophysiology of abdominal pain in patients with IG.

To carry on our discussion on the symptomatology of IG, several important differences exist when comparing IG and DG patients. Registry data from 416 patients enrolled in the NIDDK GpCRC were reviewed and similarities and differences between DG and IG were extracted: 254 had IG and 137 had DG (78 had T1DM, 59 had T2DM) . Results showed that patients with T2DM were heavier and older compared to patients with IG. Patients with T1DM had worse gastric retention on GES, required more hospitalizations (that included possible diabetic ketoacidosis (DKA) and other diabetic complications accompanying renal failure), took more prokinetics, and had gastric electric stimulators placed more often compared to patients with IG. Symptoms triggering medical evaluation included abdominal pain in patients with IG and vomiting in patients with DG. IG had more severe postprandial fullness and early satiety compared to more severe retching and vomiting in DG. Overall, the authors summarized that patients with IG have more abdominal pain with less severe delayed gastric emptying whereas DG have more nausea and vomiting with more severe delayed GE . Additionally, patients with IG had less comorbidities and hospitalizations and reported a higher quality of life (PAGI-QOL) compared to DG. Playing into these clinical outcomes is the chronicity of DG which is lifelong, whereas in IG degrees of resolution of the gastroparesis syndrome are observed.

An interesting clinical feature in IG patients is that a subset of these patient report a preceding viral infection. For example, in a case series reported by McCallum et al. with 146 gastroparetic patients, 23% with IG had a pre-viral prodrome . In another large study with 394 patients enrolled in the GCrC, an infectious prodrome was reported in 17%, most commonly gastroenteritis or food poisoning . In a large NIH GpC study describing the clinical characteristics of 243 patients with IG, 50% of patients had acute onset of symptoms and 19% reported an initial infectious prodrome such as gastroenteritis or upper respiratory flu-like symptoms . In another large GpC study comparing clinical characteristics, symptoms, and gastric emptying in patients with IG and DG, an initial prodrome was reported at the start of symptoms in approximately 15% in all three etiologies of gastroparesis (IG, T1DM, T2DM) raising the possibility that an infectious viral prodrome might not just occur in IG but could exacerbate symptoms in DG . Bityutskiy et al. reported that 23% of 143 patients with IG were suspected to have postviral gastroenteritis . Patients with post-infectious gastroenteritis typically improved over the course of a year with slow resolution toward an eventual satisfactory quality of life. Some viruses that have been associated with post-viral gastroparesis include rotavirus, herpes zoster, cytomegalovirus, Norwalk, Epstein-Barr, rotavirus, and varicella zoster . Therefore, a viral cause should be considered preceding a case of gastroparesis of unknown etiology.

Interestingly, transient jejunal inflammation induced in animal studies let to decreased myenteric neuronal NOS (nitric oxide synthase) expression which resulted in persistent post-inflammatory dysmotility . This suggests that nitrergic dysfunction is a possible mechanism that underlies postinflammatory dysmotility and that gut motor disturbances can persist even after resolution of inflammation. In another animal study, acute induced ileitis resulted in motility disturbances that persisted 1 week after the initial insult, and both the inflamed areas and the distant non-inflamed areas were affected . Again, this study supported the possibility of nitrergic dysfunction or loss of inhibitory neurotransmission being responsible for persistent post-inflammatory dysmotility seen with post-viral gastroenteritis, sometimes labeled as IG.

Pathophysiology

The pathophysiology of IG is less well understood compared to other less common causes of gastroparesis which include autoimmune, connective tissue disorders, paraneoplastic, demyelinating diseases, postsurgical, postviral, CNS degenerative diseases, among others. Hence, IG may be an overused or fallback diagnosis when the etiology of unexplained gastrointestinal symptoms is not fully explored and all possible diagnostic entities are not explored. Several mechanisms have been proposed including impaired vagal function, loss of enteric neurons, altered myoelectric activity, visceral hypersensitivity, and degeneration of interstitial cells of Cajal (ICC) .

Advancements and refinements in histopathological techniques over the past decade have allowed us to gain a better understanding of the pathological basis of gastroparesis, specifically from The London Classification published in 2010 which classifies gastrointestinal pathology with recommendations on histological techniques (i.e. safe acquisition of tissue) . Several molecular, pathophysiological and histological similarities and differences exist between DG and IG . Differences include vagal nerve dysfunction most notably in patients with DG but not in patients with IG . Also, neuronal nitric oxide synthase (nNOS) expression is decreased to a greater degree in patients with IG compared to patients with DG, specifically 40% in IG by visual grading (compared to light microscopy) compared to only 20% in DG . Similarities between both DG and IG are involved at the cellular level. When comparing full thickness gastric body biopsies in patients with IG compared to DG, similarities include loss of ICC, decreased number of nerve fibers, and an abnormal macrophage inflammatory infiltrate . These histological changes lead to the disruption and impairment of gastric electrical activity resulting in disturbed gastrointestinal motility. Other histological similarities between both IG and DG included empty nerve endings and markedly increased connective tissue stroma on electron microscopy. Additionally, greater and more diffuse ICC damage was noted in IG compared to DG under transmission electron microscopy (TEM) .

Proposed etiologies

By definition, no primary etiology is identified in idiopathic gastroparesis (IG). An antecedent infectious viral illness has been proposed as a preceding event with resultant injury to ICC, enteric nerves, and/or enteric smooth muscle cells via immune activation with good prognosis for spontaneous recovery . Several viral pathogens associated with gastroparesis have been recognized including rotavirus, Epstein-Barr virus, cytomegalovirus, and herpes varicella-zoster . In a case series of 11 children who developed post-viral gastroparesis, 8 tested positive for rotavirus, and all children eventually completely recovered from their illness . Another case series of ten children suffering from rotavirus gastroenteritis with accompanied post-viral gastroparesis also experienced complete clinical recovery within 6–24 months of their diagnosis indicating excellent prognosis in post-viral gastroparesis . A significant delay in gastric emptying has also been observed in the adult population after a bout of viral illness. Ten healthy adult volunteers developed diarrhea and were found to have marked delay in gastric emptying after the ingestion of the parvovirus-like agents of Norwalk and Hawaii viruses . In another cohort study of 143 patients with gastroparesis, 23% reported an acute onset viral illness with preceding flu-like symptoms including fever, nausea, and vomiting . Four of these patients with preceding viral illness had CMV antibody titers and two of these patients had EBV antibody titers. In a group of seven patients who met criteria for post-viral gastroparesis with ages ranging from 3 months to 47 years, specific viruses were identified in two patients during their acute illness, one CMV and one EBV . Seven of these patients had an antecedent infectious illness preceding their GI symptoms. Symptoms resolved spontaneously in four of the seven patients within 4 weeks to 12 months suggesting that most cases of post-viral gastroparesis are self-limiting. Several other studies have shown that the associated prognosis in patients with IG with a preceding viral illness is excellent and many patients eventually have slow resolution of their symptoms . One proposed mechanism by which a viral infection may result in gastroparesis is that it may lead to degeneration and loss of ICC and the enteric nervous system by either T-cell mediated or antibody-mediated damage. In a case report of a young patient with viral-like symptoms who then developed severe IG necessitating placement of a gastrostomy feeding tube to maintain nutrition, gastric wall biopsies revealed CD4 and CD8T lymphocyte infiltration along with ganglion cells . In another case report of a 32 year old woman with severe IG who underwent a subtotal gastrectomy after many failed therapies, histological and immunohistochemical evaluation revealed marked reduction in both myenteric and intramuscular ICC, neuronal dysplasia, and hypoganglionosis . In summary, a viral etiology should be considered in patients with IG who present with a history of acute onset of symptoms. However, when these patients are usually seen by a gastroenterologist, it is months later and therefore identifying a virus will be unlikely. Moreover, it is not uncommon for patients with IG to be misdiagnosed with a functional gastrointestinal disorder or a sole diagnosis of a viral illness after a negative exhaustive workup missing the diagnosis of post-infectious GP.

Diagnostic imitators of idiopathic gastroparesis

Many causes of gastroparesis have been identified including diabetes mellitus, post-viral/post-infectious, idiopathic, connective tissue disorders, autoimmune diseases, paraneoplastic syndromes, demyelinating disorders, CNS degenerative diseases, eating disorders, among other etiologies. Hence the term IG may be overused if due diligence is not undertaken to exclude secondary causes. Table 20.1 lists diagnostic imitators of idiopathic gastroparesis. We will now further discuss each of these medical entities that might mimic IG and highlight diagnostic clues that might help differentiate IG from these other diagnoses.