interstitial cells of Cajal
Gastroparesis Cardinal Symptoms Index
Patient Assessment of Upper Gastrointestinal Disorders-Quality of Life
gastric emptying scintigraphy
wireless motility capsule
Gastroparesis ( GP ) is defined as delayed gastric emptying of fluids and/or solids in the absence of a mechanical obstruction . GP has been extensively studied in the adult population with resultant treatment guidelines developed to address adult GP . Unfortunately, the same cannot be said about pediatric GP. Knowledge regarding pediatric GP with respect to its incidence, prevalence, and natural history is sparse. Nevertheless, limited data, often from retrospective studies, suggest marked differences between adult and pediatric GP in several areas. These marked differences may serve as a caution when extrapolating adult data for use in children and underscore the need for GP-specific research in children. The purpose of this review is to summarize key data related to pediatric GP and highlight the uniqueness of pediatric GP relative to adult GP. As such, this review may serve as a platform to identify what is known about pediatric GP and potentially direct research into areas which require further study.
Due to a lack of epidemiological studies, both the incidence and prevalence of pediatric GP remain unknown. However, based on a nationwide pediatric hospital database, the rate of pediatric GP hospitalizations increased significantly from 2004 to 2013 at a rate of 130 additional hospitalizations per year . The number of unique patients hospitalized with a diagnosis of GP increased from 174 to 723 during those same years . Females and adolescents appeared to have more repeat hospitalizations . Median age was 8 years (range 2–14 years), with children younger than 5 years of age having the most hospitalizations, followed by those 11–15 years of age, 6–10 years of age, and 16–21 years of age . When divided by regions, GP hospitalizations were most common in the South. The reasons why hospitalizations increased are unknown; possibilities include an increase in the prevalence of GP and/or increase in overall severity of GP .
Interestingly, hospitalizations for dyspepsia and other “specified disorders of function of the stomach” (ICD-9 536.8) also increased during this time . Though hospitalizations provide insight into the potential increase in incidence of more severe pediatric GP and/or other functional disorders of the stomach, further studies are needed to determine the overall incidence and prevalence of pediatric GP. It may be of note that cisapride was removed from the market in July 2000, some years before the rate of hospitalizations began to increase . However, whether this is a potential contributor to the increased hospitalizations is unclear.
In adults, most cases of GP are idiopathic with diabetes mellitus, drugs, and postsurgical causes following in frequency . Other etiologies include Parkinson’s disease and connective tissue diseases . It is estimated that as many as 25%–55% of adult patients with type 1 diabetes have GP . Most idiopathic cases are believed to occur as a post-viral syndrome, since at least 23% of patients with idiopathic GP had a viral illness prior to GP presentation . Females are more likely to have idiopathic GP, while nonwhites are more likely to have diabetic GP .
One of the proposed mechanisms for GP in adults is macrophage driven loss of, or functional abnormalities in, the interstitial cells of Cajal ( ICCs ), which can lead to gastric dysmotility . Histologic specimens from patients with diabetic and idiopathic GP showed: a decrease in the number of ICCs, with remaining ICCs showing injury; an abnormal immune infiltrate containing macrophages; and a variable decrease in nerve fibers on immunohistochemistry and electron microscopy . Whether these same pathophysiologic findings are present in children with GP is unknown. Studies using full thickness biopsies will be required to address this question.
Different pediatric retrospective studies have assessed possible GP etiologies. Table 22.1 contrasts the two largest studies to date. In contrast to adults, diabetic GP was one of the least common causes for GP in both studies. Given the small relative contribution of diabetes mellitus as an etiology of GP in children, current research paradigms directed at diabetes mellitus related GP in adults may not translate well to the pediatric GP population. Furthermore, given the extremely large predominance of an idiopathic etiology (or more likely, etiologies) in pediatric GP, particular efforts in this area are sorely needed.
|Waseem et al.||Rodriguez et al.|
|Etiology||From most common to least common:||From most common to least common:|
|Predominant Symptoms||Age dependent:||Age dependent:|
|Sex||Age dependent:||Age dependent:|
|Psychiatric Comorbidities||Found in 28% of children (n=239):||Not evaluated|
|Outcomes||Factors independently associated with improved outcomes:|
Whether diabetes mellitus even is responsible for delayed gastric emptying in children is unsettled. A number of studies have evaluated gastric emptying in children with diabetes mellitus but interpretation is limited by differences in methodology for measuring emptying, the test meal administered, and ages of the children studied . Two studies were able to show slower gastric emptying time in children with diabetes mellitus type 1 when compared to controls; however, the meal ingested for these studies was not the standardized meal (Tougas) routinely used to assess emptying and methodology was varied . Multiple hypotheses have been postulated to identify the cause of emptying delay in diabetic children, including deficiency in amylin peptide hormone, uncontrolled glucose, and autonomic dysfunction; unfortunately results have been mixed . In adolescents a positive correlation has been found between the duration of diabetes type 1 and gastric emptying . Indeed, there is evidence it can take approximately 10 years for a patient with diabetes mellitus to develop GP . This delay between diabetes onset and the onset of GP may be one of the reasons why delayed gastric emptying is more prevalent in adults than in children.
Connective tissue disorders and postural tachycardia syndrome
GI symptoms are among the most common complaints in patients with connective tissue disorders including benign joint hypermobility syndrome, Ehlers-Danlos syndrome ( EDS ), Marfan Syndrome, as well as postural tachycardia syndrome ( POTS ) . Several theories have been proposed as to why patients with connective tissue disorders have increased GI complications; including abnormal gut wall connective tissue that allows increased visceral compliance, which promotes hollow organ distention and interferes with gut mechanoreceptors .
The available literature on GP and connective tissue disorders/POTS is scarce. Connective tissue disorders such as EDS have been associated with multiple GI complaints. A large retrospective study found abdominal pain, nausea, and constipation were the most common complaints . In the same study, 378 patients (n=687 with a diagnosis of EDS) complained of GI symptoms, of which 76 had a gastric emptying scan and 9 had delayed gastric emptying . Interestingly, 8 had accelerated gastric emptying, alluding to a possible different mechanism accounting for this, such as decreased gastric accommodation . To date, no studies are available in the pediatric population assessing EDS and GP. Further prospective studies are needed to understand the pathophysiology of GP and what mechanisms may be responsible for causing dysmotility in children.
Multiple retrospective studies have assessed the presence of dysmotility in patients with POTS, with most showing more rapid rather than delayed emptying . Unfortunately these studies are small, and thus larger cohorts are needed in order to confirm these results.
POTS often is associated with hypermobility joint syndrome and in fact, GP has been reported in around 20% of patients with POTS, with an even greater increase in prevalence in GI symptoms in patients with both POTS and autonomic neuropathy . However, it seems that rapid emptying is more common than delayed emptying in this subset of the population . A recent study by Velasco-Benitez et al. found that children with functional GI disorders have a greater prevalence of symptoms of orthostatic intolerance; however, children with joint hypermobility did not have a greater prevalence of orthostatic intolerance or POTS when compared to children without joint hypermobility .
As noted in Table 22.1 , there are a number of conditions associated with pediatric GP. Because of significant overlap in symptoms between GP and functional dyspepsia, these disorders can be difficult to differentiate based on clinical grounds alone . Both disorders can manifest as postprandial bloating or fullness and/or pain, nausea, emesis, anorexia, and/or early satiation . Presumably, measurement of gastric emptying should differentiate the two, with the strong caveat related to the use of gastric emptying scintigraphy in children (see below).
Gastroesophageal reflux disease and peptic ulcer disease should be considered and usually can be excluded based on endoscopy, and in the case of reflux, by esophageal pH probe and/or impedance monitoring . Cyclic vomiting can be distinguished on clinical grounds given its characteristic of symptom-free periods . Rumination syndrome should be suspected based on clinical history and physical examination .
Of particular relevance to pediatrics are mitochondrial disorders and chronic intestinal pseudoobstruction as these more commonly present in childhood. Mitochondrial disorders such as mitochondrial encephalopathy, lactic acidosis, and stroke like episodes (MELAS) and mitochondrial neurogastrointestinal encephalopathy (MNGIE) can present with vomiting with the latter also associated with delayed gastric emptying .
Most cases of pseudoobstruction in children are idiopathic but they also can occur secondary to systemic diseases such as muscular dystrophy, hypothyroidism, hypoparathyroidism, and Chagas disease . Symptoms may overlap with GP and result in delay in diagnosis. The term pseudoobstruction can subsume a number of etiologies such as degenerative neuropathies and immune mediated disorders .
Symptoms and symptom assessment
In children, age significantly influences GP symptom expression ( Table 22.1 ) , a finding that is seen in a number of other pediatric disorders . Vomiting is most common in infants and young children whereas abdominal pain predominates in older children and adolescents ( Table 22.1 ) . The most common symptoms that appear to prompt medical evaluation are vomiting (42%), abdominal pain (35%), and nausea (29%) . It appears symptom duration was longer in patients with mitochondrial disease and shorter in patients with post-viral GP when compared with subjects not presenting with these conditions . Presenting symptoms between girls and boys appear to be similar . These data highlight that symptom assessment in children should be tailored to the age of the child.
Objective symptom assessment
In adults, the Gastroparesis Cardinal Symptom Index ( GCSI ) is a validated, albeit retrospective, patient-reported instrument often used to assess GP severity . It consists of three subscales: nausea/vomiting, postprandial fullness/early satiety, and bloating. Pain, which is reported by a large number of adults and older children is not included in the GSCI . GSCI scores unfortunately do not appear to correlate with gastric emptying rate .
In children, a GP-specific patient-reported symptom measure has not been developed. In an attempt to address this deficiency, one pediatric study trialed a modified version of the adult GCSI, with a reduction from the 6-point Likert-type scale to a 5-point Likert-type scale and the addition of abdominal pain (although not included in the total GCSI calculation) . Similar to adult studies, no association was found between the total modified GCSI score and degree of emptying delay . When evaluating how well the children understood the symptoms, no symptom consistently achieved “complete” understanding . Given this, and that the 5-point scale has not been validated in children with GP, development of such a scale is greatly needed. Such a scale would provide objective and reliable means to assess the presence and severity of current GP symptoms and could be used to assess outcomes of interventions used to treat and/or manage pediatric GP. Whether a prospective GP-specific patient-reported symptom measure will correlate with the degree of delayed gastric emptying needs to be assessed.
A current goal of the pediatric component of the Gastroparesis Consortium is to develop both a GP specific patient reported symptom measure as well as a quality of life measure for children. A potential starting point for the development is the PedsQL™ ( www.pedsql.org ). This measure has been validated for use in a number of pediatric diseases ranging from gastrointestinal disorders to diabetes . A specific gastrointestinal symptoms module within the PedsQL™ has been validated as well . However, the gastrointestinal symptom module does not include gastroparesis.
In adults, GP is more predominant in females . In contrast, the female: male ratio in pediatric GP varies by age ( Table 22.1 ) . It is unknown why these age-related gender differences exist in the pediatric population, but they may suggest different pathophysiologic factors playing a role in the development of GP, particularly prior to adolescence. A large retrospective pediatric study (n=239) had an almost equal distribution between female and male patients (51.5% and 48.5%, respectively), as well as similar etiologies for each sex. However, it seemed that as age increased, female predominance increased suggesting a potential hormonal involvement as is postulated in adults ( Table 22.1 ) . Whether gender influences etiology and/or symptom severity in children with GP, to our knowledge, remains to be clarified.
Rodriguez et al. found that the most common comorbidities in children with GP (n=320) were viral (18%), gastroesophageal reflux (14%), mitochondrial disease (8%), diabetes mellitus (2%), and hypothyroidism (0.4%) . When divided by age group, the most common associated disease in the infant group (n=36) was cow’s-milk protein allergy (33%), followed by viral (17%) . Children aged 1–12 years of age (n=109) reported viral (19%), gastroesophageal reflux (9.6%), and mitochondrial disease (9%) as the most common co-morbidity. In contrast, children aged >12 years of age (n=85) more commonly presented with post-viral symptoms (18%), irritable bowel syndrome (11%), and depression (8%) .
Another study by Waseem et al. found that 38.5% (n=92) of pediatric patients with GP had comorbidities as well . These included, in no particular order, seizure disorder, cerebral palsy, developmental delay, and prematurity .
Psychiatric comorbidities are less common in children with GP compared with adults. In one study (n=239) they were found in 28% of pediatric patients with GP ( Table 22.1 ) . No differences were observed with regard to frequency and prevalence between sexes. Whether the decreased frequency of psychiatric comorbidities in children with GP leads to better outcomes is unknown. Similarly, how psychiatric comorbidities in children may or may not evolve over time with continued GP symptoms is unclear.
Although the data are limited, one study in children found that only 2% of patients with GP were taking narcotics (n=239) . Thus, pediatric patients (vs. adults) with GP appear to generally have significantly less exposure to opioids.
In adults with GP several factors (e.g., male sex, age ≥50 years, post-viral etiology) have been found to be independently associated with reduced symptoms at 48 weeks . Additional factors include no to mild abdominal pain, mild gastroesophageal reflux disease severity, and no to mild depression . Unfortunately, these characteristics encompass less than a third of patients with GP, implying the majority of adults with GP may be at risk for significant disease burden .
Clinical outcomes have been evaluated retrospectively in children with GP and suggest a higher likelihood for improvement when compared to adults. Rodriguez et al. reported 52% of 204 children with GP (outcome data available on 204/230 subjects, mean age of 9.1 years) had resolution of their symptoms . Of those who had resolution, 42% reported resolution within 6 months, 84% within 12 months, and 100% by 36 months . Factors associated with symptom resolution included: younger age (infants>children>adolescents); male gender, shorter duration of symptoms at time of presentation, absence of mitochondrial dysfunction, post-viral etiology, and a favorable response to prokinetic drugs ( Table 22.1 ) . Out of 41 patients diagnosed with post-viral GP, 73% responded well to prokinetic drugs and 63% had resolution of their symptoms within 2 years. Fifty-five percent of patients (n=90) reported a positive response to prokinetic drugs regardless of age, with 58% of those reporting symptom resolution . When divided by age group in terms of prokinetic response, 40% of infants (n=32) reported a response, 64% of children aged 1–12 years of age (n=97) reported a response, and 53% of children >12 years of age (n=28) reported a response. Given that the available data on outcomes is retrospective, there are certain limitations. For example, patients with presumed post-viral GP tend to be diagnosed based on history, which often includes a presumed but not confirmed acute viral illness prior to the diagnosis of GP . Large, prospective studies are needed to assess the specific pathophysiologic causes of GP in children, and long-term outcomes of these patients.
In another study of children with GP (mean age 7.9±5.9 years) by Waseem et al., 60% of 239 patients reported significant improvement in nausea, vomiting, abdominal pain, early satiety, bloating, and weight loss at 2-year follow up regardless of sex, age, or degree of emptying delay . Interestingly, weight loss improved even though 74% of the patients received dietary modification but not supplemental nutritional support . With the exception of abdominal pain (reported more often by girls), no other significant difference in symptom outcomes was found comparing girls to boys . Patients in the 11–16 year age range had the most improvement in all symptoms, whereas the >17-year-old group reported the least improvement . It should be noted, however, that a large proportion (~40–50%) of children with GP did not have symptom resolution within 2–3 years . A subgroup of patients had repeat GES (n=28), which showed normalization of gastric emptying in 3 girls and 4 boys (25%). Whether differences in etiology, gender, or other factors contribute to the more favorable outcomes in children compared with adults with GP remains to be determined.
Waseem et al assessed the potential complications resulting from GP . GERD was the most common (67%), followed by esophagitis (40%), gastritis (36%), dehydration (9%), and anemia (6%). Interestingly, malnutrition was found in only 3% .
The average cost of hospital care for adults with GP has increased from $13,350 per patient in 1997 to $34,585 in 2013 . The number of hospitalizations also has increased, with one study finding an 18-fold rise from 1993 to 2009 and another finding a 4-fold increase from 1997 to 2013 (from 3978 to 16,460) . Despite this, no increase in GP incidence has been identified. Admissions for GP in adults tend to be longer and are associated with higher charges when compared to other GI conditions .
A recent study found that the national cost of hospital care for children with GP is rising dramatically as well, with an increase at a rate of about $3.4 million per year from 2004 to 2013, representing a 5.8-fold increase in cost . The increase was related to the higher number of pediatric GP admissions and not increases in cost per hospitalization . This rise in hospitalizations could be due to either an increase in the diagnosis and awareness of GP, increase in overall severity of GP, or both. Whether the increase in hospitalizations in children with GP is related to an increase in incidence/prevalence remains unknown.
Gastric emptying scintigraphy
The gold standard for diagnosing GP is gastric emptying scintigraphy ( GES ) using a radiolabeled standardized solid meal mixed with Tc-99m sulfur colloid. This should be consumed within a short period of time, usually no more than ten minutes . The percentage of radioactivity, which correlates with the amount of food remaining in the stomach is calculated. Methodology varies among different centers, but the recommended technique is the 4-hour imaging protocol with scans taken at 0, 1, 2, and 4 hours after ingestion of the meal . Several adult and pediatric studies have shown patients with normal GES studies at two hours ultimately resulted in a diagnosis of GP when extended to four hours . In fact, one pediatric study showed that out of the 95 participants with a normal study at 2 hours, 14 (15%) had an abnormal study at 4 hours . GES is considered delayed if retention is greater than 60% at 2 hours postprandially and/or greater than 10% at 4 hours .
Because a GES exposes children to radiation, normal values for healthy children have not been established, and thus adult values have been extrapolated for use in children. This extrapolation of values is a limitation of previous and current studies of pediatric GP. Previous studies from our group suggest this may be most problematic in infants and children aged 7–10 years of age, as these groups had more difficulty completing the GES meal . Compared with older children, they less frequently tolerated the standard meal (i.e., vomited) or could consume the entire meal . Similar findings were found by Edwards et al., as 40% of the children in their study were unable to complete the entire meal (there was a significant relationship between older age and completing the meal, P =0.011) . In addition, we found that children with delayed gastric emptying were significantly younger and smaller than those without. This may be due to younger and smaller children truly having slower emptying compared to older children and/or that the meal size relative to stomach size is greater than in older children and adults resulting in greater time needed to empty the meal relative to older children .
Stable isotope breath test
Breath tests using the stable, nonradioactive isotope 13 C given in a substrate such as octanoic acid or the plant Spirulina platensis have been used successfully in adults and children to assess gastric emptying . Their cost is similar to scintigraphy . Advantages of using breath tests include the capability of doing these tests at the bedside and avoidance of radiation which allows studies in children and pregnant or breastfeeding women . The 13 C- Spirulina platensis breath test is approved by the FDA for measuring gastric emptying in adults. When digested, S. platensis is rapidly absorbed in the duodenum and afterwards metabolized in the liver, giving rise to CO2 enriched in 13 C . The 13 CO2 abundance over time can be measured in breath and reflects the gastric emptying rate of the meal.
The 13 C-acetate breath test and GES carried out simultaneously have been used to measure gastric emptying in children (n=29). The 13 C-acetate breath test had a sensitivity of 100% and specificity of 85% when using a cut-off of T ½ breath >90 min (P<0.00001) . The 13 C-octanoic breath test also has been carried out simultaneously with GES in children (n=25), showing a good correlation between T ½ 13 C and T ½ GES (r=0.92) . Given its safety, the 13 C breath test provides an opportunity to define normal values of gastric emptying in healthy children and can be carried out simultaneously with GES in children suspected of GP. Validation of the 13 C breath test versus GES would then allow extrapolation of normal (and abnormal values) to children undergoing GES.
Wireless motility capsule
The wireless motility capsule ( WMC ) is a non-digestible, orally ingested capsule that measures intraluminal pressure, temperature, pH, and allows calculation of transit times in the GI tract . It has been approved by the FDA for the evaluation of gastric emptying and colonic transit in adults . One advantage of the WMC in GP is that it can measure gastric emptying time without exposing patients to radiation. Data on the WMC in pediatrics is very limited. One single-center pediatric study (n=21) found the WMC had a sensitivity of 100% and specificity of 50% for detecting GP when compared to the GES at two hours (the protocol at the institution was a 2-hour rather than a 4-hour GES) . The authors argue that allowing an additional 2 hours might have identified more cases of GP.
Esophagogastroduodenoscopy ( EGD ) is commonly used as part of the GP evaluation as it can help rule out other causes of delayed emptying (e.g. gastric outlet obstruction due to pyloric stenosis, active ulcer disease) and allows visualization of fasting gastric contents, including bezoars, excess bilious fluid, and/or retained food . It is recommended that adults who present with postprandial upper abdominal symptoms undergo an EGD prior to a GES . Depending on EGD findings (e.g., retained food contents, no significant pathology in the presence of upper GI symptoms), motility or other testing can be pursued afterward as indicated.
A recent pediatric retrospective case-control study by Altepeter el al. found similar clinically significant endoscopic findings amongst controls (35%, n=44) and children with GP (43%, n=30), with gastritis and esophagitis being the most common histological findings in both groups . Similar findings were noted by Thakkar et al. at our center in children undergoing EGD evaluation for abdominal pain . Endoscopy provided a diagnostic yield of 38% with reflux esophagitis (23%) being the most common diagnosis, followed by H. pylori (5%), peptic ulcers (3%), and erosive esophagitis (2%). These two studies argue for endoscopy in these patients given the relatively high prevalence of positive histological findings. In contrast, Wong et al. found that children with GP had fewer abnormal histological findings (27%, 19/70) when compared to children with normal gastric emptying (42%, 23/55), with gastritis and esophagitis also being the most common findings in the latter group . Whether the abnormal findings in children with GP contribute to the delay in gastric emptying is unclear. However, given the relatively high likelihood of positive histologic findings in children presenting with GP, endoscopic evaluation may be warranted. That said, preliminary studies from our group show that only 54% (n=103) of children with diagnosed GP underwent EGD and/or upper GI evaluation; this may be related to the fact that GES may be ordered by non-gastroenterologists.
EGG allows non-invasive measurement of gastric myoelectrical activity by placing electrodes on the abdominal surface and has been applied to evaluate a number of gastrointestinal disorders including GP . However, its role remains to be defined. Studies have assessed whether results of the EGG vary by age. In a large cohort (n=114) of healthy children (6–12 years of age) studied by Riezzo et al., age, gender, and obesity did not appear to affect the EGG parameters of mean dominant frequency ( DF ) or the power ratio ( PR ) . In contrast, the dominant frequency instability coefficient ( FIC ) was lower in girls <9 years of age postprandially than in the older girls . In a later study, Levy and colleagues studied children 6–18 years (n=55) and reported similar findings, except they could not reproduce the age/gender related response in FIC . Finally, Friesen et al. compared 28 healthy children (ages 8–17 years) to 15 previously evaluated healthy adults . They confirmed the previous studies and suggested that American Motility Society consensus adult criteria for a normal EGG are appropriate to apply to children and adolescents .
Despite the establishment of normal EGG values in children and adolescents, few studies have evaluated EGG findings in the context of pediatric GP. Barbar and colleagues carried out EGG recordings in nine children with symptoms of vomiting and/or abdominal pain (ages 5–16 years) for 30 minutes while fasting and then while undergoing GES; four children had normal gastric emptying (T1/2 on GES ≤90 minutes) and five had delayed gastric emptying . No differences were noted between groups in EGG parameters but the small number of patients and the short duration of the GES make interpretation difficult .
As noted previously, type I diabetes mellitus is associated with GP. A few studies have examined EGG recordings in children with diabetes. Cucchiara et al. conducted EGG recordings and gastric emptying measurements (by ultrasonography of the gastric antrum) in 40 children with type I diabetes (26 of whom had delayed emptying) and 15 healthy controls . There was a higher percent of tachygastria and a lower fed-to-fasting power ratio in those with delayed emptying, but it is not clear if these were pre- or post-prandial measures (or both) . Delayed emptying was associated with higher levels of HbA1c . Franzese and colleagues (from the same research group as Cucchiara) compared the effects of domperidone to cisapride in a randomized trial on symptoms, gastric emptying (by ultrasound), and EGG patterns in an 8-week trial, although it is not clear if the study was blinded . Tachygastria was present in approximately 24% of participants at baseline, and the baseline fed-to-fasting power ratio was significantly less than in historical controls . Domperidone was superior to cisapride in terms of improving symptoms, gastric emptying, and EGG parameters . In a study from Poland, 172 children with type I diabetes underwent EGG measurements and were compared with 35 healthy controls . The mean percent of bradygastria preprandially was greater, and post-prandially smaller, in those with diabetes . There were no differences in EGG results between those with <5 vs >5 years of diabetes . Confirming the Polish study, Posfay-Barbe et al. studied 49 children with type I diabetes and 17 healthy controls; preprandial bradygastria was more frequent in those with diabetes . When tested one year later, 83.8% of those previously normal remained normal; 13.5% of those previously abnormal were normal; and 2.7% previously normal were now abnormal . With advances in EGG technology (e.g., high resolution EGG), further insights into the role of myoelectric dysfunction in pediatric (and adult) GP is likely .
Dietary and lifestyle modifications
The first line of treatment for GP in both adults and children is dietary along with lifestyle modifications regardless of disease severity. Since these patients can experience early satiety, they are encouraged to eat small, frequent meals and avoid fatty foods as well as high-fiber content meals, as these delay gastric emptying . If patients cannot tolerate solid meals, they can consume their required calories in liquid form, given liquid emptying is often preserved despite the presence of solid-phase delayed emptying .
Even though it is considered first-line therapy, one study showed diet modification was used as primary therapy in only 4 patients (n=230), with 3 responders and 1 non-responder . However, when it was used as an adjunctive treatment, there seemed to be more responders (11/230) .
Proton pump inhibitors (PPIs)
PPIs are commonly used as first line medications for patients who presents with symptoms suggestive of GP. One large pediatric retrospective study showed that PPIs were used in 79% (n=230) of patients and were the first-line drugs in 77% . Mean duration was 9.1±10.2 months . Follow up was available for 96% of patients and showed only 3% had resolution of symptoms .
Prokinetics promote motility in the GI tract and can improve gastric emptying . Macrolide antibiotics at reduced antimicrobial dosages, such as erythromycin, increase motility by acting on the motilin receptor. This is believed to regulate phase III of the migrating motor complex . Unfortunately, patients usually develop rapid tolerance and tachyphylaxis to erythromycin, requiring that the drug be stopped for a period of time and then restarted which can make it difficult to use long term . In adults, larger doses have been shown to elicit non-propulsive spastic contractions or prolonged periods of antral hypermotility, which are hypothesized to be the cause of tachyphylaxis . Thus, a dose of no more than 250 mg TID commonly is used in adults to minimize this side effect . Clinical response usually decreases after four weeks; however, some patients may experience some benefit for a longer period .
Adult studies have demonstrated the effectiveness of erythromycin over other motility agents . A systematic analysis of 36 studies with 514 patients comparing prokinetic agents showed erythromycin was associated with greater symptom improvement when compared to other prokinetic agents . However, a more recent systematic review noted that the available data is very limited given that the existing studies consist of small sample sizes, uncontrolled designs, short duration, and inadequate symptom assessment, calling for more well-designed trials to assess symptom relief in these patients . Regardless, the limited data suggests erythromycin can be a potent prokinetic agent in adults.
In children the dose for erythromycin is weight based and individualized to the patient regardless of tachyphylaxis, and usually ranges between 3 mg/kg QID to 10 mg/kg QID . However, erythromycin might not have the same beneficial effect in children when compared to adults. Rodriguez et al. showed that 51% of patients (19/37) had a significant response to erythromycin, but only 5% of patients reported complete resolution of symptoms, the lowest when compared to other prokinetics (metoclopramide, domperidone, and tegaserod) . However, a randomized, double blind trial comparing erythromycin and metoclopramide showed that erythromycin might be as effective as metoclopramide, thus making it preferable due to the absence of extrapyramidal side effects . Erythromycin has been trialed in low birth weight and premature infants to assess improvements in feeding tolerance, but the results are conflicting ( Table 22.2 ) . Consequently, a recent systematic review recommended using erythromycin only for high risk preterm neonates with persistent feeding intolerance . Of note, it is thought that the migrating motor complex is not observed until 32 weeks of gestation, thus erythromycin might not be of benefit for preterm infants <32 weeks of age . Table 22.2 lists the available pediatric studies.