Nausea Predicts Delayed Gastric Emptying in Children

Nausea Predicts Delayed Gastric Emptying in ChildrenNausea Predicts Delayed Gastric Emptying in Children
Hilary Jericho, MD1, Papa Adams, BA2, Gang Zhang, PhD3, Karen Rychlik, MS3, and Miguel Saps, MD2
Objectives To assess whether the gastroparesis cardinal symptom index (GCSI), or any individual symptom, is associated with delayed gastric emptying in children, and to assess understanding of symptoms associated with delayed gastric emptying. Study design Fifty children (36 F), 5-18 years of age, undergoing gastric emptying scintigraphy (GES) at Lurie Children’s Hospital in Chicago, Illinois, completed Likert-type GCSI and symptom comprehension questionnaires. Correlation of GES results (normal or abnormal) with questionnaire results using the Wilcoxon rank sum test. Results Seventy percent of subjects had a normal GES. Children reported understanding most terms of GCSI (average score 2.59, range 0-3). The GCSI was not associated with delayed gastric emptying. Nausea was asso- ciated with delayed gastric emptying only (numerical P = .04, word P = .02). Results were not altered when poorly understood terms were excluded. Conclusions The GCSI is not associated with delayed gastric emptying in children. Lack of association does not seem to be related to lack of understanding. Nausea alone was the only symptom that showed an association with delayed gastric emptying on GES. (J Pediatr 2014;164:89-92).
G astroparesis is defined by a delay in gastric emptying in the absence ofmechanical obstruction. The clinical presentation of gastroparesis is variable and symptoms can include nausea, vomiting, early satiety, fullness, bloating, and abdominal pain, which are often also present in other common disorders. In pediatrics, gastroparesis is frequently idiopathic or
follows an infection or surgery. Postinfectious gastroparesis tends to improve spontaneously over the course of several months,1
whereas idiopathic disease can be more severe and intractable.2 Treatment is usually symptomatic and can include frequent, small, low-fat, low fiber meals, prokinetics, transpyloric feeds, botulinum toxin injections to the pylorus, or gastric electrical stimulation.
Gastric emptying scintigraphy (GES) is considered the gold standard for diagnosing gastroparesis,3,4 but exposes patients to radiation, is expensive, time-consuming, and is not readily available at the doctor’s office. The identification of one or more symptoms associated with gastroparesis could optimize the use of this test. To date, no studies have investigated the ability of symptoms commonly associated with gastroparesis to predict its diagnosis in children.
The gastroparesis cardinal symptom index (GCSI) is a validated symptom severity score designed to assess the impact of gastroparesis in adults.1,2,5-8 No such patient-reported symptom severity scale currently exists for use in the pediatric population. Determining whether the GCSI can predict gastroparesis in children would provide clinicians with a readily available, inexpen- sive, and noninvasive way to select patients for GES and monitor for symptom improvement.
This cross-sectional survey study evaluated modified versions of the adult GCSI in children with gastrointestinal-related symptoms suggestive of gastroparesis. The main objectives were to identify whether a pediatric version of the GCSI, or any specific gastrointestinal-related symptoms, could be associated with gastroparesis in children, as defined by the results of their GES.
BMI Body mass index
GCSI Gastroparesis cardinal sympt
GES Gastric emptying scintigraphy
Methods
FromMarch 2011 through July 2012, all children ages 5-18 years, undergoing a GES for presumptive gastroparesis at Children’s Memorial Hospital of Chicago (CMH) (currently Ann and Robert H. Lurie Children’s Hospital of Chicago) were invited to participate in a prospective pilot study investigating the relationship between symptoms consistent with delayed gastric emptying and results of their GES. Patients were included if they could read and understand English and provide consent/ assent. Pregnant, non-communicative, and non-ambulatory patients, and those with a psychiatric disorder or cognitive impair- ment were excluded.
Potential participants were identified from the nuclear medicine clinic schedule in the CMH electronic database (EPIC). Families were contacted by
From the 1Department of Pediatrics, Comer Children’s Hospital, Chicago, IL; 2Department of Pediatrics; and 3Department of Biostatistics Research Core; Ann and Robert H. Lurie Children’s Hospital of Chicago, Chicago IL
The authors declare no conflicts of interest.
0022-3476/$ - see front matter. Copyright ª 2014 Mosby Inc.
All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2013.09.019
Gastric emptying
7 (19)
1 X X 2 X 3 X X 4 X 5 X X 6 X X 7 X X
Male, n (% total sex)
6 (43)
1 X X 2 X 3 X X 4 X X 5 X 6 X X
THE JOURNAL OF PEDIATRICS www.jpeds.com Vol. 164, No. 1
phone to assess their willingness to participate. Those who expressed interest were met upon arrival for their GES and consent/assent was obtained. Participants completed a series of surveys that included a numerically-based GCSI, a word- based GCSI, and a short questionnaire on symptom termi- nology comprehension. The study protocol was approved by the Institutional Review Board of CMH.
GCSI Children’s symptoms were scored using amodified version of the adult GCSI, a validated symptom severity scale (0-5) that utilizes reports of patients’ symptoms over the preceding 2 weeks. The GCSI consists of 3 symptom clusters: post- prandial fullness/early satiety cluster (4 subscale items): stomach fullness, inability to finish a normal-sized meal, feeling excessively full after meals, loss of appetite; nausea/ vomiting cluster (3 subscale items): nausea, vomiting, retch- ing; and bloating cluster (2 subscale items): bloating, stom- ach or belly visibly larger.6,7 The GCSI score is calculated as the average of all the subscale scores (each subscale score is calculated by averaging the items within the subscale). We used a modified version of the GCSI thought to be more child-friendly. The modified version was identical to the adult GCSI with the exception of a reduction from the 6-point Likert-type scale used in adults (0 = none, 1 = very mild, 2 = mild, 3 = moderate, 4 = severe, and 5 = very severe) to a 5-point Likert-type response scale (0 = none, 1 = mild, 2 = moderate, 3 = severe, and 4 = very severe) with the con- cerns that children would have a difficult time differentiating between the categories “very mild” and “mild.” Scores ranged from 0-4, with higher scores reflecting perceptions of worse symptom severity. Although the GCSI does not include the symptom of abdominal pain, we assessed this individually as it is a frequent complaint in children. However, we did not include it in the GCSI calculation. Children completed 2 separate GCSI formats, numerical and word, to identify if they could better classify their symptoms with one over the other.
Symptom Comprehension Owing to the adult-focused design of the GCSI, we also found it important to identify whether children were able to under- stand the wording in the questionnaire. We asked children to rank their understanding of the meaning of each individual symptom to assess this further with a score of 0 indicating “no” understanding, 1 “a little” understanding, 2 “mostly” understanding, and 3 “complete” understanding of the term. To assess for deviation of results secondary to poorly understood terms, data was rerun only using results from children who claimed to have a clear understanding of the terms (comprehension scores of 2 or 3).
Gastric Emptying Protocol All patients were required to be nil per os for 6 hours prior to the GES in accordance to the multi-institutional, standardized protocol established by Tougas et al in 2000.8 Following the completionof the survey, childrenwere providedwith the stan-
90
dardized GESmeal consisting of 2 large eggs (Eggbeaters; Con- Agra Foods, Omaha, Nebraska) prepared with Tc99m Sulfur Colloid 0.05 mCi/kg (minimum 0.5 mCi, maximum 1 mCi), 2 slices bread and jam, and water (children with egg allergy were offered oatmeal). The patient was required to eat the meal within 10 minutes. After the meal was given, the patient was placed upright and images of the stomach and bowel were obtained at 0, 30, 60, 90, 120, and 240 minutes using a low-energy high-resolution collimator. Patient results were classified as normal or abnormal based on results at 1, 2, and 4 hours after the meal. Gastric emptying scans were classified as “delayed” when there was greater than 90% retention at 1 hour, 60% retention at 2 hours, and/or greater than 10% reten- tion at 4 hours, according to the 2007 Consensus Recommen- dations forGES fromThe Society ofNuclearMedicine andThe American Neurogastroenterology and Motility Society.8,9
Statistical Analyses Statistical analysis was conducted using SPSS 20 (IBM Corp, Armonk, New York) and SAS 9.3 (SAS Institute Inc, Cary, North Carolina). Linear correlation between word and nu- merical charts was assessed using Pearson correlation statis- tics. The statistical significance of differences in clinical features within each of the GCSI subgroups and total GCSI scores from both charts were analyzed using the Wilcoxon Rank-Sum test. All P values were 2-sided if not otherwise stated; a level <.05 is considered as statistically significant.
Results
A total of 50 children (36 females, 72%) completed the sur- veys and underwent GES. Patients ranged in age from 5-18 years. The mean patient age was 13.2 years with a SD of 3.6 years. The mean body mass index (BMI) was 19.6 with a SD of 4.7 and was not associated with delayed gastric emptying (P = .99).
Jericho et al
Table III. Correlation between symptoms and delayed gastric emptying scans (n = 46)*
Symptom Number chart P value
Word chart P value
Nausea, Mean (SD) 2.20 (1.2) .04 1.87 (1.2) .02 Vomiting, Mean (SD) 0.96 (1.3) .65 0.85 (1.3) .24 Retching, Mean (SD) 0.80 (1.3) .23 0.72 (1.2) .09 Stomach fullness, Mean (SD) 2.22 (1.4) .50 1.91 (1.3) .67 Unable to finish meal, Mean (SD) 2.35 (1.3) .75 2.00 (1.3) .62 Loss of appetite, Mean (SD) 2.02 (1.3) .34 1.76 (1.3) .20 Feeling excessively full after
meals, Mean (SD) 2.13 (1.5) .10 1.98 (1.3) .37
Bloating, Mean (SD) 1.72 (1.6) .24 1.48 (1.5) .17 Stomach or belly visibly larger,
Mean (SD) 1.50 (1.6) .39 1.22 (1.6) .56
Pain, Mean (SD) 2.87 (1.1) .36 2.61 (1.2) .44 GCSI
With retching, mean (SD) 1.72 (0.8) .86 1.46 (0.7) .87 Without retching, mean (SD) 1.80 (0.8) .76 1.57 (0.8) .82
*Severity of symptom: 0 = none, 1 = mild, 2 = moderate, 3 = severe, 4 = very severe.
Table IV. Correlation between severity of mostly or completely understood symptoms and delayed gastric
January 2014 ORIGINAL ARTICLES
Thirty-seven children (74%) were found to have normal gastric emptying (average BMI 19.45), of which 2 children had rapid gastric emptying (<30% gastric retention at 1 hour). Thirteen children (26%) had delayed emptying (average BMI 19.85) as per the consensus report guidelines. Four of the 37 children with normal or rapid scans failed to complete their meals as per protocol and were excluded from the statistical analysis. Of the 13 children with delayed gastric emptying, 2 had delayed emptying at 1 hour only, 4 had delayed emptying at both 1 and 2 hours, 4 had delayed emptying at both 2 and 4 hours, and 3 had delayed emptying at 4 hours only (Table I).
All 46 children completed the survey on symptom termi- nology comprehension. Children overall appeared to under- stand most terms used in the survey with the exception of the term retching (average score of 1.70, range 0-3). Bloating received the second lowest comprehension score of 2.32. All other terms received scores averaging from 2.57-2.89 (Table II), with nausea at 2.70. There was a strong correlation between the symptom severity scores from the numerical and word charts (r = 0.84-0.96, P = .01 for a 2-tailed test). Owing to the poor understanding of the symptom retching, total GCSI scores were calculated with and without the inclusion of this term to look for differences. GCSI scores, with and without the inclusion of retching, showed no association with delayed gastric emptying (Table III).
Nausea was the only individual symptom statistically associated with delayed gastric emptying (P = .04, numer- ical scale and P = .02, word scale). Clinically, 54% of sub- jects with delay in gastric emptying had the highest severity scores for nausea (3-4), 23% had moderate (2) and low scores (0-1), and 43% of subjects with normal scans had the lowest scores for nausea (33% had moderate and 21% had the highest scores). All other subscale symptoms were not found to be associated with delayed gastric emptying (Table III).
When only “understood” symptoms were analyzed (comprehension scores of 2 or 3), nausea continued to be the only symptom with a statistically significant association with delayed gastric emptying (P = .06, numerical scale and P= .03, word scale). Though bloating gained statistical signif- icance (P = .05, numerical scale and P = .03, word scale), this
Table II. Terminology comprehension*
Symptom Patients, n = 46
Nausea, mean (SD) 2.70 (0.7) Vomiting, mean (SD) 2.77 (0.8) Retching, mean (SD) 1.70 (1.2) Stomach fullness, mean (SD) 2.57 (0.8) Unable to finish meal, mean (SD) 2.72 (0.7) Loss of appetite, mean (SD) 2.74 (0.7) Feeling excessively full after meals, mean (SD) 2.70 (0.6) Bloating, mean (SD) 2.32 (1.0) Stomach or belly visibly larger, mean (SD) 2.68 (0.7) Pain, mean (SD) 2.89 (0.5)
*0 = none, 1 = some, 2 = most, 3 = complete comprehension.
Nausea Predicts Delayed Gastric Emptying in Children
association was for having a normal gastric emptying scan. All other subscale symptoms remained the same (Table IV).
Discussion
Our study investigated whether the GCSI, or any individual symptom used to calculate the GCSI, showed an association with delayed gastric emptying in children. We found that the total GCSI score was not associated with delay in gastric emptying, although there was an association between delay in gastric emptying and the individual symptom of nausea. In order to minimize the risk of unreliable results secondary to poor terminology comprehension, we also asked subjects to rank their understanding of each GCSI term (0-3: no to complete understanding). Only those with scores of 2 or 3 (most to complete understanding) were included in a repeat analysis. The GCSI continued to show no statistically signif- icant association with delayed gastric emptying after this exclusion and nausea continued to be the only symptom with a statistically significant association with delayed gastric
emptying (n = 46)*
Word chart P value
Nausea, mean (SD) 2.25 (1.2) .06 1.91 (1.2) .03 Vomiting, mean (SD) 1.02 (1.3) .82 0.91 (1.3) .34 Retching, mean (SD) 1.08 (1.3) .25 0.92 (1.2) .16 Stomach fullness, mean (SD) 2.43 (1.3) .46 2.10 (1.3) .80 Unable to finish meal, mean (SD) 2.40 (1.3) .90 2.10 (1.3) .84 Loss of appetite, mean (SD) 2.05 (1.3) .35 1.79 (1.3) .24 Feeling excessively full after
meals, mean (SD) 2.24 (1.5) .12 2.05 (1.4) .68
Bloating, mean (SD) 2.06 (1.6) .05 1.82 (1.6) .03 Stomach or belly visibly larger,
mean (SD) 1.46 (1.6) .32 1.27 (1.7) .36
Pain, mean (SD) 2.91 (1.1) .44 2.64 (1.2) .53
*Severity of symptoms: 0 = none, 1 = mild, 2 = moderate, 3 = severe, 4 = very severe.
91
THE JOURNAL OF PEDIATRICS www.jpeds.com Vol. 164, No. 1
emptying. Though bloating did show an increased associa- tion with normal gastric emptying after this exclusion, we feel that this may have been secondary to a low total sample size (n = 33) and poor comprehension (comprehension score of 2.32) and requires repeat analysis for verification.
The low ratio of abnormal radionuclide emptying scans found in our study questions the utility of this test for the majority of children who present with clinical findings similar to the children in our sample. Scintigraphy is an expensive test associated with radiation exposure. A cost analysis study found that the charges of evaluating a child for gastroparesis exceeded $9000, with $2300 of those charges resulting from the costs of the scintigraphic testing.10 Our study provides relevant data that could increase the yield of using nuclear medicine testing, reduce direct and indirect health care costs associated with foregone earnings by parents to accompany the child to the test, and reduce school absen- teeism and radiation exposure. Limiting the use of gastric emptying scans to children with the chief complaint of nausea may result in a more efficient approach to testing.
We cannot exclude that some of the children in our study who were tested for presumptive gastroparesis had functional dyspepsia, as the symptoms of gastroparesis are unspecific and may be present in both disorders.11 Early satiety, post- prandial fullness, and epigastric pain are the most common symptoms in dyspeptic children (approximately 90%),12
which was the same frequency found in our study popula- tion. Additionally, 20% to 26% of children diagnosed with functional dyspepsia have delayed gastric emptying,12,13
again the same percentage found in our sample. Unfortu- nately, the design of our study did not utilize the Rome criteria for diagnosis, which would have provided a greater understanding of the possible overlap in diagnosis between children with functional dyspepsia and gastroparesis.
Limitations of the study include the small sample size and the lack of standardized data from the child’s initial consultation to whenhe or shewas seenby the pediatric gastroenterologist. The design of children’s questionnaires is frequently hampered by the lack of information on symptom comprehension. Our study further supports that the terms bloating and retching may be poorly understood by children and, subsequently, may not be reliable predictors of delayed gastric emptying. The varied response to the comprehension portion of the sur- vey suggests that children were attentive to the questions and were truthful in answering to the best of their abilities.
The lack of available information on a numerically-based questionnaire, which was designed to be used in adults, prompted us to investigate whether the use of different vari- ations of the GCSI (word vs numerical) could demonstrate associations with delayed gastric emptying in children. We
92
found no difference in the ability of either of these forms to predict a delay in gastric emptying in children over the other. In an additional attempt to provide a friendlier questionnaire to children, we limited the items of the severity scale to 5 categories, though we have no indication that the use of a limited scale was beneficial in increasing the predictive role of the GCSI. n
Submitted for publication Mar 20, 2013; last revision received Jul 31, 2013;
accepted Sep 6, 2013.
Hepatology, Nutrition, Comer Children’s Hospital, MC 4065, Wyler C-499,
5839 S Maryland Ave, Chicago, IL 60637. E-mail: [email protected].
uchicago.edu
References
troparesis: presentation, treatment, and outcome. J Pediatr 1997;131:751-4.
2. Gariepy CE, Mousa H. Clinical management of motility disorders in
children. Semin Pediatr Surg 2009;18:224-38.
3. Maurer AH, Parkman HP. Update on gastrointestinal scintigraphy.
Semin Nuclear Med 2006;36:110-8.
4. Waseem S, Islam S, Kahn G, Moshiree B, Talley NJ. Spectrum of gastro-
paresis in children. J Ped Gastroenterol Nutr 2012;55:166-72.
5. Leidy NK, Farup C, Rentz AM, Ganoczy D, Koch KL. Patient-based
assessment in dyspepsia: development and validation of Dyspepsia
Symptom Severity Index (DSSI). Dig Dis Sci 2000;45:1172-9.
6. Revicki DA, Rentz AM, Dubois D, Kahrilas P, Stanghellini V, Talley NJ,
et al. Development and validation of a patient-assessed gastroparesis
symptom severity measure: the Gastroparesis Cardinal Symptom Index.
Aliment Pharmacol Ther 2003;18:141-50.
7. Revicki DA, Rentz AM, Dubois D, Kahrilas P, Stanghellini V, Talley NJ,
et al. Gastroparesis Cardinal Symptom Index (GCSI): development and
validation of a patient reported assessment of severity of gastroparesis
symptoms. Qual Life Res 2004;13:833-44.
8. Tougas G, Eaker EY, Abell TL, Abrahamsson H, Boivin M, Chen J, et al.
Assessment of gastric emptying using a low fat meal: establishment of in-
ternational control values. Am J Gastroenterol 2000;95:1456-62.
9. Abell TL, Camilleri M, Donohoe K, Hasler WL, Lin HC, Maurer AH,
et al. Consensus recommendations for gastric emptying scintigraphy: a
joint report of the American Neurogastroenterology and Motility Soci-
ety and the Society of Nuclear Medicine. Am J Gastroenterol 2008;103:
753-63.
10. Chogle A, SapsM. Gastroparesis in children. The benefit of conducting 4
hours scintigraphic gastric emptying studies. J Pediatr Gastroenterol
Nutr 2012.
motility…