PEDIATRIC HELICOBACTER PYLORI ISOLATES DIFFER IN ANTIBIOTIC RESISTANCE RATES THAN ADULT STRAINS
Benjamin D Gold, Marilyn Owens, Daphne Pierce-Smith, Stephanie Hawthorne, Steven J Czinn, Philip M Sherman, Qunsheng Song, Division of Pediatric Gastroenterology, Emory University School of Medicine, Atlanta, GA, USA; Division of Pediatric Gastroenterology, Rainbow Babies and Children's Hospital, Cleveland, OH, USA; Division of Pediatric Gastroenterology, The Hospital for Sick Children, Toronto, ON, Canada
Little is known about the prevalence and risk for antibiotic resistance occurring in H. pylori (Hp) strains infecting children. Diagnostic EGD was performed on 69 children at 6 different institutions: 5 cities in the US; Cleveland (N=27), Atlanta (N=12), Miami (N=6), Arizona (N=1), Philadelphia (N=1) and Toronto, Canada (N=22). Biopsies were taken from the stomach antrum and body, and Hp was isolated using primary culture techniques. Demographic and clinical data was collected on all subjects. Susceptibility of each isolate to amoxicillin, clarithromycin, tetracycline and metronidazole was tested by agar dilution using NCCLS standards. An ATCC Hp type strain was used as quality control. Breakpoints of resistance used in this study are shown in table 1.
Of the study cohort evaluated, 37 (54%) were male, 19 Caucasian (28%), 17 black (25%), 16 white (23%), 5 Hispanic (7%), 4 Asian (6%), 8 (12%) were designated as other or unknown. Median age was 12 years (range 3 – 18 yrs). Duodenal and/or gastric ulcer were observed in 18 of the subjects. The antibiotic resistance profiles are shown in table 2.
Metronidazole and clarithromycin resistance are common in pediatric Hp strains, whereas, amoxicillin or tetracycline resistance was not present. Metronidazole resistance was similar to that observed in adult Hp isolates in comparable populations. However, clarithromycin resistance was substantially higher than previous studies of Hp strains.
* NUTRITION AND LUNG TRANSPLANT IN CYSTIC FIBROSIS
Eyal Shteyer, Maite deLa Morena, Stuart C Sweet, Charles H Huddleston, Eric N Mendeloff, Rosemary Nagy, Paul S Hmiel, Ross W Shepherd, Division of Gastroenterology & Transplantation Services, Washington University School of Medicine, St. Louis, MO
Outcomes of major technologies in medicine and surgery such as lung transplantation require systematic study to define potentially treatable adverse pre-existing factors and evaluate areas of ongoing concern which might affect quality of life following the procedure. Investigations of the clinical relevance and effects of malnutrition in CF lung transplant recipients have been limited to date, although malnutrition and its adverse effects are well documented in CF, and potential nutritional therapies are available. We hypothesized that malnutrition in CF prior to lung transplantation is common, and has adverse effects on the post transplant outcome.
We retrospectively analyzed our lung transplant database of all CF lung transplant patients between 1991–2002. Nutritional status at time of listing, time of transplant, and at one year post-transplant were recorded. Malnutrition was defined as Z-score weight (WAZ) <-2.0. Nutritional growth failure was defined as Z score ht (HAZ) <-2.0. Outcome measures (death while waiting, death post-transplant, rejection, infection and OB) were recorded. Statistics included Kaplan –Meier survival, chi-squared analyses, and Cox proportional analysis.
One hundred CF children were studied(58 females and 42 males, age range 6–20 years, median 13y). At listing, 60% and 40% had WAZ & HAZ <-2.0 respectively (mean HAZ −1.6± 1.04 (range −3.99±0.73), mean WAZ −2.48±1.66 (range −8.8± .8)). One year post transplant, HAZ was −2.1± 1.06 (range −5.2–0.49) and WAZ was −2.1±1.5 (range −6.7±0.53), which was not significant before vs after transplant as a group, but paired t-test for survivors was significant for decreased HAZ and increased WAZ (p 0.0002 and 0.01, respectively).The overall 1,2, & 5 year survival was 88%, 81% and 54% respectively. Better 1 and 2 year (early) survival rates were observed in those with WAZ > −2.0 at the time of transplant (94% vs 82% at the first year, and 89% vs 78% at the second year). No significant effect in Cox proportional hazardous model for six varients.
Malnutrition is both common and important in CF children requiring lung transplants appearing to affect early post-operative mortality. There was only minor improvement in WAZ a year after transplant, but HAZ had decreased, possibly due to steroid therapy. These data emphasize that, pre-operative malnutrition and post-operative growth failure, both of which are potentially reversible, deserve wider study. We suggest a greater role for gastroenterology and nutritional involvement in the of management these patients.
* DESIGN, IMPLEMENTATION, AND EFFECTS OF COMPUTERIZED CALCULATION OF PEDIATRIC PARENTERAL NUTRITION REGIMENS
Robert Rothbaum, Matthew Wickline, Gerald Krausz, Kristina Kloos, Kara Kniska, Melissa Heigham, Judy Ho, Joyce Chen, Rosemary Nagy, Department of Pediatrics, St. Louis Children's Hospital, St. Louis, MO; Pharmacy Services, St. Louis Children's Hospital, St. Louis, MO; Health Administration Program, Washington University School of Medicine, St. Louis, MO
We designed and implemented a computerized system for calculating and ordering pediatric parenteral nutrition (TPN). Based on accepted formulas and published guidelines, the TPN calculator determined fluid, energy, protein, lipid, electrolyte, mineral, vitamin, and trace element amounts, concentrations, and infusion rates after input of patient age, weight, length/height, and sex. The calculator is accessible at http://tpn.wustl.edu. Any nutrient concentration or amount can be altered, and related constituents are automatically re-calculated. We analyzed these parameters just before and then six months after TPN calculator implementation: number of orders per day, % late orders each day, and physician and pharmacist time devoted to order writing
Next, we analyzed adequacy and accuracy of TPN prescriptions from the same time periods. All reviewed PN orders (n=110) met essential fatty acid and amino acid requirements. We also reviewed % of orders providing adequate calories to meet resting energy expenditure (REE) and calculated total daily energy expenditure (TDEE) and % of orders with sodium and potassium infusion within 10% of desired amounts.
The computerized design and ordering of pediatric TPN reduced time requirements of physicians and pharmacists. Even with a decreased time commitment, the adequacy and accuracy of TPN orders improved significantly. Prescribed calories more frequently fulfilled REE and TDEE; fewer errors occurred during calculation of infused amounts of sodium and potassium. The timeliness of order submission was delayed by factors other than time required to design the orders.
* EFFECTS OF FRUCTO-OLIGOSACCHARIDE SUPPLEMENTED INFANT CEREAL, A DOUBLE-BLIND RANDOMIZED TRIAL
Nancy Moore, Cewin Chao, Li-Ping Yang, Heidi Storm, Maria Oliva-Hemker, Jose Saavedra, Pediatrics, Johns Hopkins School of Medicine, Baltimore, MD, Nestle USA, Glendale, CA
Fructo-oligosaccharides (FOS) may have potential benefits since they exhibit many soluble dietary fiber-like properties, in addition to acting as a prebiotic agent, with the capability of beneficially modifying intestinal flora. Commercially available foods for weaning infants generally are low in fiber content and lack these potential benefits. Few studies document ranges of acceptable intakes of FOS. The purpose of the study was to assess acceptability, consumption and GI tolerance in infants of weaning age of a cereal supplemented with FOS.
57 healthy, non-breast feeding infants mean age of 7.5 ± 2.0 months, were randomly assigned to receive either FOS supplemented cereal (0.03 g FOS/gm cereal) or placebo (same cereal with 0.03 g Maltodextrin/ gm cereal) for 28 days (provided by Nestlé, USA, Nutrition Division). Primary outcome was tolerance, assessed by daily parental reporting of gastrointestinal functional for 28-days, (stool frequency, color and consistency; flatus, presence of vomiting, spitting up, crying, and colic/abdominal cramps). Secondary outcomes included cereal intake (g cereal/day and g FOS/day), changes in anthropometric measurements and adverse events.
Average daily total intake per infant and was similar in both groups (35 ± 22.5 g /day supplemented cereal and 35± 27.5g/day control). FOS consumption was as high as 3.0 g FOS/day/infant with a mean intake of 1.05 g FOS per day. Infants receiving FOS had more frequent stools (p=<0.01). The average number of stools per day were 2.0±0.96 (FOS supplemented) and 1.3±1.0 (control). For infants receiving FOS supplemented cereal, we found that stool consistency was more frequently reported as “soft” or “loose”, but not “watery” and less frequently to be described as “hard”. Frequency of “soft” stools were 28% (FOS supplemented) and 19% (control) for the first two stool observations/infant/day (p=<.001). Of all stools reported as “watery” (0.8%), 95% occurred in the control group. The frequency of reported “hard” stools were 2.5% (FOS supplemented) and 5.5% (control) for the first two stool observations/infant/day (p=<.001). In addition, the supplemented group reported significantly fewer days of missed stooling than the control group ( 1.02% and 3.14% respectively). We found no differences in the mean number of days that these symptoms were “more than usual” for crying, spitting up, vomiting, flatus or colic.
FOS-supplemented cereal was well tolerated in doses of up to 3 grams/day. FOS consumption led to more regular and softer stools, without diarrhea, as well as less-reported frequency of hard stools or days without stool. This study is one of few studies documenting tolerance to increased fiber intake in infants in the form of FOS in weaning food. Additional studies are needed to documents further gastrointestinal and/or other systemic benefits.
* INFANTILE GERD MAPS TO CHROMOSOME 9
Susan R Orenstein, Amitabh Suman, Theresa M Shalaby, David C Whitcomb, M Michael Barmada, Pediatric Gastroenterology & Gastroenterology, University of Pittsburgh School of Medicine, Pittsburgh, PA; Human Genetics, University of Pittsburgh, Pittsburgh, PA, Children's Hospital of Pittsburgh, Pittsburgh, PA
Increasing evidence points to a genetic predisposition for infantile gastroesophageal reflux disease (GERD).A pediatric GERD phenotype in older children was previously mapped to chromosome (chr) 13q14, but this locus was excluded in our five well-characterized kindreds with an autosomal dominant pattern of infantile GERD. 1,2 To identify an infantile GERD locus in our families, we subjected 22 members of two of them to genome-wide scan.
The two pedigrees (Families B & D) comprised 61 (38 & 23) individuals in 3 generations. The probands' symptoms, histological morphometric esophagitis 3, and diagnostic score on a validated questionnaire 4 confirmed their phenotypes. Phenotyping of the probands' generation via questionnaire identified 11/29 (38%:8/18 & 3/11) with positive phenotype, consistent with the autosomal dominant, variable penetrance mode of inheritance previously postulated. Individuals in the non-proband (older) generations were designated as phenotype unknown or unaffected. Blood was obtained for DNA analysis and submitted for genome-wide scanning on all 11 positive members of the probands' generation and 11 of the most informative members of the other generations. We genotyped 253 microsatellite markers from the ABI mapping set 2 panel. Nonparametric linkage analysis on trimmed pedigrees comprising 16 and 10 individuals was performed using the Allegro program 5.
A highly significant multipoint nonparametric LOD score of 2.4 was obtained for markers in the 9q22-q31 region (p=0.0002). Fine mapping using more densely spaced markers in this region is currently in progress. These data confirm heterogeneity in pediatric GERD and suggest a new locus for infantile GERD on chromosome 9.
1. Hu FZ, et al. JAMA 284:325, 2000
2. Orenstein SR, et al. Gastroenterology 120:A211, 2001
3. Black DD, et al. Gastroenterology 98:1408, 1990
4. Orenstein SR, et al. Clin Pediatr 35:607, 1996
5. Gudbjartsson DF, et al. Nat Genet 25:12, 2000
* TRYPTASE IS A POTENTIAL DIAGNOSTIC MARKER FOR EOSINOPHILIC ESOPHAGITIS AND EOSINOPHILIC GASTROENTERITIS WITH ESOPHAGEAL INVOLVEMENT
Mirna Chehade, Russell Castro, Margret Magid, Keith Benkov, Audrey Birnbaum, Nanci Pittman, Hugh Sampson, Kirsten Beyer, Pediatric Gastroenterology; Allergy; Pathology, Mount Sinai School of Medicine, New York, NY
Eosinophilic gastroenteritis (EG) is characterized by eosinophilic infiltration of the gastrointestinal mucosa. Current evidence suggests a Th2-mediated allergy to certain foods in a subset of patients. Eosinophilic esophagitis (EE) is believed to be a subset of EG. Patients with EE present with reflux symptoms, mimicking acid-induced GER. Mucosal esophageal eosinophil numbers are shown to be more elevated in patients with EE than those with GER, but no consensus on cut-off numbers exists. Investigating other cells important in allergic inflammation may provide a way to differentiate those patients.
To compare the number of mucosal mast cells among esophageal biopsies of patients with EE, EG with esophageal involvement, GER, and normal subjects.
Patients with biopsy confirmed EE or EG with esophageal involvement (mean>10 eosinophils/HPF) were included in the study. In addition, non-atopic patients biopsied for abdominal pain, emesis or reflux symptoms were identified. Those with negative esophageal biopsies served as controls; and those with esophageal biopsies consistent with reflux, and responding to standard anti-reflux therapies served as the GER group. Immunohistochemical staining for tryptase, a mast cell-specific mediator, was performed on paraffin-embedded esophageal tissue using anti-tryptase monoclonal antibody.
The median number of mast cells was significantly higher in patients with EE/EG than in controls or those with GER
Mast cells were degranulated to a variable extent in patients with EE/EG, as evidenced by tryptase staining. In marked EE, eosinophils were concentrated in the superficial portion of the epithelium; while mast cell infiltration was more prominent in the deeper areas.
Esophageal mast cells, as determined by tryptase staining, are potential diagnostic markers for eosinophilic esophagitis and eosinophilic gastroenteritis with esophageal involvement.[table 1]
* HETEROTOPIC GASTRIC MUCOSA IN PEDIATRICS - ANOTHER CONTRIBUTOR TO SUPRAESOPHAGEAL SYMPTOMS?
Suhasini Macha, Sushma Reddy, Raja Rabah, Vasundhara Tolia, Pediatric Gastroenterology, Hepatology & Nutrition, Childrens Hospital of Michigan, Wayne State University School of Medicine, Detroit, MI
A prospective study was performed to assess the incidence of heterotopic gastric mucosa (inlet patch-IP) in the proximal esophagus in children undergoing a diagnostic esophagogastroduodenoscopy (EGD) for usual indications over a 17 month period from 12/1/00-04/30/02.
All patients had recurrent abdominal pain, nausea and vomiting as indications for performing an EGD. A comparison of supraesophageal symptoms between children with IP and a randomly selected group of control patients with otherwise similar indications undergoing EGD was done.
IP was identified as a salmon-pink or yellowish plaque-type lesion seen in the proximal esophagus, usually within 5 centimeters of the cricopharyngeus muscle, more commonly seen during withdrawal of the gastroscope. While distal esophageal and stomach biopsies were routinely taken for all patients undergoing diagnostic EGD, an additional biopsy was obtained from the IP in those patients identified to have it. Biopsies were blindly reviewed for histology and assessment of inflammation by a pathologist using modified Sydney Classification.
A single endoscopist performed a total of 256 EGDs and an IP was identified in the proximal esophagus in 10 of those patients. IP was suspected in five additional patients, however it was not confirmed on histology.Single IP's were found in 9 patients, double IP in one. Patients with IP demonstrated a higher incidence of ENT symptoms, specifically recurrent otitis media, sinusitis, pharyngitis, and dysphagia (p=0.010) and respiratory symptoms, specifically cough, wheeze, and asthma (p=0.032) compared to the control group. There was no statistical difference in symptoms of nausea/vomiting (p=0.473), choking/gagging (p=0.303) or recurrent abdominal pain (p=0.303).
Interestingly, the presence and degree of inflammation in the gastric mucosa of IP was always greater than in the stomach (p=0.001) and all contained oxyntic and cardiac mucosa. However, there was no difference in the degree of inflammation between the squamous epithelium adjacent to the IP and the epithelium in the distal esophagus (p=0.531). The incidence of Barrett's changes in patients with IP was 20.0%.
The incidence of IP was 3.8% in our population undergoing EGD. The prevalence of IP in the adult population has been reported to vary from 0.1 to 10%.
We speculate that IP may contribute toward supraesophageal symptoms in children. Although usually considered a benign incidental finding at EGD, the presence of intestinal metaplasia suggests that long term follow up may be needed in these patients as esophageal adenocarcinoma has been reported from IP in adults.
* THE INFLUENCE OF GAVISCON INFANT™ ON GASTRO-ESOPHAGEAL REFLUX IN INFANTS
Raffaele Del Buono, Glenda Ball, Mike Thomson, Paediatric Gastroenterology, Royal Free Hospital, London, United Kingdom
Gaviscon Infant™ is part of the therapeutic approach for pathological gastro-esophageal reflux (GER) in infants. Its efficacy has so far not been examined with a physiologically-appropriate denominator to define the degree of GER.
The main aim of this placebo-controlled double-blind study was to investigate the influence of Gaviscon Infant on GER in infants by using a combined pH and intraluminal impedance measurement.
Twenty infants (mean age 163.5 days, range 34–19 days) exclusively bottle fed, with symptoms of GER underwent 24 hour studies of intra-oesophageal 6 channel impedance and dual channel pH monitoring. During the 24 hour impedance/pH measurement, six random administrations (3+3) of Gaviscon Infant (625mg in 225ml milk) or placebo (mannitol and Solvito N, 625mg in 225ml milk) were given in a double blind fashion. Impedance data were recorded and analysed blind by one observer for numbers of reflux events, pH of each reflux, height reached by refluxate and acid clearance time.
The median number of reflux events per hour (1.58 after Gaviscon Infant versus 1.68 after placebo), median number of acid reflux events per hour (0.26 versus 0.43) were all lower after Gaviscon Infant than after placebo, but the within treatment differences were not statistically significantly different from zero. The average reflux height was considerable lower after Gaviscon Infant (median 1.90) compared with placebo (median 1.31) (note that `higher' values represent `lower' reflux heights) and the within subject difference was statistically significantly different from zero (p<0.001), indicating that reflux after Gaviscon Infant tends not to travel as far up the oesophagus compared with that after placebo. There were no statistically significant differences between Gaviscon Infant and placebo in average minimum distal or proximal pH. The median total acid clearance time per hour (time with pH below pH 4) and the median total reflux duration per hour were both lower after Gaviscon Infant than after placebo, but the within treatment differences were not statistically significantly different from zero.
This trial reveals a statistically significant difference between Gaviscon Infant in one of the parameters which measures post feeding reflux (average reflux height). Several of the remaining parameters reveal a tendency for `better' results following Gaviscon Infant compared to placebo, but do not achieve statistical significance within a study of 20 patients.