An Investigation of the Association Between Helicobacter Pylori Infection and Abdominal Pain in Cystic Fibrosis Patients

Abdominal pain and Cystic Fibrosis Cystic fibrosis is the most common lethal autosomal
recessive disease among Caucasians. CF is caused by a single gene mutation on chromosome 7,
which encodes for a membrane protein, the cystic fibrosis transmembrane conductance regulator
(CFTR). CFTR channel dysfunction results in progressive pulmonary disease, which is the
primary cause of the morbidity and mortality associated with the disease. Therapies directed
to slow the progression of the lung disease have increased the USA median survival to 32.2
years.2 Abdominal complaints are very common among CF patients. Littlewood et al. reported
that 31% of children had regular complaints of abdominal pain.3 Ravilly et al. performed a
retrospective chart review on CF patients referred for treatment to the pain service and
those that expired.4 They determined that pain was very common in CF and that 19% of the
population had chronic complaints of abdominal pain, which was third behind chest pain and
headaches.4 This study likely underestimated the prevalence of abdominal pain in CF, as it
was evaluating an end stage lung disease group of patients that were frequently not eating
and had severe pain from other disease processes (i.e. headaches from hypercarbia and/or
hypoxia, and chest pain from rib fractures).

Known causes of CF related abdominal pain are: 1) malabsorption due to pancreatic
steatorrhea, 2) gastroesophageal reflux, 3) esophagitis, 4) gastritis, 5) peptic ulcer
disease, 6) pancreatitis, 7) Crohn's disease, 8) distal intestinal obstructive syndrome
(DIOS), 9) biliary duct disease including cholocystitis and cholangitis, 10) intussception,
11) acute or chronic gastrointestinal infection, 12) acute appendicitis, and 13)
constipation. Given this long list of potential causes, CF patients frequently undergo
intensive examinations to determine the etiology of the abdominal pain.4 The testing
unfortunately does not always reveal the etiology.

Helicobacter pylori World-wide, Helicobacter pylori (Hp) is increasingly being recognized as
the etiology of peptic ulcer disease and other upper and lower gastrointestinal tract
diseases.1 With the CF population, The role that Hp plays in abdominal pain has not been
elucidated. Two studies have attempted to look at the epidemiology of Hp in CF; however, the
approach utilized in those studies was determined to be inadequate. Johansen et al. and
Israel et al. both attempted to utilize Hp serology to determine the prevalence in a CF
population and determined that there was significant cross-reactivity between Hp and
Pseudomonas aeruginosa antigens.10, 11. They concluded that different modes of testing are
required if Hp is being evaluated in the CF population. To our knowledge, there have been no
further evaluations of Hp in CF using the newer diagnostic tests available.

Testing for Helicobacter pylori Since the isolation of spiral urease-producing Helicobacter
pylori bacteria (H. pylori) in 1983 by Drs. Marshall and Warren5, a significant body of
evidence has accumulated indicating that the bacteria is an important pathogen in upper GI
tract of humans 6,7. The causal relation ship between H.pylori and chronic active gastritis,
duodenal ulcer, and gastric ulcer is well documented 8, 9. Methods available for detecting
current infection of human stomach by Hp are generally divided into two general types:
Invasive and Non-invasive.

Invasive methods include esophageogastroduodenoscopy (EGD) with collection of gastric
biopsies. These biopsies are then examined by one or more detection methods: histological
examination of stained tissue, microbiological culture of the organism, or direct detection
of urease activity in the tissue (for example, the CLO test). Biopsy based methods are
expensive, entail greater patient risk and discomfort than non-invasive tests, and may give
false negative results due to sampling errors when colonization of the gastric mucosa is
patchy 6.

The non-invasive tests include several serological tests that detect serum antibodies to Hp.
A positive result with these tests can not distinguish between current infection and past
exposure to HP and, therefore, is not a conclusive indicator of current gastrointestinal
colonization.

The urea breath test is based on the principal that the urease enzyme is not present in human
tissue. Urease presence is indicative of a urease producing organism in the stomach, which is
most likely HP. The test has a sensitivity and specificity of 90% and a negative predictive
value greater that 95% and a positive predictive value of 90%.12 A third non-invasive test is
the Hp stool antigen test also has a sensitivity and specificity of about 90%, and high
positive and negative predictive values (>95%).13 Major confounding issues with HP testing
are systemic antibiotic exposure, which can lead to false negative tests and partial
treatment.14 This is especially problematic in the CF population, who are frequently on
systemic antibiotics for chronic suppurative bronchitis. Thus, as part of the inclusion
criteria, we will only include subjects who have been off antibiotics for at least 14 days
prior to enrollment. The CF patients will be allowed to continue nebulized antibiotics and
prophylactic dose of Azythromycin, and analysis will be performed to determine if this
results in false negatives by comparing the results to the stool antigen test.

Clinical measures of abdominal pain This study is also designed to determine whether the
prevalence of Hp is related to a decreased quality of life, which if documented suggests a
pathological role of Hp in CF. Our objective is to use two validated quality of life
instruments in this study: The PedsQL (Pediatric Quality of Life Inventory) and the
Gastrointestinal Symptom Rating Scale (GSRS).15 The PedsQL is a modular instrument for
measuring health-related quality of life (HRQOL) in children and adolescents ages 2 to 18.
The PedsQL 4.0 Generic Core Scales are multidimensional child self-report and parent
proxy-report scales developed as the generic core measure to be integrated with the PedsQL
Disease-Specific Modules. The PedsQL 4.0 Generic Core Scales consist of 23 items applicable
for healthy school and community populations, as well as pediatric populations with acute and
chronic health conditions. The GSRS questionnaire involves 15 items using the Likert scale
focusing on gastroesophageal reflux, abdominal pain, constipation, indigestion and diarrhea.
It uses a scale from1 to 4.

Inclusion Criteria:

1. Confirmed diagnosis of CF by a positive sweat test (more than or equal to 60 mEg/L by
quantitative pilocarpine iontophoresis and/ or genotype with two identifiable
mutations consistent with CF, accompanied by one or more clinical features consistent
with CF phenotype)

2. Abdominal pain during last 6 months

3. Age > 5 and <18 years

4. Either gender

4. Ability to do urea breath test which requires drinking water based solution through a
straw and breathing into a breath sample bag 5. Must be off any oral antibiotics and proton
pump inhibitors or sucralfate for 14 days.

Subject may continue their prophylactic (X 3/ week) dose of Azithromycin). Patients who are
currently on antibiotics or have been on antibiotics within 14 days will be re-screened for
inclusion criteria at their next clinic visit (i.e. no wash out).

Exclusion Criteria:

1. Females of child bearing potential with a positive urine pregnancy test The decision
not to include pregnant females is based on: 1) the quality of life screens are not
designed for pregnancy, 2) their abdominal complaints may be related to pregnancy, 3)
safety of the breath test has not been established during pregnancy.

2. Inability to do urea breath test which requires drinking water based solution through
a straw and breathing into a breath sample bag.