Grantee Research Project Results

Final Report: Environmental Tobacco Smoke, Biomarkers, and Childhood Asthma

EPA Grant Number: R830826
Title: Environmental Tobacco Smoke, Biomarkers, and Childhood Asthma
Investigators: Klonoff-Cohen, Hillary , Platzker, Arnold
Institution: University of California San Diego
EPA Project Officer:
Project Period: July 1, 2003 through June 30, 2009 (Extended to June 30, 2010)
Project Amount: $750,000
RFA: Biomarkers for the Assessment of Exposure and Toxicity in Children (2002) RFA Text |  Recipients Lists
Research Category: Children's Health , Human Health

Objective:

OBJECTIVE#1:

The first objective of this study was to determine if serum Eosinophil Cationic Protein (sECP) and urine Eosinophil Protein X (uEPX) would be useful biomarkers in predicting the presence of disease, clinical severity, and treatment efficacy in children diagnosed with asthma, in order provide a comprehensive health risk assessment in children.    

The specific aims used to accomplish this objective were to:  1. Establish baseline values and distribution of sECP and uEPX in infants and young children at their initial presentation of asthma.  These levels were compared to those of healthy children, matched with respect to gender, age, race/ethnicity, hospital, and clinic, and    2. To create a time profile of sECP and uEPX levels from the initial visit through the treatment period, based on a minimum of two time values for each child (e.g., at baseline, at 3-4 months) during a 12-month period. 

OBJECTIVE #2:

The second objective was to determine if in utero and postnatal environmental tobacco smoke (ETS) exposure (based on infant's urine cotinine measurements at the time of the clinic visits) would result in higher levels of sECP and uEPX. 

The specific aims used to accomplish this objective were to: 1. Examine the association between ETS exposure and baseline sECP and uEPX, and assess any dose‑response effect between numbers of cigarettes and cotinine, ECP, and EPX levels;  2. Determine whether asthmatic infants exposed to ETS in utero have the highest ECP and EPX values, whether unexposed asthmatics and ETS-exposed healthy children have intermediate values, and whether healthy unexposed infants have the lowest ECP and EPX values at baseline.

Summary/Accomplishments (Outputs/Outcomes):

Overview

At present, one of the great obstacles in pediatric pulmonology is to identify beginning stages of asthma, given that loss of lung function occurs within the first 3 years of life. Problems in diagnosing or monitoring childhood asthma may occur because of non-specific symptoms, as well as the inability to obtain reliable pulmonary function tests in children <5 years of age.  Invasive studies such as bronchoscopy, bronchial biopsies, and pediatric lung function tests are not easily performed.  Even venous blood sampling can be difficult.  An ideal inflammatory marker for diagnosing and monitoring asthma should: 1) reflect spontaneous changes in disease activity; 2) mirror improvements due to therapy; 3) have high sensitivity and specificity; 4 ) have prognostic significance; 5) be reproducible in steady state; 6) be fast, easy to obtain, and non-invasive; 7) be inexpensive; and 8) not be influenced by confounding factors.  

Project activities over the entire period of funding (achievements with respect total goals)

The activities over the entire study period consisted of identifying several pulmonology or pediatric sites in Northern and Southern California, obtaining IRB approval, recruiting asthmatic and healthy infants after obtaining informed consent from parents, collecting baseline blood and urine samples prior to treatment (for asthmatics) or medical care of healthy infants, conducting a baseline telephone interview, administering a diary, conducting follow-up telephone interviews collecting follow-up blood and urine samples.  Samples were batch analyzed and statistical analyses were performed.   The majority of the time for this project was devoted to identification and vigorous enrollment of eligible patients in order to fulfill the desired sample size.

Study design and recruitment

This was a prospective cohort study. There were several hospitals that agreed to participate in the study including University California San Diego, Rady Children's Hospital San Diego, University California San Francisco, Stanford, Children's Hospital Orange County, Children's Hospital Oakland, and Navy Medical Center San Diego.   We obtained Institutional Review Board approval from each institution (including designing individualized consents and attending meetings), which was an exceedingly time-consuming and lengthy process.  The majority of sites found the project too labor intensive to manage on a day to day basis, thereby requiring continual replacement of existing non-viable sites with new project sites.  Hence, there were multiple centers scheduled to enroll patients, however, there were insufficient funds available for doctors and staff at all institutions. In the end, the hospital sites that provided the majority of patients were from USC-LA children's hospital and Naval Medical Center San Diego.

We achieved our desired total sample of Caucasian, African-American, Hispanic and Asian/Pacific Islander children who were diagnosed with new-onset asthma between 2003-2010.  In this study, the pulmonologists made the diagnosis based on a history of recurrent episodes of wheezing, a physical examination (e.g., respiratory rate, alertness, dyspnea, skin color, ausculatory findings), and physiological parameters (peak expiratory flow rate). 

The healthy children were recruited while attending pediatric clinic appointments for baby well checks, vaccinations, fractures, other injuries, or minor surgeries.   This group was matched to the asthmatics on the basis of age, race/ethnicity, gender, and clinic site. 

There were extremely strict inclusion and exclusion criteria for this study.  The recruitment process was greatly lengthened by the decision to strictly adhere to these standards.  For example, we required patients with previously undiagnosed asthma symptoms, who were never treated for asthma which we called "virgin asthmatics."  The pulmonologists greatly overestimated the number of cases who were virgin asthmatics at the pulmonary clinics.  To counteract this problem, we broadened our recruitment efforts to emergency rooms to increase our sample, but this proved to be too difficult, since these patients were too ill, so were treated emergently with medications, making them ineligible for the study.  Initially, patients treated for airway hyper reactivity with beta adrenergic or cromoglycate sodium medications within the past 2 weeks prior to study entry were not included, since these medications cause a transient 2-4 hour reduction in ECP.  In addition, patients who have received systemic or inhaled steroids within 24 hours of blood or urine collection were also not included, since these medications can also alter ECP and EPX.  However, eliminating these patients excluded severe asthmatics.  These two particular criteria created several difficulties with recruitment, since there were hundreds of eligible children that were not included because they were on medications and not classified as virgin asthmatics.  Therefore, over time, and after consultation with the co-investigators, it was decided that those patients using or currently on medications would be collected and analyzed separately.

Additionally, subjects with a history of documented respiratory syncytial virus (RSV) or influenza in the past 3-6 months were not eligible. Likewise, healthy children who had a personal history of atopy, cough, shortness of breath, and an intercurrent respiratory infection in the past month were also excluded.  This further shrunk the size of the pool of eligible subjects.  The pulmonologists subsequently modified the criteria for the healthy children to "no personal or family history of asthma, wheezing or atopy."   

There were several other limitations with recruitment that require discussion.  There was a paucity of wheezing infants in several years within the recruitment period because the disease is seasonal.  Furthermore, parents were not comfortable with their infants providing blood, and sometimes urine samples, even though these specimens were collected by the research nurses, and the visits were coordinated to regular office visits to desensitize future visits.  Fortunately, the refusal rate at the military hospital was minimal.

Finally, there was a loss to follow-up, even with continued phone contact and incentives, for both the asthmatic and healthy groups. There were also a number of no-shows for asthmatic families for follow-up visits. This is because once the child started to respond to the medications, the parents often discontinued them, without the knowledge and approval from the physician.  Reimbursement for participation was minimal, so it was not a great motivator for parents to consent their young child to the study.  

Data Collection

Both objectives one and two were based on collecting blood and urine from all asthmatics and healthy children at each clinic visit.  The final sample size consisted of 300 subjects (160 asthmatics and 140 healthy children).   The cost of the tests and amount of labor required that the samples be batch analyzed at the end of the study.  Serum ECP was compared to urine EPX to determine which measure was superior in predicting asthma in young children.

Measurement of serum ECP

The clinical usefulness of ECP has been questioned because of problems with sample processing and the great variability of measured values.  We employed a standardized sample method to minimize variations in measured levels. 

Blood samples were carefully collected and stored, since the method of blood collection can substantially change the level of serum ECP.  It can be spuriously increased by high ambient temperatures, if excessive time is allowed for clotting, or by insufficient centrifugation and diurnal variation.

We collected whole blood specimens, allowed them to clot for the specified time, and then centrifuged them within 60 minutes from collection.  The samples were usually stored in a refrigerator for several hours before being transferred to a -80 degree freezer.  The day that the samples were delivered to the lab for batch analyses, they were sitting on dry ice; however, it was an exceedingly hot day, and they were first put in a -40 degree freezer and then transferred to a -80 degree freezer.  This could have negatively affected the quality of the samples.

Measurement of urine EPX and cotinine

The second inflammatory marker that was measured in this study was urine EPX.  This measure was chosen for several reasons.  First, the marker appears in urine.  Urine collection is thought to be superior to blood in young children because it is simple, non-invasive, and easily accomplished.  Moreover, EPX has a higher release than ECP in asthmatics and is not influenced by allergy.  It also differs in symptomatic and asymptomatic individuals.  Urine was collected at each clinic visit (i.e., initial and follow-up visits) and kept on ice until refrigerated at -80 degrees. 

Cotinine, a sensitive byproduct of nicotine, and measure of environmental tobacco smoke, was also measured in urine.  Because parents may not truthfully report smoking behavior (e.g., number of cigarettes/cigars smoked/week) at each clinic visit, particularly in families who have an asthmatic child, cotinine was used as the corroborating measure at each visit. 

Data collection instruments

All patients' parents completed a 45-60 minute telephone interview, which focused on the following topics: medical history, pregnancy history, living conditions, and environmental exposures.  Parents of asthmatic patients also completed a monthly diary to record changes in medication/treatment, exacerbations, hospitalizations, and other illnesses.  Morning and evening peak expiratory flow rates (PEFRs) were recorded at home for asthmatic children >3 years.  Peak expiratory flow rates were also conducted around a scheduled visit (1 week prior and post visit) for healthy children. 

RESULTS

Descriptive characteristics of asthmatic and healthy children

A total of 160 children were diagnosed with possible childhood asthma. The sample consisted of 33% females and 67% males, ranging in age from 3 months to 59 months. The racial/ethnic breakdown of the sample was 44% Caucasian, 15% Hispanic, 8% African-American, 9% Asian, 14% Other, and 10% Unknown.

RESULTS FOR OBJECTIVE 1

Serum ECP levels (ng/dL) in asthmatic children versus healthy children

The mean (SD) of serum ECP in patients and healthy children at the initial baseline visit were 117.47 ng/mL (32.23) and 73.91 ng/mL (10.60), respectively.  Serum ECP levels were not significantly higher in symptomatic patients than in asymptomatic patients.

The sECP levels were not significantly higher for asthmatic patients at baseline (initial clinic visit) compared to the healthy children.  In addition, at three months after treatment there was no difference between the treated asthmatic children and the healthy children.

Urine EPX levels (ug/L) in asthmatic children versus healthy children

The means of urine EPX in asthmatic patients and healthy controls did not differ at initial baseline visit (9.22 ug/ml (13.06) vs. 13.63 ug/mL(20.14) respectively.  After treatment, urine EPX values did not decrease.

RESULTS FOR OBJECTIVE 2

Parent-reported ETS and urinary cotinine levels for asthmatic and healthy children

Cotinine values in asthmatic and healthy patients were 1.83 ng/mL and 1.48 ng/mL, respectively.  There was no association between parent reported tobacco smoke exposure and baseline ECP/EPX.   The parent-reported values still require validation with urine cotinine measures.

Family history and Serum ECP, Urine EPX and asthma

A total of 50 percent of patients diagnosed with possible asthma had a family history of childhood asthma or atopy.  Patients who had a family history of childhood asthma had a mean serum ECP level of 119.42 ng/mL (21.76), whereas patients with no family history of childhood asthma had a mean serum ECP of 11.72 ng/mL (1.95).

Family history, parent-reported ETS, cotinine levels, Serum ECP, Urine EPX, and  childhood asthma

Those asthmatic patients who had a reported family history of childhood asthma and exposure to tobacco smoke had sECP levels of 44.87 ng/mL (3.92) and urine EPX values of 4.36 ug/ml (0.40) whereas those healthy patients who had no family history of childhood asthma and no exposure to tobacco smoke had sECP and uEPX values of 73.91 ng/mL (10.16) and 13.63 ug/ml (2.14), respectively.  The parent-reported tobacco smoke exposure still requires confirmation with cotinine measures.

Conclusions:

These results reveal the possibility of a relationship between serum ECP levels and family history of childhood asthma. This particular finding could add to the understanding and/or benefits to human health.  Identification of asthma in infants and children is extremely difficult to achieve because of nonspecific symptoms and the inability to obtain reliable pulmonary function tests.  If this finding was confirmed in future studies, particularly using genetic measures, it could be used as a marker for future asthmatics. 

In this study, there appeared to be no relationship between exposure to tobacco smoke, family history of childhood asthma and the inflammatory markers.  The tobacco smoke exposure reported by parents must still be validated with cotinine levels to confirm whether there is truly no effect on inflammatory markers. This is the first project to examine the relationship between inflammatory markers, environmental tobacco smoke, family history of childhood asthma, and childhood asthma, so these results must be confirmed in future studies.

We also tested whether these non-invasive markers could be used to assess asthma severity, and monitor treatment efficacy.  There was no significant change between baseline and post-pharmacotherapy values for sECP and uEPX.  This suggests a lack of efficacy in using these biomarkers to predict clinical severity and treatment efficacy in children diagnosed with asthma.  These negative findings need to be replicated in additional studies.

Journal Articles:

No journal articles submitted with this report: View all 3 publications for this project

Supplemental Keywords:

Biomarkers, inflammatory markers, environmental tobacco smoke, pediatric asthma, infants, children, serum eosinophil cationic protein, urine eosinophil protein X, epidemiology

Progress and Final Reports:

Original Abstract

2004 Progress Report

2005 Progress Report

2006 Progress Report

2007 Progress Report

2008 Progress Report

2009