Longitudinal Follow-up of Bronchial Inflammation, Respiratory Symptoms, and Pulmonary Function in Adolescents after Repair of Esophageal Atresia with Tracheoesophageal Fistula

Subjects 

This study was a longitudinal study consisting of 31 patients with EA and distal TEF. The patients underwent primary anastomosis during 1986 to 1995 in the Hospital for Children and Adolescents, Helsinki University Central Hospital, and they were observed by a pediatric surgeon (H.L.). Patients with EA have a high incidence of long-term esophagitis, gastric metaplasia, and chronic bronchitis, and therefore, it has been the practice in our clinic to perform endoscopic and bronchoscopic follow-up examinations during the preschool and school-age of patients with repaired EA.14, 15

The study was approved by the Ethics Committee at the Helsinki University Central Hospital, and informed consent was obtained from the patients and their parents before the study.

Study Design 

In March 2005, the 31 patients with repaired EA were contacted by means of a letter and invited to participate in the study, which included a clinical examination, PF analyses, and a skin prick test (SPT). The study visits took place from April to June 2005. A questionnaire on asthma and allergy symptoms, validated for use in the International Study of Asthma and Allergies in Childhood, was sent to the patients.16 Additional questions about pneumonia, dysphagia, GER, and esophagitis were formulated and enclosed. The case notes of all children were reviewed, with particular reference to respiratory and esophageal symptoms and findings. Current symptoms were defined as those present in the 12 months before the study.

Exhaled Nitric Oxide, Pulmonary Function, and Histamine Provocation 

As a marker of airway inflammation, the fractional concentration of exhaled nitric oxide (FENO) was measured according to American Thoracic Society guidelines,17 by using a chemiluminescence analyzer (Niox, Aerocrine AB, Sweden). The FENO was considered increased when more than +1.96 SD higher than the height-adjusted reference value.18 PF was analyzed by using flow volume spirometry (Spiromaster, Medikro Oy, Finland). Ventilatory function was defined as restrictive when forced vital capacity (FVC) was <80% and as obstructive when forced expiratory volume in 1 second (FEV1)/FVC% was <87% of predicted.19 Bronchial responsiveness to histamine was measured by using a dosimetric bronchial provocation test.20 The dose producing a fall of 15% in FEV1 (PD15FEV1) was determined from the dose-response curves. According to PD15FEV1, bronchial hyper-responsiveness was categorized as mild (0.41-1.6 mg), moderate (0.11-0.4 mg), or severe (<0.1 mg).20 Before the PFT, the patients were not allowed to use β2-agonists for 24 hours or antihistamines for 5 days. Moreover, the patients had to be free of respiratory infection symptoms during the previous 2 weeks.

Atopy 

SPTs were used to test the patients for atopy. Patients who reacted with at least 1 wheal ≥3 mm in diameter to the tested allergens, in the absence of a response to the negative control solution, were defined as being atopic. Standard solutions from ALK (Allergologiska Laboratorium, Copenhagen, Denmark) were used to test for birch, timothy, cat, dog, horse, Dermatophagoides pteronyssinus, and Cladosporium herbarum. Latex was tested with a solution from Stallergenes (Antony, France).

Bronchoscopy 

The bronchoscopies were performed with general anesthesia, by using a rigid 3.5-mm bronchoscope and biopsy forceps (No. 10378L; Karl Storz, GmbH and Co, Tuttingen, Germany). The endobronchial biopsies were taken from the main carina. The severity of tracheomalacia was based on a macroscopical estimation, and according to Filler et al,21 it was considered severe when the anteroposterior collapse was ≥75% with cough or expiration. Tracheomalacia was considered moderate when the collapse was 50% to 75%, and mild when the collapse was <50%.

Biopsy Processing and Quantification 

The 3-μm paraffin sections were stained with the Herovici method as described.22 Appropriate areas (ie, where the bronchial epithelium was intact) were selected in each sample and sectioned perpendicular to the surface. Ten measurements were taken for RBM thickness in 3 different areas by using a computer-aided image analysis program (Image J; National Institutes of Health, Bethesda, Maryland). The mean value and SD is given for each sample. The number of inflammatory cells was assessed in the following manner. Different inflammatory cells were identified with immunostaining by using these antibodies: T-lymphocytes (CD3), B-lymphocytes (CD20), plasma cells (CD138), eosinophils (ECP), macrophages (CD163), and dendritic cells (CD1a). Neutrophilic leukocytes were identified on the basis of morphology. The number of different inflammatory cells was assessed semiquantitatively by counting the number of positively stained cells per high power field (J.L.). Samples with no positive cells were given a score of 0, samples with <10 positive cells were given a score of 1, samples with 10 to 30 positive cells were given a score of 2, samples with 30 to 50 positive cells were given a score of 3, and samples with >50 positive cells were given a score of 4. In addition to quantitation of inflammatory cells, a morphological diagnosis of bronchitis (mild, moderate, or severe) was given on the basis of the presence of increased inflammatory cell infiltrate and changes in bronchial epithelium and stromal edema. The bronchial biopsies were divided in 3 groups according to the age of the patient during the bronchoscopy: <3 years, 3 to 7 years, >7 years.

Statistical Analysis 

The Mann-Whitney U test was used for comparisons among the questionnaire data, PFT, and biopsy findings. The increase of RBM thickness was tested with paired t test. The differences in proportions were tested with the Fisher exact test. Cell count associations were tested with linear-by-linear trend test. The correlations between esophageal and bronchial histological findings were calculated with the Sperman correlation test. Two-sided P-values <.05 were considered to be statistically significant.

Subjects 

A total of 27 patients (87%) returned the questionnaire, and data are given for these patients. Esophageal anastomosis had been performed within 2 days of birth, and the primary outcome of the children is summarized in (Table I; available at www.jpeds.com). Tracheomalacia was diagnosed as severe in 3 patients, moderate in 9 patients, and mild in 9 patients. Repair of tracheomalacia with aortopexy was performed in 7 patients at a median age of 0.25 years (range, 0.07-8.25 years), and repair of GER with fundoplication was performed in 10 patients at a median age of 0.29 years (range, 0.16-6.35 years).

Table I. Primary outcome of children with esophageal atresia with distal tracheoeosphageal fistula

		All patients (n = 27)
		n	%
	Wide gap	6	22
	Tracheomalacia	20	74
	Aortopexy	3	11
	Fundoplication	6	22
	Aortopexy and fundoplication	4	15
	Mechanical ventilation, days	3
	Primary hospital stay, days	30
	Associated anomalies	18	67
	Multiple anomalies	11	41
	VATER association	5	19
	CHARGE association	2	7
	Chromosomal abnormality	1	4

VATER, Vertebral defects, anorectal malformation, tracheoesophageal fistula, renal anomaly, and radial dysplasia; CHARGE, coloboma, heart disease, atresia choanae, retarded growth and development, genital hypoplasia, and ear anomalia or deafness.

Questionnaire 

The results from the questionnaire are summarized in Table II. Eleven of 27 patients (41%) reported current respiratory symptoms, 12 of 27 patients (44%) reported current esophageal symptoms, and 7 patients (26%) reported both symptoms. Children with current respiratory symptoms had significantly more “current dysphagia” (P = .033, Fisher) and “wheeze ever” (P = .021, Fisher) than children without current respiratory symptoms. Current esophageal symptoms correlated with the “presence of difficult esophagitis ever” (P = .008, Fisher).

Table II. Demography and self-reported prevalence of asthma, atopy, and symptoms

		All patients (n = 27)
		n	%
	Age, years (median, [range])	13.8(9.7-19.4)
	Sex, male	15	56
	Doctor-diagnosed
	Asthma	6	22
	Allergic rhinoconjunctivitis	1	4
	Atopic eczema	6	22
	Respiratory symptoms
	Current respiratory symptoms	11	41
	Wheeze	6	22
	Attacks of wheezing	3	11
	Sleep disturbed by wheezing	3	11
	Attacks of dyspnea	5	19
	Cough lasting >4 weeks	3	11
	Dry cough between colds	7	26
	Wheeze ever	14	52
	Pneumonia ever	14	52
	Esophageal symptoms
	Current esophageal symptoms	12	44
	Heartburn	9	33
	Dysphagia	8	30
	Esophagitis ever	23	85
	Difficult esophagitis ever	7	26
	Atopic symptoms
	Allergic rhinoconjunctivitis ever	4	15
	Atopic eczema ever	9	33
	Current condition of health
	Excellent	11	41
	Good	14	52
	Satisfactory	2	7
	Medication in past year
	For asthma	6	22
	For esophagitis	4	15
	Smoking
	Patient	2	7
	Family member	10	37
	Family
	Atopy (atopic eczema, rhinoconjunctivitis)	11	41
	Asthma	4	15
	Pets	10	37

Doctor-diagnosed asthma (DDA) was reported by 6 patients (22%), 5 of whom had used asthma medication (inhaled corticosteroids, n = 4; beta-2-agonists, n = 5) during the past year. Two patients with DDA had current respiratory symptoms, and 3 patients had current esophageal symptoms. Fourteen of 27 patients (52%) with a history of pneumonia did not sustain more current respiratory or esophageal symptoms than patients without a history of pneumonia. Neither did the number of pneumonias correlate with the severity of tracheomalacia, current respiratory or esophageal symptoms, or with the PFT.

Clinical Visit 

Twenty-five of 31 patients (81%) participated, and the demographics and the test results are presented in Table III. Two patients had scoliosis, 2 patients had mild atopic eczema, and 2 patients were developmentally delayed.

Table III. Demographics, pulmonary function, and atopy at the clinical visit

		Result	Abnormal, n (%)
	Age, n = 25, years⁎	13.7(9.7-19.4)
	Sex, male	15(60%)
	Spirometry, n = 23
	FVC, % predicted⁎	85(15-112)	8(35%)
	FEV1/FVC%⁎	91(74-100)	7(30%)
	Histamine provocation, n = 23
	PD15FEV1, mg⁎	0.51(0.023->1.6)	6(26%)
	FENO, n = 22, ppb⁎	9.6(6.7-47.8)	5(23%)
	Skin prick test +, n = 24	13(54%)

FVC <80% and FEV1/FVC% <87% were considered abnormal. PD15FEV1 was considered abnormal when ≤0.4 mg. FENO was considered abnormal when ≥1.96 SD of the height-adjusted reference value.

⁎ Median (range).

Pulmonary Function and Atopy 

PF was measured with spirometry in 23 patients, and the median values were within the reference range for all measured variables. Eight patients (35%) had a restrictive PF pattern (FVC <80%), 7 patients (30%) had an obstructive defect (FEV1/FVC% <87%), and 10 patients (43%) had normal PFT results.18 The PFT were similar in patients with or without current respiratory or esophageal symptoms.

Increased BR to histamine was found in 18 of 23 patients (78%): severe in 2 (9%), moderate in 4 (17%), and mild in 12 (52%) patients. Children with severe or moderate BR tended to have more restrictive PFT results than those with mild or no BR (median FVC, 72% versus 89%; P = .059, Fisher). Patients with current respiratory symptoms tended to have more severe BR than patients who had no symptoms (PD15FEV1, 0.51 mg versus 1.04 mg; P = .053, Fisher). There was no significant relationship between the measured PFT or BR and the number of associated anomalies, severity of tracheomalacia, aortopexy, fundoplication, number of pneumonias, DDA, or current esophageal symptoms.

Positive SPT was found in 13 of 24 patients (54%): hay (7), dog (7), birch (6), cat (4), latex (4), horse (1), Cladosporium herbarum (1). Although increased BR was observed in 8 of 11 patients (72%) who were SPT+ (severe/moderate in 4 patients and mild in 4 patients), no significant association between SPT and BR was found (P = .999, Fisher).

The median FENO (9.6 ppb) was within the predicted range,18 whereas it was considered increased in 5 of 22 patients (23%), all of whom were atopic (eg, had positive SPT results). The difference of mean FENO in SPT+ (32.2 ppb) versus SPT– (9.3 ppb) was statistically significant (P = .012, Mann-Whitney). Furthermore, patients with high FENO tended to have increased BR to histamine when compared with patients with normal FENO, although the difference was not statistically significant (P = .29, Fisher).

RBM Thickness and Inflammatory Cells 

The endobronchial biopsies from the carina were taken from 1986 to 2004. The mean RBM thickness increased with age (Table IV), and the mean increase between the ages <3 years and >7 years was 27% (P = .032, paired t test). Inflammatory cells were found in each age group (Table IV). There was a significant association between the T- and B-lymphocyte counts within the biopsy at each time point (P = .015; P = .007; P = .005, linear-by-linear trend test). The histological findings did not correlate with the current respiratory or esophageal symptoms, DDA, PFT, BR, FENO, or atopy.

Table IV. Thickness of reticular basement membrane and inflammatory cell counts in bronchial biopsies

		Age at the biopsy
		<3 years	3-7 years	>7 years
	Number of patients	11	21	21
	RBM, μm	2.2(0.3)	2.3(0.4)	2.7(0.6)†
	Eosinophil, score⁎	0.9(1.2)	1.1(0.7)	1.0(0.7)
	Mast cells, score	2.5(0.9)	2.5(0.9)	2.5(0.7)
	Neutrophils, score	2.2(0.6)	2.4(0.7)	1.8(0.6)
	Macrophages, score	3.4(0.7)	3.2(0.7)	2.6(0.8)
	T-lymphocytes, score	3.6(0.5)‡	3.5(0.6)‡	3.1(0.7)‡
	B-lymphocytes, score	2.7(0.9)‡	2.9(0.8)‡	2.2(0.9)‡
	Plasma cells, score	2.4(0.9)	2.4(0.9)	2.6(0.9)
	Dendritic cells, score	0.5(0.5)	0.3(0.5)	0.3(0.5)

Results (except number of patients) given as means (SD).

⁎ Score = positively stained cells, score 0-4 (0 = none; 1 = <10 cells; 2 = 10-30 cells; 3 = 30-50 cells; 4 = >50 cells). † Statistically significant increase of RBM between the first and last biopsies in 11 children (P = .032) with the paired t test. ‡ Significant association between the T- and B-lymphocyte counts in each biopsy at all points (P = .015; P = .007, P = .005) with the linear-by-linear trend test.

The histological findings of all 53 carina biopsies were analyzed retrospectively (J.L.). They were considered normal in 11 biopsies (21%), whereas mild bronchitis was found in 38 biopsies (72%) and moderate bronchitis was found in 4 biopsies (7%). The histological diagnosis did not correlate with DDA, PFT, FENO, SPT, or BR. The histological findings of simultaneous 49 esophageal biopsies were available from the case notes, and they were considered normal in 28 biopsies (57%). These histological findings did not correlate with histological findings from carina biopsies (Spearman correlation test).

The usefulness of the histological report on an endobronchial carina biopsy to the clinician was evaluated from the patient charts. This was based on the comments and actions taken by the clinician, usually a pediatrician (eg, the use of inhaled corticosteroids, request for PFT, and a decision to stop the follow-up visits). An endobronchial carina biopsy was considered to have been clinically useful in 27 of 53 cases (51%).