INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: cystic fibrosis; lung pathology; clinical findings

It has been more than 60 years since Dorothy Andersen first described the pathologic features of cystic fibrosis (CF) (1). In CF, lung involvement is universal beyond the first few months of life, and is the most important cause of morbidity and mortality (2, 3). The lungs of CF patients older than 4 months are reported to have bronchial mucus plugging, inflammation, and bronchiectasis at autopsy (4). Most pathologic descriptions of the CF lung have focused on the presence or absence of structural abnormalities (2, 4). Few studies defined the range of the various pulmonary alterations in CF, especially with regard to the degree of bronchiectasis and small airways disease (5, 6).

The severity and progression of CF pulmonary disease is commonly assessed with pulmonary function tests and chest X-rays. Several studies showed that chest radiograph scores correlate with parameters of lung function, exercise tolerance, and general health in CF (7-10), but the correlation between the noninvasive tests and lung pathology is not known. Until recently, this correlation could be investigated only in CF patients who died. We now have the opportunity of examining the explanted lungs of CF patients who have undergone lung transplant surgery. To investigate the relationship between clinical findings and lung pathology, we reviewed the preoperative clinical findings and lung pathology of 21 children with CF who had lung transplantation at our hospital. We hypothesized that (1) patients with worse pulmonary function tests and chest X-ray scores will have worse airways inflammation and fibrosis, and (2) hypercapneic children with CF will have worse airways disease.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The medical records of 21 patients with CF who underwent lung transplantation at Childrens Hospital Los Angeles between 1994 and 1998 were reviewed. The study was approved by the Institutional Review Board (CCI) at Childrens Hospital Los Angeles. Data collected included demographics, method of chest physiotherapy, medications (including nebulized antibiotics, rhDNase, inhaled steroids), pulmonary function tests, arterial blood gases, and chest X-rays prior to transplantation. Hypercapnia was defined as PCO₂ > 50 mm Hg. Chest X-rays were scored using the Brasfield method by the same pediatric radiologist (11).

Pathology Scoring

Explanted lungs were examined by the same pediatric pathologist. The lungs were inflated with 3.7% neutral buffered formalin at a pressure of 24 cm of water for 24 hours. The lungs were cut along the hilar plane at the major segmental bronchi; multiple levels were exposed. Representative sections were submitted for routine histology, two sections from each lobe, each section to include airway, vasculature, and alveolated parenchyma. Sections were routinely processed for paraffin embedding and 5-µm sections were stained with hematoxylin and eosin. Recut slides obtained from selected blocks for this study were also stained with trichrome and Verhoef-van Gieson stain. Sequential fields from upper lobes containing pleural surface were examined until 14-15 small airways were sampled. Airway diameters were measured using a calibrated eyepiece reticle. Small airways were defined as airways less than 2 mm in diameter. Mean small airway diameter, percentage of small airways less than 0.35 mm in diameter, and small airway density per square centimeter were measured as previously described (6, 12). Small airways were evaluated for neutrophilic lumen inflammation, neutrophilic and lymphocytic mural inflammation, and fibrous narrowing with each feature graded 0 to 4 (Table 1 and Figures 1 and 2). Large airways were defined as airways more than 2 mm in diameter. Significant plugging was defined as plugging of more than 50% of the airway lumen with inflammatory cells and/or mucus.

View larger version (162K): [in this window] [in a new window]   Figure 1.   + 4 Neutrophilic inflammation of a small airway (hematoxylin and eosin stain) (original magnification ×40).

View larger version (167K): [in this window] [in a new window]   Figure 2.   Fibrous narrowing of a small airway (hematoxylin and eosin stain) (original magnification ×40).

Statistical Methods

All results are expressed as mean ± SD when appropriate. Parameters were compared between children with CO₂ retention and without CO₂ retention using the independent Student's t test. Linear regression analysis and calculation of Spearman's coefficients of correlation were performed to assess relationship between age, pulmonary function tests, arterial blood gases, and chest X-rays, plotted against lung pathology scores. Statistical significance was set at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Characteristics

There were 12 males and 9 females. Mean age was 15.0 ± 3.8 years (range 9.6 to 24.1 years). Five patients (24%) had reactive airways disease, five patients (24%) had history of allergic bronchopulmonary aspergillosis, and three patients (14%) had Mycobacterium avium intracellulare. Eight patients (38%) were on nebulized antibiotics (five nebulized tobramycin, two high-dose nebulized tobramycin, and one nebulized colistin). All of the patients were using nebulized rhDNase. Six of the patients (28%) were also treated with inhaled steroids. None of the patients were on systemic steroids for at least 1 month before transplantation. Nine patients (43%) were using high-frequency chest compression, four patients (19%) were using a flutter valve, and two patients (10%) were using a positive expiratory pressure valve for chest physiotherapy. The remaining patients (28%) were receiving chest physiotherapy via percussion and postural drainage.

Prior to transplantation, all patients had severe changes on chest X-rays (Brasfield scores of 9.0 ± 3.2; range 4 to 15) and severe airways obstruction (FEV₁ = 25.6 ± 5.6% predicted, FEF_25-75% = 11.0 ± 4.5% predicted), and hyperinflation (residual volume [RV] = 341.8 ± 75.8% predicted). As might have been predicted, the number of patients and their severe lung disease did not allow for any correlation of lung pathology with the pulmonary function tests or chest X-ray scores. At the time of transplantation, arterial blood gases showed that all patients were hypoxemic (Pa_O2 = 64.2 ± 8.2 mm Hg) on room air, and five patients (24%) were hypercapneic.

Pathologic Features

The lungs showed a variety of abnormalities as described previously in CF patients (3, 13). Atelectasis, mucoid impaction, acute and chronic inflammation, bronchiectasis, cyst formation, and fibrosis were widespread. Vascular changes of pulmonary hypertension were commonly noted. In our study group, the small airway density decreased with age (Figure 3). Patients were grouped based on the presence and absence of hypercapnia. There were no significant differences in age, airways obstruction, hyperinflation, or chest X-ray scores between these two groups (Table 2). The proportion of small airways less than 0.35 mm in diameter in all small airways was significantly lower in the hypercapneic patients (25.8% versus 36.9% in hypercapneic and nonhypercapneic patients, respectively, p = 0.04) (Figure 4). There were no significant differences in proportion of lumenal neutrophils, neutrophilic and lymphocytic mural inflammation, or fibrous narrowing of the small airways between the two groups (Figure 5). When we compared the hypercapneic patients to the nonhypercapneic patients with regard to significant large airway plugging, there was also no significant difference between the two groups.

View larger version (12K): [in this window] [in a new window]   Figure 3.   Small airway density plotted against age. Correlation coefficient is shown. Small airway density decreased with age.

View larger version (11K): [in this window] [in a new window]   Figure 4.   Percentage of smallest airways (< 0.35 mm in diameter) in hypercapneic and nonhypercapneic patient groups. Results are reported as mean ± SEM.

View larger version (13K): [in this window] [in a new window]   Figure 5.   Pathology scores of small airway inflammation and fibrous narrowing in hypercapneic and nonhypercapneic patient groups. Results are reported as mean ± SEM. NLI = neutrophilic lumen inflammation, NMI = neutrophilic mural inflammation, LMI = lymphocytic mural inflammation, FN = fibrous narrowing. There is no significant difference between the two groups for any of the parameters shown.

In three patients (14%), fungal hyphae were identified within mucopurulent exudate in ectatic airways. Of these patients, one had clinical and laboratory findings of allergic bronchopulmonary aspergillosis, which was consistent with the pathologic findings, and the other two patients were considered to be colonized because they did not have any evidence of inflammation associated with hyphae or invasive fungal infection. One patient had budding yeasts and a small focus of fungal infection.

Of the three patients with positive preoperative cultures for atypical mycobacterial infection, one patient had necrotizing pulmonary granulomas that were negative for fungal or acid-fast organisms. One of those patients had rare granulomas within the parenchyma, and in the central peribronchial and in the central peribronchial regions, Mycobacterium avium intracellulare polymerase chain reaction was negative for this patient.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In our study group, small airway density decreased with age. The loss of small airways caused by obliterative fibrosing bronchiolitis may be the cause for decreased small airway density in older patients. Sobonya and Taussig have studied the lungs of nine CF patients and did not demonstrate loss of small airways by the technique of quantitating the number of small airways per square centimeter of tissue (6). Hogg and coworkers, using a different technique, studied the lungs of three children with CF, and demonstrated a loss of small airways in two of the three (14). The decreased small airway density in older patients could also be explained by the presence of more severe bronchiectasis. Bronchiectasis is described in virtually all patients with CF who survive infancy (2), but these studies have not documented the severity of bronchiectasis in different age groups. In one study of CF in adults, the severity of bronchiectasis was variable (4). In the study conducted by Sobonya and Taussig, bronchiectasis was not more severe in the older patients than in the younger patients (6). However, the youngest patient in our study was older than the three patients Sobonya and Taussig studied. We also did not expect this finding in our study group because these patients had lung disease severe enough to require lung transplantation.

The number of small airways per square centimeter of tissue in our study group exceeded the values previously reported for CF patients (6). This could be secondary to particularly tortuous small airways in our patient group with end-stage CF lung disease.

In patients with CF, thickened airway secretions with bacterial infection and mucus hypersecretion, airway mucosal edema and inflammation, and respiratory muscle weakness and fatigue, all contribute to development of respiratory failure. Obstruction of airflow in small and large conducting airways results in hypoxemia due to altered matching of ventilation and perfusion, and this results in hypercapnia due to alveolar hypoventilation. All patients in our study group had hypoxemia on room air, as would be expected for these patients who have received lung transplantation. When patients were grouped based on the presence and absence of CO₂ retention, the percentage of smallest airways (airways less than 0.35 mm in diameter) was significantly lower in CO₂ retainers. This may reflect dilatation of small airways in this patient group. Sobonya and Taussig showed that younger patients in their study group had dilated small airways evidenced by decreased percentages of smallest airways compared with those in the control children (6). In that study, older patients showed evidence of narrowing of small airways, either by a decreased mean small airway diameter or an increased percentage of smallest airways. Although our patients were sick enough to receive lung transplantation, they were not terminally ill. The lack of obliteration of small airways in our patient group compared with Sobonya and Taussig's study, maybe due to the fact that those results were obtained from biopsy specimens as opposed to the explanted lungs in this study.

Our sampling was from the upper lobes. In CF, lung lesions may be irregularly distributed. On chest radiographs of patients with CF, the upper zones frequently appear more severely involved (15, 16). Lobar atelectasis, mucoid impaction of bronchi, and air cysts are specific lesions reported to occur more frequently in the upper lobe of CF patients (17, 18). Tomashefski, and colleagues showed that CF in the lung is unevenly modeled, and demonstrated that chronic lung disease is quantitatively more severe in the upper lobe (18). The causes of this irregular distribution are unknown, but may be related to mechanical factors that favor mucus stasis and progressive dilatation of bronchi and cysts in the upper lung zones. Therefore, we feel that our samples would be representative of the lung of our patients with the end-stage CF lung disease.

In summary, in our patient group with end-stage lung disease, the degree of airway inflammation did not differ with age, and inflammation was no greater in patients with CO₂ retention. There was decreased small airway density in older patients, which may be caused by an increased number of dilated small airways as the patients get older. Patients with CO₂ retention also had more dilated small airways compared with non-CO₂ retainers, as evidenced by a decreased percentage of smallest airways.

It is clear that inflammation centered in the airways rather than the alveoli is the real Achilles heel of CF. Three processesairway obstruction, chronic endobronchial infection, and inflammationinteract continually, and ultimately result in lung destruction. Variations in the pathologic features of the CF lung are the result of a complex interaction of many factors, and noninvasive methods do not accurately predict the pathologic changes in the lungs. In Tomashefski's study, there was no significant correlation between volume proportion of parenchyma and age, duration of pulmonary symptoms, or chest radiograph score (19). Hence, there was no correlation between noninvasive clinical tests and lung pathology findings in children with end-stage CF lung disease. Severe airway inflammation was present despite the use of currently available therapeutic modalities. Thus, our findings support the need for effective antiinflammatory therapy and airway clearance at the earliest possible stage of the disease to delay CF lung damage.