INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Previous reports have documented the occurrence of bronchial hyperreactivity in patients who have undergone lung or heart-lung transplantation (1). However, studies have produced conflicting results as to the incidence and significance of this finding. Problems with the existing published series include small sample size, comparison of patients undergoing different transplant procedures, and failure to compare at a uniform time posttransplantation. Examination of repeated measurements of bronchial hyperreactivity over time with long-term follow-up has not been described to date, nor have attempts been made to correlate the presence of hyperreactivity with significant clinical outcomes such as chronic rejection.

Lung allografts frequently develop clinical signs of chronic dysfunction resulting from graft rejection. A working formulation has been developed to establish clinical and pathological criteria indicative of the occurrence and severity of this (11). The categorization of sustained declines in forced expiratory volume in 1 s (FEV₁) with or without histologic evidence of bronchiolitis obliterans is referred to as bronchiolitis obliterans syndrome (BOS) and represents an important clinical and prognostic end point in lung transplant recipients.

We hypothesized that changes in lung physiology such as the presence of bronchial hyperreactivity may indicate a pathologic change with potential adverse consequences. We therefore undertook to examine the natural history of bronchial hyperreactivity and its relationship to the development of BOS in a cohort of lung transplant recipients at The Toronto Hospital (Toronto, Canada).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Population

Data were reviewed from all patients undergoing double lung transplantation at The Toronto Hospital between February 1988 and May 1994. Patients receiving single lung or heart-lung transplants were excluded. All transplants were performed by the sequential single transplantation technique with the exception of the first 12, for whom transplantation predated the introduction of this technique.

Surveillance Protocol

All patients recorded daily home FEV₁ from a portable electronic spirometer. Laboratory-based spirometry was requested monthly and when clinically indicated. Bronchoscopies with bronchoalveolar lavage and transbronchial biopsies were performed within 2 wk posttransplant and at 6 wk and 3, 6, 9, 12, 18, and 24 mo, as well as yearly thereafter. Bronchoscopies and biopsies were also performed whenever clinically indicated (i.e., 10% decline in FEV₁, new chest X-ray finding, or progression of respiratory symptoms and signs).

Methacholine Challenge Testing

Methacholine inhalation challenge tests were performed according to the method described by Hargreave and co-workers (12). Results were expressed as the provoking concentration of methacholine required to decrease baseline FEV₁ by 20% (PC₂₀). A PC₂₀  8 mg/ml was interpreted to represent a positive test and to indicate the presence of nonspecific bronchial hyperreactivity (NSBHR).

Methacholine challenge testing was requested at 3, 6, 9, 12, 18, and 24 mo, and annually thereafter, as part of routine posttransplantation follow-up. Testing was not performed if patients were deemed to be clinically unstable at the scheduled time. Methacholine challenge testing has not been routinely requested since 1994.

Bronchiolitis Obliterans Syndrome

The diagnosis of bronchiolitis obliterans syndrome (BOS) was established according to the system of Cooper and colleagues (11).

Data Analysis

Time from transplantation to occurrence of a positive methacholine challenge and time to development of BOS were determined from actuarial survival curves according to the method of Kaplan and Meier (13). Comparison between groups was made with a Cox proportional hazards model. The Student t test was used to compare patient characteristics between groups. PC₂₀ was evaluated as a predictor of BOS at prespecified time intervals using the ² statistic; sensitivity, specificity, and positive and negative predictive values were calculated. Where applicable, values are expressed as means ± standard error. Results were deemed to be significant at p < 0.05 in all cases.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Demographics

One hundred eleven consecutive patients undergoing double lung transplantation were identified. Methacholine challenge data were available for 85 patients; 23 of the remaining 26 died in the first few months after transplantation and thus were never tested. Of the 85, 51 were men and 34 were women; their mean age was 41.7 ± 12.3 yr. The most common underlying disease for which transplantation was performed was cystic fibrosis (35%), followed by emphysema (27%) and ₁-antitrypsin deficiency (11%). The remaining preoperative diagnoses included bronchiectasis, pulmonary fibrosis, primary pulmonary hypertension, Eisenmenger's syndrome with atrial septal defect, bronchiolitis obliterans, lymphangioleiomyomatosis, sarcoidosis, and eosinophilic granuloma.

Methacholine Challenge Testing

Results of methacholine challenge tests performed at each time interval between 3 and 48 mo are shown in Table 1. Many patients completed testing at some, but not all, requested time intervals; the overall number completed ranged from one to nine for each individual, with 75% undergoing between three and seven tests. Patient noncompliance to the request for methacholine challenge testing was the most common reason for incomplete follow-up. A positive methacholine challenge occurred in 18 of 60 patients (30%) tested at 3 mo and in 14 of 59 (24%) at 6 mo, with similar results observed at later testing intervals.

The majority of patients showed substantial individual variation in PC₂₀ results over time. Many produced a positive test at one session and a negative test on the following visit or vice versa. Among 70 patients completing at least 3 separate tests, 10 (14%) had a PC₂₀  8 on all occasions, 7 (10%) did so more than 50% of the time, 21 (30%) did so less than 50% of the time, and 32 (46%) never had a positive challenge. For those who manifested bronchial hyperreactivity on more than one occasion, wide variation was also seen in the severity of individual hyperreactivity as reflected by the PC₂₀ value.

Further analysis was therefore performed with data at each time interval included only if the patient had not missed a prior scheduled test (Table 2). The observed proportion of patients having exclusively negative methacholine challenges was 61.9% at 12 mo, with similar results observed among patients monitored to 18 and 24 mo. Among 21 patients with a complete set of four tests at the end of 12 mo, 13 had no positive challenges, 4 had only one, 2 had two, and 2 had three; no patient had positive challenges on all four occasions.

The true proportion with persistently negative challenges, however, would tend to be overestimated by the above-described analysis since data could be collected only from patients able to be present for testing at each time period. The probability of having exclusively negative challenges was therefore determined by the Kaplan-Meier method (Figure 1). By this approach, it was estimated that 43.9% of patients would never have a positive methacholine challenge; of the remainder who had at least one positive challenge, this occurred by 3 mo in 53% and by 6 mo in an additional 23%.

View larger version (7K): [in this window] [in a new window]   Figure 1.   Kaplan-Meier plot of methacholine challenge results over time in patients after lung transplantation. The probability of a patient having exclusively negative challenges up to each given time period is indicated.

Methacholine Challenge and BOS

The pattern of methacholine challenge responses in patients with a complete set of tests at 12 mo was compared with subsequent development of BOS. Of the two patients with three of four challenges positive, one developed BOS at 16 mo posttransplant; for the other, only 12 mo of follow-up data were available. Of the two patients with two of four challenges positive, one developed BOS at 14 mo, while the other did not develop BOS within 25 mo of follow-up. Among the 13 patients with exclusively negative challenges, 4 developed BOS (all between 22 and 25 mo posttransplant) while 9 did not develop BOS within the entire duration of follow-up.

The entire cohort of patients was divided into two groups based on the occurrence of a positive or a negative methacholine challenge at 3 mo posttransplantation (Table 3). The positive methacholine group had a somewhat higher proportion of women than did the negative group. As well, mean baseline FEV₁ was lower in the positive group. This finding persisted when FEV₁ was expressed as a percentage of the predicted value (81 versus 110%, p = 0.0002) and thus was not accounted for by differences in sex or body size.

Acute rejection was defined as present if evidence of acute rejection grade 1 or higher was found on transbronchial biopsy at any time in the 3 mo before the first methacholine challenge or on the next biopsy after the challenge study if this occurred within 30 d. Accordingly, acute rejection had been documented on biopsy in 85% of patients with positive methacholine challenge but was also seen in 58% of patients with negative challenges; the difference was not significant (p = 0.77). These results did not differ when analysis was limited to biopsies performed within 30 d of methacholine challenge. Other findings on biopsy included cytomegalovirus (CMV) infection in four patients, lymphocytic bronchitis or bronchiolitis in two, and posttransplant lymphoproliferative disorder in one. Only one of these patients had a positive challenge. Matching of pretransplant CMV antibody status in donor and recipient was not different between groups.

Medication use was similar in both groups. Only one patient was using an inhaled -agonist at the time of methacholine challenge; her challenge was positive. Two other patients, both with cystic fibrosis, received a -agonist in conjunction with inhaled tobramycin therapy; both had negative challenges. Although our methacholine challenge protocol routinely instructs patients to avoid -agonists within 12 h of the test, confirmation that this had in fact occurred in these cases was not available from the records obtained. One patient was taking an inhaled steroid and an oral antihistamine and had a positive challenge. Of five patients using nasal steroids, one had a positive challenge; one additional patient was using an anticholinergic nasal spray and had a negative challenge.

Patients in the positive group were significantly more likely to develop BOS than those with negative challenges at 3 mo (mean time to BOS, 16.9 ± 2.2 versus 43.9 ± 5.8 mo; p < 0.006), with all 18 of the former progressing to BOS during the study period (Figure 2). Results were still significant after adjustment for sex (p < 0.02). For patients in the positive group, the proportion of overall tests that were positive (100%, more than 50%, or less than 50%) did not significantly influence time to BOS.

View larger version (8K): [in this window] [in a new window]   Figure 2.   Time to development of bronchiolitis obliterans syndrome (BOS) in patients with positive and negative methacholine challenges 3 mo after transplantation. The solid line indicates patients with a positive challenge (PC20  8). The broken line indicates patients with a negative challenge (PC20 > 8). Differences between the curves are significant (p < 0.006).

Correlation between methacholine challenge result at 3 mo and development of BOS by 12, 18, and 24 mo was assessed (Table 4). Positive challenge was a significant predictor of progression to BOS for all of the above-described time intervals. This test displayed only intermediate sensitivity but had intermediate to high specificity for the outcome of interest.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study represents, to our knowledge, the largest series to date describing methacholine challenge testing after lung transplantation and is the first to examine the results of serial testing over time. Several findings are particularly noteworthy. A significant proportion of patients manifested bronchial hyperreactivity to methacholine posttransplantation, with an observed incidence of 30% at 3 mo; however, many patients manifested no hyperreactivity at any follow-up interval. Among patients who had at least one positive methacholine challenge, the majority did so early, with more than half positive at 3 mo and three-quarters positive by 6 mo. For a given individual, significant variability in the PC₂₀ was seen on repeat testing over time. All patients who completed four tests as scheduled within the first 12 mo had at least one negative challenge during that period. The mean time to development of BOS was 27 mo earlier in the group with positive methacholine challenges at 3 mo compared with those with negative challenges.

Ten previous studies have reported the results of methacholine challenges in lung transplant patients (Table 5). The earliest six of these all found increased bronchial hyperreactivity to methacholine in the majority of patients, most of whom underwent combined heart-lung transplantation (1). In contrast, Herve and colleagues (8) found methacholine hyperreactivity in only 1 of 13 patients tested. The remaining, more recent, studies (9, 10), including the largest reported series preceding ours, showed intermediate results, with 40% of patients demonstrating a positive methacholine challenge; our findings are most consistent with the incidence seen in these latter series.

The incidence of positive methacholine challenges did not appear either to increase or decrease significantly with increasing time posttransplantation. For a given individual, however, substantial variability in the PC₂₀ value was seen, although there was no consistent pattern to this variability. This suggests that bronchial hyperreactivity is a phenomenon related to the individual patient and not a general consequence of the lung transplantation procedure. Previous data on individual responsiveness to methacholine over time is extremely limited, consisting of three trials that include reports of small numbers of patients tested on two separate occasions. Variable responses were observed in three patients (4) and two patients (7) studied in this fashion. Glanville and colleagues (6) reported seven patients who were challenged twice, 6 to 24 mo apart; one patient showed no hyperreactivity at 2 mo but had a positive methacholine challenge at 14 mo. These results are consistent with our observations.

Possible explanations for posttransplant methacholine hyperreactivity have included denervation hypersensitivity (1, 3- 6), epithelial damage due to inflammation or rejection (1, 3), modification of pulmonary noncholinergic nonadrenergic innervation (1, 3, 5), changes in the properties and clearance of mucus (1), decreased baseline airway caliber (1, 7), disruption of lymphatic channels and blood supply to the lungs (5), and effect of immunomodulatory agents or other drugs (1, 5). The mechanism to which the most attention has been paid is the suggestion that the loss of tonic vagal stimulation to the bronchi results in hypersensitivity of muscarinic receptors. Histologic evidence of denervation is seen in the lung allograft posttransplantation (14) and these changes appear to be persistent over time (4). As well, certain pulmonary reflexes have been observed to be abolished (2, 10, 15).

However, an in vitro study of bronchial tissue from heart- lung transplant recipients has revealed normal cholinergic responsiveness, no change in receptors, and normal function of postganglionic cholinergic nerves (16). Furthermore, if bronchial hyperreactivity results from denervation hypersensitivity, it is not clear why this should occur in some patients and not others, since all transplanted lungs are denervated. Neither is this mechanism consistent with our observations of variability in individual PC₂₀ results from one testing period to the next. Higenbottom and colleagues (4) have attributed the nonuniformity of methacholine responsiveness to individual differences in susceptibility to denervation. Evidence of such differences in lung transplant patients is lacking, however, and no pathophysiologic explanation as to why such differences might occur has been proposed. In addition, the possible clinical implications of variable susceptibility to denervation remain unexplored; if this is indeed the mechanism by which the observed bronchial hyperreactivity is mediated, it could merely represent an epiphenomenon of another process in the lung allograft.

The finding that patients with a positive methacholine challenge 3 mo after transplantation progress to bronchiolitis obliterans syndrome more than 2 yr sooner on average compared with those with a negative challenge is particularly striking. Such a correlation suggests that posttransplantation hyperreactivity is a manifestation of a significant pathologic process in the lung allograft and represents an association of bronchial hyperreactivity with important long-term clinical consequences. It is noteworthy that in a previous study, a response to bronchodilator at low lung volumes, which suggests the presence of small airways hyperreactivity, also predicted the development of BOS (17).

In the past, methacholine challenge results have been compared with lung biopsies at the time of challenge, but not with the development of subsequent abnormalities. Maurer and colleagues (5) noted chronic inflammatory changes in one patient who had a concomitant positive methacholine challenge, although this did not differentiate the patient from five others with normal biopsies who also had positive challenges. The Papworth group has published three studies (4, 9, 10) in which transbronchial biopsies were performed close to the time of methacholine challenge and in each case has concluded that no correlation existed between pathology and hyperreactivity. However, in the most recent of these studies (10), five of six patients with a positive challenge had evidence of rejection on the accompanying biopsy, compared with only three of nine patients with a negative challenge; performance of a ² test on these data yields a p value of 0.057, which approaches the level of conventional significance. The same trial included a biopsy from a patient with a positive methacholine challenge that revealed bronchiolitis obliterans. It is noteworthy that the study of Herve and colleagues (8), in which hyperreactivity was present in only 1 of 13 patients, explicitly selected only individuals whose biopsies revealed normal histology. Finally, it must be noted that transbronchial biopsy is subject to sampling error and thus cannot rule out the presence of rejection with certainty, a consideration with which the previous studies must be interpreted.

Most patients who developed bronchial hyperreactivity in our series did so in the first few months after transplantation. It is possible that in these patients, early changes of graft rejection may be occurring to produce this hyperreactivity. Nearly all the patients with positive methacholine challenges had shown prior evidence of acute rejection on biopsy. However, acute rejection was also common among those with negative challenges. While this does not rule out the possibility that acute rejection may have contributed to the bronchial hyperreactivity, it clearly cannot be the sole explanation.

Patients in the positive methacholine group had significantly lower baseline FEV₁ measurements on average at the time of testing. It could be argued that these patients might have been more likely to progress to BOS on this basis alone. However, baseline FEV₁ and PC₂₀ were subsequently examined by least-squares linear regression and were found to correlate poorly (r² = 0.13). In addition, a PC₂₀  8 was found to be a significant predictor of BOS in our study, although the sample size did not permit stratification for level of baseline FEV₁. Nonetheless, we cannot exclude the possibility that some of the observed differences in outcome between the two groups may be attributable to the difference in FEV₁. If so, decreased FEV₁ and increased bronchial hyperreactivity may both be measurements of the same pathophysiologic phenomenon in lung transplant recipients. However, such a phenomenon should also be related to subsequent development of BOS. In our study, the difference in FEV₁ could not be attributed to known risk factors for BOS such as acute rejection, CMV infection, or lymphocytic bronchiolitis, although other factors such as ischemic airway injury or HLA mismatch were not evaluated.

Other causes of nonspecific bronchial hyperreactivity, such as asthma, must be considered in interpreting our findings. The development of clinical asthma in the previously nonasthmatic recipient of lungs from an asthmatic donor has been reported (18). In most cases, information regarding the medical history of the donors was not sufficient to rule out the presence of asthma reliably. However, even in previous series in which donors with asthma were excluded, bronchial hyperreactivity still occurred (4, 9). Also, the incidence of bronchial hyperreactivity in our cohort is significantly higher than the expected prevalence of asthma in the population of donors. Thus, although an asthmatic predisposition of the lung allografts cannot be excluded, it is highly unlikely to account for our results. Furthermore, this would not explain the occurrence of early BOS in these patients, as there has been no indication to date that lungs from asthmatic donors are prone to developing BOS at a faster than expected rate.

Our study differed from those previous in that it exclusively incorporated patients who had undergone double lung transplantation. All previous series combined include a total of only 12 recipients of double lung as opposed to heart-lung transplants. Double lung transplantation involves a different operative technique with less mediastinal dissection compared with the heart-lung procedure. As has been suggested by Ernst and colleagues (7), this may have some impact on bronchial hyperreactivity. However, the more recent series of heart- lung transplant patients report an incidence of positive methacholine challenges similar to ours, suggesting that differences in technique are not responsible.

In summary, we have described the results of long-term serial methacholine challenge testing in a large cohort of patients receiving double lung transplantation. Only a portion of lung transplant patients develop bronchial hyperreactivity to methacholine and in such patients, it appears to be a transient or variable phenomenon. Patients with a positive methacholine challenge at 3 mo appear to be at significantly increased risk of developing early BOS. If intervention for early BOS is clinically important, methacholine challenge testing soon after lung transplantation could be considered in all patients as a component of routine postoperative surveillance. Further investigations or increased treatment may be indicated in patients with a positive challenge. Additional studies to elucidate the significance of and appropriate response to posttransplantation hyperreactivity are warranted.