INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: nitric oxide; lung transplantation; bronchiolitis; organ rejection.

Lung transplantation now provides a realistic opportunity for prolonged survival and a better quality of life in selected patients with end-stage pulmonary disease (1). Survival rates beyond 5 yr remain disappointing compared with those achieved by recipients of other solid organ transplants. The development of fixed small airways obstruction in the form of obliterative bronchiolitis (OB) is the commonest cause of late graft failure and contributes to poor long-term survival (1). OB is a manifestation of chronic allograft rejection and there is evidence that repeated episodes of early acute rejection in the graft predisposes to the subsequent development of OB (2). The histology of acute rejection shows infiltration of the graft with predominantly CD8-positive lymphocytes. This infiltration may be localized to the perivascular region or involve the bronchial and bronchiolar epithelium and submucosa to produce a lymphocytic bronchiolitis (LB) (3). There is increasing evidence that the lymphocytic airway inflammation component of acute rejection contributes to the development of OB in animal models (4) and in humans (5).

Nitric oxide (NO) is implicated in the pathophysiology of airways disease (8). The upregulation of the inducible form of NO, synthase (iNOS) in airway epithelial cells is associated with the increased production and prolonged release of NO, which can modify immunologic and inflammatory responses occurring in the airway (9). The concentration of NO in exhaled breath (eNO) has been shown to reflect levels of NO found in the lower airways (10). Compared with healthy subjects, eNO is increased in diseases associated with airway inflammation such as asthma (11, 12). Furthermore it has been shown that the elevated eNO levels seen in asthma decrease after treatment with inhaled anti-inflammatory drugs (12). We have recently shown in a cross-sectional study of 104 lung transplant recipients, significantly higher concentrations of eNO in recipients with a concurrent diagnosis of LB compared with those who were clinically well whereas patients with acute vascular rejection alone had no increase in eNO measurements (13). Current management strategies for OB include reduction of acute rejection rates or alteration of immunosuppression when there is established OB. We hypothesized that identification of those at risk of OB and early intervention with targeted therapy may help reduce the impact of OB on our recipient population. Early detection of subclinical graft airway inflammation may allow prompt treatment potentially preventing the subsequent progression to OB. In this study, we aimed first to determine whether an increase in an individual's eNO measurement provides a useful noninvasive marker of LB. Second, we assessed in recipients who developed LB, the effect of inhaled corticosteroids as a topical anti-inflammatory treatment on eNO concentrations and graft function.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Groups

Between June 1987 and May 1997, the Freeman Hospital Cardiopulmonary Transplant Unit had performed 216 lung transplants in 212 recipients. eNO measurements were performed on all recipients who attended for routine or urgent review between June 1997 and July 1998. The health authority ethical committee approved this study.

Recipients were considered clinically stable if they remained well throughout the study period without a significant change in lung function and no evidence of infection or rejection at bronchoscopy. Bronchoscopy, bronchoalveolar lavage (BAL), and transbronchial biopsies were performed routinely at 1 wk, 1, 3, 6, and 12 mo posttransplant, and whenever clinically indicated. NO measurements were made before each biopsy. Lavage specimens were cultured for the presence of bacteria, viruses, and fungi. Patients with active infection were excluded. Transbronchial biopsy specimens were graded for the presence of acute vascular rejection, LB, or OB using current guidelines (14). Acute rejection was considered significant if it was graded A2 (mild) or higher.

The development of new clinical symptoms, a decrease in pulmonary function, or the development of radiographic changes was investigated with BAL and transbronchial biopsy. Recipients who had an active infection, proven on BAL or throat swabs, were excluded owing to the effect infections have on eNO values (15). Recipients with characteristic histologic changes of OB on biopsy or who developed bronchiolitis obliterans syndrome (BOS) (14) were not considered to have reversible graft dysfunction and were excluded from the conclusion.

NO Analyzer and NO Measurement

All laboratory measurements were in accordance with British Thoracic Society guidelines on lung function testing (16). eNO was measured using a method as previously described (17). We used a rapid, highly sensitive chemiluminescence analyzer (LR2000, version 2.2; Logan Research, Rochester, U.K.) with a resolution of 0.3 part per billion (ppb) of NO and response time (0-95% rise time) of 0.4 s. The analyzer also measured CO₂ (resolution 0.1% CO₂, response time 0.2 s) by single-beam infrared absorption, mouth pressure, exhaled flow, and exhaled volume. The sampling rate was 250 ml/min for all measurements. At this flow rate the delay time was 1.4 s for the CO₂ analyzer and 1.8 s for the NO analyzer. The analyzer was calibrated daily using medical grade NO at a concentration of 106 ppb in nitrogen (BOC special gases; Surrey Research Park, Guilford, U.K.) and certified 7.9% CO₂ (Cryoservice; Worcester, U.K.). Mouth pressure and flow were calibrated using a water manometer and timeter (Timeter Instruments Corporation, St. Louis, Missouri) respectively. The technique for single-breath NO measurements used has been previously validated as a reproducible marker of lower airway NO production in lung transplant recipients over a wide range of lung function (17). Briefly, after full inspiration seated subjects were asked to exhale slowly from total lung capacity through a narrow Teflon coated tube. Recordings were made of exhaled NO (ppb), CO₂ (%), mouth pressure (cm H₂O), and exhaled volume (liters). Subjects wore a nose-clip and maintained mouth pressure at 4 to 5 cm H₂O by using a biofeedback visual display. At this pressure, the internal nasal route is closed off by the soft palate, eliminating nasal NO contamination of the exhaled gas mixture (18). A constant expiratory flow (250 ml/s) was obtained as eNO concentrations are highly flow-dependent (19). The procedure was repeated until three technically acceptable measurements were obtained. This method has been shown to be reproducible and a valid marker of lower airway NO production in transplant patients (17). The measurement technique was performed in agreement with recommendations on exhaled NO measurements, published by the European Respiratory Society since this study was completed (20). eNO measurements were recorded at end-exhalation at the point where exhaled CO₂ reached a plateau value. The mean of the two closest measurements was recorded as the eNO value in accordance with guidelines (16).

Pulmonary Function Measurements

FEV₁ and FVC were performed, in each of the lung transplant recipients immediately before each eNO measurement. Forced expiratory flow between 25% and 75% of the vital capacity (FEF 25-75%) was derived from an expiratory flow-volume curve using a pneumotachograph (Sensormedics, Yorba Linda, CA) from three repeatable recordings.

Immunosuppression

All patients received combination triple therapy immunosuppression as tolerated, comprising prednisolone at a dosage of  0.2 mg/kg/d, azathioprine, and cyclosporine. Acute rejection of grade 2 or greater was treated with pulsed methyl prednisolone followed by oral prednisolone 1 mg/kg/d reducing to baseline levels over 30 d. Patients with isolated LB were treated with inhaled budesonide 800 µg twice daily via a turbuhaler and received no empirical antibiotic therapy. Recipients with both LB and acute rejection were treated as for isolated acute rejection.

Statistics

Obtained eNO concentrations were not normally distributed even after log transformation and therefore nonparametric testing was used. Group results are expressed as median and range. eNO levels during episodes of acute graft dysfunction were compared with baseline and posttreatment values for the individual by Wilcoxon signed-rank test. Serial NO measurements in patients who remained clinically well were compared by the method of Bland and Altman to assess reproducibility (21), and this was expressed as the coefficient of reproducibility.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A total of 120 lung transplant recipients attended for review on more than two occasions; 43 remained clinically well throughout the study period. The range of eNO values in this group was 1.25 to 9.7 ppb, median 4.65 ppb, coefficient of reproducibility = 2.36 ppb, suggesting that an increase above 3 ppb is outside assay variability. Seventy-seven recipients experienced deterioration in their graft function during the study period. Nine patients had infections identified as the cause of their graft dysfunction and were excluded from analysis. Five patients had previously been treated with inhaled steroids for LB and were also excluded. A further eight patients were excluded from analysis because of prior diagnoses of neoplasia (posttransplant lymphoma n = 4, other n = 1), recurrent A2 rejection (n = 1), or nonspecific pneumonitis or cryptogenic organism, pneumonia (COP) (n = 1 each). Thirty-four patients developed OB or BOS and were considered to have irreversible graft dysfunction. The remaining patients had histologic evidence of reversible graft dysfunction, 14 with isolated LB, 2 with LB and acute rejection, and 5 with isolated acute rejection.

Fourteen recipients developed isolated LB during the study period and had eNO measurements performed before the diagnosis of LB (nine had two or more premorbid eNO levels). This allowed changes associated with the diagnosis of LB to be compared with each patient's own baseline values. Nine patients had LB diagnosed at routine biopsy whereas five patients were found to have LB at symptom-driven biopsy. Patients with isolated LB demonstrated a significant increase in eNO at the time of diagnosis compared with when well (Figure 1). Median (range) eNO ppb when well was 4.33 (1.0 to 10.76), rising at diagnosis to 10.9 (4.6 to 48), p = 0.008. The median increase in eNO was 4.75 (range 0.17 to 34 ppb). The recipients with isolated LB had NO measurements performed at review after treatment with inhaled steroids for at least 1 mo (range 1 to 2.5 mo). They demonstrated a decline in eNO back to baseline levels, 4.85 (2.6 to 15.9), p = 0.002 (Figure 1). Pre-LB and posttreatment eNO concentrations had no statistical differences (p = 0.19). There was no trend toward higher eNO concentrations in those who had symptom-driven biopsy. There was also no trend toward higher eNO concentrations in those patients who had suffered the greater declines in FEV₁. No significant differences in cyclosporine trough concentrations or oral prednisolone doses were noted comparing prediagnosis levels and those at diagnosis (p = 0.4). Median cyclosporine concentration at diagnosis of LB was 280 ng/ml and prednisolone dose 10.5 mg, which was not significantly different from prediagnosis, or posttreatment levels (p > 0.05).

View larger version (23K): [in this window] [in a new window]   Figure 1.   Change in eNO in 14 recipients with isolated LB after treatment with inhaled steroid therapy (n = 14). Recipients showed a significant increase associated with the diagnosis of LB (p = 0.008) and after treatment with inhaled steroids showed a significant decrease in eNO (p = 0.002).

One patient did not have an increase in eNO at the diagnosis of LB. On reviewing this patient's series of eNO measurements, the prediagnosis level was taken 2 wk before the biopsy diagnosis of LB and was higher than preceding eNO levels. The failure of eNO to rise in this patient concurrent with the diagnosis of LB may, therefore, be related to a delay in the diagnosis of LB.

Deterioration in lung function was associated with the diagnosis of LB. A decrease from baseline FEV₁ liters 2.28 (1.36 to 4.45) to 1.91 (0.92 to 4.55) was noted (p = 0.004). After treatment with inhaled steroids a significant increase in FEV₁ (liters) from diagnosis of LB was noted 2.32 (1.06 to 4.61), p = 0.02 (Figure 2). There was no statistical difference in FEV₁ prediagnosis to posttreatment p = 0.79. Tests of small airway function using FEF_25-75% displayed a similar pattern with a decrease in FEF from 2.25 L (0.84 to 8.12) to 1.71 (0.46 to 5.69). After treatment there was an increase in FEF_25-75% although this did not return to baseline 1.73 (0.66 to 6.76), p = 0.08. 

View larger version (21K): [in this window] [in a new window]   Figure 2.   Change in FEV1 in recipients with isolated LB after treatment with inhaled steroid therapy (n = 14). Recipients showed a significant decrease in FEV1 associated with the diagnosis of LB (p = 0.004) and after treatment with inhaled steroids showed a significant increase in FEV1 (p = 0.02).

It is noteworthy that in view of the multiple statistical testing between the time points of prediagnosis, at diagnosis and posttreatment the highly significant p values remain significant after statistical adjustment for testing, for example, using the Bonferroni correction.

In the two recipients with LB and acute rejection, measures of eNO were obtained after systemic augmented immunosuppressant therapy. There was no significant change in their exhaled NO concentrations after treatment (mean NO 5.4 ppb compared with at diagnosis 6.9 ppb, p = 0.5) nor after treatment with oral corticosteroids (p = 0.25) (Figure 2). The five patients with isolated acute vascular rejection had no change in their exhaled NO at diagnosis or after oral augmentation compared with baseline (p = 0.38 and 0.12 respectively).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

LB has been suggested as a cause of acute deterioration graft function (22) and as a precursor to subsequent OB (5). In this study, the development of LB was associated with a significant increase in concentrations of exhaled NO, supporting previously described evidence that exhaled NO is a marker of airway inflammation (13). Our data also suggest that exhaled NO levels show little variation in lung transplant recipients during periods of clinical stability. Changes in exhaled NO reflect the development of pathology in the graft and may be used as an aid to clinical management. Increased expression of iNOS, an inducible isoenzyme responsible for NO production, has been demonstrated during rejection episodes affecting other solid organ transplants (23). The upregulation of iNOS expression in the airway epithelial cells in response to the host T lymphocytes would be a possible explanation for the increased exhaled NO in LB. The specificity of increased NO in exhaled air in reflecting airway inflammation is confirmed by the lack of an increase in patients with acute vascular rejection alone and correlates with previous results (17). This contrasts to previous work suggesting an increase with eNO associated with acute rejection although exclusion of coexistent LB or airway inflammation in this publication was not clear (24). All eNO measurements and lung function measurements were taken contemporaneously. In some patients with serial eNO measurements taken before this study, we have noted a rise in eNO before a decline in lung function associated with LB.

A decrease in lung function was associated with the diagnosis of LB and correlates with previous work (22). Previous findings also demonstrated the failure of high-dose systemic steroids to improve lung function in isolated LB-affected patients. The sole addition of inhaled steroids to systemic immunosuppression resulted in a return to baseline eNO values and reversed the lung function changes associated with the diagnosis of LB. This suggests that inhaled corticosteroids have sufficient topical anti-inflammatory effects to reduce airway NO production, as it does in asthma (12, 25). In two patients who underwent serial testing within 2 wk of the commencement of inhaled therapy, normalization of exhaled NO concentrations fell to normal levels and predated the recovery in pulmonary function. NO measurements may therefore be more sensitive than pulmonary function in reflecting the presence of airway inflammation resulting from LB (data not shown). There was no trend toward a higher eNO concentration in those patients who were symptomatic or those with the greatest decrease in lung function.

Recipients who had both LB and acute rejection received oral immunosuppression, but failed to show a significant decrease in their exhaled NO levels after this. Although the numbers are small it suggests that oral immunosuppression may not be as effective as inhaled steroids in controlling the airway inflammation of LB after lung transplantation. These findings are in agreement with a previous study in heart-lung transplant recipients which showed that nebulized budesonide appears to offer protection against the development of OB in recipients with recurrent episodes of acute rejection (26).

Serial single-breath eNO measurements may provide a simple, noninvasive method of early detection of LB. This would allow augmented anti-inflammatory treatment to be commenced with the hope of delaying or preventing the progression to OB. Lung transplant recipients already receive oral prednisolone; however, the targeting of anti-inflammatory treatment to the airways may have an additional benefit over and above that achieved by systemic treatment. The addition of inhaled steroids to the immunosuppressant regimen may have two benefits. First, it may reduce the risk of graft from the development of LB and subsequent progression to OB. This strategy, used in association with other ongoing attempts to modify immunosuppressant regimens, such as the role of nebulized cyclosporine, could improve the way we protect the lung from rejection. Larger studies involving randomized placebo-controlled use of inhaled steroids in transplant recipients are required to validate our findings.

The efficacy of inhaled steroids in improving clinical symptoms, lung function, and lowering exhaled NO concentrations in this group of recipients with LB has important implications reaching beyond transplantation. This study lends support to the concept that targeting treatment to the airway in conditions manifest by airway inflammation has a significant benefit over that achieved by systemic treatment.

The development of OB remains the single most important long-term complication limiting the survival and quality of life of lung transplant recipients. Although further studies are required, our findings suggest a potential role for exhaled NO measurements in guiding immunosuppressant therapy and aiding in the early detection, prevention, and management of OB.