After diagnosis of CLAD, we found no difference in survival according to the underlying disease before transplantation (P=0.56), neither between single versus sequential single lung recipients (P=0.52) nor between patients on different immunosuppressive regimes. All but four patients were on azithromycin treatment, because NRAD had been previously diagnosed as stated earlier (n=15) or because azithromycin therapy was initiated when CLAD was diagnosed (n=52), which is a common practice in our center. However, none of these 52 patients experienced a sustained FEV1 response more than 10%, excluding NRAD (3). Of the 71 patients, 17 were switched from azathioprine to mycophenolate mofetil at diagnosis and 67 were already on tacrolimus treatment, 4 remained on cyclosporine treatment; 55 patients were additionally treated with an oral steroid taper when CLAD was diagnosed. There was no survival difference neither in the mycophenolate mofetil group (P=0.17) nor in the tacrolimus group (P=1.0).
A mean of 4.0±2.5 (range 1–13) pulmonary CAT scans was performed per patient from time of CLAD diagnosis to death or most recent follow-up visit (up to June 2010). Of the 71 patients, 22 patients had persistent parenchymal infiltrates on CAT scan (31%). After CLAD was diagnosed, a mean of 12.8±11.3 pulmonary function tests per patient was found. In 51 of 71 patients (72%), the pulmonary function remained obstructive, compatible with BOS, and 20 patients (28%) developed a restrictive pulmonary function, suggestive for RAS, of whom 17 (85%) also developed persistent parenchymal infiltrates on CAT scan, whereas this was only detected in 5 patients (9.8%) in the BOS group (Fig. 1B).
Of these 20 RAS patients, 15 were diagnosed based on persistent TLC decline of more than 10%, whereas in 5 not enough TLC data were available to comply with the RAS definition as outlined by Sato et al. (8). However, these five patients experienced a decline in FEV1 of more than 20% and a simultaneous decline in forced vital capacity (FVC), so that the FEV1:FVC ratio remained normal or even increased above normal, which is also compatible with a restrictive pulmonary function defect (12).
When we analyzed the cellular differentiation in the broncho alveolar lavage (BAL) fluid at diagnosis of CLAD, we found no significant differences between patients with RAS or BOS. The median BAL neutrophilia and lymphocytosis was 4.2% and 6.0% in RAS versus 12.2% and 6.2% in BOS patients (P=0.35 and 0.41, respectively).
The actuarial survival post-CLAD (n=71) is shown in Figure 2A (median 26 months). There was no clear survival difference between males and females (P=0.82); however, there was a significant survival difference between patients who developed early (within 24 months after transplantation, n=35) compared with late CLAD (n=36) with a median survival of 20 versus 39 months (P=0.038, Fig. 2B).
Patients with RAS had a significantly worse survival compared with the ones with BOS (median 8 vs. 35 months, P=0.001, Fig. 2C). When we compared post-CLAD survival according to the presence or absence of persistent parenchymal infiltrates on CAT scan, there was a significant difference with a median survival of 14 versus 35 months (P=0.0058; data not shown). Patients who had previously been diagnosed with NRAD before their current diagnosis of CLAD (n=15/71) had a significantly worse survival (median 11 vs. 35 months, P=0.0086, Fig. 2D).
In univariate cox regression, the previous development of NRAD and development of a restrictive pulmonary function pattern were both associated with a significantly worse survival, whereas the early development of CLAD and the development of persistent parenchymal infiltrates on CAT scan showed a strong trend for a worse survival. Gender, age, type of transplantation (single vs. sequential single), and immunosuppressive treatment were not significant. In multivariate analysis, including all parameters with a P value less than 0.1 in the univariate analysis, early development of CLAD, previous NRAD, and a restrictive pulmonary function pattern were correlated with a worse survival (P=0.0002 for the whole model; Table 2). If we included persistent parenchymal infiltrates on CAT scan in the multivariate regression instead of restrictive pulmonary function (both together is statistically not allowed because of the strong correlation between restrictive pulmonary function and persistent parenchymal infiltrates on CAT), the model remained significant (P=0.0006), with persistent parenchymal infiltrates on CAT having a hazard ratio of 3.35 (95% confidence interval 1.46–7.68, P=0.0043) for mortality.
In this retrospective single-center study, we corroborated the existence of different phenotypes of CLAD, and we further analyzed the risk factors for mortality after diagnosis. First, we found that 51 of 71 CLAD patients (almost 72%) developed an obstructive pulmonary function, which is compatible with the current definition of BOS (2), whereas 20 (28%) became restrictive, which is rather compatible with RAS (8). Most of the patients with a restrictive pulmonary function also developed persistent parenchymal infiltrates on CAT scan (17/20), whereas only 5 of 51 patients with an obstructive pulmonary function developed persistent parenchymal infiltrates on CAT scan. Despite these CAT scan abnormalities, these latter patients did not meet the TLC criteria for restriction as outlined by Sato et al. (8), which may be explained by the fact that in the initial phase of CLAD, they developed the BOS phenotype, whereas only later on during the evolution of CLAD, persistent parenchymal infiltrates appeared on CAT scan. This suggests that if follow-up had been longer, they might have qualified for RAS instead of BOS. This phenomenon was also reported in the study by Sato et al. (8). What it specifically means for survival remains unanswered as this group of five patients is too small to perform statistical comparisons.
In contrast to Sato et al. (8), this article differs in at least two ways: (1) in addition to a TLC decline of more than 10% to define pulmonary restriction, we have also used the FEV1:FVC ratio, which is indeed an accepted parameter to describe restrictive pulmonary function (11); (2) in our series, we also included patients with a single lung transplantation, which were deliberately excluded in the analysis by Sato et al. This may also explain why we had more patients with persistent parenchymal infiltrates in the BOS group, especially when a single lung transplantation was performed for emphysema. Single lung transplant patients with confounding factors that might have affected the pulmonary function pattern, such as hyperinflation or progressive interstitial lung disease of the native lung, were excluded for further analysis (Fig. 1A).
The present data confirm the data of the Toronto group, in which the concept of RAS was introduced as a new CLAD phenotype. In their series of 156 patients with CLAD, 47 (30%) developed the RAS phenotype, which is in line with the 28% in the current series. Moreover, we confirmed the worse survival in the RAS patients, which indeed significantly differed from the survival in patients with the classical BOS phenotype (median of 8 vs. 35 months).
Woodrow et al. (7) recently demonstrated that 35% of their patients with CLAD developed persistent parenchymal infiltrates on CAT scan, which they called nonspecific BOS, without giving further specification as to their pulmonary function. In their series, however, there was no survival difference, which may be explained by their use of different criteria to define persistent infiltrates as patients with pleural effusions were also included in this analysis and by the fact that their cohort was rather small (7).
As a consequence, development of a nonobstructive pulmonary function decline after lung transplantation seems to be a consistent finding in different series, and therefore, it seems more appropriate to use other terminology besides BOS, which may better reflect all patterns of persistent decline in FEV1. In that case, CLAD may be the preferred terminology, whereas BOS and RAS may then be a subdivision of CLAD, not only characterized by the type of pulmonary function decline but also characterized by CAT scan findings, mostly demonstrating persistent parenchymatous involvement in the RAS phenotype. Whether these parenchymatous changes are consistent with chronic rejection remains to be further investigated, although Sato et al. could demonstrate the presence of diffuse alveolar damage and extensive fibrosis in the alveolar interstitium, visceral pleura, and interlobular septa with or without scattered bronchiolitis obliterans lesions, which suggests that chronic rejection may indeed be involved in the RAS phenotype (8). Martinu et al. (13), investigating the pathology of explant lungs in patients who were retransplanted for BOS, found at least 2 of 12 patients with extensive interstitial changes with septal widening. Burton et al. (14) also found an increased prevalence of cryptogenic organizing pneumonia and interstitial fibrosis in their transbronchial biopsy study. These findings do not prove that chronic interstitial changes are related to chronic rejection, but at least they are suggestive for it. In our patient group, transbronchial biopsies did not specifically contribute to the differential diagnosis between BOS and RAS, as they were mostly nonspecific and were rather used to exclude acute rejection and infection. Also, BAL cell differentiation was not different between these two phenotypes, whereas for instance in the BOS subphenotypes, NRAD, and fybrotic bronchiolitis obliterans syndrome, this is indeed clearly different (3, 10).
Another, rather unexpected result from our study was the fact that previous development of NRAD seemed to be a risk factor for mortality after later CLAD diagnosis. NRAD has indeed been described as a specific phenotype of BOS, characterized by an obstructive FEV1 decline, neutrophilic BAL, and more than 10% FEV1 reversibility on azithromycin treatment (3). In this study, although most of the patients were started (52/71) or were already on azithromycin for a previous NRAD (15/71), none of them had a sustained FEV1 response more than 10% after CLAD was diagnosed, excluding current NRAD as a potential bias. It has recently been shown that some 30% of the patients who were previously diagnosed with the NRAD phenotype of BOS will redevelop CLAD during further follow-up despite still being treated with azithromycin (15, 16). Although persistent BAL neutrophilia is a recognized risk factor for the development of BOS (17–19), it is now also accepted that BAL neutrophilia (as in NRAD) can be adequately treated with macrolides (20). This may favor an early-onset preventive treatment of NRAD with macrolides. In fact, in a recently published double-blind, placebo-controlled trial with azithromycin, we showed that the BOS prevalence at 2 years after transplantation was indeed significantly lower in the active treatment group, which was mainly due to the prevention of the development of NRAD (11). Whether this approach may lower the prevalence of CLAD afterward remains to be investigated.
Some of our findings are in line with other published data such as the mean time to onset of CLAD (31 months) (21), the 5-year survival after diagnosis of CLAD (26% in Duke  and 32% in this study), and no influence of gender (21), which may contrast with the data of Lama et al. (22) who showed that females with BOS had a worse lung function trajectory (although the effect on survival was not established).
Half of our patients developed an early-onset CLAD (i.e., within 24 months after transplantation), with an inherent worse prognosis that corroborates previous data (20–22). Because most of our patients (84.5%) were diagnosed with CLAD in stage 1 disease (i.e., >20% decline in FEV1), we found no difference in survival of higher CLAD stages at diagnosis, although it is accepted that patients with a fast decline in FEV1 have indeed a worse survival, whether this develops early or late after transplantation (21, 23, 24).
In conclusion, our study clearly confirms the existence of the RAS phenotype, which occurred in approximately 28% of patients with CLAD after lung transplantation and has an inherent worse prognosis compared with patients who develop BOS. Also, early CLAD (<24 months after transplantation) and previous NRAD were identified as other predictors for a worse survival after CLAD diagnosis. What this means in terms of prophylactic treatment with for instance azithromycin remains to be further investigated.
MATERIALS AND METHODS
We retrospectively investigated the medical records of all patients with a first lung transplantation in our center between January 2002 and December 2009. We choose these dates because our follow-up database started in 2002 and because we wanted to have an individual follow-up period of at least 6 months.
The pulmonary function of these patients was longitudinally evaluated at several time points after transplantation (spirometry 1–2 times per week during the first 3 months, and then at least once per month for the first postoperative year, thereafter at least once every 3 months). Full pulmonary function tests using body plethysmography was performed at least twice per year (from 2005 on). At the same time, patients were instructed to use a home spirometry device on a daily basis and to contact the transplant center in case the FEV1 decreased by more than 10% at two consecutive occasions. Patients with a persistent FEV1 decline of more than 20% were defined as having CLAD and were included in the study population. This FEV1 decline was calculated as a 20% drop from the mean of the two best postoperative values, with at least 3 weeks in between the two measurements, without another identifiable cause for this decline (2). The CLAD population was then subdivided according to the development of an obstructive (BOS) or restrictive pulmonary function pattern (RAS). Similar to Sato et al. (8), we defined a restrictive pulmonary function as a decline in TLC of at least 10% from baseline (8), but in case not enough TLC measurements were available, a FEV1:FVC index that remained normal or increased above normal with an FVC decline of at least 20% from baseline was also considered restrictive, whereas a FEV1:FVC index of less than 0.7 was considered obstructive (12). Pulmonary function tests were performed according to the American Thoracic Society guidelines on a Masterscreen IOS system or a Masterscreen PFT system (Jaeger, Acertys, Aartselaar, Belgium) (25).
All consecutive pulmonary CAT scans from the time of diagnosis of CLAD were evaluated by G.M.V. and W.D.W. until consensus was reached and further categorized according to the presence or absence of persistent (at least 3 months) parenchymal involvement (such as interstitial opacities, ground glass opacities, upper lobe dominant fibrosis, and honeycombing). The radiologic findings on the last performed CAT scan after CLAD diagnosis were used for final analysis. CAT scans were performed using four different scanners: Siemens Sensation 16, Siemens Sensation 64, Siemens Definition Flash, or Philips Brilliance 64. A spiral computed tomography scan was performed during inspiration and full expiration, with settings as follows: 120 kV, 140 mAs+dose modulation with 5- and 1-mm sections for inspiration and 120 kV, 150 mAs, 1-mm sections and feed=30 mm for expiration.
The local hospitals Ethics Committee approved this retrospective study. Statistical analysis was performed using GraphPad PRISM version 4.0 (GraphPad, San Diego, CA) and SAS (SAS software, version 9.1, SAS Institute, Cary, NC) for Mann-Whitney U test and Spearman rank correlations. Contingency tables were constructed using chi-square test or Fisher's exact test where appropriate. Kaplan-Meier survival curves were analyzed using the log-rank test. Patient who were retransplanted because of CLAD (n=7) were considered dead at the time of retransplantation.
Cox proportional hazard survival regression analysis (P<0.1 entry and P>0.1 removal) was used to estimate hazard ratios and their 95% confidence intervals of several parameters possibly predicting post-CLAD survival: gender, pretransplant diagnosis, single versus sequential single lung transplantation, postoperative day of CLAD diagnosis (early vs. late), parenchymatous infiltrates involvement on CAT scan findings, previous development of NRAD, immunosuppressive regimen, and pulmonary function pattern (obstructive vs. restrictive). Data are presented as mean±standard deviation or median with IQR; P less than 0.05 was considered as significant.
1. Christie JD, Edwards LB, Kucheryavaya AY, et al. The Registry of the International Society for Heart and Lung Transplantation
: Twenty-seventh official adult lung and heart-lung transplant report–2010. J Heart Lung Transplant
2010; 29: 1104.
2. Estenne M, Maurer JR, Boehler A, et al. Bronchiolitis obliterans syndrome 2001: An update of the diagnostic criteria. J Heart Lung Transplant
2002; 21: 297.
3. Vanaudenaerde BM, Meyts I, Vos R, et al. A dichotomy in bronchiolitis obliterans syndrome after lung transplantation
revealed by azithromycin therapy. Eur Respir J
2008; 32: 832.
4. Vos R, Vanaudenaerde BM, De Vleeschauwer SI, et al. Follicular bronchiolitis: A rare cause of bronchiolitis obliterans syndrome after lung transplantation
: A case report. Am J Transplant
2009; 9: 644.
5. McManus TE, Milne DG, Whyte KF, et al. Exudative bronchiolitis after lung transplantation
. J Heart Lung Transplant
2008; 27: 276.
6. Pakhale SS, Hadjiliadis D, Howell DN, et al. Upper lobe fibrosis: A novel manifestation of chronic allograft dysfunction in lung transplantation
. J Heart Lung Transplant
2005; 24: 1260.
7. Woodrow JP, Shlobin OA, Barnett SD, et al. Comparison of bronchiolitis obliterans syndrome to other forms of chronic lung allograft dysfunction after lung transplantation
. J Heart Lung Transplant
2010; 29: 1159.
8. Sato M, Waddell TK, Wagnetz D, et al. Restrictive allograft syndrome (RAS
): A novel form of chronic allograft dysfunction. J Heart Lung Transplant
2011; 30: 735.
9. Glanville AR. Bronchoscopic monitoring after lung transplantation
. Semin Respir Crit Care Med
2010; 31: 208.
10. Verleden SE, Vos R, Mertens M, et al. Heterogeneity of chronic lung allograft dysfunction: Insights from protein expression in broncho alveolar lavage. J Heart Lung Transplant
2011; 30: 667.
11. Vos R, Vanaudenaerde BM, Verleden SE, et al. A randomized placebo-controlled trial of azithromycin to prevent bronchiolitis obliterans syndrome after lung transplantation
. Eur Respir J
2011; 37: 164.
12. Pellegrino R, Viegi G, Brusasco V, et al. Interpretative strategies for lung function tests. Eur Respir J
2005; 26: 948.
13. Martinu T, Howell DN, Davis RD, et al. Pathologic correlates of bronchiolitis obliterans syndrome in pulmonary retransplant recipients. Chest
2006; 129: 1016.
14. Burton CM, Iversen M, Carlsen J, et al. Interstitial inflammatory lesions of the pulmonary allograft: A retrospective analysis of 2697 transbronchial biopsies. Transplantation
2008; 86: 811.
15. Gottlieb J, Szangolies J, Koehnlein T, et al. Long-term azithromycin for bronchiolitis obliterans syndrome after lung transplantation
2008; 85: 36.
16. Vos R, Vanaudenaerde BM, Ottevaere A, et al. Long-term azithromycin therapy for bronchiolitis obliterans syndrome: Divide and conquer? J Heart Lung Transplant
2010; 29: 1358.
17. Riise GC, Andersson BA, Kjellström C, et al. Persistent high BAL fluid granulocyte activation marker levels as early indicators of bronchiolitis obliterans syndrome after lung transplantation
. Eur Respir J
1999; 14: 1123.
18. Devouassoux G, Drouet C, Pin I, et al. Alveolar neutrophilia is a predictor for the bronchiolitis obliterans syndrome, and increases with degree of severity. Transpl Immunol
2002; 10: 303.
19. Neurohr C, Huppmann P, Samweber B, et al. Prognostic value of bronchoalveolar lavage neutrophilia in stable lung transplant recipients. J Heart Lung Transplant
2009; 28: 468.
20. Verleden GM, Vanaudenaerde BM, Dupont LJ, et al. Azithromycin reduces airway neutrophilia and interleukin-8 in patients with bronchiolitis obliterans syndrome. Am J Respir Crit Care Med
2006; 174: 566.
21. Finlen Copeland CA, Snyder LD, Zaas DW, et al. Survival after bronchiolitis obliterans syndrome among bilateral lung transplant recipients. Am J Respir Crit Care Med
2010; 182: 784.
22. Lama VN, Murray S, Lonigro RJ, et al. Course of FEV(1) after onset of bronchiolitis obliterans syndrome in lung transplant recipients. Am J Respir Crit Care Med
2007; 175: 1192.
23. Brugière O, Pessione F, Thabut G, et al. Bronchiolitis obliterans syndrome after single-lung transplantation
: Impact of time to onset on functional pattern and survival. Chest
2002; 121: 1883.
24. Jackson CH, Sharples LD, McNeil K, et al. Acute and chronic onset of bronchiolitis obliterans syndrome (BOS
): Are they different entities? J Heart Lung Transplant
2002; 21: 658.
25. Standardization of spirometry, 1994 update. American Thoracic Society. Am J Respir Crit Care Med
1995; 152: 1107.
Keywords:© 2011 Lippincott Williams & Wilkins, Inc.
Lung transplantation; CLAD; BOS; RAS; Phenotyping