Bronchiectasis diagnosed after renal transplantation: a retrospective multicenter study

Clinical characteristics at the time of diagnosis of bronchiectasis are summarized in Table 1. The mean age was 58.2 years. Twenty-four patients (52.2 %) were men. Chronic cough and sputum (50 %) were the main symptoms leading to chest CT scan. The mean duration of symptoms before diagnosis was 1.5 years. Most patients had undergone only one renal transplantation (89.1 %). Autosomal dominant polycystic kidney disease (ADPKD) (32.6 %) was the main underlying renal disease. Immunosuppressive drugs used prior to the diagnosis of bronchiectasis are listed in Table 1. All patients received steroids and all but three of the patients received MPA (93.5 %).

Few patients exhibited respiratory symptoms before renal transplantation, including chronic cough (n = 3), dyspnea (n = 2), annual bronchitis (n = 2) and pneumonia (n = 1). No patient exhibited chronic sputum. Eighteen patients (40.9 %) were current or former smokers and two had a diagnosis of chronic obstructive pulmonary disease. Significant pre-existing extrarenal medical conditions were unusual, including mycobacterial infection (n = 1) and rheumatoid arthritis (n = 1) (Table 2). No case of chronic respiratory bacterial colonization was identified.

Bronchiectases were cylindrical (100 %) and usually extensive (84.8 %). Table 3 describes the characteristics of the chest CT scan. The mean total bronchiectasis score was 7.4 ± 5.5 with a significant gradient from apex to bases. Bronchiolar nodules (41.3 %) and mucoid impactions (21.7 %) were the most common associated CT scan findings.

Chest X-ray prior to renal transplantation was available for 36 patients and was considered to be normal. Chest CT scan was performed before (n = 3) or just after (n = 9) renal transplantation in 12 patients and demonstrated no signs of bronchiectasis.

Spirometry, laboratory and microbiological examinations were performed inconstantly at the time of diagnosis of bronchiectasis (Table 4). For patients with available data (n = 36), mean lymphocyte count was 1295 cells/mm³. Hypogammaglobulinemia, defined as gammaglobulins less than 9 g/dL and/or IgG less than 7 g/L, was observed in 15 patients (46.9 %) associated with lymphopenia in all but one case. Mean MPA area under the concentration-time curve (AUC) evaluated in 21 patients was 42.1 mg.h/L for a mean MPA daily dose of 1.5 g/day. Spirometry was available for 22 patients and showed an obstructive disorder in 8 cases (36.4 %). Microbiological data were obtained in 34 cases from sputum analyses (n = 4) or fiberoptic bronchoscopy procedure (n = 30). Microorganisms were identified in 22 cases. Haemophilus influenzae was the most common pathogen. Concomitant pathogens were associated with Haemophilus influenzae in 7 patients, including Streptococcus species (n = 2), Aspergillus fumigatus (n = 2), Escherichia coli (n = 2), and Pseudomonas aeruginosa (n = 1) (data not shown).

MPA was stopped at the time of diagnosis of bronchiectasis in 3 patients and during follow-up in another 8 patients. Immunoglobulin replacement therapy was initiated in 6 patients either at the time of diagnosis (n = 3) or during follow-up (n = 3). The main symptoms reported after the diagnosis of bronchiectasis were chronic sputum (43.5 %) and recurrent respiratory tract infections (37.0 %). None of the six deaths (13 %) was attributable to bronchiectasis (Table 5).

At least one chest CT scan was performed after the diagnosis of bronchiectasis in 28 patients (60.9 %). The last CT scan available for each patient was performed 2.6 ± 2.5 years after the diagnosis of bronchiectasis. A trend towards worsening of the total bronchiectasis score was observed at follow-up for these 28 patients (7.7 ± 6.2 vs. 10.9 ± 6.7; p = 0.07). Bronchiectasis was more extensive in 19 patients (67.9 %), while stabilization was observed in 3 patients (10.7 %) and improvement was observed in 6 patients (21.4 %). Bronchiolar nodules were more frequent (p = 0.01).

To our knowledge, this study reports the largest series of bronchiectasis diagnosed after renal transplantation. The main finding of this study is that the clinical and microbiological characteristics of bronchiectasis are similar to those usually described in non-cystic fibrosis (CF) bronchiectasis [12]: (1) chronic sputum or recurrent bronchitis are the most common symptoms; (2) an obstructive disorder is frequently associated; (3) Haemophilus influenzae is the most common micro-organism identified. However, some results need to be highlighted. First, bronchiectasis can be diagnosed very late after renal transplantation, as illustrated by the two previous reports in adults with a mean time to diagnosis ranging from 3 to 11.7 years [6, 7]. Second, a CT scan gradient was observed from the apex to the bases. Previous reports have described a predominance of bronchiectasis in lower lobes in 3 patients, while the distribution of bronchiectasis was not specified in the other 12 cases [6‐8]. Interestingly, other forms of non-CF bronchiectasis also exhibit such predominance in the bases [13, 14]. Third, the bronchiectasis score frequently deteriorated during follow-up. As observed in our study, some previous reports have shown that patients with non-CF bronchiectasis had persistent or worsening symptoms on long-term follow-up [15], but very limited data are available on the course of bronchiectasis based on follow-up CT scan score [16, 17]. Finally, the overall mortality in our series was 13 % with a median follow-up of 3 years, similar to the results of recent studies that have reported mortality rates ranging between 16.3 % at 4 years and 29.7 % at 13 years in non-CF bronchiectasis [18‐20]. In a long-term prospective study, Loebinger et al. found that the primary cause of death was respiratory (70.4 %), especially respiratory infection or failure [19]. In contrast, despite frequent respiratory infections, no death was related to a respiratory cause in our series.

Prior to the study by Pijnenburg et al. in 2004, no case of bronchiectasis occurring after renal transplantation had been reported in the literature [8]. In our study, only two patients had a diagnosis of bronchiectasis before 2004 (data not shown). Bronchiectases were then mainly diagnosed within the ten last years, which may be related to expanded indications of CT-scan and changes in immunosuppression strategy. The absence of systematic chest CT scan before or just after renal transplantation does not allow pre-existing bronchiectasis to be formally excluded. A chest CT scan without bronchiectasis was available for only 12 patients (three before and nine after renal transplantation). The presence of asymptomatic bronchiectasis before transplantation in some cases therefore cannot be formally excluded. In particular, one patient had rheumatoid arthritis and another had a history of mycobacterial infection, two potential causes of bronchiectasis [12, 21]. Moreover, patients with ADPKD could also be at increased risk of bronchiectasis, as recent studies have demonstrated an increased prevalence of mild-to-moderate cylindrical bronchiectasis with bilateral lower lung predominance in ADPKD [22‐24]. ADPKD is associated with defective primary ciliary function in renal epithelial cells. Functional abnormalities in polycystin-1 and 2, two membrane regulatory proteins expressed in the cilia of both human airway epithelial and airway smooth muscle cells, may result in radiological bronchiectasis due to decreased mucociliary clearance or impaired airway injury repair [22, 24]. However, these studies present a number of limitations including the absence of data on comorbidities associated with bronchiectasis [22] and the inclusion of transplant recipients [23]. At least, the minimal interval 0.4 year between first renal transplantation and diagnosis of bronchiectasis, probably too short to develop bronchiectasis, argues for undetected pre-existing bronchiectasis.

Bronchiectasis can be induced by primary or secondary immunodeficiency [12, 21, 25, 26]. Drug-induced immunosuppression following transplantation predisposes to recurrent lung infections and increases the risk of bronchiectasis after bone marrow, heart or lung transplantation [27‐29]. Low immunoglobulin and mannose binding protein levels after renal transplantation are associated with infectious complications [30]. In our series, 15 patients exhibited hypogammaglobulinemia, which may contribute to the development of bronchiectasis [7].

Three recent studies have suggested that MPA may be a causative factor of bronchiectasis [6‐8]. MPA is a relatively new immunosuppressive drug commonly used in renal transplantation for the prevention and treatment of allograft rejection. According to the results of these three studies, the potential involvement of MPA is based on the following arguments: 1) some patients did not have any respiratory symptoms despite long periods with other immunosuppressive drugs [8]; 2) an improvement of symptoms after MPA withdrawal was observed in some patients [6]; 3) some cases of bronchiectasis have been described in children who have received renal transplantation with MPA, whereas bronchiectasis is usually very uncommon at this age in the absence of CF or immunodeficiency [8]. In our series, five patients not treated by MPA for a first renal transplantation developed symptomatic bronchiectasis after a second or third renal transplantation with MPA treatment. Two mechanisms have been proposed to explain the role of MPA in bronchiectasis. First, MPA, as a potent inosine monophosphate dehydrogenase inhibitor, inhibits purine synthesis and severely depresses both cell-mediated and humoral immunity by inhibiting T- and B-cell proliferation. The resulting hypogammaglobulinemia has been shown to be more frequent and more severe in patients receiving MPA compared to other immunosuppressive drugs [31]. Only six of the 15 patients with hypogammaglobulinemia in our study received immunoglobulin replacement therapy. Second, MPA may directly affect bronchial epithelium by altering mucociliary clearance [32].

Our results should be interpreted in the context of the several limitations of this study. 1) The retrospective identification of the cases of bronchiectasis revealed after renal transplantation is a major limitation. Even if the study design was as rigorous possible, we can not exclude that a significant number of cases have not been identified by the centres. 2) The retrospective design of this study does not allow estimating the prevalence of the disease. 3) Thoracic CT scans were analysed in all the cases by a panel of pulmonologists, and radiologist. Cases were included in the study in presence of defined bronchiectasis. However, since HRCT scan was not available for each patients, we cannot rule out that distal bronchiectasis should not be visualized in some patients. 4) Observation bias may exist, even if it has been minimized by using a standardized data collection form. In particular, information on respiratory medical history may have been underreported. 5) The spirometry, biological and microbiological data were very heterogeneous. Moreover, the role of potentially pathogenic micro-organisms in the clinical course of bronchiectasis cannot be determined in absence of repeated respiratory samplings, micro-organisms quantification and characterisation of associated clinical status. 6) No conclusion can be drawn on the interest of MPA withdrawal or immunoglobulin replacement therapy due to the retrospective design and the small number of patients. Despite these limitations, we believe that our study provides important findings on the characteristics of bronchiectasis occurring after renal transplantation.

Finally, it should be stressed that, even after intensive investigation, one or more causative factors are identified in only 47 % of cases of bronchiectasis [21]. It can therefore be hypothesized that multifactorial mechanisms, including underlying diseases, immunosuppressive drugs and respiratory tract infections, could contribute to the pathogenesis or clinical emergence of bronchiectasis after renal transplantation. Due to the retrospective design of our study and the small number of patients, we were not able to perform subgroup analyses in order to determine the burden of causal variables such as ADPKD, MPA and immunoglobulin deficiency. Similar studies would be interesting in other solid organ (liver and heart) transplantation.