The Effect of Reflux and Bile Acid Aspiration on the Lung Allograft and Its Surfactant and Innate Immunity Molecules SP-A and SP-D

Authors


* Corresponding author: Shaf Keshavjee, shaf.keshavjee@uhn.on.ca

Abstract

Gastro-esophageal reflux and related pulmonary bile acid aspiration were prospectively investigated as possible contributors to postlung transplant bronchiolitis obliterans syndrome (BOS). We also studied the impact of aspiration on pulmonary surfactant collectin proteins SP-A and SP-D and on surfactant phospholipids—all important components of innate immunity in the lung. Proximal and distal esophageal 24-h pH testing and broncho-alveolar lavage fluid (BALF) bile acid assays were performed prospectively at 3-month posttransplant in 50 patients. BALF was also assayed for SP-A, SP-D and phospholipids expressed as ratio to total lipids: phosphatidylcholine; dipalmitoylphosphatidylcholine; phosphatidylglycerol (PG); phosphatidylinositol; sphingomyelin (SM) and lysophosphatidylcholine. Actuarial freedom from BOS was assessed.

Freedom from BOS was reduced in patients with abnormal (proximal and/or distal) esophageal pH findings or BALF bile acids (Log-rank Mantel-Cox p < 0.05). Abnormal pH findings were observed in 72% (8 of 11) of patients with bile acids detected within the BALF. BALF with high levels of bile acids also had significantly lower SP-A, SP-D, dipalmitoylphosphatidylcholine; PG and higher SM levels (Mann-Whitney, p < 0.05). Duodeno-gastro-esophageal reflux and consequent aspiration is a risk factor for the development of BOS postlung transplant. Bile acid aspiration is associated with impaired lung allograft innate immunity manifest by reduced surfactant collectins and altered phospholipids.

Introduction

Long-term survival after lung transplantation is limited by chronic graft dysfunction or bronchiolitis obliterans syndrome (BOS). Pathologically this has been associated with bronchiolitis obliterans, a progressive fibroproliferative process that accounts for more than 30% of deaths occurring after the third postoperative year (1,2). Five-year survival after the onset of bronchiolitis obliterans is only 30–40%, and survival at 5 years after transplantation is 20–40% lower in patients with bronchiolitis obliterans (3).

Many risk factors including recurrent acute rejection, ischemia-reperfusion injury, and cytomegalovirus infection have been identified for bronchiolitis obliterans and its clinical correlate, BOS, which is defined as a persistent drop in lung function following transplantation (2). Recently gastro-esophageal reflux (GER) and aspiration have been suggested to also be a potential contributing factor to BOS (4–12). GER in lung transplant patients may potentially be induced by immunosuppressive drugs (e.g. calcineurin inhibitors) or iatrogenic vagal nerve injury at surgery or it may even be present as a preoperative condition. In fact, interstitial lung disease secondary to connective tissue disorders, idiopathic pulmonary fibrosis and cystic fibrosis have all been associated with GER (13–24). We have previously documented the association between BOS and bile acid aspiration along with the association between alveolar neutrophilia and IL-8. The bile acids that were found within the bronchoalveolar lavage fluid appear to be a marker of retrograde aspiration (12). We, therefore, sought to further investigate whether GER and aspiration in lung transplant patients predicted the development of BOS.

While traditionally thought of in terms of injury related to antiallogeneically directed adaptive immunity, recent lines of evidence suggest that BOS is associated with impaired innate defenses within the lung (25,26). Bile acid aspiration in particular may function as a barrier breaker for pulmonary surfactant and have cytotoxic effects on the alveolar epithelium, thus affecting surfactant protein and phospholipid production and homeostasis. Therefore, we studied the impact of GER and bile acids on pulmonary phospholipids and the surfactant proteins A and D, the latter being specific components of the lung allograft's innate immunity (27–30). Phospholipids are involved within the tension active function of the surfactant but also function as the bronchiolo-alveolar lipid membrane barrier. Surfactant proteins A and D belong to the collectin super-family and serve as opsonins, but also appear to have a regulatory role toward macrophage cytokine production and lymphocyte proliferation.

Methods

Starting from January 2003, lung transplant patients were prospectively evaluated for GER and its relationship to posttransplant lung function. The protocol was approved by the Research Ethics Board of the Toronto General Hospital, University Health Network and informed consent was obtained from each patient for use of excess BALF and serum in adherence to the principles set forth in the Helsinki declaration.

GER assessment

Patients were scheduled for esophageal manometry, 24-h ambulatory pH-testing and nuclear medicine solid and liquid gastric emptying studies at 3 and 12 months after transplantation.

Esophageal manometry was performed by stationary technique using a 6-channel probe that has a Dent Sleeve distally (one perfusion port at the proximal end of the sleeve and four more ports at 5-cm intervals proximally (Dentsleeve Pty. Ltd, Wayville, Australia), that is perfused by a perfusion pump (Mui Scientific, Mississauga, Ontario, Canada).

Esophageal 24-h pH monitoring was performed using a two-channel pH-probe with 15-cm spacing between sensors connected to a continuous pH-recording device (Comfortec, Sandhill Scientific, Highlands Ranch, CO, USA). Prior to the study, the patients were instructed to stop proton pump inhibitors for at least 7 days and were kept for 5 days on H2-blocking agents that were stopped for 48 h prior to testing. The study was carried out only if the gastric pH was below 4. The probe was positioned with the sensors 5 and 20 cm above the lower esophageal sphincter as determined by the esophageal manometry. Continuous pH recordings were obtained for 24 h. Patients were asked to maintain normal activity and a normal diet excluding acidic foods and drinks. They were instructed to record the occurrence of episodes of heartburn, cough, time and duration of meals and time and duration of supine and upright position. The DeMeester score was applied for the distal pH findings, and it considers six components of physiological relevance: percent total time pH < 4; percent upright time pH < 4; percent supine time pH < 4; number of reflux episodes in 24 h; number of reflux episodes >5 min; longest episode (31). The proximal pH findings were analyzed according to Castell's method (32).

Gastric emptying studies were scheduled at 3 and 12 months posttransplant and were considered prolonged when the half-time of the radiotracer in the stomach was greater than 45 min for liquids and greater than 120 min for solids.

Clinical information and biological samples

Clinical information and excess biological samples of broncho-alveolar lavage fluid (BALF) were collected in all patients at routine follow-up times. BALF was collected following two repeat lavages of 50 mL of normal saline solution in the right middle lobe or the lingula in lung transplant recipients.

Bronchoscopies with BALF collection are routinely performed at 2 and 6 weeks following transplantation, and every 3 months for the first year, every 6 months for the second year and as clinically indicated.

Data were collected regarding BOS development using routine pulmonary function testing. BOS status was determined according to the International Society of Heart and Lung Transplantation grading criteria (2): BOS-0 absence of the syndrome; BOS-1-3 presence of one of the three grades of the syndrome. For the purpose of this study, patients were required to have been out from transplant at least 6 months prior to calculation of the BOS grade so that they could be classified according to presence or absence of the syndrome. The interval of time from transplant to development of BOS was calculated in months.

Aliquots of the BALF were collected and immediately snap-frozen at −80°C. After thawing, protease inhibitors (Complete Mini tabs, Boehringer-Mannheim, Germany) were added to the samples that were then clarified by centrifugation at 5000g for 10 min. The resulting supernatant was assayed for bile acids by a spectrophotometric assay (Bile Acid kit, Sigma Diagnostics, Inc. St. Louis, MO, USA) and for surfactant proteins A (SP-A Test/Kokusai-F, Sysmex Corporation, Kobe, Japan), and D (SP-D Kit/Yamasa EIA, Yamasa Corporation Choshi, Japan).

Mass spectral analysis of phospholipids

BALF samples were clarified by centrifugation at 2000g for 5 min and the supernatant collected and ultracentrifuged at 50 000g for 40 min at 4°C in order to separate large and small aggregate surfactant phospholipids. The large aggregate pellet was resuspended in 1.5 mL of 0.15 M NaCl. BALF samples were spiked with 1 μg of deuterated 16:0/16:0 phosphatidylcholine (PC), deuterated 14:0/14:0 phosphatidylglycerol (PG), and 15:0 lyso-phosphatidylcholine (lyso-PC) (Avanti polar lipids, Alabaster AL, USA) as internal standards. Samples were extracted by a two-step procedure as described by Bligh and Dyer (33). The chloroform layers were removed and dried under nitrogen gas and reconstituted in 200 μL 3:1 chloroform/methanol that was acidified with 1 μL of 5 mM ammonium acetate per liter and 30 μL of sample was injected by auto-sampler into an API4000 triple-quadruple mass spectrometer (MDS SCIEX, Concord, Ontario, Canada) equipped with an ion turbo spray. Individual lipid species were assayed for PC and lysoPC at M + 1 and M + Na and in positive mode and for PG at M − 1 in negative mode. Quantification was carried out using multiple reaction monitoring for the common phosphocholine daughter ion at 184 M/z (193 M/z for the internal standard). PG was detected by Q1 scan. Spectral data was analyzed using MDS/Sciex Analyst 1.2 software (33).

Statistics

Statistical analysis was performed using StatView 5 software (StatView, SAS Institute Inc., NC, USA). A nonparametric statistical analysis (Mann-Whitney test) was used for comparison of continuous variables between groups. Receiver operating characteristic methods were used to identify the BALF bile acid cutoff value for best accuracy. Categorical data were analyzed using the chi-Square test or the Fisher's Exact test. Nonparametric actuarial curves along with the Breslow-Gehan-Wilcoxon test and the Mantel-Cox Log-rank test were used to test differences in patients grouped by esophageal pH findings and the presence of bile acids with regard to freedom from BOS. Statistical differences were considered significant when the p value was less than 0.05.

Results

To date esophageal pH testing has been performed in 70 transplant (67 double lung Tx, 2 single lung-Tx and 1 heart-lung Tx) patients (M/F = 40:30, median age 52 years, range 21–70). Fifty patients were tested at 3 months after lung transplant and 30 at 12 months. To date only 12 were tested at both time points and 18 entered the study after the 3 months time point. The underlying lung diseases for which patients required lung transplantation were idiopathic pulmonary fibrosis 18, emphysema 25, cystic fibrosis 15, scleroderma 2, Eisenmenger's syndrome 2, lymphangioleiomyomatosis 3, bronchiolitis obliterans 2, Langerhan's cell histocytosis 1, eosinophilic granulomatosis 1, and broncho-alveolar carcinoma 1.

Tables 1 and 2 show the median values for the pH parameters and the prevalence of patients with abnormal findings at 3 and at 12 months, respectively. Abnormal distal and/or proximal pH findings were observed in 32% (16 of 50) patients at 3 months and in 53% (16 of 30) at 12 months.

Table 1.  pH findings at 3 and 12 months after TX
Median (range)NormalAll patientsCOPDCFIPF
3 months (50)12 months (30)3 months (18)12 months (10)3 months (11)12 (7)3 months (14)12 months (9)
Total time %<4%1.2 (0–15)4.2 (0–17)1.3 (0–15)3.2 (0.7–16)0.8 (0–6)4.4 (0.1–7)2.5 (0–10)4.4 (0–10)
Upright T %<6%1.2 (0–12)4.6 (0–18)3.5 (0–10)4.2 (1.5–18)0.4 (0–8)1.3 (0–10)5.5 (0–10)4.3 (0–12)
Supine T %<1.5%0.4 (0–24)0.5 (0–21)0 (0–24)0.4 (0–18)0 (0–11)0.2 (0–12)2.5 (0–13)1.9 (0–6)
No. episodes<5022 (0–165)48.5 (0–229)18 (0–133)49 (10–229)22 (1–101)40 (2–163)29 (0–165)43 (0–129)
No. episodes >5 min<30 (0–7)1.5 (0–12)0 (0–7)1 (0–12)0 (0–1)0 (0–5)1 (0–5)2 (0–7)
Longest episode<9 min4 (0–60)6 (0–39)4 (0–42)5.3 (3–39)3.3 (0.3–45)4 (1–28)7.4 (0.7–61)10 (0–18)
De Meester score<155 (0–59)14.6 (0–63)5.5 (0–59)10.6 (3–60)4 (0.2–24)16.5 (1–27)10.6 (1–36)16 (0–28
Proximal upright T %<1.3%0 (0–1)0 (0–7)0.5 (0–0.3)0 (0–3)0 (0–0.5)0 (0–7)0 (0–0.1)0 (0–3)
Proximal supine T %=0%0 (0–2)0 (0–7)0 (0–1.8)0 (0–0.1)0 (0–1.2)0 (0–0.3)0 (0–0.7)0 (0–1.7)
Table 2.  Abnormal pH findings at 3 and 12 months after TX
Abnormal tests (%)All patientsCOPDCFIPF
3 months (50)12 months (30)3 months (18)12 months (10)3 months (11)12 months (7)3 months (14)12 months (9)
Total time%24% (12)53% (16)11% (2)40% (4)18% (2)57% (4)42% (6)56% (5)
Upright T%12% (6)37% (11)5% (1)30% (3)9% (1)29% (2)21% (3)33% (3)
Supine T%36% (18)43% (13)28% (5)40% (4)9% (1)29% (2)57% (8)56% (5)
No. episodes26% (13)47% (14)17% (3)50% (5)9% (1)43% (3)43% (6)33% (3)
No. episodes >5 min10% (5)40% (12)11% (2)40% (4)043% (3)14% (2)33% (3)
Longest episode28% (14)43% (13)11% (2)40% (4)18% (2)29% (2)42% (6)56% (5)
DeMeester score26% (13)50% (15)17% (3)40% (4)18% (2)57% (4)46% (6)56% (5)
Proximal upright T%010% (3)010% (1)014% (1)011% (1)
Proximal supine T%14% (7)17 % (5)17% (3)10% (1)18% (2)14% (1)7% (1)11% (1)

At 3 months posttransplantation, gastric emptying studies for liquids were completed in 22 patients and were found to be prolonged in 36% (8 of 22), while at 12 months it was completed in seven patients and in 57% (4 of 7) it was prolonged. The solid emptying study was completed in 43 patients at 3 months and 91% (39 of 43) were found to be prolonged, at 12 months it was completed in 21 patients and 81% (17 of 21) were prolonged. As shown in Table 3, findings were uniform across all subgroups of primary lung disease.

Table 3.  Gastric emptying findings in lung transplant patients by type of pulmonary disease end stage: prevalence of abnormality for each group
 3 months after TX
All patients (43)COPD (15)CF (9)IPF (13)
% Prolonged91% (39)80%89%100%
 12 months after TX
All patients (21)COPD (6)CF (5)IPF (7)
% Prolonged81% (17)83% (5)80% (4)86% (6)

Medical therapy was implemented in patients with abnormal pH findings and/or abnormal gastric emptying using proton pump inhibitors and/or domperidone. No surgical intervention was adopted.

In 31 patients, the 3 or 12 months posttransplant BALF was collected prior to pH testing and gastric emptying studies, thus no medical therapy had been implemented. Abnormal proximal and/or distal esophageal pH testing was documented in 72% (8 of 11) of patients with presence of bile acids within the BALF. Bile acids were detected in the BALF of 53% (8 of 15) patients with abnormal pH testing. In particular, as shown in Table 4, a significant relationship was noted between increasing levels of BALF bile acids and the proximal esophageal pH abnormalities. The proximal pH testing was abnormal in eight patients and in the majority of them (7 of 8) the abnormal findings were documented only in the supine position with a median time percent of pH < 4 of 0.6% with a range of 0.1–6.7 (normal is 0).

Table 4.  Frequency distribution for proximal esophageal pH testing and levels of BAL bile acids: χ2= 6; p = 0.04
 Abnormal proximal pH (8)Normal proximal pH (23)
High BAL bile acid75% (3)25% (1)
Low BAL bile acid29% (2)71% (5)
No BAL bile acid15% (3)85% (17)

BALF was collected at various time points in 98 patients. In 39 patients BALF samples were collected at 3 months from transplant and bile acids were detected in 21 patients (54%), at 12 months BALF was collected in 35 patients and bile was detected in 18 (50%).

Time to BOS diagnosis was calculated for patients that had pH testing at 3 months and for patients that had BALF samples collected at 3 months and were out from transplant for at least 6 months. Figure 1 shows the actuarial curves documenting a significantly reduced freedom from BOS in patients with abnormal pH findings or elevated levels of bile acids within the BALF at 3 months after transplantation. BALF bile acid levels were determined high with cutoff ≥3.5 μmol/L by ROC methods for best overall accuracy.

Figure 1.

Interval from transplant to development of BOS. (A) The actuarial curves show a significantly reduced time from transplant to BOS development in patients with abnormal pH findings at 3 months after transplantation (Mantel-Cox Log-rank test p < 0.01). (B) A significantly reduced time from transplant to BOS development was observed for increasing levels of bile acids within the BALF collected at 3 months after transplantation (Mantel-Cox Log-rank test p < 0.05). The Mantel-Cox Log rank test showed p < 0.01 for the comparison between the group with no bile acid in the BALF and the group with high bile acid levels.

Patients with abnormal pH testing either at 3 or at 12 months after transplantation had significantly lower levels of SP-A within the BALF as tested at all time points and averaged for each patient (Figure 2). SP-A was 1315 ng/mL median (range 30–5829) in patients with abnormal pH testing, and 2538 ng/mL median (range 844–6348) in normal patients (Mann-Whitney test p < 0.05). SP-D was 420g/mL median (range 17–1132) in abnormal pH tested patients, and 475ng/mL median (range 140–1258) in patients with normal pH testing.

Figure 2.

Patients with abnormal pH findings had significantly reduced level of SP-A. No difference was noted for SP-D. *Mann-Whitney test p < 0.05.

A BALF bile acid dose effect was noted with regard to SP-A and D both at the 3-month time point, and for all time point BALF data averaged for each patient (Figure 3). SP-A at 3 months after transplantation was 1979 ng/mL median (range 342–5534) for patients with no bile acids within the BALF, 3522 ng/mL (1005–12 512) for patients with low levels of BALF bile acids, and 1129 ng/mL (728–1678) for patients with high levels of bile acid (Kruskal-Wallis test p < 0.005; Mann-Whitney test between no bile vs. low bile groups and vs. the high bile group showed a p < 0.01 and p < 0.05, respectively, and between low and high bile group showed a p < 0.005). Similarly, in all 98 patients, SP-A average was 1589 ng/mL median (range 336–5534) in patients with no BALF bile acid, 2150 ng/mL (148–6348) in patients with low BALF bile acids, and 1174 ng/mL (30–5193) in patients with high levels of BALF bile acids (Kruskal-Wallis test p < 0.05; Mann-Whitney test between high bile vs. the no bile and the low bile groups showed a p < 0.05 and p < 0.01, respectively, no difference was noted between no and low bile groups). SP-D at 3 months after transplantation was 657 ng/mL median (range 127–1402) for patients with no bile acids within the BALF; 714 ng/mL (155–2434) for patients with low levels of BALF bile acids, 362 ng/mL (53–654) in patients with high levels of bile acids within their BALF (Kruskal-Wallis test p = 0.06). In all 98 patients, SP-D median average was 790 ng/mL median (range 136–1788) in patients with no BALF bile acid, 529 ng/mL (28–1438) in patients with low BALF bile acids, and 246 ng/mL (14–1330) in patients with high levels of BALF bile acids (Kruskal-Wallis test p < 0.0001; Mann-Whitney test between no bile vs. the low bile and the high bile groups showed a p < 0.005 and p < 0.0001, respectively, and between low and high bile groups p < 0.005).

Figure 3.

Bile acid levels within the BALF are associated with SP-A and SP-D at 3 months after transplantation but also at all time points of BALF collection with data averaged for each patient. The Kruskal-Wallis test showed p < 0.005 for the 3-month BALF collection and the Mann-Whitney test between no bile versus low bile groups and versus the high bile group showed a p < 0.01 and p < 0.05, respectively, and between low and high bile group showed a p < 0.005. Similarly in all 98 patients, Kruskal-Wallis test p < 0.05; the Mann-Whitney test between high bile versus the no bile and the low bile groups showed a p < 0.05 and p < 0.01, respectively, no difference was noted between no and low bile groups. For SP-D at 3 months the Kruskal-Wallis test p = 0.06. In all 98 patients, the Kruskal-Wallis test p < 0.0001; the Mann-Whitney test between no bile versus the low bile and the high bile groups showed a p < 0.005 and p < 0.0001, respectively, and between low and high bile groups p < 0.005. The boxes indicate the 25th, 50th and 75th percentile and the bars the 10th and 90th percentile.

Single samples from 18 consecutive patients with abnormal pH testing and from 17 consecutive with normal pH findings were selected for phospholipids analysis by mass spectrometry. For each patient the BALF sample with the highest level of bile acid was chosen. Bile acid was present in 16 samples. A significantly lower level of PG and sphingomyelin (SM) was noted in samples from patients with abnormal pH findings (Figure 4). A dose-dependent effect was also noted for BALF bile acid levels and dipalmitoylphosphatidyl-choline (DPPC), PG and SM (Figure 5). Dipalmitoylphosphatidylcholine in BALF samples with no bile acids was 51% of total phosphatidylcholine (median, range 43–59), was 49% (39–56) in BALF samples with low bile acid levels, and was 38% (16–53) in BALF samples with high bile acid levels (Kruskal-Wallis test p = 0.01; Mann-Whitney test between high BALF bile acid vs. the no and low bile groups showed a p < 0.01 and p = 0.01, respectively, no difference between the low and the no bile acid groups). PG was 17% of total lipids (median, range 13–44) in BALF samples with no bile acids, was 18% (11–35) in BALF samples with low bile acid levels, and was 12% (6–47) in BALF samples with high bile acid levels (Kruskal-Wallis test p = 0.01; Mann-Whitney test between high BALF bile acid vs. the no and low bile groups showed a p < 0.05 and p < 0.01, respectively, no difference between the low and the no bile acid groups). SM was 0.4% of total lipids (median, range 0.3–7) in BALF samples with no bile acids, was 0.5% (0.2–4) in BALF samples with low bile acid levels, and was 2.7% (0.5–11) in BALF samples with high bile acid levels (Kruskal-Wallis test p < 0.005; Mann-Whitney test between high BALF bile acid vs. the no and low bile groups showed a p < 0.005 for both comparisons, no difference between the low and the no bile acid groups was observed). A significant negative correlation was noted between bile acids and DPPC r=−0.5; PG r=−0.5; and a significant positive correlation with SM, r= 0.6 with p < 0.05 following multiple comparison correction.

Figure 4.

Patients with abnormal pH findings had a significantly reduced level of phosphatidylglycerol and increased levels of sphingomyelin within the BALF. No difference was noted for the other phospholipids. Phosphatidylglycerol in patients with normal pH testing was 17% of surfactant total lipids (median, range 13–48) and in patients with abnormal pH testing 13% (6–32) (Mann-Whitney test p < 0.05). Sphingomyelin in patients with normal pH testing was 0.4% of total lipids (0.2–9), and in patients with abnormal pH testing 1.6% (0.3–11) (Mann-Whitney test p < 0.01). PC = phosphatidylcholine; DPPC = dipalmitoylphosphatidylcholine; PG = phosphatidylglycerol; PI = phosphatidylinositol; SM = sphingomyelin; Lyso-PC = lysophosphatidylcholine.

Figure 5.

An association was noted between increasing levels of BALF bile acids and dipalmitoylphosphatidyl-choline (Kruskal-Wallis test p = 0.01; Mann-Whitney test between high BALF bile acid versus the no and low bile groups showed a p ≤ 0.01, no difference between the low and the no bile acid groups); phosphatidylglycerol (Kruskal-Wallis test p = 0.01; Mann-Whitney test between high BALF bile acid vs. the no and low bile groups showed a p < 0.05 and p < 0.01, respectively, no difference between the low and the no bile acid groups); and sphingomyelin (Kruskal-Wallis test p < 0.005; Mann-Whitney test between high BALF bile acid versus the no and low bile groups showed a p < 0.005 for both comparisons, no difference between the low and the no bile acid groups was observed). PC = phosphatidyl-choline; DPPC = dipalmitoylphosphatidylcholine; PG = phosphatidylglycerol; PI = phosphatidylinositol; SM = sphingomyelin; Lyso-PC = lysophosphatidylcholine.

Discussion

This study demonstrates that aspiration secondary to duodeno-GER is a risk factor for earlier development of BOS after lung transplantation. A possible pathophysiologic mechanism involved with bile acid aspiration lies in the impairment of the pulmonary specific innate immunity. Indeed, patients with elevated BALF bile acid levels and abnormal esophageal pH findings had lower levels of pulmonary surfactant proteins SP-A and SP-D and alterations of the surfactant phospholipids.

Aspiration has been suggested to be potential contributors to allograft dysfunction. The impaired cough reflex and mucociliary clearance posttransplantation may promote a prolonged contact time of aspirated gastric contents with greater lung injury and possible development of bronchiolitis obliterans (9,34).

pH findings were abnormal in 30% of patients at 3 months and 50% at 12 months. Aspiration, as per bile acid detection within the BALF, was documented in 50% of lung transplant recipients both at 3 and 12 month after transplantation. In 70% of patients with documented BALF bile acids, the pH testing was abnormal. In contrast, 50% of patients with abnormal pH testing and 20% of those with normal findings had BALF bile acids. Unfortunately, the standard pH testing apparatus does not test for alkaline (pH > 7) or alkalacid (4 < pH < 7) refluxate (35). In our series 90% of the recipients had delayed gastric emptying supporting the need to address nonacid GER.

Bile aspiration has been associated with pulmonary injury (36–39) with dose-dependent cytotoxicity ranging from alteration of cellular cationic permeability to disruption of the cellular membrane as observed in type II pneumocytes (40). Of the various components of the duodeno-GER, the acid per se may play a limited role in the injury related to chronic, silent aspiration (40,41). Bile acids are barrier breakers of the gastric mucosa, disrupting its protective surfactant phospholipid layer (42,43). They appear to down-regulate the innate immune system via specific receptors (TGR5), expressed in monocytes and macrophages with significantly reduced phagocytic activity and cytokine production (44); and via interferon-mediated signal transducer, activator of transcription 1 phosphorylation, important in the antiviral innate response (45).

Liver, kidney and heart transplantation have a recipient and graft 5-year survivals exceeding 70% while lung and intestine linger at 50%. Both the latter organs suffer the disadvantage of having ongoing exposure to the environment, thus they strongly rely on their organ-specific innate immunity for protection from the environmental noxious agents (46–49). Important components of the lung-specific innate immune defense mechanisms are provided by pulmonary surfactant phospholipids and the pulmonary surfactant associated proteins. In this study, we investigated the impact of bile acid aspiration on these defense mechanisms.

SP-A and SP-D belong to the collectin superfamily and function as opsonins for bacteria, fungi and viruses; regulate macrophage and neutrophil cytokine production and provide direct or indirect modulation of lymphocyte proliferation. In fact, are considered ‘crosstalk’ proteins between the innate and adaptive immune systems (28–30).

Lower BALF SP-A. and SP-D has been described in children with abnormal pH findings (50). In our study, similar observations was documented for SP-A. Although, a dose-related negative correlation was noted between the BALF bile acid and the BALF levels of SP-A and SP-D both at 3 months—and for the average of all samples. Significantly, lower levels of SP-A and SP-D were observed in patients with elevated BALF bile acids. In contrast, at low levels there appears to be a stimulatory effect on the SP-A production suggesting an irritant/stimulatory effect from low bile acids levels and a noxious effect at high levels on the type II pneumocytes (the major source of surfactant proteins within the pulmonary parenchyma). Bile acids have been shown to induce intracellular cAMP, and in turn cAMP has been shown to induce surfactant protein A gene expression (45,51) although at high doses, they disrupt cell membranes and induce apoptosis (40,52,53).

BALF bile acids have been previously identified by our group as markers of BOS, associated with alveolar neutrophilia and elevated levels of interleukin-8 (12,54). Moreover, a significant relationship between bile acids and the presence of bacteria and fungi within the BALF and with inflammation on transbronchial biopsy was documented, thus suggesting an effect on the lung's innate immunity (12).

Phospholipids contribute to the protective physical barrier of the pulmonary epithelium, in addition to the alveolar surface tension. BALF bile acids had negative dose-dependent effect on the tension active surfactant phospholipids, DPPC and PG (27) and a positive one on the membrane-related phospholipid, SM, thus supporting their cytotoxic role (40).

In conclusion, our study confirms that duodeno-GER aspiration is a risk factor for BOS development. Patients with abnormal pH testing or BALF bile acid develop BOS earlier and have impaired pulmonary surfactant phospholipids and surfactant-associated proteins, key players within the lung organ-specific innate immune defense mechanisms. Bile acids were investigated as markers of retrograde aspiration, although, a temporal link along with a dose-response effect was observed with the earlier development of BOS. Furthermore, bile acids carry a plausible biologic activity toward lung injury (38,39,42,43), thus this study also supports their active role, whether primary or contributive to other agents, toward the development of chronic lung allograft dysfunction.

GER is but one of many potential contributing factors in the multifactorial process leading to BOS. Although initially thought to be simply ‘chronic rejection’ and synonymous with the pathological entity ‘bronchiolitis obliterans’—it is clear that BOS is a more complex process involving interactions between the innate and acquired immune systems. The term ‘Chronic Graft Dysfunction’ hence is more appropriate to describe this phenomenon. It is hoped that an improved understanding of the mechanisms underlying these complex interactions will lead to strategies to prevent and treat chronic graft dysfunction in the future.

Acknowledgment

F. D'Ovidio was supported by Canadian Institutes of Health Regenerative Medicine Research Training Award, Department of Surgery Intensive Care and Organ Transplantation, University of Bologna, Bologna, Italy. This work was presented as a ‘Poster of Distinction’ at the 2005 American Transplant Congress.

Ancillary