Shana Wauters

Long-term azithromycin therapy for bronchiolitis obliterans syndrome: Divide and conquer?

BACKGROUND Azithromycin may reverse or halt the decline of pulmonary function (FEV(1)) in bronchiolitis obliterans syndrome (BOS). In this study we investigated the effects of long-term azithromycin treatment in lung transplant recipients with BOS. METHODS A retrospective, observational, cohort study was performed on 107 patients with BOS (Stages 0p/1/2/3, n = 23/62/20/2), who were treated with azithromycin for 3.1 ± 1.9 years. Patients were evaluated 6.3 ± 3.8 years after transplantation and assessed for evolution of FEV(1), bronchoalveolar lavage neutrophilia and overall survival after initiation of azithromycin. Survival curves were analyzed using the log-rank test. Cox proportional hazard survival regression analysis was performed to estimate hazard ratios of clinical variables predicting outcome. RESULTS FEV(1) increased ≥ 10% after 3 to 6 months of treatment in 40% of patients, of whom 33% later redeveloped BOS. FEV(1) further declined in 78% and stabilized in 22% of the remaining non-responders. Pre-treatment neutrophilia was higher in responders: 29.3% (9.3% to 69.7%) vs 11.5% (2.9% to 43.8%) (p = 0.025), in whom it significantly decreased to 4.2% (1.8% to 17.6%) (p = 0.041) after 3 to 6 months of azithromycin. Responders demonstrated better survival compared with non-responders (p = 0.050), with 6 and 21 patients, respectively, dying during follow-up (p = 0.027). Multivariate analysis identified initial azithromycin response and earlier post-transplant initiation of azithromycin to be protective for both BOS progression/relapse (hazard ratio [HR] = 0.12 [95% confidence interval 0.05 to 0.28], p < 0.0001; and HR = 0.98 [95% confidence interval 0.97 to 0.98], p < 0.0001, respectively) and retransplantation/death during follow-up (HR 0.10 [95% confidence interval 0.02 to 0.48], p = 0.004; and HR 0.96 [95% confidence interval 0.95 to 0.98], p < 0.0001, respectively). CONCLUSIONS Long-term azithromycin benefits pulmonary function and survival in BOS, particularly in patients with increased lavage neutrophilia.

Early Outcome After Lung Transplantation From Non–Heart-Beating Donors is Comparable to Heart-Beating Donors

BACKGROUND The use of non-heart-beating donors (NHBD) to overcome organ shortage is moving into the clinic. In 2007, 5 of 51 lung transplantations (LTx) in our center were performed with lungs from controlled NHBD. METHODS Our aim was to describe these 5 NHBD LTx recipients and compare early outcome (<or= 6 months) with a cohort of 10 heart-beating donor (HBD) LTx recipients matched for age, gender, underlying disease, and time of surgery. Clinical outcomes were assessed, including ischemic times, gas exchange, primary graft dysfunction, time to extubation, time of discharge from intensive care unit/hospital, and pulmonary function. Airway and systematic inflammation were evaluated by bronchoalveolar lavage, total and differential cell counts, and plasma C-reactive protein levels, respectively. RESULTS Early outcome in the NHBD group was comparable to the HBD group at the clinical and inflammatory level. The NHBD group showed a trend for earlier extubation (p = 0.054), greater increase in forced expiratory volume in 1 second (p = 0.054), and a significantly lower number of infections (p = 0.01). The NHBD group also had lower numbers of total cells (p = 0.04) and macrophages (p = 0.03) in bronchoalveolar lavage on day 21. CONCLUSIONS Outcome after LTx in NHBD recipients is not inferior to HBD recipients during the first 6 months. Late results and effect on chronic rejection should be further awaited. Controlled NHBD may offer a valid source of lungs to overcome organ shortage in LTx.

Medium-term outcome after lung transplantation is comparable between brain-dead and cardiac-dead donors

BACKGROUND Donation after cardiac death (DCD) to overcome the donor organ shortage is now moving into the clinical setting, but the medium-term outcome after DCD lung transplantation (LTx) remains largely unknown. METHODS In this retrospective study, DCD LTx recipients (n = 21) were compared with a cohort of donation-after-brain-death (DBD) LTx recipients (n = 154) transplanted between February 2007 and July 2010. Immediate (post)operative outcome was evaluated by assessing need for peri-operative extracorporeal membrane oxygenation (ECMO), time to extubation, hospital discharge and primary graft dysfunction (PGD) within the first 48 hours. Survival, incidence of bronchiolitis obliterans syndrome (BOS), acute rejection (AR) and inflammatory markers in blood and in bronchoalveolar lavage (BAL) were assessed and compared over a median follow-up of 327 days for DCD and 531 days for DBD, showing no statistically significant difference (NS). RESULTS There were no differences between groups with regard to patient characteristics except for a higher number of patients transplanted for obliterative bronchiolitis in the DCD group (4 of 21 vs 7 of 154; p < 0.05). In the DCD group, 2 of 21 patients died, vs 23 of 154 patients in the DBD group (NS). Actuarial survival rates at 6 months, 1 year and 3 years are 95%, 95% and 71% for the DCD group and 96%, 91% and 75% for the DBD group (NS). Three patients (14%) in the DCD group developed BOS vs 15 patients (10%) in the DBD group (NS). Survival and freedom from BOS were not different between the groups. AR, inflammatory markers and immediate (post)operative outcome also did not differ. CONCLUSIONS In our experience, both early- and medium-term outcome in DCD lung recipients is comparable to that of DBD lung recipients. Use of lungs recovered from controlled donors after cardiac death is a safe option for expansion of the donor pool.

Chronic rejection pathology after orthotopic lung transplantation in mice: the development of a murine BOS model and its drawbacks.

Almost all animal models for chronic rejection (CR) after lung transplantation (LTx) fail to resemble the human situation. It was our attempt to develop a representative model of CR in mice. Orthotopic LTx was performed in allografts receiving daily immunosuppression with steroids and cyclosporine. Controls included isografts and mice only undergoing thoracotomy (SHAM). Allografts were sacrificed 2, 4, 6, 8, 10 or 12 weeks after LTx. Pulmonary function was measured repeatedly in the 12w allografts, isografts and SHAM mice. Histologically, all allografts demonstrated acute rejection (AR) around the blood vessels and airways two weeks after LTx. This decreased to 50–75% up to 10 weeks and was absent after 12 weeks. Obliterative bronchiolitis (OB) lesions were observed in 25–50% of the mice from 4–12 weeks. Isografts and lungs of SHAM mice were normal after 12 weeks. Pulmonary function measurements showed a decline in FEV0.1, TLC and compliance in the allografts postoperatively (2 weeks) with a slow recovery over time. After this initial decline, lung function of allografts increased more than in isografts and SHAM mice indicating that pulmonary function measurement is not a good tool to diagnose CR in a mouse. We conclude that a true model for CR, with clear OB lesions in about one third of the animals, but without a decline in lung function, is possible. This model is an important step forward in the development of an ideal model for CR which will open new perspectives in unraveling CR pathogenesis and exploring new treatment options.

Lymphocytic Bronchiolitis After Lung Transplantation Is Associated With Daily Changes in Air Pollution

Acute rejection represents a major problem after organ transplantation, being a recognized risk for chronic rejection and mortality. Recently, it became clear that lymphocytic bronchiolitis (LB, B‐grade acute rejection) is more important than previously thought, as it predisposes to chronic rejection. We aimed to verify whether daily fluctuations of air pollution, measured as particulate matter (PM) are related to histologically proven A‐grade rejection and/or LB and bronchoalveolar lavage (BAL) fluid cellularity after lung transplantation. We fitted a mixed model to examine the association between daily variations in PM10 and A‐grade rejection/LB on 1276 bronchoscopic biopsies (397 patients, 416 transplantations) taken between 2001 and 2011. A difference of 10 μg/m3 in PM10 3 days before diagnosis of LB was associated with an OR of 1.15 (95% CI 1.04–1.27; p = 0.0044) but not with A‐grade rejection (OR = 1.05; 95% CI 0.95–1.15; p = 0.32). Variations in PM10 at lag day 3 correlated with neutrophils (p = 0.013), lymphocytes (p = 0.0031) and total cell count (p = 0.024) in BAL. Importantly, we only found an effect of PM10 on LB in patients not taking azithromycin. LB predisposed to chronic rejection (p < 0.0001). The risk for LB after lung transplantation increased with temporal changes in particulate air pollution, and this was associated with BAL neutrophilia and lymphocytosis. Azithromycin was protective against this PM effect.

A Model of Ex Vivo Perfusion of Porcine Donor Lungs Injured by Gastric Aspiration: A Step Towards Pretransplant Reconditioning

BACKGROUND Evidence of aspiration remains a major reason for declining donor lungs contributing to current organ shortage. The aim of the present ex vivo lung perfusion (EVLP) study was to compare lungs injured by gastric juice (GJ) with normal lungs. METHODS Pigs (32.3 ± 11.2 kg) were divided into two groups. Study group [GJ; n = 6] was intratracheally instilled with GJ, while sham bronchoscopy was performed in control group [C; n = 6]. Graft function was assessed during EVLP for 2 h. Oxygenation, aerodynamic, and hemodynamic parameters were recorded every 30 min. Wet to dry weight ratio (W/D) was calculated. Bronchoalveolar lavage was performed. Tissue samples were collected. RESULTS Pulmonary vascular resistance was higher and pulmonary flow was lower in [GJ] versus [C] at T120; (P < 0.05). Mean airway pressure was higher in (P < 0.05) and compliance was lower (P < 0.001) in [GJ]. No differences in oxygenation were seen between groups. W/D of left lung in [GJ] after EVLP was slightly (P < 0.05) higher compared with the nonperfused right lung. More neutrophils were present in [GJ] before (P < 0.01) and after EVLP (P < 0.05). Histologic alterations were more prominent in [GJ], but did not worsen after EVLP. CONCLUSION EVLP of injured lungs is possible for 2 h despite increased edema. This model could help to investigate ways to repair caustic lung injury during EVLP.

A Porcine Model of Acute Lung Injury by Instillation of Gastric Fluid

BACKGROUND About 15% of donor lungs are declined because of aspiration contributing to current organ shortage. The aim was to develop a porcine lung injury model by gastric juice (GJ) instillation to study different pretransplant treatment strategies. MATERIALS AND METHODS Pigs (n = 6/group) were anesthetized and monitored. At T0 bronchoalveolar lavage (BAL) was performed followed by instillation of 4 mL/kg GJ or saline solution (SAL). Hemodynamics, aerodynamics and oxygenation were recorded for two hours. Serum samples were collected. At T120 a second BAL was performed. CT scans of explanted, inflated lungs were taken, tissue samples were collected and wet/dry weight ratio (W/D) was calculated. Pepsin and bile acids were measured in BAL. IL-8, CRP and MMP-9 were measured in serum and in BAL. RESULT Oxygenation and lung compliance was lower in [GJ] versus [SAL] (P < 0.01 and P < 0.001, respectively). More consolidation areas were noticed on CT in GJ versus SAL (P < 0.01). Hemorrhage, edema and neutrophil inflammation were seen on histology in [GJ] (P < 0.01, P < 0.001, P < 0.001, respectively). BAL neutrophils, pepsin, bile acids, and IL-8 (P < 0.05) increased in [GJ]. W/D was higher in [GJ] versus SAL (P < 0.001). CONCLUSION Instillation of GJ in pig lungs caused acute lung injury with impaired oxygenation and increased inflammation in BAL, on histology, and on imaging. This model holds promise to assess the efficacy of a broad range of treatment strategies including ex vivo lung perfusion (EVLP).

Preemptive therapy with steroids but not macrolides improves gas exchange in caustic-injured donor lungs.

BACKGROUND Donor lungs are susceptible to aspiration in the period before and after brain death. We hypothesized that preemptive anti-inflammatory treatment may result in better graft performance as evaluated during ex vivo lung perfusion (EVLP). METHODS Pigs (27.4 ± 8.3kg) were divided into four groups (n = 6). Group [S] was treated with steroids (methylprednisolone 10 mg/kg), group [M] with macrolides (clarithromycin 250 mg), and group [P] with placebo (saline). In all groups, treatment was given i.v. 25, 18, and 1 h before instillation of 4 mL/kg gastric juice (GJ). Group [C] without GJ instillation served as a control group. Two hours after the onset of acute lung injury, lungs were flushed, explanted, and stored cold for 75 min. Graft performance was then assessed during EVLP for 2 h. Aerodynamic and hemodynamic parameters and oxygenation capacity (∆PO(2)) were recorded every 30 min (T30-T120). BAL samples were collected and analyzed for total and differential cells, IL-6, IL-1β, 8- isoprostane, and CRP levels. Wet-to-dry weight ratio [W/D] was measured and tissue samples were collected for histology. RESULTS ΔPO(2) in [S] was comparable to [C] and was higher versus [M] (P < 0.05) and [P] (P < 0.01). No differences were observed in pulmonary vascular resistance, lung compliance, and mean airway pressure. No additional edema was noticed after 2 h of EVLP. Cellular and biomolecular changes in BAL fluid and histologic alterations were comparable among the three study groups. CONCLUSION Preemptive treatment of donors with steroids but not macrolides improves gas exchange in a porcine lung injury model independently from its anti-inflammatory activity.

Evaluating lung injury at increasing time intervals in a murine brain death model.

INTRODUCTION Only 15%-25% of brain death (BD) donors match the ideal donor criteria for lung transplantation. Lung injury may evolve in the hours after onset of brain death, but the evolution over time has not been well studied in lung. The aim of this study was to evaluate lung injury at different time points after BD using a murine model. MATERIALS AND METHODS Male C57BL6/J mice (8-10 wk) were anesthetized, tracheotomized, and mechanically ventilated. Mice were randomly assigned to six groups (n=8/group): 1 h, 3 h, and 6 h sham ([SH1], [SH3], [SH6]) and 1 h, 3 h, and 6 h brain death ([BD1], [BD3], [BD6]). BD was gradually induced by a subdural balloon catheter. Heart rate and mean arterial pressure were continuously monitored. At the end of the experiment, bronchoalveolar lavage was performed and the left lung was excised for histopathologic analysis. RESULTS The Cushing reflex was characterized by a rapid increase in heart rate and mean arterial pressure after balloon inflation in BD animals. An increase in percentage of neutrophils was seen with a longer follow-up period (P<0.05). Interleukin 6 and interleukin 10 levels in bronchoalveolar lavage progressively increased with longer time intervals after BD ([BD1] versus [BD6]; P<0.01). Histologic signs of lung injury (congestion, hemorrhage, and neutrophilic influx) were more pronounced in [BD3] and [BD6] compared with the other groups; however, this difference did not reach statistical significance. CONCLUSION Three hours after brain death, significant signs of inflammation and lung injury were seen compared with sham-operated animals. This murine BD model gives us opportunities for further mechanistic studies regarding treatment of BD-related donor lung injury.

The need for a new animal model for chronic rejection after lung transplantation.

The single most important cause of late mortality after lung transplantation is obliterative bronchiolitis (OB), clinically characterized by a decrease in lung function and morphologically by characteristic changes. Recently, new insights into its pathogenesis have been acquired: risk factors have been identified and the use of azithromycin showed a dichotomy with at least 2 different phenotypes of bronchiolitis obliterans syndrome (BOS). It is clear that a good animal model is indispensable to further dissect and unravel the pathogenesis of BOS. Many animal models have been developed to study BOS but, so far, none of these models truly mimics the human situation. Looking at the definition of BOS, a good animal model implies histological OB lesions, possibility to measure lung function, and airway inflammation. This review sought to discuss, including pros and cons, all potential animal models that have been developed to study OB/BOS. It has become clear that a new animal model is needed; recent developments using an orthotopic mouse lung transplantation model may offer the answer because it mimics the human situation. The genetic variants among this species may open new perspectives for research into the pathogenesis of OB/BOS.