We hypothesized that the presence of interstitial pneumonitis (IP) on bronchoscopic lung biopsy is associated with an increased future risk of bronchiolitis obliterans syndrome (BOS)and death in lung transplant recipients.
We reviewed the lung biopsies of 146 transplant recipients followed at Brigham and Women's Hospital between January 1, 2000 and December 31, 2007. The presence of IP was defined by pathologic identification of interstitial inflammation and/or fibrosis on two or more biopsy specimens in the absence of organizing/infectious pneumonia. We developed Cox Proportional Hazard models using IP as a time-varying covariate to assess the hazard ratio (HR) for BOS, as defined by the International Society for Heart and Lung Transplantation, and death in patients with and without IP.
41 transplant recipients developed IP during the study period. Patients with and without IP were similar with respect to age at transplant, gender, underlying lung disease, transplant procedure type, and CMV status. Patients with IP had significantly shorter follow-up time than those without IP. Crude analysis showed a significant association between the presence of IP and subsequent development of BOS (HR 2.046, p=.0038). This association remained significant in a multivariable model (HR 2.105, p=.0048). The incidence rate of BOS in patients with IP was 0.475 per person year compared to 0.234 per person year in patients without IP. The development of IP did not significantly impact mortality (HR 0.930, p=.8122).
Our data suggest that lung transplant recipients with IP on bronchoscopic lung biopsy are twice as likely to develop BOS as those without IP. Discrepancies in length of follow-up between groups limits conclusions about the impact of IP on mortality.
The finding of IP on post-transplant lung biopsy may be a marker for severe or recurrent lung injury, such as primary graft dysfunction or recurrent aspiration, which influences graft survival. Further studies are needed to validate these findings and investigate the underlying causes of IP development in this setting.
Hilary Goldberg, No Financial Disclosure Information; No Product/Research Disclosure Information