Birgitta Versluijs

164 Predictors for allo-LS direct progression of RSV-disease in HCT is a ground for delay of transplant or antiviral treatment. Somewhat counterintuitive we believe it is an indication for prolonged immu- nosuppression to prevent immune mediated lung disease later on. Donor immunity is thought to be the main culprit in lung complications shortly after transplant. Respiratory viruses and the respiratory microbiome. More and more is known about the role of microbiota in human health. So far research has largely focused on the gut microbiota, also in the context of GvHD. But the micro- bial ecosystem at other body sites, including the respiratory tract is under growing at- tention. Host and environmental factors influencing the respiratory microbiota include genetics, microbial exposure (birth mode, feeding type, day care), vaccination, infections and antibiotics. 32 Viral infection interacts with the microbiome by disrupting the airway epithelial barrier facilitating bacterial adhesion, liberating host derived nutrients and de- creasing muco-ciliary clearance. In addition RV can modulate innate and adaptive im- mune responses promoting bacterial colonization. 32 The role of a disturbed respiratory microbiome/virome in lung disease is postulated for asthma and chronic obstructive pulmonary disease (COPD). 33 Respiratory viruses and Allo-LS The definition criteria for Allo-LS 34,35 describe the clinical, radiologic and functional as- pects of lung pathology, with exclusion of other evident causes of this phenotype, like heart failure and infection, including RV infection. One can argue if this holds true for RV detected by PCR. The detection modes have become much more sensitive over time, so the impact of positive findings on the disease criteria should be reevaluated. In our cohort of HCT recipients we show a high RV prevalence, with RV persisting for weeks after HCT because of low immunity, with the onset of respiratory symptoms only after a median of 8.5 weeks after HCT. So, at the time of Allo-LS diagnosis, RV is still present in the majority of patients, but should not be regarded as the direct cause of the lung disease, and therefore must not be seen as an exclusion criterium for IPS nor BOS. An interesting paper in this matter was recently published by Seo et al. 36 In 69 patients with IPS, they went back to BAL samples at time of diagnosis, and applied more sensitive diagnostics for microbial pathogens. In 56% of patients an occult pathogen was found, 36% being a RV. All patients were treated with steroids because of IPS. Overall mortality was higher in the group of patients with an occult pathogen, than in the group without. The authors conclude that these patients had had to be excluded as IPS patients, that they had infectious pneumonia and that steroid treatment had adversely influenced their outcome. 9