Birgitta Versluijs

66 Materials and methods hematopoietic cell transplantation because, especially in this period, it is important to make a distinction between allo-LS and nonallo-LS. 16 At the start of the hematopoietic cell transplantation procedure, all parents and patients more than 12 years old signed informed consent for data collection and analysis in accordance with national and insti- tutional ethical board protocols. Characteristics of the patients included in the study are shown in Table 1. High-resolution CT HRCT examinations were performed using a single-detector row scanner (Tomoscan EVT, Philips Healthcare). In infants and young children, the anesthesiologist controlled the airway during the scanning procedure. Depending on the preference of the anes- thesiologist, the child was either intubated or a laryngeal mask was used. Studies were performed at 25-cm H2O pressure (inspiration) and 0-cm H2O pressure (expiration). In older children (from approximately 5 years old and older), studies were performed using a breath-hold technique at suspended inspiration and at expiration. Inspiration images were obtained using fixed 90 kVp and 18–60 mAs (depending on body weight). For ex- piration images, we used 90 kVp and 11 mAs. We made 1-mm slices every centimeter in inspiration and at five levels in expiration. FOV was adapted to patient size. High-resolution CT scoring For the purposes of this study, a scoring system for hematopoietic cell transplant re- cipients was developed on the basis of common HRCT findings in this patient group (Table 2). Relevant morphologic abnormalities (bronchiectasis, bronchial wall thicke- ning, tree-in-bud, nodules, consolidation, ground-glass pattern, and septa thickening) and expiratory airtrapping were scored for each lobe of the lung, including the lingula, on a scale ranging from 0 (no abnormality) to 3 (severe). Abnormalities were defined accor- ding to the Fleischner Society criteria. 17 Scores for each abnormality and the composite scores were expressed on a 0–100 scale as a percentage of the maximum score. After the first statistical analyses, we defined an allo-score on the basis of the individual items that showed to be significantly different between alloimmune and non-alloimmu- ne lung disease. All HRCT scans were scored by two independent observers (9 and 14 years of experience in HRCT interpretation) who were blinded to patient characteristics. One observer repea- ted the scoring after more than 1 month. 4