RT Journal Article
SR Electronic
T1 Late Breaking Abstract - Predicting long term lung function outcomes of prematurely born infants using cluster analysis
JF European Respiratory Journal
JO Eur Respir J
FD European Respiratory Society
SP OA273
DO 10.1183/13993003.congress-2019.OA273
VO 54
IS suppl 63
A1 Harris, Christopher
A1 Lunt, Alan
A1 Peacock, Janet
A1 Greenough, Anne
YR 2019
UL http://erj.ersjournals.com/content/54/suppl_63/OA273.abstract
AB Background: We postulated identification of multidimensional subgroups might improve prediction of long term outcomes and identify efficacious subgroup specific treatments.Aim: To detect if differences in lung function and exercise tolerance occurred in subgroups of 16-19 year olds born prematurely.Methods: Cluster analysis of 11 key pre and postnatal characteristics for 330 infants born at less than 29 weeks was performed. Decision tree analysis was used to derive a predictive algorithm for clustering based on a minimal subset of variables.Results: Four clusters were identified (data displayed as mean (SD) or n (%)). BPD, gestational age and sex were primary predictors of cluster membership with 96% accuracy. Lung function shown as z-scores, and exercise distance (ED) differed between clusters.In cluster D, lung function was better for those who received high frequency oscillation, eg FEV1 (0.19 versus -0.65, p=0.033).View this table:Conclusion: These results suggest possible subtype-specific treatment effects and may inform directed therapy approaches.FootnotesCite this article as: European Respiratory Journal 2019; 54: Suppl. 63, OA273.This is an ERS International Congress abstract. No full-text version is available. Further material to accompany this abstract may be available at www.ers-education.org (ERS member access only).