3-Year Outcomes
At 36 months post-tracheostomy placement, 44 children (26.8% of all children; 33.1% of alive children) were still using respiratory support and 100 children (61.0% of all children; 75.2% of alive children) were still cannulated with their tracheostomy (Figure 1b ,Table 3 ). Nearly one-fifth (18.9%) of children died within 3 years post-tracheostomy. Six children were lost to follow up by 3 years post-tracheostomy; these children were lost to follow up at median 8.9 months post-tracheostomy (IQR: 7.6-11.3 months).
In unadjusted analysis, fewer children with P. aeruginosaidentification were decannulated at 3 years post-tracheostomy as compared with children without P. aeruginosa (18.2% vs. 35.0%, p=0.07, Table 3 ) though this difference did not reach statistical significance. No differences in respiratory support, decannulation, or mortality outcomes were identified for children with any pathogen identification vs. no pathogen identification or for children with chronic P. aeruginosa identification vs. no chronic identification.
In adjusted analysis, having any P. aeruginosa identification was associated with lower probability of decannulation by 3 years post-tracheostomy (adjusted risk ratio [aRR] 0.48, 95% confidence interval [95% CI] 0.23-0.98, Figure 2 ). Furthermore, when analyzing count of P. aeruginosa isolations during the timeframe, each identification of P. aeruginosa was associated with a 35% lower probability of achieving decannulation by 3 years post-tracheostomy (aRR 0.65, 95% CI 0.44-0.98). P. aeruginosaidentification was not associated with ongoing use of respiratory support or mortality at 3 years post-tracheostomy on adjusted analysis. Any pathogen identification was not associated with any of the measured 3-year outcomes.
In further examination of covariates, increasing number of CCCs per child was found to be significantly associated with the primary outcome of ongoing use of respiratory support at 3 years post-tracheostomy for all models regardless of the predictor group (Appendix Table ). Non-White, non-Black race (classified as “Other” race) was identified to be a significant covariate in the association of culture predictors with mortality by 3 years.