Results
Ninety-two infants were discharged from our NICU; 71 were discharged with home oxygen. Our referral group included 54 patients discharged from other NICUs that were seen in pulmonary clinic for home oxygen management. Of the 125 infants evaluated in our home oxygen clinic, five patients were lost to follow up between NICU discharge and 1 year, leaving 120 patients as our final cohort for analysis. The median gestational age of the cohort was 27 weeks (IQR 23-30); 56 were females and 64 were males. The corrected median gestational age at discharge was 38 weeks (IQR 37-40) for infants with Grade 1 BPD; 43 weeks (IQR 40-46) for infants with Grade 2-3 BPD, and 39 weeks (IQR 38-41) for infants with BPD of unknown grade referred to our clinic from other NICUs.
The median pCO2 at NICU discharge was 55 mm Hg (IQR 51-58), with no significant differences between the NICU group (median 55, IQR 52-59) or the referral group (median 55, IQR 50-58; p=0.344). Only 18 infants from either the NICU group (n= 11) or the referral group (n= 7) were discharged home with pCO2 ≥ 60 mmHg. Table 1A displays associations between pre-discharge pCO2 and NICU clinical illness characteristics. Infants with higher discharge pCO2 were born at an earlier gestational age and had more days of mechanical ventilation; at discharge they were prescribed higher liter flow of oxygen and more bronchodilators. Otherwise, there were no other significant associations between discharge pCO2 and NICU variables, either in the total cohort, or in the NICU and referral groups when examined separately.
The median pCO2 at 36 week corrected gestational age for the NICU group was 59 (IQR 53-63). Table 1B displays associations between 36-week pCO2 and NICU illness characteristics. Higher 36-week pCO2 was associated with earlier gestational age, lower birth weight, patent ductus arteriosus ligation, severe BPD, more days of mechanical ventilation, and more days of supplemental oxygen while in the NICU. At discharge, a higher 36-week pCO2 was associated with later corrected gestational age at discharge and discharge with diuretics.
For our primary outcome of readmissions, 23% of infants experienced at least one respiratory readmission in the year after NICU discharge. There was no significant association between either pre-discharge or 36-week pCO2 and respiratory readmission. Even after logistic regression adjusting for other significant measures of illness severity including gestational age, ductus arteriosus ligation, BPD severity, and diuretic use, neither pre-discharge nor 36-week pCO2 were associated with readmissions (Table 2).
Associations between pre-discharge and 36-week pCO2 and secondary outcomes are shown in (Table 3). There was no association between either pre-discharge or 36-week pCO2 and secondary hospital encounter outcomes of emergency department visits, number or duration of admissions, intensive care unit admissions or intubation. For respiratory medical management, a higher pre-discharge pCO2 was associated with receipt of systemic steroids and being prescribed new or increased inhaled corticosteroids.
For home oxygen management, higher 36-week pCO2 was associated with later corrected gestational age at which oxygen was discontinued (Figure 2), longer duration of home oxygen in weeks, failing more clinic room air trials and home overnight oximetry studies (Table 3). Higher pre-discharge pCO2 was only associated with failing clinic room air trials. Infants with more clinic visits in the year after NICU discharge had higher 36-week and pre-discharge pCO2, but there were no differences in missed appointments or in unsupervised oxygen discontinuation.
The median time to discontinuing home oxygen for all infants was 21 weeks following NICU discharge; 85% of patients were off oxygen by 1 year following NICU discharge (Figure 3). For the 19 infants still receiving oxygen at one year following NICU discharge, 12 had experienced hospital admissions, ER visits and other significant respiratory complications. Four of those infants also had pulmonary hypertension that required medical management. The other 7 still requiring supplemental oxygen at one year following NICU discharge had poor outpatient follow up.