Discussion
Although support for outpatient chronic diuretic use is lacking, many preterm children with BPD are prescribed diuretics after initial hospital discharge.(18) (19) In this study, we characterized clinical features associated with outpatient diuretic use and timing of diuretic weaning, in children < 32 weeks gestation with the diagnosis of BPD. We found that children diagnosed with severe BPD in the NICU were more likely to be on a diuretic at their first pulmonary visit compared to those who were not. Additionally, lower birthweight and earlier gestational age were associated with outpatient diuretic use in children with BPD. In the outpatient setting, thiazides were the most commonly prescribed diuretic, with most children prescribed either a thiazide alone (46.4%) or a thiazide and a potassium sparing agent (44.8%). Interestingly, children on a potassium sparing agent were just as likely to be prescribed potassium supplementation as those on non-potassium sparing diuretics. Children who were prescribed loop diuretics or greater than two diuretics at their first outpatient pulmonary visit were more likely to have comorbidities, including tracheostomy, home ventilator use and pulmonary hypertension. Healthcare providers guided diuretic weaning in the majority of patients and most children in this study were weaned off diuretics by two years of age. Diuretic weaning was slower in children born at an earlier gestational age, who had comorbidities, and who were prescribed loop diuretics. Children on public insurance also had slower diuretic weaning, suggesting that socioeconomic barriers to weaning may exist in this group of children. There was no difference in supplemental oxygen weaning between those on one or no diuretic, suggesting that home use may not be necessary in some children with BPD weaning off home supplemental oxygen.
In our study, thiazide monotherapy or dual therapy with a thiazide and potassium sparing agent were the most commonly prescribed diuretic(s) in children with BPD at their first pulmonary outpatient visit. This finding supports other studies which found that these diuretics are commonly used in the outpatient setting to treat children with BPD.(20) In our study we found no significant differences in gestational age, birthweight, BPD severity, respiratory support requirements or pulmonary hypertension diagnosis between children on thiazide monotherapy or dual therapy with a potassium sparing agent. Furthermore, those on a potassium sparing agent were just as likely to be on electrolyte supplementation as those on other diuretics. Another study that examined diuretic use in NICUs across the United States also found only modest reductions in use of potassium chloride supplementation in infants with BPD on thiazides plus potassium sparing agents compared to those on thiazides alone.3 Our findings indicate that the combination of a thiazide plus a potassium sparing agent may not provide additional benefit over a thiazide alone, with regard to electrolyte imbalances.
In the NICU, loop diuretics are commonly used to treat infants with severe BPD. 2,19 Slaughter et. al., found that furosemide was the most commonly used diuretic in the NICU, however thiazides in the NICU were used for the longest duration of use.19 In our study, we found that only 8.4% of children were on a loop diuretic at their first outpatient pulmonary visit. This finding supports a study by Bamat et. al., in which reported use of loop diuretics at NICU discharge ranged between 13.5% and 5.1% in children with grade 2 or 3 BPD.2 Although infrequently used in our outpatient population of BDP children, those prescribed loop diuretics (n=62) were more likely to be technology dependent and to be older at NICU discharge compared to those on non-loop diuretics (n=675). Those on loop diuretics also had a higher likelihood of home oxygen dependency, gastric tube use and the diagnosis of pulmonary hypertension. These children were also more likely to require electrolyte supplementation. Our study indicates that loop diuretics, although commonly used in the NICU, are less likely to be prescribed in the outpatient setting in children with BPD; with their use primarily limited to those with significant disease co-morbidities and technology dependence. Our study also found that BPD severity in the NICU at 36 weeks post menstrual age (PMA) did not correlate with timing of diuretic weaning in the outpatient setting, suggesting that NICU severity scores at 36 weeks gestation are not predictive of outpatient diuretic weaning. Not surprisingly however, we found that slower diuretic weaning was associated with children on home supplemental oxygen, loop diuretics, inhaled corticosteroids and technology dependence.
Of further interest, we found that children with BPD on public insurance in the outpatient setting were more likely to wean slower from diuretics, compared to children on private insurance. We previously found that children with BPD on public insurance were more likely to utilize acute care and have nighttime respiratory symptoms during the first 3 years of life, suggesting that these children may be sicker as outpatients and/or have impaired access to timely healthcare.(21) We also previously found that children on public insurance had delayed weaning of home supplemental oxygen compared to those on private insurance.(22) Social determinants of health may be a major factor that influences rate of weaning diuretics in the outpatient setting in children with BPD, causing prolonged and possibly unneeded exposure to diuretics. Although public insurance provides a safety net for children, those on public insurance may also have greater challenges due to poverty, access to care, or food insecurity. There is an ongoing need to understand the relationship between social determinants of health and outpatient outcomes and medication use in children with BPD.
Interestingly, in the subset of children on home supplemental oxygen we found no difference in oxygen weaning between those on no diuretics versus one diuretic. This suggest that children precribed one diuretic may be unnecessarily exposed to a diuretic when weaning off home supplemental oxygen. However, we did find that weaning of home oxygen was slower in those children prescribed 2 or more diuretics. Our findings suggest that children prescribed two or more diuretics and who are on home supplemental oxygen, may either have greater lung disease severity and/or that healthcare providers are more reluctant to wean supplemental oxygen due to a perception of more severe lung disease.
Our study has limitations including the retrospective nature of the study and inclusion of only two sites. However, our study included both children both prescribed or not prescribed diuretics and those on and off home supplemental oxygen, providing us an opportunity to study differences in patient characteristics between these groups and to examine diuretic weaning practices in children on or off home supplemental oxygen. Although we noted practice variations between the two sites, thiazides were found to be the most frequently prescribed diuretic in the outpatient setting, in contrast to loop diuretics which have been reported to be the most commonly prescribed diuretic in the inpatient setting. The outpatient preference to use thiazides may reflect the better safety profile of thiazides compared to loop diuretics. Future prospective studies are needed to address outpatient practice variation across diverse geographic areas and to more comprehensively phenotype patients in the outpatient setting to further distinguish those who may benefit most from outpatient diuretic usage.
In summary, diuretic use is common in children with BPD in the outpatient setting. Patient characteristics are similar between those on one or two diuretics, whereas children on loop diuretics are more likely to be technology dependent and to have comorbidities, including pulmonary hypertension. Interestingly, there were no differences in weaning off home supplemental oxygen between those on one or no diuretic. Finally, we noted that children on public insurance are more likely to be weaned slower from diuretics in the outpatient setting. Further studies are needed to understand how outpatient diuretic use influences long-term respiratory outcomes in BPD and how social determinants of health can influence diuretic weaning. Understanding these associations will help optimize health outcomes and potentially lessen side-effects of outpatient diuretic use in children with BPD.