4-DISCUSSION
This randomized controlled study involving children aged between 1 and 24 months diagnosed with moderate and severe bronchiolitis requiring supplemental oxygen determined that HFNCO therapy provided a notable improvement in HR, RR, and CRS over time compared with LFO therapy. This treatment modality offers a heated, humidified, and high-flow oxygen concentration regardless of the patient’s effort;24 additionally, it optimizes the gas exchange by decreasing dead space volume and maintaining mean airway pressure.24,25 The improvement in mucociliary clearance,25 with increased alveolar ventilation, could positively affect high HR and RR baseline values at admission. Assessing the disease severity is essential for the clinical management of patients and clinical research. Although many scoring scales exist in the literature, there are conflicting results regarding the effect of HFNCO therapy on CRS.11,14,15 Those contradictory outcomes might be associated with variable clinical score systems used in studies. Because of the presence of many care providers for these patients, we used the CRS reported by Liu et al.,19 an excellent inter-rater agreement (82-88%), to minimize evaluation bias. However, comprehensive studies focusing on valid and reliable clinical tools involving objective parameters are needed.
A recent study conducted in children with moderate-to-severe bronchiolitis found no notable difference in the length of oxygen requirement and LOS between therapy groups, suggesting that early use of HFNCO does not modify the underlying disease process.13 Another randomized, controlled study demonstrated a lower rate for escalation of care in HFNCO therapy but no evidence for shorter LOS and the length of oxygen requirement.12 Similar to these studies, our secondary results showed a comparable LOS and a lower rate for escalation of care in HFNCO therapy. Surprisingly, we determined that HFNCO therapy might shorten the length of oxygen requirement. While many studies showed no differences in the length of oxygen requirement,12,13,27 a study conducted in the ICU suggested that children treated with HFNCO had a shorter duration of oxygen than those treated with LFO.28 Besides, we note that secondary results of trials should be assessed carefully because of the power analysis calculated for primary outcomes. In our study, we administered HFNCO, at a flow rate of 2 L*kg/min, to the patients admitted to the general ward and did not identify any severe AEs. Although recently published reports did not determine any severe AEs,10-13,26,28 critical air leak syndrome has been described in children treated with HFNCO.29 Therefore, HFNCO therapy, a relatively new mode of respiratory support and increasingly being used, may be performed by closely monitoring the patients’ clinical status.
This study had some limitations. First, it was performed in a single center, and we could not reach the number of participants calculated with the power analysis because of the financial support problem. Although we used effect size for each of the primary outcomes, the generalizability of the recommendations reduced. Second, we did not blind the allocation of oxygen therapies to introduce the risk of performance bias. Finally, the patients were not allocated to receive therapy in a ratio of 1:1. Therefore, the results may have been affected by this diversity.
One of our trial’s strengths was using a clinical scoring system for children with bronchiolitis, a valuable tool for monitoring the clinical severity of bronchiolitis. Although there may be a potential bias in any clinical scoring system, CRS used in this study has been identified to have an excellent inter-observer agreement. Second, we performed the parent interview questionnaire by contacting parents 15 days after discharge to increase the accuracy of the results for each treatment arm.
In conclusion, HFNCO may provide enhanced respiratory support with a notable improvement in HR, RR, and CRS than LFO. Comprehensive studies are needed to assess the clinical efficacy of HFNCO therapy.