Discussion:
Our study demonstrates that a RT-driven HFNC management protocol can be safely implemented for pediatric critical asthma patients in the PICU. When used concurrently with a continuous albuterol weaning protocol, it can reduce HFNC duration, PICU LOS and hospital LOS. This was done without increased use of NIV and IMV or a sustained increase in 7-day PICU and hospital readmission rates.
The use of HFNC has become more common in PICUs over the course of the last two decades, and its efficacy in other similar disease processes such as bronchiolitis has been well established (23,24). In critical asthma, it has been shown to be safe, but limited data has not shown significant clinical benefits or decreased length of stay with its use (25,26). Many physicians utilize HFNC as a delivery mechanism for aerosolized medications such as continuous albuterol, which is shown to be effective at lower levels of HFNC (27). It is possible that despite clinical improvement, patients remained on HFNC to facilitate continuous albuterol delivery, which could help explain why there was not a decrease in duration of HFNC in PDSA 3 and 4.
There are other therapies used in the treatment of pediatric critical asthma aside from inhaled beta agonists. Steroids have long been a mainstay in treatment (28,29). Other therapies, such as magnesium sulfate, aminophylline, and terbutaline have had mixed results (30-32), and because of this, institutions tend to have different protocols for management of critical asthma. In our study, there was no increase in the utilization of other adjunct medications which might affect our outcome measures. It would be useful to have more definitive guidelines on the use of adjunctive therapies in pediatric critical asthma in the future to improve outcomes.
There is concern that with widespread use of HFNC in other disease processes, hospitalization costs have increased (7,8). In many hospitals, HFNC is used exclusively in the PICU which could be a driver of the increased cost. Concerns have been raised regarding the environmental impact associated with increased use of HFNC related to carbon emissions (33). While we recognize HFNC as an important respiratory support modality in many disease processes, it is important to be judicious in its use. Often, patients in respiratory distress are placed on HFNC in the emergency department as it is a quick and relatively easy way to provide respiratory support. However, a patient may rapidly improve in the PICU and no longer require HFNC treatment, but this goes unrecognized by the care team because of higher acuity patients, which leads to PICU and hospital stays, higher costs and increased environmental impact. Standardization of care, such as RT-driven HFNC management protocols, can help mitigate these issues.
There was no increase in adverse events in patients with critical asthma with the implementation of the HFNC management protocol and its subsequent modifications. During PDSA 1, there was a small but statistically significant increase in PICU and hospital readmission with return to baseline low levels in subsequent PDSA cycles. The rate of IMV decreased during the study. A similar study on the utilization of a HFNC management protocol in patients with bronchiolitis showed a similar pattern (12). This could be due to increased provider comfort with utilizing HFNC and other noninvasive respiratory support modalities or related to improvement in the care of patients with critical asthma in general. While previous studies have shown that HFNC decreases IMV rates in other disease processes such as bronchiolitis (23,24), this is less clear in critical asthma.
This study has several limitations. This was a single center quality improvement project, which may limit its generalizability to other centers that have different practices for treatment of critical asthma. While the PASS is a validated measure, the Riley Hospital Respiratory score is not, though it is similar to other scoring systems used in other studies (34-36). Like many other PICUs, RT staffing was limited at times which may have led to increased length of HFNC due to other more acute needs. In addition, other factors can influence PICU and hospital LOS, such as ward bed availability, nurse staffing, and social issues that prevent timely discharge. These factors are difficult to monitor and are outside the scope of this study. Finally, our protocol did not include a standardized criteria to start HFNC. Given the national trend of trading conventional oxygen therapy via facemask with HFNC in pediatric critical asthma (37), future interventions can aim to standardize the HFNC initiation in emergency rooms and PICUs.