The opportunity to present my research today, my study that I'll be presenting is entitled hyphal nasal cannula use in hospital length of stay for pediatric asthma. Disclosure ahead is that this work is funded by an internal Indiana University research grant through the Division of Critical Care. The heated humidified hyphal nasal cannula is a common pediatric respiratory support device that's used in the ICU, and it has several proposed mechanisms of action. It works by facilitating washout of nasopharyngeal dead space, minimization of oxygen dilution, reduction of energy expenditure by helping with inspiratory flow, and possibly providing a small measure of continuous positive airway pressure. It can also be used to deliver nebulized medications. Thus, for these reasons, it may be beneficial in patients with, pediatric patients suffering from asthma. Several studies have looked at the use of hyphal nasal cannula in pediatric asthma with several different outcomes that have been measured. These include improvements in vital signs and laboratory values, such as decreased respiratory rate, heart rate, and CO2 levels. Other studies have looked at length of stay in the ICU and the hospital and seen in quality improvement studies, improvement in these. Some studies have shown decreased time needed for continuous albuterol for these patients, and some studies have used validated asthma symptom severity scores and showed improvements in these symptoms within the first two hours of use. Now, while we have seen some benefit, there is a potential debate regarding the high flows effectiveness as far as delivering nebulized aerosolized medications. In contrast to other common pediatric respiratory diseases, such as viral bronchiolitis and pneumonia, the first line evidence-based therapy for asthma is inhaled bronchodilator medications. While these can be delivered through the high flow device, there is some debate as to how well the medications are deposited in the lower airways using these devices. Thus, we may see some benefits from the physiologic mechanisms of the high flow, but how well does it deliver the evidence-based therapies needed for asthma? Several recent studies have used high-fidelity mannequins and looked at the mechanistic delivery of albuterol to these mannequins. Overall, most of these studies have shown poorer delivery of the nebulized medications through the high flow nasal cannula device, especially at higher flow rates. One of the more recent studies showed poorer aerosol delivery in antiflow rates above 4 to 6 liters per minute, which is concerning because the vast majority of pediatric patients admitted to the ICU are receiving flow rates above 6 liters per minute. Thus, this conflicting evidence leads to variability in institutional practice regarding the use of high-flow nasal cannula for pediatric asthma, and this provides the foundation for this study. The hypothesis that we had for this study was that pediatric asthma patients treated with high-flow nasal cannula have increased hospital length of stay compared to mass treatment. Moving on to the methods, this was a retrospective matched cohort study, and the inclusion criteria for the study population was all patients ages 2 to 18 years old admitted to the hospital with a diagnostic code for asthma between January of 2010 and December of 2021. The data for this study was collected through the Indiana Network for Patient Care, and this is a regional health information exchange that collects data from over 90 percent of the hospitals in Indiana, and this includes Riley Hospital for Children, the Academic Health Center, as well as all pediatric-specific hospitals in the state. The changes using the high-flow nasal cannula were defined by having an electronic health record order and a hospital charge for the device. No specific flow rates were used. Variables included in this data set and used in this study include admission vital signs, demographics, data regarding respiratory support modalities, medications, and geocoded data describing the social and environmental determinants of health. Patient matching for this study was done using logistic regression-based propensity score matching techniques. Each patient that met inclusion criteria had a calculated propensity score for their propensity to receive treatment with high-flow nasal cannula that ranged from 0 to 1. Each patient that met criteria for high-flow nasal cannula use was matched on a 1-to-1 ratio with the one potential control that was closest to their propensity score with a maximum caliper adjustment of 0.01, which means that any patient that was treated with high-flow nasal cannula but did not have a potential control that matched these criteria was not matched, was not included in the study. The statistical test used for comparison testing was the Mann-Whitney U for continuous variables and Chi-Square Fisher's exact test for categorical variables, depending on the sample size. This is the study flow sheet. At the top is the inclusion criteria, of which, in our data set, 23,659 patients met inclusion criteria. Of these, 2,197 were treated with high-flow nasal cannula, which left 21,462 as potential controls. Each of these patients underwent propensity score matching and was matched according to the previously described methods. And after matching was done, we were left with a total of 1,766 cases with an equal number of match controls. This figure is called a jitter plot, and what it describes is the propensity score distribution as well as the matching categories. The propensity scores are on the x-axis from 0 to 1, and the four potential matching categories are on the y-axis. The potential matching categories, starting from the top, is unmatched treated units, and these were patients that had propensity scores all above 0.5, and these treated units did not have a sufficient control to match to, so they were not included in the analysis. On the bottom is the unmatched control units, which all had propensity scores less than 0.5, which represent control patients that didn't have an adequate control to match to. In the middle two rows represent the matched treated and matched control units, and as you can see, the distribution of propensity scores are higher in the lower ranges that stretch to 1 and are fairly even to the eye. And what we can also see is, if you prefer overlapping histograms, this also displays the distribution of propensity scores between the high-flow case group and the controls. So this figure illustrates the hospital length of stay in days on the y-axis, the box plot for the high-flow nasal cannula group on the right, and the matched controls on the left. What we can see is that the high-flow nasal cannula group had a median total hospital length of stay of 3.65 days or 87 hours, and the matched control had a hospital total length of stay of 2.75 days or 66 hours, which was a statistically significant difference of about a 21-hour median difference. So this table compares the patient characteristics between the high-flow nasal cannula group and matched controls. All the variables included in this table and the next two tables were included in the matching algorithm. As we can see from this table, there was no significant difference in the sex, race, ethnicity, oxygen saturation, temperature, or frequency of pneumonia diagnosis between the groups. There was a significant difference in the heart rate and respiratory rate for the adolescent age group, with the high-flow nasal cannula group having significantly higher heart rates and respiratory rates in the 12- to 18-age group, but this difference was not seen in the lower-age groups. This table illustrates the social and environmental determinants of health that was used in this study. The social determinants of health included aggregate, or sorry, environmental determinants of health included aggregate particulate matter, air pollution risk score, and tobacco access score. All these were obtained through geocoding at either the zip code, census tract, or block group level, whichever was most available. There was no significant difference in the environmental determinants of health. The social determinants of health include life expectancy, median household income, percentage living in poverty, percentage living in a food desert, urban residence, transit score, and a composite score known as the social vulnerability index. And overall, there was no significant difference in the social determinants of health between the groups either. This final table illustrates the medications used, the respiratory support devices used, as well as the frequency of chest X-rays between the groups. The adjunctive asthma medications that were studied include ipratropium, methylprednisolone, magnesium, aminophiline, epinephrine, and terbutaline. The only significant difference between the groups was that the high-flow group had a higher frequency of ipratropium use compared to the matched controls, but there were no differences in any of the other adjunctive medications. Looking at the respiratory support devices, there was no difference in the rate of intubation and mechanical ventilation. However, the use of noninvasive was significantly higher in the high-flow group at 21% compared to 6.7%, and there was no significant difference in mortality. So because of the significant and substantial difference in noninvasive use between the groups, we did a sub-analysis looking at only the patients that didn't receive any positive pressure in the study. And overall, the boxplots are pretty similar to the previous boxplots. The difference is that the median length of stay in the high-flow group in this sub-cohort was 77 hours, whereas the matched controls were 63 hours. So the absolute difference did decrease from 21 hours in the whole cohort to 14 hours in the sub-analysis. So moving to the discussion, there's two potential explanations for the differences that we saw in the study. The first is that, in fact, the high-flow nasal cannula may not deliver nebulized bronchodilator therapy adequately as the mask does, which may lead to continued symptom prevalence, which may lead to a longer length of stay. The other potential explanation is that despite patient matching, it's possible that the high-flow nasal cannula group still just had a higher degree of disease burden, which led to a prolonged length of stay. We do have two potential avenues that we're moving towards to try and validate these findings. The first is we're developing an enhanced dataset that will include several important variables that weren't included in this initial study. These include longitudinal vital signs, asthma severity scores, as well as flow rates. This will also allow us to hopefully have better patient matching as well as more granular patient outcomes besides length of stay. The other step we're doing is we're working on obtaining similar datasets from other centers to further validate these findings. So in conclusion, the use of high-flow nasal cannula for pediatric asthma is associated with a longer length of hospital stay, but further studies are needed to validate these findings. This represents a list of everyone that was on the study team. I appreciate all the help that they've given to make this possible. Thank you very much.

high-flow nasal cannula

pediatric asthma patients

hospital length of stay

ICU

nebulized medications