Chapters Transcript Video High-Flow Oxygen Therapy for Bronchiolitis: Steps to Wiser Use in Pediatrics Good morning everyone. Thank you for attending Graham rounds this morning. My name is Irma Seabrook, the director of respiratory care services, pulmonary function laboratory and sleep medicine. Today's lecture will be presented by DR Manti chata, pick your intense office and the Medical director for respiratory care Dr gang, Dr dane Barringer and Dr Sarah Co pediatric Hospitalist and heather esparza management coordinator for respiratory care services. The respiratory therapists were vital partners to the success of this project and I would like to take a few minutes to acknowledge heather for her exceptional leadership and for the clinical guidance and support she gave to this team. We couldn't have done this without her. These esteemed presenters will share qualitative and quantitative results from our high flow project and I and will expound on the inter professional collaboration between physicians, respiratory care practitioners, nurses and other clinical stakeholders who came together to enhance the delivery of care research has shown that inter professional collaboration if enriched in the forethought of what we do and the care we provide for our families and patients will positively of impact health outcomes. This project was a great testament to that. I will turn it over now to dr mandy chatter who will be our first presenter. Good Morning everyone. Uh thanks so much for the introduction and also to the grand rounds committee for allowing us to join you this morning and and share our work. Um I wanna say that all the presenters today, my co presenters are all seasoned professionals in their own right. But this project required all of us to venture a little bit outside of our respective domains of expertise. And so we have a lot of thanks uh to start with not only two people in the BCH community who mentored us through this quality improvement experience um and the quality and safety team here at B. C. H. And of course all of our colleagues and our respective divisions um in helping us complete this largest Q. I. Effort to date for our group. I also want to highlight the collaboration that we had with our cross bay colleagues especially dr model um and um David Wolsey from restaurant therapy um at Mission Bay. There is lots of room for that collaboration to grow. But today my co presenters and I wanted to really highlight um the fabulous team effort that we had here in Oakland especially by our multidisciplinary um heated high flow nasal cannula committee, the R. T. Department and others who made this work possible. So so thanks to all of you um this is our disclosure and uh you know the objectives of our talk are to describe briefly the path of physiology and clinical characteristics of Children with bronchiolitis. Uh to list the evidence based therapies used in the management of Children admitted with bronchiolitis and acute restaurant failure and to describe how quality improvement methodology can be used to address overuse of heated high flow nasal cannula therapy for bronchiolitis and reduce length of stay and cost it would seem obvious to to begin with a simple definition for bronchiolitis, but that's not a simple task, in large part in pediatrics. When we talk about bronchiolitis, we're talking about a basic pattern of disease and there are a few things that most practitioners can agree on in this definition, it usually is used to describe the first or second episode of restaurant distress in infants or young Children that's associated with a viral respiratory illness. And typically uh, there are symptoms of lower restaurant tracked involved involvement, including things like wheezing or crackles, although different diagnostic criteria, especially from different international sources. Uh use different criteria for lower restaurant tract involvement. Um, there are aged related differences in the severity of disease, in symptomology, in the response to therapy and in the progression of the disease, we now know that there are differing molecular immune signatures among the broad population of patients diagnosed with bronchiolitis. There are differences among the positive viruses and of course there is a continuum with other disease states including chronic disease and a topic disease. So a question uh, bronchiolitis is an infection of the restaurant track characterized by which of the following brain area, wheezing, hypoxia, mia, takinmia or all of the above. And if you have your answer in mind, hopefully you recognize that we see all of these things in this broad entity called bronchiolitis. Just a moment to talk a little bit about the path of physiology of bronchiolitis. Um you can see on the left uh the histological appearance of a healthy bronchial. You see a nice patent lumen with a healthy blood supply and patent Albiol. I and you can see in the lumen um um continuum of epithelial cells doing their vital functions. And then the lower left, you can see uh the same image in a patient with bronchiolitis that looks quite different. You can see all of the the classic path pathologic signs of this disease. We know that with spread of infection from the nasopharynx to the lower rest, retract epithelial cells are infected and become deranged. They may be slaughtered and aspirated. And as the infection continues to move down into the lower airways there is a significant increase in inflammatory cell infiltration, edema, increased mucus secretion and impaired cilia reaction. And ultimately, uh we see that the accumulation of this debris and the degree of edema and bronco spasm can lead to intra Luminal obstruction and air trapping and eventually some degree of ad electrolysis. And and so in that respect, um gas exchange is significantly impaired. You can see also in all of these critical steps uh the the ultimate manifestation of the symptoms that we learn to recognize in bronchiolitis including brian Arria and fever. Um restaurant distress and of course wheezing or crackles, these are the typical players for bronchiolitis and uh respiratory syncytial virus. Rsv um uh comprises the majority of cases of classic pediatric bronchiolitis. But we see an abundance of other viruses that are implicated in this disease process. And I'll note here that while coronavirus is certainly associated with a classic phenotype of bronchiolitis in some cases um I think what we haven't observed during the pandemic is uh you know, a particular preponderance of covid 19 associated bronchiolitis, although we see lots of other manifestations of that disease in our hospital. And it's worth noting here also um that as we understand the molecular immunology of these infections more we know that there is an interesting association between certain viruses especially RSV and uh uh and immune dis regulation uh that ultimately leads to a more a topic state. Right with eosinophilic activation, basic bill activation and I. G. Production. And what's not clear is whether uh patients who demonstrate who are on this continuum have some underlying predisposition to develop those disease states or whether the virus is actually modify respiratory and immune function that lead to these disease states or some combination of those things. But I think this accounts for a lot of the significant heterogeneity in bronchiolitis and also the challenge for us as pediatricians and differentiate what what we think might be a simple viral infection versus um uh signs or the first episode of more chronic um uh disease state that is associated with recurrent wheezing. We know that bronchiolitis um is a big player in our lives as pediatricians and as restaurant therapists and nurses um and it has a pretty significant uh cost associated with it. There isn't a lot of recent data on this, but this was a fairly large sample. Um hospital discharges that was published in 2006. And at the time and recent estimates are not Very different from this. It's estimated that about 150,000 patients in a single year were hospitalized with Bronchiolitis in our country. And the frequency of hospitalizations was higher among certain categories, such as Children, less than a year of age. Uh It is the leading cause of hospitalization in this age group. Um Male gender. And of particular note in non white patients. We we do recognize that there is a disproportionate impact of of this disease state in our underrepresented communities, especially here in the East Bay. The mean length of stay for these hospitalizations was typically 3.3 days. And that will be important when we talk a little bit about the data in our project and the cost is substantial, up to around $4,000 per hospitalization. Those numbers are certainly higher now in the current economic state and they can increase dramatically with a co diagnosis of asthma or pneumonia, especially when I. V. Antibiotics are required frequently. More than double that amount. So bread and butter pediatrics. Um We've all seen patients like this and they tend to get our attention pretty quickly. Um You can see many of the classic symptoms of uh you know an ill appearing child with significantly increased work of breathing and sub costal and super sternal retractions. And you could hear the sound, you know, you would hear um wheezing and cough and and and maybe some crackles on consultation. So how do we treat this? Um As you all know, mostly we are um we have supportive measures um and I'm not talking at all about the prophylactic regimens that exist for trying to reduce the burden of RsV and other viral infections in our at risk patients. Um But once you have the disease, you know, generally the mainstay of our approach is to improve option delivery. Um We can support gas exchange in a variety of ways. Um There are a lot of ways that we can um improve the comfort of the patient. We can reduce metabolic demands, such as with treating fever, um uh positioning and suctioning Children, um airway clearance, um have as a role and of course very common indication for hospitalization is the need to provide additional hydration. Either ivy or central hydration and nutritional support. And then ultimately for our most severe patients, we do end up needing to support their gas exchange mechanically in some form. And of course we should remember to treat co morbid conditions. So, another question uh evidence based therapies for the initial management of previously healthy Children admitted with uncomplicated bronchiolitis include low flow auction delivery, high flow auction delivery, nebulizer, albuterol steroids and nebulizer, hyper tonic sailing. If you have in your mind, what you would initially do for patients. The only real evidence based therapy here is to use low flow auction delivery. We're fortunate in that we do have um updated consensus guidelines from the ap for the diagnosis, management and prevention of Bronchiolitis. They were initially put out in 2006 and they were updated in 2016. And I'll highlight from a from a summary of those of those recommendations in New England journal that again, the only two real recommendations are for supplemental oxygen Uh for some degree of hypoglycemia. And generally we we have um landed on a definition of hypoglycemia of 90% auction saturation on room air and uh nutrition and hydration. Um and these guidelines were not without some controversy, I think many people accused the ap of of an overly nihilistic approach to management. Um and uh and I think also there were a lot of frustration with the fact that as for front line providers for example, in the emergency room or in the clinic that it may not be, you know, you're seeing a phenotype of patients and it may not be evident what the underlying disease processes. And so the recommendation against even a trial bronchodilators, I think some found irksome but we also have some evidence that that following the guidelines does pay some dividends in terms of reducing cost uh and not affecting the outcome of patients. So this is some data from from an interesting paper looking at the trend over time of non recommended test. And I just highlighted the inpatient group data. There's also E. D. Group data um from this paper but you can see that after the implementation and then revision of the guidelines in 2016 that we did increase the rate of decline in the use of these non supportive therapies without any negative impact on outcomes in patients admitted for bronchiolitis. So I think there is some reasonable rationale here in being restrained in using some of these. Um Not not well evidence based therapies. I'm an intensive so I can't get away with talking a little bit about the meaning of life according to me you know much of what we do in pediatrics and in medicine is balancing auction delivery and auction consumption. Um And it's important to say out loud the words respiratory failure because we see it a lot in our hospital. Um And and it's not always the dramatic presentation that we sometimes think about in the I. C. U. With the patient is intubated and on the oscillator we see restaurant failure every day in our clinics and in the emergency room it occurs when the restaurant system is in a state of dysfunction and inadequate gas exchange occurs. So you know a patient with bronchiolitis who's working hard to breathe and has some degree of hypoglycemia is in respiratory failure. A question which of the following may be indicators of acute restaurant failure in a child presenting with bronchiolitis, Room air saturation of 90 labored breathing hypercar bia measured by blood gas analysis uh and organ dysfunctions such as altered mental status or poor perfusion or all of the above for for for pediatrics. This these are hopefully gimmes. And I think we all recognize that we can see all of these things in patients who present with restaurant failure and our approach to restaurant failure in our hospitals. Uh you know, becoming um a little more consistent and easier to predict. And this is kind of an example of an acute restaurant failure algorithm that we use when we're managing transports and are kind of evaluating our own process in the hospital and in the ICU. And you can see the mainstays of this include auction supplementation and after initial efforts at auction supplementation that heather is going to talk about in much more detail. Um You can see that we move on to typically heated high flow nasal cannula. Um and then onto more um uh more sophisticated forms of of support which include positive pressure ventilation with non invasive ventilation and then of course intubation and other modalities and at the end of the road, although rare, we do sometimes have patients with viral bronchiolitis who end up on VV ECMO a brief word about about the history of high flow, which heather will expand on. Um the idea of using high flow rates of compressed air or oxygen is not a novel strategy. But what what can happen is that with high flow rates you may get significant desiccation of the mucosa that leads to impairment of new distillery function, thickening of secretions, increased inflammation and this can be a real burden in terms of caloric expenditure and heat loss for patients and increases their insensible losses. So once we developed reliable um techniques for humid ification and temperature regulation, that's when we really saw dissemination of heated high flow nasal cannula therapy in both adults and Children and their although there have been some technologic issues, there has been some evidence of benefit for heated high flow including improved comfort, decreased work of breathing and improved oxygenation. What we have lacked um is really high quality evidence on impact on outcomes for this therapy. In Oakland. We were fairly early adopters for bronchiolitis uh for using heated high flow for bronchiolitis. We developed our own um bronchial hi flo rigs um um uh to use that therapy and in primarily the I. C. U. But we have expanded the use of high flow to alternate indications. So patients for example with hypoglycemia due to other causes um like pneumonia. Uh And we have also expanded the use of high flown out from out of the ICU into the wards into the emergency room. And our hope has always been in the past to reduce the necessity for transferring patients to the ICU. And to reduce the need for escalation and intubation. We do now have increased flow limits and a weight based algorithm which heather will go over. And one of our former residents and chief residents, Margaret Nguyen did a really um um lovely um um retrospective review um of this um expansion up heated high flow around 2019. And what she found is that we largely succeeded in what we were trying to accomplish the rate of transfer to the I. C. U. Um either from the emergency room or from the words after high flow was initiated did reduce over time. And so we seem to have been where we wanted to be in the utilization of this therapy. Um But I think for some time it's also been apparent. Uh And we've been concerned about the use of high flow in a variety of indications without a lot of evidence base. And there has certainly been um a lack of clear demonstration of benefit for patients with bronchiolitis which is the original indication for this therapy in our hospital. Um Our utilization of high flow and the length of state for bronchiolitis were at or above national averages. And so then potentially also costs we had somewhat inconsistent practice high flow was was used and tie traded in slightly different ways and different parts of the hospital. And while we did have an R. T. Driven weaning protocol that may have also extended the duration of therapy and many of us have wondered whether in the use of high flow, whether in some instances there might be potentially a delay in escalation of care for patients and delaying intubation that might be necessary and therefore prolonging the duration of hospitalization. And we've always been strapped for resources in terms of personnel and time to actually do comprehensive evaluation and quality improvement work in this domain. So I will hand it over to heather, who will talk about the actual therapy. Good morning. I want to start my presentation by sharing an interesting fact. Horses, similarly to neo needs are obligate nose breathers. Hypo was initially started by delivering oxygen therapy to racehorses. Racehorses have massive, massive lungs and they can pull in 100 and 60 liters per minute and so they're prone to develop pulmonary hemorrhage on the racetrack. So, heated high flow was delivered to these horses to precondition their cardiopulmonary system before a race. And the race track is actually pretty dusty and some of these horses were developing lung infections. So we started infusing antibiotics within the heated high flow systems for the horses. And this began in the early 1980s and the late 1980s, we began to use heated high flow with human beings And over the last few decades we've really enhanced our usage of heated high flow and our system are heated high flow systems have enhanced and the way that we deliver heated high flow with it. Different type of interfaces have enhanced as well, heated high flow is a really simple system. And if you see the picture I have here on the left hand side, it really starts with just a simple blender that has two km attached to it. So we can now blend gas, oxygen and air to provide therapy for our patients. The oxygen air comes out of the flow meter and there's that little green oxygen tubing that will connect to a pressure relief valve. The gas flows through that pressure release valve into a heater pot chamber which is heated to relative body temperature at 37 degrees Celsius. That heated humidified enriched oxygen gas will then go through the patient circuit the blue patient circuit into a heated high flow nasal cannula. And if you note that there is a water bag attached to the pole on the upper part of the pole and that's how we provide our heated humidified cation through the heater chamber, I have a picture of the different sizes of the optic flow nasal cannula. I think this is really important. We have an array of sizes and each size is associated with a weight and a leader flow. So our smallest size is blue and that's extra small and the max leader flow is eight liters per minute and our largest size is the very bottom and those are the optic flow pluses and those are meant for adult patients and the highest leader flow that they accommodate is about 50 liters per minute. So we can provide really high leader flow rates initially when high flow was first used in the human being in the 19 eighties late 19 eighties. The maximum leader flow is 20 liters per minute. So we now have interfaces that will accommodate more leader flow, which is great. Also, another important aspect of these nasal cannula says is the way that they fit a fit really matters. Um it should include about 50% of the nares, no more. So gas should be able to escape past the nasal passage. And so you need about 50% occlusion for these nasal cannula as next slide acute respiratory failure algorithm. I think it's important to discuss this, especially in regards to the escalation tree where high flow really lies within the escalation tree. So, I know man deep touched on this a little bit, but we typically for respiratory failure, we will start a page on simple oxygen devices and these oxygen devices are non fixed. So what that means is that a nasal cannula. A simple mask and a non rebreather are not close systems that provide a fixed precise oxygen percentage for our patient and a high flow. I'm sorry, a nasal cannula. The patient can breathe in room air and dilute the amount of oxygen that we're trying to provide. The same with the simple mask, as many of you have seen with the ports on the side of the mask. Non re breathers do have valves but they do not sit perfectly on a patient's face. It doesn't sit anatomically correct. So when the patient turns and whatnots, there are cracks within the system that allow the patient to n train room air, therefore diluting the amount of oxygen we're trying to provide. So these are non fixed 02 devices heated high flow on the other hand, is more of a fixed oh two device when providing heated high flow for neonatal because they are obligate nose breathers. You can have lower flow rates than five, but in pediatric patient population they do breathe through their mouth as well. So leader flows of five and higher will create an atomic reservoir within the patient's upper airway and will not allow the patient to entrained room air. So therefore it's a fixed 02 device After a heated high flow. The escalation tree really requires. And there's a lot of information here. I won't go into all of it, but it really requires a mechanical ventilation ventilator, whether it's noninvasive or invasive therapy. After heated high flow, we must escalate to a mechanical ventilator. Next slide. So there was a lot of evidence initially on how he did high flow help patients who were either in acute respiratory failure or in respiratory failure. And these um indications really focused on reducing the patient's respiratory rate by meeting the patient's minute ventilation demand. High flow provides high flow rates that's able to meet or exceed the patient's minute ventilation demand. Therefore reducing work of breathing. High flow, although not a ventilator can reduce C. 02 by the large high flow rates, washing out C. 02 in the dead space, particularly in the back of the old bearings and within the lung that is not participating in gas exchange. Those high flow rates will wash the C. 02 out, therefore reducing the C. 02 on blood gasses. High flow also creates a back pressure when the patient excels during exhalation phase which acts as a. C. Path and will increase increase fRC for the patient and by increasing F. R. C. We are increasing overall title volume and this helps with oxygenation, heated high flow also provides heated, humidified gas which helps keeps the mucus Ilary blanket moving forward. So it essentially helps with secretion removal. Whether the patient is able to cough the secretions out after the cilia has pushed the mucus blanket forward or we might need to suction the secretions out so it really helps with heating and humid if I ng um to provide better mucus Ilary clearance for our patients. High flow also provides a provide precise oxygen delivery because it will not allow I mentioned it before but it doesn't allow the patients to n train room air. So whatever you are delivering on that blender I showed you earlier that is what the patient will receive. There's no one training of room air so it will improve the patient's oxygenation. Next slide this video will compare low flow oxygen to heated high flow oxygen versus is how much disseminating I guess the dissemination of the oxygen within the lung for heated for low flow oxygen and this is high flow oxygen. This is how much is disseminated throughout the lung. Okay so this is a demonstration of how the high flow creates an atomic reservoir within the oral pharynx and how C. 02 is washed out of the patient's mouth again patients can and train roomier but can release through the patient's mouth next time this is our acute care pathway. There's three arms and I'm going to focus mostly on the left arm which is the winning arm. So if a patient is initiated in the heat of high flow, there's indications on the very top of this pathway but when a patient is initiated, unheated high flow, if we are able to provide less than 50% F. I. 02 and the patient saturation is greater or equal to 92%. We can begin the weaning arm of this pathway and the winning arm is really really slow. We wean leader flow by 1 to 2 liters every two hours is tolerated. And we could also win F. I. 02 by 20.1 or more every two hours as tolerated to maintain saturation of 92 or greater. So this is a slow weaning pathway. Um It has worked for us in the past but it it demonstrates how slowly high flow can you know be weaned from a patient. So we have patients who are on heated high flow for several days at a time. So high flow evidence the use of high flow and management of Bronchiolitis has increased dramatically in the last 10 years. Initial single centered observational studies suggested improved outcomes but over time these initial findings have not held true. So our committee has analyzed the last our, I'm sorry our committee analyzed randomized controlled trials that were written within the last five years regarding high flow use in Bronchiolitis patients. One of the studies had a sample size of 2900 patients who had mild to moderate bronchiolitis. These patients were randomized to either high flow or standard nasal cannula therapy. Nearly 70% of the patients who were randomized to low flow nasal cannula did not require an escalation in therapy. There was no difference within groups in the icu utilization intubation rates length of stay and the duration of the overall use of 0.2. This was likely in part responsible for the increased medical costs. There are wide varieties in the initiation and weaning practices of heated high flow and the inability to predict which patients might benefit. Rapid discontinuation of high flow has been shown to be saved, both within the I. C. U. And the wart settings, decrease duration of hypo treatment and decreased length of stay. This is there's a rare or no adverse events or needs of escalation. When a rapid discontinuation of high flow nasal therapy has been initiated next night. Have a question. Rapid discontinuation of heated high flow nasal cannula therapy for bronchiolitis may result in a successful weenie be intubation, see reduced length of stay. D. Mid stat or code blue or A. And C. So give you a minute to think about it. But I hope you chose A. And C. Successful leaning and reduce length of stay. Thank you very much. And I would like to introduce the next speaker dr cho thanks heather um in the next few minutes um I'll provide an overview of our Q. I. Project. Um So we partnered with the A. P. Value in inpatient pediatrics and participated in the high flow interventions to facilitate less overuse the high flow Q. I. Project is one of many with the V. I. P. Network which has a longstanding history focusing on improving in patient care with various projects Over the last 10 plus years. Many of Children's hospitals have participated in several of these and so some of these projects may look familiar to you. Our particular project stands just under one year with initial planning. Beginning in June of 2021. Just last year the intervention was implemented during the typical respiratory viral season from November to March of this year, the Ap provided several resources for the project, including education and marketing materials, small group coaching and database support. Found the global aim of the project was to reduce over utilization of high flow nasal chemotherapy and the treatment of infants with bronchiolitis. Now there were two separate arms to the interventions with each hospital who was enrolled, assigned to one or the other. So one intervention was the high flow nasal cannula initiation pause and the other was referred to as the high flow nasal cannula. Holiday, which was involved which involves the rapid weaning process. Our center was assigned to the holiday intervention. Next slide please, with the specific aim for the holiday being to reduce the use of high flow nasal cannula by 30% and that was reflected in the primary outcome with, with the total hours of treatment with high flow nasal cannula being reduced. Our process measure looked at the percent of patients who were treated with high flow nasal cannula that undergo a holiday attempt and our balancing measures were inpatient length of stay and then an unplanned escalation of care. Within one hour of holiday attacked. And we began our Q. I. Initiative by bringing together a stellar team um that other folks have already alluded to. Our team met on a regular basis throughout the project. Um It was a cross bay multidisciplinary team consisting of intensive ist from the ICU and Hospitalist in the queue care units and of course our hard working residents, respiratory therapists including leadership, was also very vital in implementation. As the intervention at our institution involves an RT driven pathway. We also had significant nursing support and buy in and our informatics colleagues also played an important role and this collaborative effort with individual expertise and enthusiasm for the project was really crucial and unique for the success of the Q. I. Initiative. So who qualifies for the intervention? The what we focus on patients ages 1 to 23 months who were hospitalized with a primary diagnosis of bronchiolitis patients who were premature or had significant cardiac disease or chronic lung disease requiring home oxygen were excluded. In addition, patients who required higher respiratory support above high flow nasal cannula with positive pressure or ineligible. In addition to patients who had high flow nasal cannula initiated an outside facility. We began our initial phase by looking at our baseline data with our high flow nasal cannula use in patients with bronchiolitis. Looking at data from the respiratory season of 2019 to 2020 which was pre covid, we all know after covid hit, nothing has followed the usual pattern. So this was thought to be more reflective of baseline data. And then prior to the official start of the implementation phase, we provided education for key players involved including M. D. S. R. N. And R. T. S, along with presenting the project at various meetings, we posted and distributed this tip sheet on all the units involved along with the RT driven clinical pathway which rides away to rapidly discontinue high flow nasal cannula. So the pathway we use looks like this. Um and I will review some of the key features. So once the patient was found to be eligible for holiday intervention providers would determine their holiday readiness based on oxygen saturation and heart rate. And then R. T. Would perform the high flow holiday. And that involves turning off the high flow device completely and removing the cannula from the base. They would monitor the patient for 5 to 15 minutes and then reassess within 30 to 60 minutes. The reassessment would then yield one of three outcomes based on the clinical picture and vital signs. So if the patient was doing great they would pass the room air if they were doing fine clinically but noted to be hypoxia mimic, then low flow cannula would be started and then if there were any signs of clinical decompensation with increased work of breathing, increased heart rate or respiratory rate, then they would be placed back on their previous high flow nasal cannula settings and then reassessed once again for another holiday attempts. And the goal was to perform these holidays two times a day And we did see patients who did not pass on the day shift with their first attempt perhaps but were able to be weaned later on in less than 2012 hours on the night shift to help us with implementing this Q. I. Initiative, we utilize various EMR tools. The high flow nasal cannula order itself was adjusted to include eligibility for high flow holiday intervention. R. T. Was also trained and educated on using a new smart phrase to document the high flow attempt and outcome. And then once we started our intervention phase, we performed real time chart review looking at our progress over two week intervals to ensure that the holiday attempts were occurring for eligible patients and tracking how we were doing and reducing our high flow nasal cannula overuse. This was done with pds A cycles and addressing needs as they were noted by reviewing our data during the implementation phase, I will now pass it on to dane to present the results from our hype Ohio Today intervention. Good morning everyone. Um I have the pleasure of talking about the results of this project, which I think we're all really pleased about. So this the numbers total sites enrolled initially, there were 86 sites across the country that enrolled in this. 50 of them were in the same arm that we were in the holiday arm where again like Sarah just talked about. We basically took patients who are on high flow nasal cannula. Had uncomplicated bronchiolitis and we just turned it off and we watched to see how they did. Um some of those sites dropped off and in the end there were 60 that provided all the information throughout the intervention process. So probably slightly less than 50 in our arm. But the number of baseline patients, this was pre covid from 2019 into 2020 was over 9500 patients. And then the intervention portion of this, which was 16 weeks. Those cycles included 4,217 patients that decrease like Sarah alluded to was a direct result of having decreased respiratory viral illness and Children during the pandemic. Alright, here we go. This is hours of high flow use. So the duration that patients were on high flow in Oakland. Um and if you look at the left side of this figure, this graph, the blue line represents Oakland. We were site number 87 the oranges red line represents all of the sites combined. Um It's kind of notable initially when you look at this that Oakland's baseline data. This is before the intervention pre covid. We obviously had patients on high flow much longer, significantly longer than the average across the country. I think there's multiple reasons for that. Um One we were early adopters to this and we expanded our use of high flow to multiple areas of the hospital, particularly the emergency department and the wards. Now there's a lot of sites across the country that only use high flow in their intensive care units or like in Mission Bay, they use them in their transitional care unit, the teacup, um in those units, there's much higher staffing levels and much more confidence to be aggressive with weaning processes so that the intensive ists are able to get patients off of high flow much more quickly than on the wards where we really stick to our protocol. And in Oakland we had a very excellent protocol, but it was a really slow wean that was Archie driven and sometimes the R. T driven protocols can be really efficient and effective but when they have a lot of patients to see and it's the middle of the night and you have a patient that is on high flow that seemed to be more sick than many of the patients on the floor. If they're stable on current settings, it was frequently seen that it's just easier to leave them on those settings throughout the night and deal with it in the morning. So I think those are all factors that played into our um longer use of high flow. But you look at the right side of this figure and you see our fantastic results. So by implementing the high flow holiday protocol, we were able to decrease the hours, the duration of high flow use by 48%. So the average patient had 78.6 hours prior to the implementation and 40.9 hours afterwards, Our goal was 30% reduction. We hit nearly a 50% reduction. And you see that our duration of high flow use after this is really equivalent to the rest of the country and that's after the rest of those sites also saw what ends up being about a 30% improvement in their duration. This is San Francisco's data. So we work closely the model, especially with the position they're in there. Um I see you and transitional unit that worked closely with us and they also saw a reduction of 27%. Their data was, it was limited because they have a lower volume of patients that are eligible. Um they see fewer bronchial itics but more especially just uncomplicated bronchial irritants are rare. The length of stay is almost as exciting as the duration of high flow use to me. Um if you see again we saw our baseline data 16 weeks 2019-2020. Um our average length of stay was 83 hours for a patient with uncomplicated Bronchi Elias. Again, that's higher than the average of all the rest of the sites combined and I think the majority of that has to do it again with our use of high flow nasal cannula. There's probably some cultural things amongst the practice and the conservative nature of us um Weaning patients and discharging patients that we continue to work on. Um Also there's some dis disposition hurdles that our patient population and our hospital deal with that we also continue to work on But after the implementation of the high flow holiday, that protocol, we saw that across all commerce. So this is all bronchial itics is not just the patients that were placed on high flow but all uncomplicated bronchial itics. We saw a decrease in the average length of stay by 24 hours which as you can imagine the cost savings and resource savings of 24 hours is tremendous. I think Mandy had said there were 100 and 50,000 admissions for bronchiolitis each year. So that's a great number. This is San Francisco again, they also saw a reduction in their length of stay by 13 hours. Again they had a lower patient this briefly looking for opportunities to improve throughout this process. Uh this this graph represents of the patients that were eligible to have high flow holidays. How many of them did we actually assessed to have the high flow holidays and we did a good job were much higher than the average. Across that the project except itself. Um there was one opportunity I think in the middle of the implementation cycle we saw a dip and with the fantastic help of our respiratory therapy leadership, they went through some education and our respiratory texts were very receptive and helpful and such an important part of this process. They responded very quickly and nearly all of our patients that were eligible were assessed. So in conclusion um we saw this as a gigantic success. We were decreased the hours of high flow nasal cannula use and length of stay on both the Mission Bay and the Oakland campuses of BCH. And we reduced the hours of high flow and length of stay significantly for patients with uncommon complicated bronchiolitis. Um I think that it was, I I just want to mention again how important it was to have a collaborative effort. Um The the nursing leadership as well as our T leadership were extremely helpful. Um Our residents were fantastic. They were charged with identifying every patient that needed to be enrolled in this study and then our nursing staff again, this was during a time where staffing shortages which we continue to deal with but at that time they were a real problem for us and every patient that we have on the floor has a nursing ratio of 3 to 1. Now if they're not on high flow nasal cannula, they have a nursing ratio of four patients to one nurse. And we were having such a hard time with staffing 4 to 1 to 3 to 1 is a it's a significant difference. So there was motivation across the board from our tea and nursing leadership to make this work and really see if it could help help us. And as you can see our goals were uh they were met very easily. I'll pass it off to dr ciara A mute. Uh this is just a brief, just recapitulation of what Dane said, the national summary. Uh the overall project across that many centers did ultimately achieve their their aim of 30% reduction and the average was a 14 hour reduction. So we far exceeded that and there was almost no clinical deterioration. It was exciting that in Oakland we didn't have any clinical deterioration um for any patient and in San Francisco as well. Um and the same was true for the initiation pause um which I think is exciting and compelling for us as we consider our our um our next steps um challenges and lessons that were learned. Uh you know, d implementation is challenging. Change is hard And there's actually ample evidence that that change is even harder when you're trying to do less. I think there's some there's some studies that site that takes on average about 17 years for groups of positions to incorporate um uh best quality evidence into their practices and hopefully we're a little ahead of that curve here. Um We have a variety of units with very different staffing experiences dane mentioned. And so again it clearly was a function of teamwork and diligence that we were able to do this so successfully in so many different parts of the hospital. We had quite outdated or inadequate data tools and um we had tremendous help and engagement from mike lang and his entire team um in the medical information services and informatics but the long there's a long lead time for developing newer tools. So a lot of what was done in this project was brute force by sarah cho and her team of residents and hospitalists that we're processing the data in real time and I can't thank her and her team enough for um keeping us on track. We were one of the few centers that um had consistent timely um up up uploading of data into the Ap database. Um and so that was a really great distinction. Um and we came to recognize important cross bay differences with our colleagues at Mission Bay, very different workflows as heather and others have have cited, especially in the way the RTs work and and again, I think it's a really it's a wonderful opportunity for us in Oakland that we do have a certain degree of autonomy and um motivation on the part of the parties. Uh you know, they're really partners um clinical partners in the management of of a lot of these patients and our incredible resource um very different volumes of uncomplicated versus complicated bronchiolitis on both campuses as you might imagine. And I think it's it's worth reminding everyone again how different our community demographics are and how access to resources really can be impactful for our our patients especially are underrepresented minorities. Um considering that we did this project during the pandemic. Um and also during this, you know, ongoing kind of growing affiliation with um UCSF in Mission Bay was a real stress tests of sorts. And I think it just it just goes to show in Oakland um that we've always been very committed to promoting high value care. I think one of the things I learned when I first came to Oakland over a decade ago is that um that we people are really proud of doing more with less because we have to and um you know, d implementing overused interventions is important. Um and I think what our work with our cross Bay colleagues has taught us is that that we're always seeking opportunities to learn and align our practice and potentially improve outcomes for patients on both sides of the bay. We have a variety of next important steps um as in addition to reporting our results as we're doing now, we have cost analysis that we'd like to do to really elaborate how much we saved by utilizing this intervention. Um we are very interested in looking at again how our communities underrepresented communities are are disproportionately affected by this disease process and our approach to its management. Um we're working on sustainability efforts were awaiting great new data tools. We want to look at the patients who were ineligible, technically right. The complicated bronchiolitis I think is another intriguing area of potential improvement. And so we'll be looking at both the initiation arm and also continuing the high flow intervention, the holiday intervention in the post covid era. And we'll be continuing to work very closely with our Cross bay colleagues and I just want to highlight, we did present this data at the sixth annual UCSF Health Improvement symposium. A total of 196 posters represented from across the UCSF health system. It was the first, it was the return of the in person symposium after the pandemic for several years and I'm delighted to report that out of all those posters, we were um um highlighted as among the 32 highest rated improvement initiatives. And we also received a blue ribbon award for high value care 18 presentations received this distinction from the UCSF Center for Health Care Value. So I think it's a wonderful testament to this incredible team effort. You've heard us acknowledge many people and there's just too many to list and we've tried to put as many down as we could and I also want to give a special shout out um to the BCH Oakland residents who again there, we had a number of people who stepped up to assist with the data analysis and who wanted to be high flow heroes for us on the wards and uh we've said many times how much we appreciated the diligent work of party leadership and the RTs and we had great support also from informatics and the BCH quality and safety team. Uh and uh that's our presentation mandy, I think we can look at a few of these questions that are up um Dean, I can go through the questions. So we have some great questions when um from Dr Spin Nazi, who is one of our champions for patients with Down syndrome and other um developmental disabilities. So she is wondering how that initial data about high flow nasal cannula applies and um strategize to patients with neuromuscular differences. Um Specifically with people with Down syndrome who have mid face hyperplasia. El valor simplification um airway and Malaysian hipaa Tonia with likely upper airway obstruction at baseline. Um I'll respond to that. The you know, one of the important aspects of this quality improvement project is that we we were focused on uncomplicated bronchiolitis. And so I'm gonna I'm going to use some of the evidence that we have reviewed to to answer dr smith Nazis question. Uh You know, there is a lot of mixed data on the utility heated high flow in in in recruiting lung and maintaining F. R. C. I think this is a really critical problem in patients with neuromuscular disease and hipaa Tonia and airway obstruction, all of which are present in our patients with Down syndrome. Um And there are some unique challenges. You know, I think most of us recognize that it doesn't really matter how much high flow you use if the mouth is gaping open. Um then you know, your opportunity for recruitment is probably pretty limited but there is nonetheless some evidence that even in patients who have your muscular disease, that high flow may improve al Viola recruitment. I think that's uh it's a simple uh It's a it's a simple uh need for the clinician to do that critical assessment and use their examination and radiographic evidence and otherwise to determine whether that's adequate. But many of those patients, I think just from from the function of their upper airway obstruction frequently need more um overt positive pressure support like uh CPAP or or N. I. B. For example. So I I don't think anyone would suggest that high flow is the answer to these more challenging groups of patients, but it probably plays some role. Thank you. Um We have a couple of questions about like other evidence based interventions for bronchiolitis. Um So one is um is there any still any role for albuterol hyper tonic ceiling or c make epinephrine nubs um in these patients? And if so are we um is that part of our management? Well, I think I love the hospitals to answer that because I think they're seeing a lot more of these patients. Yeah, I can just say a few things. There are definitely when it comes to strictly bronchiolitis. No. There's not a lot of evidence that albuterol hypersonic salient with any of those other Nibs or systemic glucocorticoids help at all. Um But not every patient is the same and we see patients very frequently that come through um that do seem to have some component of reactive airways in the setting of viral bronchiolitis and a lot of patients that come through the emergency department, they're put in a difficult situation, are they wheezing or they not? Do they have a family history of hp or not? And they do get steroids, systemic steroids and trials of bronchodilators and show some response. So we get caught in this middle place or do we treat them as if they have reactive airway disease? Or do we not? Because the majority of the evidence points towards it being not significantly helpful. Um and we have that conversation every single day with ourselves and with our residents for the most part, No, there's not a role for those in either complicated or uncomplicated bronchiolitis, but if patients have some amount of underlying reactive airway disease or they show us that they are very responsive to bronchodilators, for example, then we will continue them on some systemic steroids. And um is that the most appropriate thing all the time? No, not all the time, but sometimes it seems to be seems to be helpful for them. And as we know that um and I think it was Dr Niyazi someone alluded to in one of their earlier questions and Amandeep spoke about the development of asthma reactive airway disease after having bronchiolitis as an infant. Um you know that there's an association there, we don't know enough about it at this point. And no Dr Niyazi, I do not know that if we know the difference between getting rsv bronchiolitis at age three months or at age 16 months makes a difference in the developed future development of asthma. I don't know if anyone else on this panel does, but I'm not sure about that. Um the that there are there are plenty of patients that are admitted with respiratory failure and broccoli is who do have some reactive airway disease and will respond. I'll just I'll just quickly add 11 point, which is a common motif in the feedback we got from this project, because in the I. C. U, you know, we tend to do things a little more comfortably and we might wean patients pretty rapidly. And one of the things we heard very consistently from our hospitalist colleagues and RTs and residents on the ward is that this project and the support they received really made them feel comfortable and kind of, they felt like they gave them permission to move patients along more swiftly than they would. And I think the same thing applies for all those other therapies. I think we can use a Q. I methodology to help us um reduce uh non non evidence based interventions. And I think so, I think having permission to not do things that we don't think have a good evidence base that's almost as much of a challenge as incorporating other best practices. So I would just encourage everyone to to think about how we can use Q. I. Methodology to to try to make those kinds of change in our practice great. Um I think just to be respectful of everyone's time, that's a great point to end on. Um thank you all so much for sharing these these results with us in this great project. Um and I hope everyone has a wonderful day. Created by
