2016–2017 Influenza Season Activity and Recommendations for Clinicians
Moderator: William Koehne
Presenters: Angela Campbell, MD, MPH; Alicia P. Budd, MPH
Date/Time: February 16, 2017, 2:00 – 3:00 pm ET
Coordinator: Welcome and thank you for standing by.
At this time all participants are in a listen only mode until the question and answer session of the call. If you’d like to ask a question during that time please press Star followed by number 1.
Today’s conference is being recorded. Any objections you may disconnect at this time.
Now I’d like to turn over the meeting to William Koehne. You may begin.
William Koehne: Thank you Angela. Good afternoon. I’m William Koehne and I’m representing the Clinician Outreach and Communication Activity, COCA, with the Emergency Risk Communication Branch at the Centers for Disease Control and Prevention.
I’m happy to welcome you to today’s COCA Call “2016-2017 Influenza Season Activity and Recommendations for Clinicians.”
You may participate in today’s presentation by audio only via webinar or you may download the slides if you’re unable to access the webinar. The PowerPoint slide set and the webinar link can be found on our COCA webpage at emergency.cdc.gov/coca.
Free continuing education is offered for this COCA call. Instructions on how to earn continuing education will be provided at the end of the call.
CDC, our planners, presenters and their spouses/partners wish to disclose that they have no financial interests or other relationships with the manufacturers of commercial products, suppliers of commercial services or commercial supporters. Planners have reviewed the content to ensure there is no bias.
Dr. Angela Campbell will be – will discuss neuraminidase inhibitor medications (antivirals) that are approved only for the treatment of uncomplicated influenza. She’ll discuss the off-label uses such as for hospital patients and ages not FDA-approved. She may also discuss investigational use of IV zanamivir in hospitalized patients.
At the end of the presentation you’ll have the opportunity to ask the presenters questions. On the phone dialing Star, 1 will put you into the queue for questions. You may submit questions through the webinar system at any time during the presentation by selecting the Q&A pod.
Questions are limited to clinicians who would like information related to influenza. And for those who have media questions please contact CDC Media Relations at 404-639-3286 or send an email to firstname.lastname@example.org. If you are a patient please refer your questions to your healthcare provider.
At the conclusion of today’s session participants will be able to describe the current status of influenza activity in the United States, explain the circulating influenza strains seen this season and implications for clinicians, discuss antiviral treatment of influenza and implications for patient evaluation, testing and treatment.
Today’s first presenter today will be Ms. Alicia Budd. Ms. Budd is an epidemiologist in the Epidemiology and Prevention Branch in the Influenza Division of CDC’s National Center for Immunization and Respiratory Diseases.
Ms. Budd received her Master of Public Health Degree at Emory University’s Rollins School for Public Health, her Bachelors of Science Degree from the College of William & Mary. She joined CDC in 1997 and worked in the National Vaccine Office on Adult Immunization and Pandemic Influenza Planning. Ms. Budd first joined the Influenza Division in 1999 and managed the day-to-day functioning of the National Influenza Surveillance System and continued her pandemic influenza planning efforts.
In 2007 Ms. Budd left CDC and spent six years with the John Hopkins Hospital in the Hospital Epidemiology and Infection Control Department and two years at the Centers for Medicaid and Medicare Services working on healthcare quality initiatives.
She rejoined the Influenza Division in 2016 and is again focusing on domestic influenza surveillance.
Our second speaker today is Dr. Angela Campbell. Dr. Campbell is a Medical Officer in the Epidemiology and Prevention Branch in the Influenza Division in CDC’s National Center for Immunization and Respiratory Diseases. Dr. Campbell completed her medical degree at the Vanderbilt University. She trained in pediatrics and pediatric infectious diseases including an MPH at the University of Washington.
Dr. Campbell first worked at CDC as an EIS Officer followed by several years in academic medicine in Seattle before rejoining CDC in 2013. Dr. Campbell’s current projects focus on studies of influenza antiviral treatment and antiviral effectiveness, vaccine effectiveness, pandemic preparedness and development of CDC clinical guidance related to treatment and prevention of seasonal and novel influenza viruses.
She is an Adjunct Associate Professor of Pediatrics at Emory University School of Medicine and has a professional staff appointment at the Children’s Healthcare of Atlanta.
At this time please welcome Ms. Alicia Budd.
Alicia Budd: Thanks Will and good afternoon everyone and thank you for joining the call this afternoon.
As you just heard there are going to be two of us presenting today so I’m going to start the call by providing an update of the 2016-2017 influenza season activity to date.
Then I’m going to turn it over to my colleague, Dr. Campbell, who’s going to talk about the interim estimates of influenza vaccine effectiveness that were just released earlier today. She’s also going to discuss influenza diagnostic testing and review influenza antiviral recommendations.
I’d like to start with a brief overview of the surveillance system itself. The U.S. Influenza Surveillance is a collaborative effort between CDC and many partners including state, local, and territorial health departments, public health and clinical laboratories, vital statistics offices, healthcare providers, clinics, emergency departments, and others.
We collect five categories of information which you can see here in the light green circles at the top of the slide from eight different data sources. And most of those data sources operate year round.
The data are reported to CDC at least weekly and each week here at CDC we analyze the data and summarize it and feed the information back to our public health partners and offer it to the general public.
CDC makes these data publicly available in both a static report that we call FluView and an interactive online application called FluView Interactive. Both of these are updated each Friday with information about influenza activity that occurred during the previous week.
And you can see here on this slide the website where you can access both of those tools.
Today I’ll be presenting information from the FluView Report that was published last Friday and covers activity that occurred through the week ending February 4.
This slide shows influenza positive tests reported to CDC by approximately 250 clinical laboratories around the country. Clinical labs-the slide just changed. One moment, sorry, let me get back, there we go.
Clinical labs primarily test respiratory specimens for diagnostic purposes. And we use these data to determine both the timing and the intensity of influenza activity.
This season activity began to increase in mid-December and has continued to increase through the week ending February 4th. And during that most recent week 21% of the tests for influenza were actually positive for the virus.
Influenza A viruses have been reported more frequently than influenza B viruses throughout the season.
This next slide shows data reported to CDC by approximately 85 U.S. public health laboratories. These laboratories primarily test specimens for surveillance purposes so that we can understand what proportion of the circulating viruses belong to each influenza A subtype, either H1 or H3, and influenza B lineage, either Victoria or Yamagata.
This season influenza A/H3N2 is the predominant virus that we’ve seen. Influenza A/H1 and both lineages of influenza B viruses have also been reported but at very low numbers.
The lower bar here on the right hand side of this screen doesn’t represent a true decrease in flu activity. It’s there because there was a lag in data that we see. This graph represents data according to the week of specimen collection so if the specimen is not collected, tested and reported within the same week we have a bit of a lag. And that usually gets backfilled in the following week.
Public health labs in turn with their own testing they will send a subset of their influenza positives to the CDC Lab for additional testing. Antigenic characterization is one type of testing that CDC performs on these viruses.
And this testing gives us some information about how the vaccine might protect against circulating viruses based on how similar the circulating viruses are to the viruses in the vaccine.
So this slide shows the results of the antigenic characterization of influenza viruses collected in the U.S. since October 1. The yellow bars show the number of viruses in each influenza subtype or lineage that were found to be antigenically similar to the reference viruses representing the vaccine component.
All of the H1 and B/Yamagata viruses and almost all of the H3 and B/Victoria viruses are similar to their respective vaccine reference virus. This means the vaccine is well matched to the currently circulating influenza viruses that have been tested at CDC.
CDC also performs genetic sequencing on the viruses that are submitted by the public health labs. All of the A/H1, B/Victoria and B/Yamagata viruses genetically characterized belong to a single genetic group within their respective subtype or lineage.
There’s a small amount of genetic variability among the H3 viruses but the vast majority, 96%, belong to what we call the 3C2A genetic group.
Another component of influenza virus surveillance is surveillance for human infection with novel influenza A virus. A novel influenza A virus is one that’s different from the currently circulating human seasonal viruses. So we expect the population to have little or no preexisting immunity to these viruses.
It’s important that these viruses be identified and characterized quickly so any necessary public health measures can be taken. So far this season we’ve had two novel influenza A virus infections reported in the U.S. The first was an influenza A/H1N2 variant virus infection in a person in Iowa. A variant virus is one that normally circulates in swine but is detected in a person.
In this case the person reported close contact with swine in the week before they became ill and had a relatively mild illness. As is the case in all novel influenza A virus infections an investigation was conducted and this investigation found no evidence of human-to-human transmission with this virus.
The second novel influenza A infection identified in the U.S. this season occurred in New York City and involved an influenza A/H7N2 virus. This person also had relatively mild illness and no human-to-human transmission was identified.
What’s interesting about this situation is that this is the first known human infection with an influenza virus likely acquired through exposure to a cat. New York City was having an H7N2 Outbreak among cats in a number of animal shelters and this infected person had prolonged unprotected exposure to some of these sick cats.
As part of the investigation that occurred, more than 350 people that had exposure to these cats were screened or tested for infection. And this was the only human infection identified.
Now I’m going to switch from virologic surveillance to morbidity and mortality surveillance. This slide shows the percent of visits to nearly 2000 outpatient providers that were for influenza-like illness or ILI. The current season is shown in red and the other lines represent selected previous seasons.
The percent ILI has been at or above the national baseline for the past eight weeks and was 4.8% during the week ending February 4. You can see that the timing and intensity of ILI activity varies from season to season and this season so far looks to be within range of what we might expect based on past seasons.
This map shows the level of ILI activity in each state for the most recent week, the week ending February 4th. Activity levels are based on a comparison of the current week’s percent ILI to the average percent ILI in that state during weeks with low or no influenza virus circulation.
For this past week, 23 states and New York City experienced high ILI activity, 10 states experienced moderate activity, 8 states and Puerto Rico experienced low activity, and 9 states experienced minimal activity.
We also track laboratory-confirmed influenza-associated hospitalizations using a multistate population-based surveillance system. This slides shows the cumulative hospitalization rate for this season in purple and selected previous seasons. The cumulative rate as of the week ending February 4 is 24.3 hospitalizations per 100,000 population. Similar to the ILI data, these data indicate that the timing and intensity of this season appears to be falling between what we’ve seen before, not the earliest or latest season and not the most intense or mildest.
But we’ll have to wait for the end of the season to know for sure where severity of activity this season will lie.
This slides shows the cumulative hospitalization rates for this season by age groups. As usual, older adults, those 65 and older, have the highest hospitalization rates. So far this season it’s 114 per 100,000. The rates for the other age groups range from 23.5 per 100,000 for the 50 to 64 year-olds to 4.8 per 100,000 for the 5 to 17 year-olds.
What’s a bit unusual, especially for an H3 season, is that the 50 to 64 year-old age group has a higher hospitalization rate than the 0 to 4 year-old age group.
This slide shows the hospitalization rates for the 50 to 64 year-olds and the 0 to 4 year-olds during this season and the previous five seasons. During this season, which is shown in the top left graph, the pink line representing the 50 to 64 year-olds is higher than the green line which represents the 0 to 4 year-old. In all the other graphs you can see that the rates are nearly the same between the two age groups, or the rate in the 0 to 4 year-olds is higher.
These data reflect what we’ve been hearing from many of the state influenza surveillance coordinators at the state health departments and what we’ve seen in some of the ILI data which is that older folks were getting hit hard by flu earlier in the season while young kids were not. And now it seems like the flu activity has really picked up in some of the younger and school age children.
We’ll have to wait till the end of the season to see if these cumulative rates for these two age groups converge or if they stay separate as they are now.
This slide shows the proportion of hospitalized cases with selected underlying conditions. And these data are based on a subset of the cases for which we currently have complete medical chart abstraction. Among the 744 hospitalized adults with complete chart abstraction, 95% had at least one reported underlying medical condition. The most commonly reported were cardiovascular disease, metabolic disorders and obesity.
Among the 52 hospitalized children with complete chart abstraction, 54% had at least one underlying medical condition. The most commonly reported were asthma, chronic lung disease and neurologic disorder.
While just over half of the hospitalized children had an underlying medical condition this of course means that nearly half of them did not. And this is a good reminder that previously healthy children can have influenza disease severe enough to require a hospitalization.
Now switching to mortality surveillance, we track pneumonia and influenza related mortality using data from CDC’s National Center for Health Statistics. This slide shows data for the past several seasons with the current season highlighted in the blue box.
The red line is the percent of all deaths occurring each week that have pneumonia or influenza listed anywhere on the death certificate. The black lines represent the seasonal baseline and the epidemic threshold for each week.
To allow appropriate time to receive an update of – to calculate a stable estimate, these data are lagged a few more weeks than the other influenza surveillance data. The most recent data we have is for the week ending January 21 and for that week 7.9% of deaths were due to pneumonia or influenza. And this is the third week that this system has been above its epidemic threshold this season.
We have a second mortality surveillance component and this one focuses on influenza associated deaths in persons less than 18 years of age with laboratory confirmed influenza. Twenty such deaths occurring this season have been reported. Nine were associated with an influenza A/H3 virus, 1 was associated with an influenza A/H1 virus, 5 were associated with influenza A viruses that have not been subtyped, 4 were associated with influenza B viruses and 1 was associated with an influenza virus for which type was not determined.
And the eighth data source that we use to track influenza activity is reports from state and territorial health departments about the geographic spread of influenza in their jurisdiction. During the week ending February 4, widespread activity was reported by 43 states and Puerto Rico. Regional activity was reported by 6 states and Guam. One state and Washington, D.C. reported local activity. And the U.S. Virgin Islands reported no influenza activity.
Now I’m going to briefly touch on what’s happening with influenza activity internationally. Data from the U.S. Influenza Surveillance System is sent into a worldwide influenza surveillance system that’s run by the World Health Organization. And they make these data publicly available on their web site.
On this slide are data for Calendar Years 2016 and 2017 for the Northern Hemisphere and the Southern Hemisphere. The blue box on the Northern Hemisphere graph indicates the current influenza season. Northern Hemisphere activity as a whole is similar to what we’re seeing in the U.S. Activity is currently at elevated levels and H3 viruses are being reported most frequently.
The Southern Hemisphere which has influenza seasonality opposite to ours is not unexpectedly seeing very low levels of activity right now.
In addition to seasonal influenza surveillance we also keep a close eye on novel influenza A infections that are occurring worldwide. Of particular interest right now is influenza A/H7N9 activity in China. Since 2013, China has had annual epidemics of human infection with this avian virus and we’re now in the fifth wave of this epidemic.
This graph reflects only officially reported cases with known onset date but recent information that’s not reflected here suggests that there might now be more cases in the fifth wave than we’ve seen in any of the earlier waves.
Most of these infections have occurred after exposure to poultry. And while there may be some limited person-to-person transmission going on within small clusters, there’s no evidence of sustained person-to-person transmission.
China has reported almost all the cases but there are a few cases that have been identified in other countries in travelers that are returning from China.
CDC is following the situation closely and working with domestic and international partners to characterize both the virus itself and also the epidemic to try to understand if anything has changed from what was seen in earlier waves.
And so far it appears that’s this fifth wave, while it may have more cases associated with, it is otherwise similar to previous waves.
We also monitor reports from other countries about sporadic cases of human infection with other novel influenza A viruses and reports of influenza viruses being found in avian species. Knowing what’s happening in avian species is important because there’s often an association between increased circulation of certain influenza A viruses in avian species and increased risk of human infection rather with those viruses.
As a frontline clinician, laboratorian and our public health partners you really are critical to the prompt identification of possible novel influenza A virus infection.
And we need you to remain vigilant for the possibility especially in people that have ILI or ARI, Acute Respiratory Illness, who’ve had recent contact with sick or dead poultry, wild aquatic birds, captive birds of prey that have had contact with wild/aquatic birds, or ill swine.
And also with the increased H7N9 activity in China, we want you to consider H7N9 virus infection as a possible etiology for severe respiratory illness if you have a traveler returning from China, especially if they had exposure to poultry.
If you have a patient that meets these criteria, please notify your public health department right away and they can provide you with appropriate guidance on specimen collection and anything else that needs to happen.
If you happen to be a clinical laboratorian and you have an influenza positive specimen that can’t be typed or subtyped using commercially available tests, you would want to contact your state public health lab as soon as possible and they can help you facilitate transport of the specimen for additional testing.
And in turn, public health labs know to immediately notify us at CDC and to send virus specimens if they can’t type or subtype an influenza virus using their standard methods.
There’s a lot more information about both Avian and swine or variant virus infections on the CDC websites that are listed there on your screen.
In terms of summarizing the influenza season activity to date, activity began to increase in mid-December and it’s still elevated at this point. We expect it to continue for several more weeks though we don’t know when the season will peak exactly.
H3N2 viruses are the predominant virus this season. And we know that H3 predominant seasons often have higher mortality and hospitalization rates among older adults and young kids than we see during H1 or B predominant seasons.
The majority of the viruses characterized to date are well matched by the current influenza vaccine. And lastly, just again another plug, another reminder to really ask you to remain alert to the possibility of novel influenza A virus infection in persons with ILI or ARI and certain recent travel history and/or exposure to sick animals.
And with that I’m going to turn it over to my colleague to talk about the next components of this call.
Angela Campbell: Good afternoon. Hi. This is Angie Campbell. And we are literally switching chairs and now switching topics.
First, I’m going to review with you a brief discussion of influenza vaccination and I’ll show you the interim estimates of this season’s vaccine effectiveness.
So since the 2010-11 season the CDC’s ACIP has recommended annual vaccination for all persons 6 months of age or older, and even before that seasonal vaccine was recommended for certain high risk groups over different years.
I’m not going to go into this in any detail. I just want to briefly mention the principal changes to the ’16-’17 ACIP Statement. (Not changes to that statement but changes that came out with that statement that was published in August of 2016.) And really the main, I think main change that I’ll comment on here was that live attenuated influenza vaccine was not recommended for this season because of concerns regarding the low effectiveness against the influenza A(H1N1)pdm09 virus, particularly during the previous ’13-’14 and ’15-’16 seasons.
So that’s probably the biggest change and the main thing I just wanted to highlight. For these other things I would refer you to the MMWR – it reviewed new and recent vaccine licensures, as well as a slight change to the egg allergy recommendation.
And also, this was discussed on a COCA call in August—in October, which also included a couple of additional vaccine licensures and so the link is provided for you to look at that as well.
So what I am going to share with you today is the release of our 2016-’17 interim estimates of seasonal flu vaccine effectiveness. And this is pretty hot off the presses. It’ll be published officially in tomorrow’s MMWR and it came out electronically at 1:00 pm Eastern today.
And so just to orient you, these interim estimates come from the U.S. Flu Vaccine Effectiveness Network which is a network of the five sites that you see here. These sites enroll patients age 6 months and older who seek outpatient medical care for an acute respiratory illness with cough during the flu season.
And this is a slide of the methods used by the network which have been published previously. Outpatients of all ages over 6 months and over with acute respiratory illness with cough for less than or equal to seven days duration are enrolled.
Enrollment begins at each site when the flu season starts using local surveillance data, so it varies by site. And these estimates are using data through February 4th from a total of 3144 children and adults enrolled.
The study design is a test-negative design that’s widely used in VE, vaccine effectiveness, studies in which the odds of being vaccinated among flu positive cases, by PCR, are compared with the negative controls.
And for the interim estimates this vaccination status is defined as receipt of at least one dose of any 2016-17 seasonal flu vaccine and that is according to a combination of medical records, registries, and/or self-report depending on each site.
For the interim estimates VE is calculated as 1 minus the adjusted odds ratio, times 100% and adjusted for all the factors that you see listed.
So this table gives you the interim estimate for the overall vaccine effectiveness against Flu A and B virus infection associated with medically- attended acute respiratory illness. I thought it might be fun to try to use the pointer function of this. The column you should really focus on is this adjusted estimate column.
And it shows that for the overall estimate it was 48%, with a 95% confidence interval of 37% to 57%. So flu vaccines are offering significant protection against flu; they are reducing a vaccinated person’s risk of getting sick and having to go to the doctor because of flu by about half.
Effectiveness against the predominant H3N2 viruses has been 43% with significant confidence intervals. And against B viruses has been 73%.
So these interim estimates indicate improved protection compared with the ’14-‘15 season when the predominating circulating virus was antigenically different from the vaccine. And they’re similar to previous VE estimates against H3N2 during the ’11-’12 and ’12-’13 seasons.
Just to put that in a little more context this is a really nice recent meta-analysis of vaccine effectiveness studies that use the test-negative design. And I’m focusing only on the tables that show the estimates against H3N2. The top part of this is showing the pooled VE against H3N2 by season, starting in 2010-‘11 and going to ’14-’15.
And you can see that before ’13-’14, these were pretty high, 32% to 46% for pooled VEs, whereas it was 10% or less in the subsequent seasons. And in this analysis the pooled VE was 33% against antigenically similar H3N2 viruses.
So it does go to show that even where there is a good match these studies suggest that flu vaccine does provide less protection against the influenza A H3N2 viruses and better protection against H1N1 viruses and B viruses. They also show estimates for those that are pooled.
I just wanted to show you the H3 estimates to say that what we’re seeing this season is in context with what’s been seen before. And we know that similar VE estimates in the past have shown that the vaccine prevented millions of flu illnesses and thousands of hospitalizations. So it’s still the best strategy to prevent flu.
And we’re going to get rid of the pointer.
All right, I did just want to sort of say how are we doing this year or how were we doing when they did the early season vaccination coverage. And this is as of early November, 2016.
And so the first set of bars for each group shows the early season estimates for the past four seasons, and it shows it for all persons, for children and for adults.
And then the second set of bars is the end of the season and you can see those are generally higher, particularly for kids because they require two doses for certain ages.
But what I want you to see is that only 37% of children age 6 months to 17 years have been vaccinated as of early November and 42% of adults. And you can actually go to the website listed at the bottom and look at the various breakdowns by age.
But the bottom line is that we likely have weeks to go in the current flu season. And all people, your patients who haven’t yet been vaccinated, really should be encouraged to do so because we’re likely to have several weeks of flu activity remaining.
I also wanted to provide a list of resources to use in talking with patients or parents of patients about the flu vaccine. I think these are always important to know about. They include information regarding common misconceptions, excuses, safety concerns, etc. We know from data that clinician recommendations and offer of vaccination are a tremendous motivator to patients. And so it really does make a difference when you recommend vaccination to your patients.
So now I’ll move on to discuss flu diagnosis. And I will say that an important reason for discussing diagnosis of flu is that we know with a vaccine effectiveness of 48%, some vaccinated people will become infected with flu. And clinicians need to maintain a high index of suspicion for influenza infection among persons with acute respiratory illness because influenza antivirals are what we have for our second-line of defense against flu. And early treatment can reduce severity and complications of influenza illness.
So unfortunately we know that the decision to initiate treatment is often delayed while waiting for lab results, and it’s often dependent on insensitive assays. And that really shouldn’t be the case, so that’s why I just want to take some time to talk about diagnostics.
So as this says and I’ve already mentioned, accurate and prompt diagnosis of flu is important for clinical decision making, both potentially to guide treatment and also to facilitate infection prevention measures and perhaps even to reduce inappropriate antibiotic use.
And there are a variety of diagnostic tests available to clinicians to detect influenza viruses in specimens. These differ in many ways. Some are point-of-care tests which are CLIA waived, and that means they’re convenient and can be performed outside of the lab. Although those have some disadvantages. And then there are others that are moderately complex and require a clinical laboratory.
So the two main types of diagnostic tests in clinical settings are antigen detection tests and molecular assays. And really all of these have pretty high specificity for flu. That means if they say it’s flu it likely really is influenza.
But I want to emphasize that the antigen detection tests, the rapid influenza diagnostic test that we call RIDTs, these have pretty low to moderate sensitivity depending on which assay when compared with viral culture or PCR as a gold standard. And they are particularly poorly sensitive in older adults and with H1N1 viruses. They are mostly immunoassays. There are 17 approved in the U.S. and about half of those are CLIA waived.
So they’re nice because they’re fast. But they really are relatively insensitive. And some don’t actually distinguish between influenza A or B.
The rapid molecular assays, they are also quite fast. And they tend to require a little more time but still not to the extent of more conventional methods. There are two of these that are CLIA waived.
So there is a lot more detail on our website. This is a table of influenza virus testing methods that also includes culture and rapid shell vial culture. But I think I’ve told you most of the highlights with the RIDTs and the rapid molecular assays.
So, there are a few caveats to keep in mind when interpreting the results of the lab test. Unfortunately it’s not just the sensitivity and specificity of the assay but can vary by a lot of other factors.
For immunocompetent patients it’s really best to collect these specimens as early as possible, certainly within three to four days from illness onset. However you can have prolonged viral replication in young infants, immunosuppressed patients, or severely ill patients. So in patients like that where it’s critical to get a diagnosis, then it could still be useful farther after illness onset.
With respect to the optimal specimens, typically if the nasopharyngeal area is sampled either with a nasopharyngeal swab or a nasal plus throat swab, that’s conventionally been the best – although there are some mid-turbinate swabs that are performing well.
And it’s just important to remember that no sample type is perfect because of all of the other caveats with the patient. In particular I want to say that you can have a very ill hospitalized patient with a negative upper respiratory tract sample who may actually be positive in their lower respiratory tract. And we’ve seen that a number of times.
And so if it’s critical to find out the cause of a patient’s respiratory failure, it’s best to get an endotracheal sample or a bronchoalveolar lavage sample to test the lower respiratory tract as well.
And then lastly, it’s important to interpret the results in the context of what’s going on in the community and consider that in particular an RIDT may give a false negative result which I’ll go into in a little more detail here.
So this is a figure provided by a colleague, Tim Uyeki, and I think it does a nice job of sort of showing that we know that predictive values of influenza tests are important, and this will vary according to the prevalence of flu that is circulating.
So if the influenza activity is low, as shown here in the beginning and end of this season in the tails of this curve here, then the positive predictive value of an RIDT or the proportion of patients with positive results who actually have influenza is lowest. In other words, false positives are more likely. And the negative predictive value of an RIDT is highest when influenza activity is low. And so in other words a negative result is most likely a true negative.
Conversely, when the influenza activity is high in the community the negative predictive value, or the proportion of patients with negative results who don’t have flu, the negative predictive value of an RIDT is lowest and false negative results are a lot more likely to occur, when this influenza prevalence is high in the community.
So I think that just has to be kept in mind. We are constantly hammering the idea that RIDTs are insensitive, and particularly so the negative predictive value is very low when there’s a lot of circulating flu.
So this is the updated webpage for our CDC flu testing guidance. And it has a number of different links to guidance and algorithms for when and when flu is and isn’t circulating in the community.
One example of this is this guidance to consider flu testing. And this is a little stretched on my screen. But I would encourage you to look at it on the website. It basically provides some questions to help decide whether testing might be useful. Is flu – does the patient have signs and symptoms suggestive of flu? Is the patient being admitted to the hospital? And will testing results influence clinical management?
It’s really meant to be a guide and not prescriptive. And it includes a number of footnotes with caveats that emphasize that you really don’t need to confirm flu virus infection by diagnostic testing to make decisions on prescribing antivirals. We would encourage those decisions to be made based on signs and symptoms and the patient and epidemiologic factors.
We really encourage that empiric treatment shouldn’t be delayed while waiting for results of testing because we know that treatment is most beneficial when it’s started as close to illness onset as possible. But still these algorithms are provided to sort of give some practical guidance.
And so I’ve already kind of talked about antiviral treatment. So we’re just going to move into this next section.
First, just to say that there are three FDA-approved antivirals, they’re all neuraminidase inhibitors. They all have a similar chemical structure. These three are recommended for use this season.
The first is oral oseltamivir. And you probably heard that it’s now available as a generic version, or also as Tamiflu. This is true for the capsules only. The suspension is not yet available generic. But the capsules were approved by the FDA in August and became available in December.
The other products available are inhaled zanamivir which is known by the trade name Relenza, and intravenous peramivir which is called Rapivab. And this was approved in 2014 for treatment of uncomplicated flu in adults. It’s a single IV infusion.
So I want to preface the section of our recommendations by saying that I’m really not going to present all of the evidence behind our guidance today. I did actually go into a lot of detail about that in a COCA call in January of 2015. And the COCA website has this wonderful archive of the old talks and slides. So if you’re interested there are a lot of data tables there.
I will just say that it has been shown in randomized clinical trials and observational studies that early antiviral treatment can shorten the duration of fever and illness symptoms and may reduce the risk of complications from flu. And so therefore anyone may benefit from treatment.
But I want to note that the focus of CDC’s treatment guidance is on prevention of severe outcomes. This is why we focus on treating people with severe disease and people at highest risk of complications. And for those indications there really aren’t any randomized clinical trials available.
And so our recommendations are based on observational studies and meta- analyses of antiviral effectiveness. And these studies have consistently shown that early antiviral treatment among hospitalized patients is associated with reduced mortality in adults and shortened length of stay in children and older adults, and also that the earlier the treatment is initiated in the patient’s illness the better the outcome.
These recs are common to CDC, ACIP, as well as other clinical societies including IDSA and AAP.
So that’s all I’m really going to sort of provide for a little bit of evidence and rationale as to why we recommend what we recommend.
And then I’ll just show you our guidance in sort of two nutshell slides. The first is that all patients in the following categories with suspected or confirmed flu should be treated as soon as possible without waiting for confirmatory influenza testing. That’s for hospitalized patients, patients with severe or progressive illness and patients that are at high risk for complications from flu, either outpatient or hospitalized. And I didn’t include a risk of high risk – I didn’t include a list of high risk patients but that is available on our website.
The second part of our recommendation is that treatment certainly can be considered for any previously healthy non-high risk symptomatic outpatient with suspected or confirmed flu on the basis of clinical judgment, in particular if treatment can be initiated within 48 hours of illness onset. This is what all the trials were based on.
And so how is the compliance with our recommendations, particularly for hospitalized patients?
This slide shows data using the multistate hospital surveillance system that Alicia referred to called FluSurv-NET. And this is showing the antiviral treatment of patients hospitalized with lab-confirmed flu over five seasons.
Overall we found that antiviral treatment increased from 72% in the 2010-11 Season to 89% in the ’14-’15 Season. And this slide shows treatment for each season by age groups. Treatment across all seasons with obviously higher for adults than children.
But we did see that among the different age groups treatment increased the most in infants less than 1 year, from 51% back in 2010-11 to 82% in ’14-’15.
And so even though this overall high compliance with recommendations was encouraging, when we look at when treatment was given, we found that only about half of patients get treated on the day of hospital admission. So this figure breaks up the patients into those who were admitted on the same day or the day after their illness began – that’s day zero or 1 – or day 2 to 4, or after 4 days of illness.
And across the board the proportion treated on the day of admission was 55%, 59% and 51% respectively. So it seems like this could be a lot higher. Either flu wasn’t suspected or treatment may have been delayed while waiting for lab confirmation. We don’t know.
So I’ve probably beat the timing enough but when indicated it really – treatment really should be started as soon as possible and shouldn’t be delayed for the results of testing.
Just a note on high risk outpatients and early treatment – during the flu season especially, providers should advise their high risk patients to call their provider promptly if they have symptoms of flu. It may actually be useful for providers to implement triage lines to enable high risk patients to discuss their symptoms and perhaps, when feasible, an antiviral prescription could be provided without testing and before a visit.
And so we have an algorithm online. Again I don’t expect you can really read this well on the slide. But the link is at the bottom.
And it’s meant to provide a flowchart to be used when flu viruses are circulating, and it may help medical office staff triage calls from patients with possible flu and identify when it might be appropriate to initiate antiviral treatment before an office visit.
I’ve kind of beat the early treatment to death but I still do want to say that antiviral treatment may actually be beneficial in some patients even when started after 48 hours.
In particular, we do recommend that for hospitalized patients. Observational studies suggest that treatment might still be beneficial when initiated four or five days after symptom onset. We’ve seen also data in pregnant women that has shown treatment can provide benefit when started three to four days after onset. And there was one randomized trial in children that showed that oseltamivir initiated at 72 hours after illness onset, among these kids that had fever and uncomplicated flu, still showed clinical benefit.
So definitely treat early if you can. But if it’s indicated and the person would be recommended to receive treatment, sometimes after 48 hours can still provide benefit.
This is another algorithm from the diagnostic website again. And it’s to assist in interpretation of flu testing results and clinical decision making when influenza viruses are circulating. I think that what’s most important about this is on the blue lane it has influenza positive results. But on the red lane it says what to do if you have an influenza negative result. And it provides this interpretation: you cannot rule out influenza virus infection, especially if the test doesn’t have high sensitivity to detect flu viruses, or if the specimen was collected more than four days after influenza onset. So at that time it recommends that you still start treatment as soon as possible. If the patient is in the high risk category, has progressive disease, or is being admitted, it also recommends considering additional testing.
So I would encourage you to look at these algorithms on the website.
So how would you treat for an outpatient? Any of the neuraminidase inhibitors are recommended. It would be a 5-day course of oseltamivir, and that can be used down to any age, any infant, or inhaled zanamivir which is for children age 7 or older. Or for some instances, perhaps when nausea or difficulty swallowing pills limits the other options, a single dose of IV peramivir can be given in persons age 18 and older.
We do recommend oral oseltamivir preferentially for pregnant women. And that’s really based on the fact that that’s the drug we have the most data for.
For hospitalized patients we recommend treating with oral or enterally-administered oseltamivir. There are studies that show if it’s given by orogastric or nasogastric tube it’s still well absorbed. We don’t recommend inhaled zanamivir just because of lack of data in patients with very severe disease. And also there are insufficient data on IV peramivir for hospitalized patients.
We also do recommend that for patients who remain severely ill in the hospital that longer treatment courses may need to be considered.
If there is concern about absorption of the oseltamivir, then that would be an indication to consider the use of IV peramivir or investigational IV zanamivir. There was a trial of IV peramivir performed in hospitalized patients. It just did not demonstrate clinical benefit when compared to the comparator arm which was oseltamivir and standard of care.
But the – that was the dose that was used. And again, likely longer dosing might be needed for patients who remain severely ill.
So this is for concerns of absorption. If there are concerns of an oseltamivir resistant virus, that is a time where we often recommend thinking about zanamivir because we know that some viruses can become resistant to both oseltamivir and peramivir, and they may remain susceptible to zanamivir. We really haven’t seen a lot of resistance this season. It’s much more common with the H1N1-predominant seasons.
And just a plug to say that antibiotics aren’t effective against influenza. I’m sure you all know that, but several reports do suggest that inappropriate use of antibiotics is given to patients with influenza. Now that said – bacterial infections can certainly be a complication of flu, and so should be considered and treated if that is suspected.
I have two slides on institutional outbreaks. The main point is to say that management of an institutional outbreak – where high risk people live, these are long-term care facilities, nursing homes or other living facilities with high risk people, not simply a school – but this type of an outbreak requires vaccination ahead of time, and then during the outbreak diagnostic testing and rigorous infection control, treatment of ill persons, and chemoprophylaxis of others.
And specifically for antiviral chemoprophylaxis, we recommend that that be given for all of the residents regardless of vaccination status, to unvaccinated healthcare personnel, and for a minimum of two weeks if the outbreak is shorter than that, or continuing at least seven days after the last known case is identified.
There’s not a lot to say about antiviral supply this year. Fortunately, there have been no current shortages and manufacturers state that they do have sufficient product on hand.
Sometimes pharmacies may experience temporary stock outs or stock shortages if there’s a local outbreak. And these are normally resolved in 24 hours.
There is a Call Center set up specifically for long-term care facilities experiencing difficulty accessing antiviral supplies if needed, for like a large outbreak setting. In that instance, CDC can coordinate with commercial supply chain partners to rapidly redirect supply of large orders of drugs if needed. And so it’s available every day, Monday through Friday, at that email address there.
So in summary – and what’s not on here is that we still recommend vaccination as the first defense against influenza. However, if someone does become ill, early empiric antiviral treatment is recommended for suspected or confirmed flu among hospitalized patients, patients with progressive illness, or patients at high risk for complications. Physicians shouldn’t wait for laboratory confirmation. And again, clinical benefit is certainly greatest when treatment is initiated early but treatment after 48 hours may still be beneficial for some patients.
This is a list of some of the main resources that were cited in the antiviral section of the talk.
This is to say that Alicia and I work with a large group of people who have provided data and support and help that went into these slides.
And we’re happy to take any questions that you might have. Thanks for your attention.
William Koehne: Thank you presenters for providing our COCA audience with a wealth of information. We’re now going to open up the lines for the question and answer session. Questions are limited to clinicians who’d like information related to influenza.
For those who have media questions please contact CDC Media Relations at 404-639-3286 or send an email to email@example.com. If you are a patient please refer your questions to your healthcare provider.
When asking questions please state your organization and also remember that you can submit questions through the webinar system as well.
Operator if you could open up the phone lines for questions and answers.
Coordinator: Thank you. We will now begin the question and answer session. If you’d like to ask a question please press Star 1, record your name clearly.
William Koehne: And I have a few questions through the webinar system that I’m going to start off with.
Our first question is are you able to comment about any resistance associated with the increased use of antivirals particularly since there appears to be some delay in treatment based on the data that you have shown us?
Angela Campbell: Thanks for that question. We really haven’t seen that. And in fact it’s interesting. So in countries like Japan where they use antivirals so much more than we do in the outpatient setting, there really hasn’t been – data haven’t shown that the viruses are increasing in antiviral resistance there either.
I think it’s a good question the way it was asked, because certainly the time when we worry about it the most is when the patient really does have influenza and they might be started on a dose that’s suboptimal. But for the most part it seems like that it hasn’t been a problem.
Now it’s certainly not a problem – one thing I feel like people get confused about is that if oseltamivir is given to someone who doesn’t have influenza – then it’s really no concern because it’s not the same as an antibiotic that could alter the, you know, microbiome and affect the bacteria that are already in you. The person has to have the virus to develop resistance.
But we have not seen that. And in fact really have seen very little resistance at all this season in particular. But not in the last few seasons either.
William Koehne: Thank you. Operator do we have any questions on the phone line?
Coordinator: One moment for the first question. Dr. Warner Hudson your line is open.
Warner Hudson: Oh, hi. This is Warner Hudson, Medical Director for Occupational Health for UCLA. We have a number of travelers and researchers that of course go to places like, you know, H6N9, is that the right one?
Anyway is the post-exposure prophylaxis regimen the same as for seasonal or is the dose twice a day instead?
Angela Campbell: Okay hi. So I think – are you – you’re thinking about H7N9, right?
Warner Hudson: H7N9, there’s been approaching 400 cases with this recent uptick.
Angela Campbell: Right.
Warner Hudson: And just interested. I get questions from the people going over there about should they take oseltamivir and what’s the post-exposure prophy dose. And if they start to get flu like symptoms should they take it longer, double it up, those kinds of questions.
Angela Campbell: Well so it’s really an excellent question. We actually have changed our website considerably in the last couple of years. So there’s an entire section devoted to novel influenza viruses.
And I can be sure that this information is put in the notes from this call.
There’s a section on treatment of patients with novel virus and a section on chemoprophylaxis. In general the chemoprophylaxis section focuses on people who are exposed to other persons with known novel A virus. And in that case you’re correct and have likely been following this situation that…
Warner Hudson: Right, right.
Angela Campbell: …we have actually in the last few years and along with WHO have recommended basically treatment dosing to be given as chemoprophylaxis.
Warner Hudson: Right.
Angela Campbell: With the idea that there can be very high levels of viral replication and the traditional chemoprophylaxis dose may be inadequate.
I wouldn’t – it’s not something we would routinely recommend people sort of take and start if they feel like they may have been exposed to a live bird market, for instance. It’s typically something that if this has come up we’ve had consultation with the health department or the person, you know, the providers involved to sort of decide if that might be necessary.
But that is why we developed that guidance. It’s not something I would recommend in general for travelers to those areas.
Warner Hudson: No. I’m talking about really researchers that are going out and in areas often collecting specimens and sending them back, things like that. We have some labs working on things too. So thank you.
Angela Campbell: Well that – it’s interesting. And I will – I’m happy to follow-up, you know, sort of specifically on your situation if that would help.
Warner Hudson: Yes. I’m at —@mednet.ucla.edu. And Tim Uyeki and I have talked over the years about H5N1 exposures and getting labs ready for research in those kind of areas.
Angela Campbell: Yes, I…
Warner Hudson: And all the vaccines that are in the pipeline.
Angela Campbell: I might just ask you to send your information to Will. And also for those, I think we, you know, may not get to all the questions I recognize. If you do have questions we don’t get to please send them to the COCA box and we can address those.
William Koehne: Yes and…
Warner Hudson: Thank you so much.
William Koehne: …thank you for that. Unfortunately we are out of time for more questions. And I do see that there’s a lot more questions on the webinar system as well as I’m sure that there’s some to the phone line.
So as was stated please if you have questions for the presenters email them to firstname.lastname@example.org. That’s C-O-C-A @ C-D-C dot G-O-V. And we’ll make sure that those questions get forwarded onto our presenters today and that you receive some answers for those questions.
And so on behalf of COCA I’d like to thank everyone for joining us today on this call. With special thank you to our presenters, Dr. Angie Campbell and Ms. Alicia Budd.
The recording of this call and the transcript will be available on the COCA web site at emergency.cdc.gov/coca within the next few days.
All continuing education for COCA calls are issued through the TCE Online System, the CDC Training and Continuing Education System at www.cdc.gov/tceonline.
Those who participated in today’s COCA call who would like to receive continuing education should complete the online evaluation by March 16, 2017 and use the Course Code WC2286.
Those who will review the call On Demand and who would like to receive continuing education should complete the online evaluation by February 15, 2018 and they will use the Course Code WD2286.
To receive more information on upcoming COCA calls please subscribe to the COCA mailing list on – by going to the COCA website at emergency.cdc.gov/coca and click on the COCA Mailing List link. Also you’re invited to like your page on Facebook. It’s facebook.com/cdcclinicianoutreachandcommunityactivity. That way you can stay up-to-date with the latest news with COCA.
Thank you again to all the participants for being a part of today’s COCA call and have a great day.
Coordinator: Thank you for your participation in today’s conference. Please disconnect at this time.