What to do about the Flu?
/
Butler, C., Velden, A., Bongard, E., Saville, B., Holmes, J., Coenen, S., Cook, J., Francis, N., Lewis, R., Godycki-Cwirko, M., Llor, C., Chlabicz, S., Lionis, C., Seifert, B., Sundvall, P., Colliers, A., Aabenhus, R., Bjerrum, L., Harbin, N., Lindbæk, M., Glinz, D., Bucher, H., Kovács, B., Jurgute, R., Lundgren, P., Little, P., Murphy, A., Sutter, A., Openshaw, P., Jong, M., Connor, J., Matheeussen, V., Ieven, M., Goossens, H., Verheij, T. (2019). Oseltamivir plus usual care versus usual care for influenza-like illness in primary care: an open-label, pragmatic, randomised controlled trial The Lancet https://dx.doi.org/10.1016/s0140-6736(19)32982-4
This was an open label (non-blinded) pragmatic randomized control trial comparing usual care and usual care plus Oseltamavir in patients 1 year old and older. 3266 patients were enrolled in 15 European counties in the primary care outpatient setting who met the inclusion criteria of having reported fever, at least one upper respiratory symptom (cough, sore throat, rhinorrhea, or congestion), and at least one systemic symptom (headache, myalgias, diaphoresis, chills, or fatigue) for less than 72 hours of duration. They were randomized into a group where they would receive either routine symptomatic care (not specifically defined) or routine care plus a 5 day course of Tamiflu. Severity was rated as mild, moderate, or severe by clinicians. Only about 50% were flu positive in each group.
The primary outcome was patient reported time to recovery and they tracked this by having patients record their symptoms daily in a diary. They found that on average people's symptoms improved one day earlier in the Tamiflu group and patients who were chronically ill and older had even larger benefit in the time to recovery. Secondary outcomes included no difference in documented Tamiflu benefit in patients with influenza A, influenza B, and influenza negative testing (which may be from PCRs poor sensitivity and false negative rate or potential cross-reactivity of Tamiflu on other viruses). They found that patients assigned to the Tamiflu with routine care group had less antibiotic prescriptions and had a lower incidence of new household infections with no change in repeat visits, hospitalizations, X-ray confirmed pneumonia or increase in over-the-counter medication use. There was an incidence of nausea and vomiting in the Tamiflu group (21% in the Tamiflu group compared to 16% in the usual care group with a NNH of 20). Serious reactions were very rare.
Hayden, F., Sugaya, N., Hirotsu, N., Lee, N., Jong, M., Hurt, A., Ishida, T., Sekino, H., Yamada, K., Portsmouth, S., Kawaguchi, K., Shishido, T., Arai, M., Tsuchiya, K., Uehara, T., Watanabe, A. (2018). Baloxavir Marboxil for Uncomplicated Influenza in Adults and Adolescents New England Journal of Medicine 379(10), 913-923. https://dx.doi.org/10.1056/nejmoa1716197
Hayden et al,’s 2018 publication in the New England Journal of Medicine describes the results of the phase 2 and phase 3 trials of baloxavir marboxil, a novel anti-influenza medication. Baloxavir inhibits a different step in the influenza replication process than the existing therapies, and it has been associated with decreased mortality and viral loads in mouse influenza models, suggesting potential benefit in humans as well.
The phase 2 and phase 3 trials were conducted similarly, though with some minor differences. Both trials were double-blinded and randomized. The phase 2 trial compared a single administration of baloxavir at three dosages (10mg, 20mg, 40mg) to placebo in Japanese adults during the 2015-2016 influenza season. The phase 3 trial compared weight-based baloxavir (40mg or 80mg) with a 4-day placebo tail to a 5-day course of oseltamivir and to a five-day course of placebo in American and Japanese adults during the 2016-2017 influenza season; the phase 3 trial also compared a single administration of baloxavir to placebo in pediatric patients. Criteria for inclusion in both phases included an axillary temperature greater than 38°C plus one respiratory symptom, with the onset of symptoms less than 48 hours prior to enrollment. A positive influenza test was required for phase 2 but not for phase 3, though only influenza positive patients were included in the primary analysis for phase 3. Notably, exclusion criteria included pregnancy, patients with underlying comorbidities, weight less than 40 kg, and symptoms that required hospitalization.
For the duration of the study, enrolled participants rated 7 influenza symptoms on a 4-point scale for 14 days, and they assessed their overall health status on a 10-point scale daily throughout this time. Participants’ temperatures were measured during the 14 days, and swabs were obtained daily until day 8-9 for viral quantification and susceptibility testing. The primary endpoint was defined as the time to alleviation of symptoms, defined as the point in time when all 7 symptoms were either absent or mild for at least 21.5 hours. Secondary endpoints included time to resolution fever and return to usual health as well as viral load and susceptibility testing.
During the phase 2 trial, 389 of 400 participants completed the trial. Participants who received baloxavir at any dose demonstrated a lower median time to alleviation of symptoms compared to placebo (49.5-54.2 versus 77.7 hours). Baloxavir recipients also demonstrated greater reductions in viral loads on days 2-3 compared to placebo. Adverse events were similar between placebo and baloxavir, suggesting that the medication is well tolerated. Notably, however, 2.2% of baloxavir recipients developed viral strains with amino acid substitutions that result in a substantially decreased susceptibility to the trial drug.
During the phase 3 trial, 1064 of 1436 enrollees were included in the intention-to-treat primary analysis. Both adult and pediatric baloxavir recipients demonstrated a lower median time to alleviation of symptoms compared to placebo. This reduction in time to alleviation was greater among patients who began baloxavir earlier after onset of symptoms. Interestingly, even among influenza negative enrollees, the baloxavir recipients still demonstrated a significant reduction in time to alleviation compared to placebo, potentially suggesting some antiviral activity beyond just influenza. Baloxavir recipients also demonstrated more rapid declines in viral load and a shorter duration fever compared to the oseltamivir and placebo treatment arms. However, baloxavir and oseltamivir demonstrated no difference in time to alleviation of symptoms, a finding which the study authors believe may be reflective of some therapeutic “ceiling in self-limited influenza illness in adults.” This phase 3 trial re-demonstrated a concerning finding from the phase 2 study, with 9.7% of baloxavir recipients developing influenza strains with mutation patterns consistent with resistance to the drug. Furthermore, enrollees who developed these resistant strains demonstrated prolonged infectious shedding duration and time to alleviation of symptoms compared to placebo. This mutation pattern was not seen in trial enrollees who did not receive baloxavir.
In summary, baloxavir is a novel anti-influenza medication that appears to have a similar reduction in time to alleviation of symptoms compared to oseltamivir, with maximum therapeutic benefit achieved if taken as soon as possible after symptom onset. The drug also has a reassuring side effect profile compared to both placebo and oseltamivir. Potential benefits of baloxavir compared to existing therapies include its one-time dosing as well as its greater reduction in viral load, though these effects have not been associated with any beneficial patient-centered outcomes. The greatest potential complication revealed by the phase 2 and phase 3 trial data is the emergence of new resistance patterns to this medication, a risk which was reproducible and associated with prolonged duration of symptoms and infectious shedding.
Kwong, J., Schwartz, K., Campitelli, M., Chung, H., Crowcroft, N., Karnauchow, T., Katz, K., Ko, D., McGeer, A., McNally, D., Richardson, D., Rosella, L., Simor, A., Smieja, M., Zahariadis, G., Gubbay, J. (2018). Acute Myocardial Infarction after Laboratory-Confirmed Influenza Infection The New England Journal of Medicine 378(4), 345-353. https://dx.doi.org/10.1056/nejmoa1702090
We reviewed the Canadian study by Kwong et al. published in the NEJM in January of 2018. This was a self-controlled case series which identified 364 hospitalizations for myocardial infarction in the province of Ontario between 2009 and 2014 that occurred either one year before or one year after a laboratory diagnosis of influenza. The authors compared the incidence of MI within the risk interval (1 week after the influenza diagnosis) to that in the two-year control interval. They found that the incidence of MI was 6.1 times higher in the risk interval compared to the control (20.0 admissions per week vs. 3.3 admissions per week). Interestingly, they also analyzed subtypes of influenza, as well as other respiratory viruses, and found that the MI incidence ratio was 10.1 for influenza B (a finding particularly pertinent this year), 5.2 for influenza A, 3.5 for RSV, and 2.8 for other respiratory viruses.
The authors propose several potential mechanisms for their findings – perhaps influenza can contribute to MI through acute inflammation, biomechanical stress of tachycardia, through platelet activation and endothelial dysfunction, or perhaps through vasoconstriction. They do mention several limitations of their study which are inherent to the self-controlled case control design — such as the uncertainty of the timing of actual onset of influenza infection and MI, as well as confounding due to time-varying factors (both influenza and MI are more common in the Canadian winter). In addition, this is not the first study demonstrating the association between influenza and MI, and most cardiologists were already well aware of this link.
I do, however, think that it is a particularly convincing study that will change my practice pattern in the Emergency Department. While I certainly encourage my ED patients with lung disease to get vaccinated yearly, I will now be as aggressive in encouraging this in my patients with a history of MI, or indeed any other cardiac risk factors. I do not know whether this study will encourage me to medicate my influenza patients with cardiac risk more aggressively, and I think that RCTs to investigate this question are very much warranted.
References
Hayden, F., Sugaya, N., Hirotsu, N., Lee, N., Jong, M., Hurt, A., Ishida, T., Sekino, H., Yamada, K., Portsmouth, S., Kawaguchi, K., Shishido, T., Arai, M., Tsuchiya, K., Uehara, T., Watanabe, A. (2018). Baloxavir Marboxil for Uncomplicated Influenza in Adults and Adolescents New England Journal of Medicine 379(10), 913-923. https://dx.doi.org/10.1056/nejmoa1716197
Kwong, J., Schwartz, K., Campitelli, M., Chung, H., Crowcroft, N., Karnauchow, T., Katz, K., Ko, D., McGeer, A., McNally, D., Richardson, D., Rosella, L., Simor, A., Smieja, M., Zahariadis, G., Gubbay, J. (2018). Acute Myocardial Infarction after Laboratory-Confirmed Influenza Infection The New England Journal of Medicine 378(4), 345-353. https://dx.doi.org/10.1056/nejmoa1702090
Butler, C., Velden, A., Bongard, E., Saville, B., Holmes, J., Coenen, S., Cook, J., Francis, N., Lewis, R., Godycki-Cwirko, M., Llor, C., Chlabicz, S., Lionis, C., Seifert, B., Sundvall, P., Colliers, A., Aabenhus, R., Bjerrum, L., Harbin, N., Lindbæk, M., Glinz, D., Bucher, H., Kovács, B., Jurgute, R., Lundgren, P., Little, P., Murphy, A., Sutter, A., Openshaw, P., Jong, M., Connor, J., Matheeussen, V., Ieven, M., Goossens, H., Verheij, T. (2019). Oseltamivir plus usual care versus usual care for influenza-like illness in primary care: an open-label, pragmatic, randomised controlled trial The Lancet https://dx.doi.org/10.1016/s0140-6736(19)32982-4
Authorship
Butler, et al. Usual Care - Written by Trevor Skrobut, MD, PGY-3, University of Cincinnati Department of Emergency Medicine
Hayden, et al. Baloxavir - Written by Nicholas Jensen, MD, PGY-3, University of Cincinnati Department of Emergency Medicine
Kwong et al - Written by Michael Gleimer, MD, PhD, PGY-3, University of Cincinnati Department of Emergency Medicine
Peer Review and Editing by Jeffery Hill, MD MEd
Conflicts of Interest
No conflicts of interest to report for any of the study authors or editors
