Influenza: spread by breathing; flu association with MI

​(second of 2 blogs)

a recent study found that influenza virus could spread just by exhaled breath (see doi:10.1073/pnas.1716561115 )​.

Details:

--355 symptomatic volunteers with acute respiratory illness in a college community, with 142 of 178 confirmed influenza infections in patients

--in those with complete data: 50% male, 22% flu vaccination in current year, 15% vaccinated in current plus past year, asthma in 21%, 15% smokers, mean age 20, BMI 22.7, median coughs/30 min= 18, no sneezes, 8% systemic symptoms, 3% lower resp tract and 7% upper resp tract sx

--218 paired nasopharyngeal (NP) samples as well as 30-minute breath samples on days 1-3 after symptom onset, with the breath samples divided into coarse (>5 mm) vs fine (<=5 mm) particles

Results:

--infectious influenza virus was recoverable from:

    --89% of the NP swabs, with 8.2x10⁸ RNA copies and up to 10⁵ infectious particles

    --39% of the fine aerosols, with 3.8x10⁴ RNA copies in 30-minutes of normal tidal breathing and up to 10³ infectious particles

    --coarse aerosol, with 1.2x10⁴ RNA copies in 30-minutes

--152 of the participants had at least one positive NP swab and 4 had only a positive aerosol sample

--in a small substudy of 23 participants who did not have a cough, 3 had coarse-aerosol samples with positive viral RNA; and 11 had fine-aerosol samples​ with detectable viral RNA/8 had positive influenza cultures

--both fine and coarse aerosol viral RNA levels were positively associated with BMI and with cough frequency, and negatively associated with increasing days since symptom onset

    --though coughing or sneezing were not necessary for potentially infectious aerosol

--fine-aerosol viral RNA was positively associated with having had the flu vaccine for both the current and previous flu seasons [unclear why???]

--NP swab viral RNA was positively associated with URI symptoms and negatively associated with age, but not associated with fine or coarse aerosol or their predictors

Commentary:

--this study suggests that fine aerosol has potentially infectious flu virus, not necessarily associated with coughing, and that they may remain suspended in air, thereby presenting a viable risk for airborne transmission

--and the discordance between the NP swabs and breathing suggests that the flu virus may independently be in the upper or lower respiratory tracts (the fine aerosol reflecting infection in the lung). not surprisingly upper resp sx were strongly associated with positive NP swabs. no specific symptoms or systemic signs were associated with aerosolized virus

--also not so surprising was that there was not much relationship between people without cough and coarse aerosol virus, since the cough tends to produce aerosols form large airways by its force, leading to coarse droplets

So, this study, though not showing actual influenza disease transmission, does raise the high likelihood that at least some influenza is spread through breathing and without coughing. So, we are probably correct in asking our staff to wear masks if they have upper respiratory infections…. [though best if the mask covers both the mouth and nose]. And reinforcing to frail patients in particular that they should avoid crowded conditions during flu season, such as buses/stores/etc. Maybe they should also wear masks???

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 

And a new Canadian study reported an association between influenza and acute myocardial infarction (see Kwong JC. N Engl J Med. 2018;378: 345).

Details:

--self-controlled case-series of 332 patients hospitalized for acute MI within 1 year before and 1 year after a positive laboratory-confirmed influenza infection

--median age 77 (26% <65yo), 52% male, 24% with prior MI, 49% diabetic, 38% dyslipidemia, 85% hypertension, 31% vaccinated

--82% influenza A (10% H1N1, 34% H3N2, 38% not subtyped), 18% influenza B

--they then looked at when during this year that the MI occurred: defining a "risk interval" as being the first 7 days after the respiratory specimen was collected, and the "control interval" being other times in the 1 year before 1 year after the risk interval

Results:

--during the risk interval, there were 20.0 admissions/week for acute MI

--during the control interval, there were 3.3 admission/week

    --the incidence ratio of admission for acute MI comparing these intervals as 6.05 (3.86-9.50)

    --no increased incidence after the 7 day interval

--comparing the risk for viral subgroups (they repeated their analyses for RSV and viruses other than flu/RSV to assess the specificity of their findings):

    --influenza B: 10.11 (4.37-23.38)

    --influenza A: 5.17 (3.02-8.84)

    --RSV: 3.51 (1.11-11.12)

    ​--other viruses: 2.77 (1.23-6.24)

--subgroup analysis: no statistical differences by >65yo vs<65yo, sex, influenza type/subtype, vaccination status, history of MI (though several of these subgroups were pretty small and hard to interpret the significance of these findings)

Commentary:

--as compared to several older studies, this study provides more specific data finding an association between seasonal flu activity and cardiovascular mortality

--the study relies on the extensive Canadian health databases, through their universal system of healthcare access to physician services, hospital care, and lab testing [if only we had such a system..... not just better health care but more useful public health and clinical data.....]

--purported mechanisms: acute inflammation leading to platelet activation and endothelial dysfunction. also perhaps increased metabolic demand in patients with borderline cardiac reserve [there is suggestive support for this: looking at the incidence of acute MI, it was much higher on days 1-7, in the 6-fold range, though actually (nonsignificantly) lower days 8-28, in the 25-40% lower range; this is consistent with a time shift earlier of vulnerable people with marginal cardiac reserve getting the stress of the flu, then having MIs earlier in the acute flu infection range of 1-7 days.  there is no info on the MI change for other times, and no real analysis was done on this]

--and the big limitation of a study like this is getting really accurate clinical data and timing, since this study is basically big database trolling. and there might have been significant biases as to which patients got viral testing done, for example. 

So, this study adds to the imperative to vaccinate the elderly and those with underlying heart disease. And per the blog yesterday, doing so annually to optimize protection. Though these people in particular often have less of a protective effect of immunization, there are also prior studies suggesting that immunizing others at home and health care workers (ie, vectors) decreases the chance of flu in the elderly/infirmed.