Skip Navigation

A Prospective Study of Influenza Shedding in the Community

By Amesh A. Adalja, MD, FACP, December 21, 2012

The efficient spread of influenza within a community is accomplished through shedding of infectious virus by people who are infected. Several investigators have detailed the dynamics of shedding in volunteer challenge studies and in natural infection. Characterizing how and when shedding occurs is vital to creation of strategies to limit the spread of influenza. A new paper from the Robert Koch Institute, published in PLOS ONE, offers important insights into viral shedding of influenza.

Sequential Influenza Seasons from 2008-2011 Studied

In this study, Suess and colleagues conducted a prospective household study during 4 sequential influenza seasons in Germany. Each season was characterized by a different dominant influenza virus; notably, the 2009-2010 season was dominated by the pandemic H1N1 strain. The researchers analyzed data from 122 households in which an index patient was diagnosed with flu, defined as having an influenza-like illness and a positive rapid antigen test confirmed with PCR (which may have biased the study toward high shedders). Once an index patient was diagnosed, the patient’s household was observed for several days, during which time nasal wash or swab specimens were obtained daily from all participating household members. Data on symptoms, antiviral use, and vaccination status were also collected.

Spread from the Index Patient

The 122 households studied included a total of 442 participants. Index patients—some of whom received influenza vaccination that season—were variously infected with seasonal H1N1, pandemic H1N1, H3N2, and influenza B. In each season, children were almost always the index patients, and transmission to contacts occurred in 67 households. Of 320 household contacts who became infected, 20 had been previously vaccinated. Of these, just 2 became symptomatic with laboratory-confirmed influenza. In the 300 unvaccinated contacts, 56 individuals met the study criteria for flu, with an additional 9 asymptomatic subjects. Interestingly, all cases of transmission to unvaccinated children were symptomatic. The transmission interval ranged from 1.9 to 2.4 days depending on the strain involved.

Shedding Characteristics

Shedding curves based on nasal specimen analysis of 180 subjects indicated the following:

  • Evidence of presymptomatic shedding was observed in 30% of samples 1 day prior to onset of symptoms

  • Shedding occurred at the highest levels on days 1-3 of illness and was followed by a decline

  • Quantity of virus shed appeared to be the same for asymptomatic and symptomatic infected subjects

  • Those infected with influenza B exhibited the highest viral shedding load

  • Antiviral therapy induced a milder clinical course and faster illness resolution but had no effect on viral shedding

  • Vaccinated and unvaccinated patients did not differ in terms of clinical course and shedding

  • Viable viral shedding, as measured by viral culture, did not persist as long as PCR positivity

Results Should Inform Policy Discussions

Several aspects of this study should be incorporated into ongoing policy discussions about how best to prepare for influenza. Of major importance are the findings regarding presymptomatic shedding and the similar quantitative shedding that occurs in people who are and are not vaccinated.

The mounting evidence that a significant amount of presymptomatic shedding occurs should be used to inform 2 specific aspects of influenza planning: healthcare worker influenza vaccination and airport fever screening. Healthcare worker vaccination policies—often controversial—should reflect the fact that workers may be shedding virus prior to illness onset. Protocols that hinge solely on requiring symptomatic workers to abstain from patient care will not fully safeguard patients. Moreover, the fact that considerable shedding can still occur despite vaccination highlights the need for appropriate infection control, though some risk of transmission from vaccinated healthcare workers will remain.

Additionally, nations that employ airport fever screening to identify travelers harboring influenza and other infections should acknowledge that presymptomatic shedding undermines the efficacy of their systems, and continued use of these systems may engender a false sense of security.

Finally, the fact that shedding rates were not decreased in vaccinated subjects should influence the assumptions applied in viral spread models, in which people who are vaccinated are often assumed to be less efficient virus spreaders.


Suess T, Remschmidt C, Schink SB, et al. Comparison of shedding characteristics of seasonal influenza virus (sub)types and influenza A(H1N1)pdm09; Germany, 2007-2011. PLOS ONE 7(12): e51653.