Analysis of Warning Process: Swine-Origin Triple Reassortant Influenza A(H3N2) Virus In USA

by James Wilson, M.D. - Biosurveillance - December 21, 2011

We have been monitoring the situation with swine-origin triple reassortant influenza A(H3N2) virus (S-OtrH3N2) inside the United State for some time now.  Recent reports of likely human-human transmission sparked debate, discussed here in this post.

1.  First, the ECDC's risk assessment was an excellent discussion of the situation.  CDC's report may be found here.

2.  The time delays in issuing statements about the evolving situation reflect the forensic process and length of time it takes to recognize novel pathogens and execute public warning inside the United States.  In this case, a joint press releaseon Sept 2nd and 5th, respectively, from CDC and Pennsylvannia state officials announced recognition of S-OtrH3N2 pediatric cases in Indiana and Pennsylvania stemming from exposure to infected swine.  The breakdown of timelines as follows:

Case A (Indiana):

• Day 0 (July 23rd).  Parental recognition of a child with flu-like illness.  Child brought to local ED, clinical laboratory testing at the hospital positive for type A influenza.   Child found to have been vaccinated the prior fall, had chronic medical condition, discharged from the ED same day without anti-virals.
• Day 1 (July 24th).  Child returns to ED with a worsened clinical picture and admitted.
• Day 4 (July 27th).  Child does well and is discharged from the inpatient ward to home.  No further issues.
• Day 25 (Aug 17th).  Indiana State Department of Health Laboratories test the sample and discover suspect swine-origin influenza A (H3N2) virus.  CDC notified and sample forwarded to CDC for confirmation and further analysis.
• Day 27 (Aug 19th).  CDC confirms findings of the Indiana State DOH Lab. 
• Day 41 (Sep 2nd).  CDC reports publicly results of genomic sequencing: swine-origin triple reassortant influenza A(H3N2) virus that now includes segments of 2009 pandemic H1N1 virus.  Report indicates patient had contact with infected swine.
• Summary:
  
  • Time delta: state public health recognition of unusual influenza virus= 25 days
  • Time delta: clinical processing of patient to CDC awareness= 25 days
  • Time delta: CDC issuance of warning to the public= 41 and 14 days from initial clinical processing of the patient and state notification to CDC, respectively
    

Case B (Pennsylvania):

• Day 0 (Aug 20th).  Parental recognition of symptoms, take child to local ED.  Nasal swab positive for influenza A.  Child was vaccinated the prior fall.
• Day 3 (Aug 23rd).  Pennsylvania State Department of Health Laboratory idenifies suspected swine-origin influenza A (H3N2) virus.
• Day 4 (Aug 24th).  Pennsylvania state officials notify CDC of the findings.
• Day 6 (Aug 26th).  CDC confirms findings.
• Day 13 (Sep 2nd).  CDC reports publicly results of genomic sequencing: swine-origin triple reassortant influenza A(H3N2) virus that now includes segments of 2009 pandemic H1N1 virus.  Report indicates patient had contact with infected swine.
• Summary:
  • Time delta: state public health recognition of unusual influenza virus= 3 days
  • Time delta: clinical processing of patient to CDC awareness= 4 days
  • Time delta: CDC issuance of warning to the public= 13 and 7 days from initial clinical processing of the patient and state notification to CDC, respectively

Questions were asked openly of CDC, USDA, and Pennsylvania state officials on Sept 12th about how these cases were recognized.  More to the point, guidance for clinical sampling was requested publicly.  This request was never answered.  At this point, the situation had migrated to the top of biological threat monitoring list.  Lack of response was not reassuring.

On Nov 23rd, a probable human-human transmission cluster of S-OtrH3N2 cases was reported by CDC, with an evident 7-10 day lag between initial clinical processing, laboratory based recognition at the state level, and CDC issuance of a public statement.  This was a much better time delta and likely reflective of surveillance grid sensitization this particular issue. 

Overall, we see a drop from a time delta between clinical processing and CDC issuance of public warning of these events from 41 to 13 to 7 days as the situation evolved.  It was not until Nov 26th did media reports indicate a seed strain had been sent to vaccine manufacturers in case S-OtrH3N2 begins to overtake current circulating strains of influenza virus in the human population.

There are several summary observations to make at this point:

1. S-OtrH3N2 is our #1 pathogen of concern for the United States and the world at this time, from the perspective of potential social disruption.  Projected impact would be similar to that seen in a vaccine drift year of seasonal influenza (e.g. 2004 in the United States).  Thus far no fatalities have been observed, but the sample size is far too small at this time to make firm assertions of case fatality rate or morbidity profiles. 
2. CDC should be praised for proactively forwarding seed stock to vaccine manufacturers.  What will be interesting to observe at this point is the time delta between initial recognition of sustained human-human transmission and an order to ramp up and distribute S-OtrH3N2 vaccine to local healthcare providers.  In this instance CDC stands in much better position to execute more efficient warning to response operational coupling.
3. The above said, we note with great concern how long it took to recognize and effectively communicate to the public the initial situation in Indiana.  From the perspective of this author, a pediatrician practicing in a rural setting, there is no healthcare provider sensitization to this issue- i.e. there is no change at all in sampling procedures for children presenting with influenza-like illness.  It is as though this season is like any other.  Especially after the 2009 H1N1 pandemic experience and the current economic recession, we observe a general reluctance to test for influenza unless there is a compelling reason to do so.
4. Not all states have the surveillance capabilities of Pennsylvania and Iowa.  There is an incomplete understanding of how pervasive S-OtrH3N2 transmission is inside the United States at this time. 
5. Abrupt community inundation of pediatric flu-like illness, especially among children previously vaccinated for the 2010-2011 influenza season, should prompt an index of suspicion at least to obtain samples for testing by clinicians.  Submission of samples that are processed by state laboratories and moved up the chain to CDC is the only way to assess this evolving situation accurately.

http://biosurveillance.typepad.com/biosurveillance/2011/12/analysis-of-warning-process-swine-origin-triple-reassortant-influenza-ah3n2-virus-in-usa.html