Thursday, 7 January 2016

Zika virus disease (ZVD): 14 cases until 2007 then Yap island, Micronesia...

Update #1: 17JAN2016
Between the discovery of Zika virus (ZIKV) in 1947 and today's review subject, an outbreak on Yap Island,[1] there had only been 14 cases of Zika virus disease (ZVD) described. The keyword there is detected, or actively sought.

Some opinions on ZIKV...

In my opinion - and as the following study supports for this outbreak it describes - the confirmed ZIKV cases we heard of during 2014 are likely the tip of a vastly bigger iceberg of human infections. Many of the countries from which we have recent reports of "first" autochthonous (local) ZIKV transmission may well have had this virus in their midst for much (much, much, much?) longer since it often causes mild illness. Unless carefully looked for, ZVD is clinically very similar to other mosquito-borne virus infections. 

I hope that on the list of things to do to understand ZIKV - probably topped by investigating the suspected link between ZIKV infection and microcephaly) - is go back to stored human sera and look for traces of ZIKV or antibodies against ZIKV in these newly announced transmission zones. This will give us an idea of how long this virus has been around.

Yap Island ZVD outbreak in 2007...

In April and May of 2007, an outbreak of rash, conjunctivitis, subjective (not measured) fever, arthralgia, and arthritis occurred on an island group (the 4 Yap Islands; population ~7,400 in 2000) that comprise Yap State in the Federated States of Micronesia. Three initial patients tested positive for IgM antibodies that suggested recent dengue virus (DENV) infection (although IgM can be non-specifically triggered as a result of another infection). However. the doctors believed the symptoms were not due to DENV infection.

Samples were sent away to the United States Centers for Disease Control and Prevention and ZIKV RNA was detected in the sera of 10 of 71 (14.1%) symptomatic people using a specific reverse transcription polymerase chain reaction (RT-PCR) assay. Other viruses were considered but could not be detected using specific assays. These viruses included dengue, chikungunya, o’nyongnyong, Ross River, Barmah Forest, and Sindbis virus. The finding implicated ZIKV as the likely cause of the outbreak, and ZVD as the disease.

The result triggered a bigger investigation during which 185 suspected cases (including 49 confirmed and 59 probable) were further investigated. 66% of the confirmed and probable casess were female and these 108 people had a median age of 36y. The earliest infections occurred 15th April. Among 31 of the confirmed cases who reported symptoms, rash lasted for a median of 6d (2-14d) and arthralgia for 3.5d (1-14d). None had travelled outside of Yap.

Enzyme-linked immunosorbent assay (ELISA), an antibody detection method, was used to look for IgM antibodies indicating recent ZIKV or DENV infections. The identification of very specific infection-neutralizing antibody to these viruses was also sought and the level determined by using a plaque reduction neutralization test with a threshold value of 90% (PRNT90). Antibodies to other flaviviruses were not sought in this study.

A case definition was created:
  • A patient with suspected ZVD had acute onset of generalized macular or papular rash, arthritis, arthralgia, or nonpurulent conjunctivitis. 
    • Blood samples were requested from acute phase ( within 10d after the symptom onset) and convalescent phase (14d later)
  • A patient with probable ZVD had:
    • IgM antibody against ZIKV
    • ZIKV PRNT90 that was still detectable at a dilution of (titer) at least 1:20
    • A ratio of ZIKV PRNT90 titre to DENV PRNT90 titre was less than 4
    • Either no ZIKV RNA detected by RT-PCR or a sample which was inadequate for the performance of RT-PCR
  • A patient with confirmed Zika virus disease had:
      • ZIKV RNA detected in their serum or 
      • IgM antibody against ZIKV (detected by ELISA) and 
      • A ZIKV PRNT90 titre of at least 20 and 
      • A ratio of ZIKV PRNT90 titre to DENV PRNT90 titre of at least 4. 
173 of 200 randomly selected households were surveyed yielding 557 blood samples of which 414 (74%) had IgM anti-ZIKV antibody detected and 156 of these recalled a relevant recent illness. Among those without any IgM detected, 27 also recalled a relevant illness. Extrapolation of these figures resulted in the authors estimating that 73% of residents were infected during this outbreak with 18% of those infected having a clinical illness likely due to ZIKV infection.

An examination of 1,366 containers holding water, frequently (43%) found that they also contained 9 species of mosquito larvae/pupae and such containers were present in 87% of households. [A simple intervention method is to tip these out and ensure they cannot refill with water]. Three other mosquito species were identified to as adult mosquitoes. Aedes hensilli was the most frequently identified mosquito species overall. 

No mosquitoes tested contained detectable infectious ZIKV (examined by cell culture methods) or ZIKV RNA (RT-PCR).

So, jumping forward from the 1940s to the naughties (2007), we see a major shift in the diagnostic arsenal which supported this study. It allowed us to see just how effectively a mosquito-borne virus can apparently consume an entire island community. I wonder what IgG results - generally interpreted to indicate older earlier infection by an agent - on these same sera would yield? Would they further support the author's conclusions that a viraemic human (infected individual with ZIKV in their blood) or a virus-infected mosquito had only recently arrived from somewhere else?

Could something have changed in the levels of the vector population/species resulting in a sudden surge in virus? 

The authors note that there had not been any previous disease outbreaks of this sort reported on Yap-but were isolated cases occurring earlier? Had the virus been slowly smoldering in rare transmissions from mosquito to human/other animal to mosquito etc, for a while before building up speed and manifesting as an outbreak?  Why did none of the tested mosquitoes contain traces of ZIKV virus? 

It is also worth noting that many ZVD "cases" described in more recent months may be better defined as "suspect" and not yet confirmed. But I stand to be corrected on that.

  1. Zika Virus Outbreak on Yap Island, Federated States of Micronesia
    Duffy MR, Chen TH, Hancock WT, Powers AM, Kool JL, Lanciotti RS, Pretrick M, Marfel M, Holzbauer S, Dubray C, Guillaumot L, Griggs A, Bel M, Lambert AJ, Laven J, Kosoy O, Panella A, Biggerstaff BJ, Fischer M, Hayes EB.
    N Engl J Med. 2009 Jun 11;360(24):2536-43. doi: 10.1056/NEJMoa0805715.


  1. Added 18% figure to clinical illness from population extrapolations in this paper