Our public health authorities (WHO and CDC) apparently ignored a Eurosurveillance Rapid Communication: "Detection of Zika Virus RNA in Whole Blood of Imported Zika Virus Disease Cases Up to 2 Months After Symptom Onset, Israel, December 2015 to April 2016." Source: Eurosurveillance, Volume 21, Issue 26, 30 June 2016
More recently (Dec. 2016): "detection of Zika virus RNA in vaginal secretions up to day 14 and in erythrocytes up to day 81" (almost 3 months in blood). Source: December 14, 2016 CDC post called Prolonged Detection of Zika Virus in Vaginal Secretions and Whole Blood.
Re: "ZIKV viremia is normally detectable only within the first week ..." is what is commonly written in medical literature.
I wonder if this is because of the highly flawed Trioplex assay (which fails to detect 40 percent of Zika cases and all four strains of dengue). The ZIKV Detect IgM Capture ELISA test, made by InBios International has been deemed "too inaccurate" by the FDA, as well.
As of November 12th, 2016, a whopping 564,571 donations have tested positive on the cobas Zika test.
Dr. Jay S. Epstein, Office of Blood Research and Review director at the FDA, said in a statement:
"Zika virus is a transfusion transmitted disease which can cause potentially severe consequences including microcephaly and Guillain-Barré syndrome. The requirement to test blood donations for Zika virus has already resulted in interdicting contaminated collections confirming the value of testing."
Sadly, in my numerous attempts to let both Canadian Blood Services and the American Red Cross know that they should extend their deferment policies of 21 days and 28 days, respectively, they informed me they would not. India imposed a 120-day wait (the only country I know of to heed the early warnings).
Source (with citations): http://www.infobarrel.com/Zika_Virus_Our_Tainted_Blood_Supply