Injections and acute flaccid myelitis: the dog that hasn't barked
The editorial and review of the US outbreaks of acute flaccid myelitis/AFM is timely, but it fails to mention a potentially important co-factor in the cause of this “mystery illness”—intramuscular injections and provocation paralysis. (Stelzer-Braid, BMJ, 19 Dec 2018)
I venture to say that few clinicians today are aware that injections are strong risk factors for paralytic polio: recent injections (e.g. antibiotics, vaccinations) have accounted for 66% to 86% of attributable risk of paralysis when polioviruses are circulating. (Hill and Knowelden, BMJ, 1 July 1950. Strebel et al, NEJM 1995;332:500. Kohler et al, Int J Epidem, 2002;32:272) Even fewer clinicians are likely to know about poliovirus receptors, which are not expressed in normal human muscle fibers but are rapidly up-regulated in muscle damaged by injections. (Dalakas et al, NEJM 1995;333:62) This enables circulating polioviruses to bind to motor end plates from where they are transported along motor nerves to the spinal cord. (Gromeier and Wimmer, J Virol 1998;72:5056. Ren and Racaniello, J Infect Dis 1992;166:747) 99% of poliovirus infections are benign and self-limited, but of the 1% of paralytic cases a substantial proportion are provoked by injections. There is a dose-response effect: in Strebel’s report of vaccine-associated paralytic polio/VAPP in Romania, a single injection within 30 days of paralysis onset increased VAPP risk 8-fold; 2 to 9 injections increase VAPP risk 27-fold; and 10 or more injections increased VAPP risk 182-fold! For the contacts of OPV recipients the peak risk occurred when injections were given 8 to 21 days before onset of paralysis, similar to Hill’s observations from the 1949 polio epidemic in the UK.
It would seem that the foregoing observations should apply to recent AFM outbreaks. Gromeier, for example, showed experimentally how IM injections provoked paralytic polio, and suggested that the same thing could happen with non-polio enteroviruses. In spite of this, CDC investigations have included no questions about injections. This is surprising since the CDC was responsible for the Strebel study of provocation paralysis in Romania. They steadfastly publicize AFM as a “mystery disease.”
As of December 17, 2018 the CDC had confirmed 165 US cases of AFM so far in 2018; another 155 possible cases were under investigation. Since August 2014 there have been a total of 491 cases of AFM confirmed by the CDC, and some experts believe that the actual number is substantially larger because many cases go unreported…..This is a devastating disease: only 8% to 18% of children with AFM fully recover and 8% to 14% are left with severe disabilities. (Gill et al, CMAJ 2018 Dec 3;190:E1418)
Mere mention of injections in connection with AFM seems to be taboo, at least in US publicity about this mystery disease. I am aware of a single exception. On November 1, 2016 the Seattle Times reported the death of a 6 year-old boy from Bellingham, Washington from what was thought to be AFM. The case provoked some controversy because the family suspected vaccinations: 14 days before the onset of his illness he had received multiple vaccines for school admission, plus a flu shot. The Washington State Department of Health and hospital authorities publicly dismissed any role for vaccinations. Eventually, the CDC decided that his case was not AFM. The cause of his illness was never announced. (Seattle Times, Nov 5 and Nov 14, 2016)
Non-polio enteroviruses are at the top of the list of suspects, but I know from private correspondence that a number of pediatric experts believe we must also consider injections and provocation paralysis as possible co-factors. AFM epidemiology is consistent with the pattern of enterovirus circulation; it also follows the timing of back-to-school shots. Why is AFM so prominent in the US? Does it have anything to do with immunization policies and practices? Where do we go from here…..?
ALLAN S. CUNNINGHAM 20 December 2018
Competing interests: No competing interests