Acute Flaccid Myelitis in 2018: The Largest Outbreak Since Surveillance Began

In brief

• Acute flaccid myelitis (AFM) is a rare, serious paralytic illness. U.S. outbreaks have come every other year since 2014, mostly in children in late summer and early fall.
• In 2018, 233 confirmed cases were reported — the most since surveillance began in 2014.
• Confirmed patients more often had only their arms affected (42%) and a respiratory illness or fever in the 4 weeks before weakness began (92%).
• Median times from the start of limb weakness to hospitalization, MRI and reporting to CDC were 1, 2 and 18 days.
• Prompt recognition, early specimens and rapid reporting speed investigations and help explain AFM.

Background

AFM was first recognized as a distinct condition in 2014, when previously healthy children across the country developed sudden limb weakness during a large outbreak of severe respiratory illness caused by enterovirus D68 (EV-D68). It has no known treatment or prevention. It is a known complication of West Nile virus, adenovirus and enteroviruses, and EV-D68 and EV-A71 have caused sporadic cases — but the recent pattern of outbreaks every two years, clustered in late summer and fall, had not been seen before, and its cause is still under investigation.

How cases were counted

Health departments sent CDC reports of anyone with sudden flaccid limb weakness. A confirmed case also had an MRI showing a spinal cord lesion largely in the gray matter; a probable case had raised white blood cells in the spinal fluid instead. Patients with other diagnoses — transverse myelitis, acute disseminated encephalomyelitis, Guillain-Barré syndrome, other spinal cord disease or stroke — or MRI findings inconsistent with AFM were non-AFM. CDC tested spinal fluid, respiratory, blood and stool specimens for enteroviruses and rhinoviruses (EV/RV), parechoviruses and, in stool, poliovirus; 31 patients were also tested for arboviruses.

What 2018 showed

Of 374 people reported, 233 (62%) from 41 states were confirmed, 26 (7%) probable and 115 (31%) non-AFM.

Figure 1. Confirmed AFM cases reported to CDC, August 2014–December 2018 (559 in all). CDC, MMWR.

Figure 2. AFM cases reported to CDC in 2018, by classification. CDC, MMWR.

Confirmed patients were significantly older than probable ones and younger than non-AFM patients; sex and race did not differ. Arms only were affected in 42% of confirmed cases versus 10% of non-AFM. A respiratory illness (cough, runny nose, congestion) or fever in the previous 4 weeks was reported in 92% of confirmed and probable cases versus 62% of non-AFM.

27% of confirmed patients needed respiratory support, most of them on a mechanical ventilator. None died during the acute illness, but two confirmed patients died months later.

Speed. Confirmed patients reached the hospital in a median 1 day and had an MRI in 2 days — a sign clinicians were recognizing AFM quickly; probable patients took 3 and 4 days. Specimens were collected 2 to 7 days after weakness began, depending on type, and reports reached CDC 18 to 36 days after onset, confirmed and probable cases sooner than non-AFM. Compared with 2016, MRI and respiratory and stool specimens came sooner in 2018, but reporting to CDC took a median 18 days, against 15 in 2016.

Testing. Among confirmed patients, respiratory specimens gave the highest yield: 44% tested positive for an enterovirus or rhinovirus, most often EV-D68. Only 2 (3%) spinal fluid specimens were positive — one EV-D68, one EV-A71. No poliovirus was found in any stool, and all arbovirus tests were negative.

What it means

The 2016 outbreak had 149 confirmed cases; 2018's was the largest, 42% of the 559 cases from August 2014 through December 2018. Fever or respiratory illness before weakness, a predominance of arm weakness, and viruses in respiratory specimens in about half of patients were all consistent with earlier years, and years of surveillance have sharpened the case definition.

• Spinal fluid testing is unlikely to explain the outbreaks: even when collected quickly, only 3% yielded a virus — perhaps because the virus does not reach the spinal fluid, or because fluid is taken only after neurologic symptoms begin.
• Respiratory specimens were still taken a median of about 3 days after weakness and 5 days after respiratory illness began, and viral shedding can be brief. CDC has added tests for antibody responses to implicated viruses.
• AFM can progress fast and may need respiratory support, so suspected patients should be hospitalized for close monitoring. With no confirmatory test, care should draw on symptoms, lab results, MRI and electromyography, with infectious disease specialists and neurologists. A 24-hour consultation service run by the AFM Physician Support Portal connects providers with transverse myelitis specialists at UT Southwestern and Johns Hopkins, and CDC's interim clinical considerations neither favor nor oppose the common treatments — immunoglobulin, steroids and plasmapheresis. Early physical rehabilitation may improve long-term outcomes.
• Reporting is still slow. In June 2019, the Council of State and Territorial Epidemiologists added MRI evidence of spinal lesions with some gray matter involvement to the reporting criteria. A physician's clinical diagnosis may differ from the surveillance definition.

Limitations: surveillance is passive and relies on clinician awareness, so cases may be missed; without a confirmatory test some may be misclassified; not every specimen type was sent for every patient; and national data on long-term outcomes are not yet available.

Sources

Based on Adriana Lopez, Adria Lee, Angela Guo, Jennifer L. Konopka-Anstadt, Amie Nisler, Shannon L. Rogers, Brian Emery, W. Allan Nix, Steven Oberste, Janell Routh and Manisha Patel, "Vital Signs: Surveillance for Acute Flaccid Myelitis — United States, 2018," MMWR Morbidity and Mortality Weekly Report, volume 68, Centers for Disease Control and Prevention; a work of the United States government in the public domain. The report was corrected by an erratum (volume 68, number 31): the median age of non-AFM patients is 8.8 years, as given here, not the 8.6 in the original results text.