The Jury is Still Out

Solid Evidence is Lacking That COVID-19 Causes Kawasaki-Like Symptoms in Children

On May 14, 2020, the Centers for Disease Control and Prevention (CDC) sent an advisory to state health departments regarding a new clinical syndrome potentially affecting children with the unwieldy name "Multisystem Inflammatory Syndrome (MIS-C) Associated with Coronavirus Disease 2019 (COVID-19)."

The advisory highlighted two recently published articles in The Lancet that described symptoms resembling Kawasaki disease in children - some of whom had tested positive for COVID-19. Kawasaki disease, first described in 1967, causes inflammation in the blood vessels of the heart, but potentially also throughout the body.

The CDC advisory could easily alarm parents who are already anxious for the well-being of their children in the pandemic, so I should stress that to date, no documented cases of Kawasaki-like illness associated with COVID-19 have been reported at the University of New Mexico Children's Hospital.

And, on a positive note, MIS-C appears to respond favorably to established treatments, and the vast majority of patients recover. UNM's pediatric subspecialty experts have all necessary tools needed to ensure that New Mexico's children receive the highest quality medical care.

An unanswered question at this point is whether this newly reported illness is in fact caused by the SARS-CoV-2 virus. Over the years, no definitive cause for Kawasaki has ever been established, although many candidates have been proposed.

One of the Lancet articles reported on a cluster of eight children seen at a hospital in London over a period of 10 days in April who showed features similar to atypical Kawasaki disease, Kawasaki disease shock syndrome or toxic shock syndrome.

These children were treated with intravenous immunoglobulin and aspirin, and seven recovered. In the hospital they all tested negative for SARS-CoV-2 via a polymerase chain reaction (PCR) test. However, all eight later tested positive by means of antibody testing.

The second Lancet article was from Italy, where a retrospective review classified children with Kawasaki-like disease into two groups. Group 1 included 19 cases diagnosed between Jan. 1, 2015, and the start of the COVID epidemic on Feb. 17, 2020.

In Group 2, there were 10 cases diagnosed between Feb. 18 and April 20, 2020. All of the patients in both groups recovered. Nose and throat swabs were COVID-positive for two of the Group 2 patients, but eight later tested positive via antibody testing.

In summary, The Lancet published two retrospective studies of cases of Kawasaki-like disease during the COVID pandemic, both of which raise the question of whether COVID is associated with a Kawasaki-like disease in children.

Due to study limitations related to sample size and the retrospective uncontrolled design, the answer to this question remains unknown. Half the patients in both studies had confirmed or suspected COVID contacts. Overall, SARS-CoV-2 was detected by PCR in only four out of 20 cases. (Positive PCR provides direct evidence of SARS-CoV-2 infection, while indirect evidence of SARS-CoV-2 infection was provided through antibody testing.)

Antibody testing for SARS-CoV-2 is reported to have a sensitivity of 95% and a specificity of 85-90% when compared with the PCR test, but it remains unknown whether it might also show cross-reactivity with other coronaviruses. The CDC recently established a SARS-Cov-2-specific antibody test that does not exhibit cross-reactivity with antibodies related to common seasonal coronaviruses. Hopefully, this test will be soon broadly available.

The CDC advisory provides a case definition for MIS-C associated with COVID. Interestingly, no evidence of infection at the individual level is necessary to diagnose such cases as, theoretically, a patient could test negative by all available methods for SARS-CoV-2 and still qualify for the diagnosis just by having history of positive exposure to COVID.

Similar cases have now been reported from all over the world. The European Centre for Disease Prevention and Control reported on May 14th, that, based on established epidemiological principles, there was only limited evidence suggesting that the virus is causing the Kawasaki-like disease.

It concluded, "To date, an association between SARS-CoV-2 infection and this new clinical entity of multisystem inflammation has not yet been established, although an association appears plausible."

Based upon current data, it is clear the risk of COVID-19 infection in children is low and their probability of developing a severe illness, such as MIS-C, is very low. It remains unknown whether the association of severe illness such as Kawasaki-like disease even exists, because data from prospective controlled studies are needed in order to establish an association.

It is hard to collect reliable data when studying rare phenomena and, furthermore, COVID-19 testing is currently imperfect. More data is needed to advance our understanding and identify children who may be at risk, as well as the optimal treatment.

We will need to develop prospective representative groups of children to gain a deeper understanding of the epidemiological behavior of the virus - which will better guide our management of this disease in the future.