The ongoing COVID-19 pandemic has marched its way to almost all parts of the world, with hundreds of thousands of deaths and over 4.5 million cases as of May 15, 2020. One saving grace has been the apparent sparing of children, though the mortality in older people is quite high.
A new paper published on the preprint server medRxiv* in May 2020 reports the occurrence of a vasculitis syndrome called Kawasaki disease (KD) in children without any other symptoms of COVID-19, but who test positive for the virus. In children and adolescents, COVID-19 causes typically mild respiratory symptoms, unlike severe pneumonia and dyspnea seen in adults.
KD is the most frequently seen vasculitic condition of childhood. It is an inflammatory disorder that affects mostly the small and medium-sized arteries, without any apparent underlying reason. It is most common in Japan, where more than 300/100,000 children above the age of 4 years may be affected by it.
In contrast, the incidence in children aged 5 years and above in North America is less than a tenth of this, at 25/100,000. KD can cause a much-feared complication called a coronary aneurysm, where one of the arteries supplying the heart develops a ballooned-out weak spot as the result of inflammation.
Another severe complication is Kawasaki Disease Shock Syndrome (KDSS), often occurring with myocarditis and requiring life-support interventions in the acute stage.
What causes KD? It is not yet clear. Some believe it is the result of a reaction to a viral infection in children whose genes predispose them to respond to such infection with arterial inflammation. Some of the viruses that have been linked to KD include seasonal coronavirus.
A recent paper reports nine children with a clinical profile suggestive of KD and who tested positive for COVID-19, in the US and England. Of these, eight had signs of hyperactive inflammation and multiple organ damage, as is seen in KDSS.
The link with COVID-19 cannot automatically be taken to imply causality, because different parts of Europe show widely varying prevalence rates for the infection with SARS-CoV-2. The current study seeks to assess the possibility that the two are linked with respect to time, in a case series of 17 patients with KD. They were admitted to the pediatrics department of a Paris university hospital between April 27 and May 7, 2020.
The study showed that the admission of 17 children with a diagnosis of KD over the period between April 27 and May 7 exceeded the usual rate of 1 patient per fortnight, by 13-fold on average. The number of males to females bore a ratio of 0.7, and the median age was 7.5 years.
Interestingly, about 60% of the children had African or Caribbean ancestry on at least one side of the family, while 12% were from Asia. About 82% were under the 97th centile for body mass index.
About half the children had incomplete KD, and the rest had incomplete KD. The most frequent criteria for the diagnosis of KD included a skin rash and redness of the eyes, in 3 of 4 patients. The skin rash had a variable appearance in different patients.
All patients had gastrointestinal symptoms that occurred quite early in the illness, followed by the primary manifestations of KD. The chief gut symptom was acute abdominal pain, often with nausea and diarrhea.
Three patients had peritoneal effusion, and two had acute abdominal pain, which led to immediate surgery, one for suspected appendicitis which turned out to be aseptic peritonitis.
In about two-thirds of patients, irritability was present, with a third manifesting headache, meningeal signs, and confusion. Lumbar puncture was carried out in three cases.
Pericardial effusions were present in almost half the patients and pleural effusion in about a fifth. About 70% had myocarditis with a low ventricular ejection fraction (10% to 57%). The ECG showed significant changes in two patients.
What Were the Imaging and Laboratory Findings?
Thirteen patients had X-rays or CTs of the chest. In five patients, the lungs showed ground-glass opacities, patchy shadowing in certain areas, and other changes. In half the cases, coronary arteries abnormalities were seen on echocardiography carried out after 5-11 days (median 7.5 days).
A third of the arteries showed dilation while three patients had thickened coronaries.
Inflammatory markers were increased in all the patients, chiefly neutrophil-predominant leukocytosis, C-reactive protein (CRP), procalcitonin (PCT) and IL-6. The median hemoglobin level was 8.1 g/dL, indicating a high prevalence of anemia. Hyponatremia and hypoalbuminemia were almost universal. Half the patients had transient renal failure.
Liver enzymes ALT and GGT showed moderate elevations in over 50% and 75% of patients, with the latter taking several days to rise. D-dimers were high in almost all patients, while troponin and B-type natriuretic peptide (BNP) was high in over 75% and 65% of patients.
How Were the Patients Managed?
All patients received high-dose intravenous immunoglobulins (IVIG) after five days (median value) of fever, along with low-dose aspirin. Almost one in three were resistant to IVIG (defined as “persistent or recrudescent fever between 36 hours and seven days after completion of first IVIG infusion.”), and these patients were given a second dose. Six patients in all received corticoids to help control the inflammation. Positive bacterial cultures in 82% of the patients required broad-spectrum antibiotics until they became negative.
Over three out of four patients needed intensive care unit admission because of cardiovascular weakening. KDSS was diagnosed in 65%, or 11 patients, who needed IV fluid resuscitation, with vasoactive drugs in 7 cases.
Myocarditis requiring inotropic support was required in 8 patients, while cardiovascular near-collapse in 9 patients required respiratory support. Systemic inflammation markers in the patients admitted in the ICU were higher.
The median duration of ICU stay was 5.5 days. Concerning the outcome, 14 of the 17 were discharged by the end of the study, with a median hospitalization duration of 8 days.
Were the Children Infected with SARS-CoV-2?
France closed all schools along with a general lockdown on March 17, 2020, as the pandemic worsened. The 17 children in the current study were at home from that date onwards. Six of them had brief bouts of fever (less than 48 hours), cough or cold, and one had anosmia.
The median interval between these symptoms and KD symptoms was 42 days, ranging from 18 to 79 days. In 9 families, the child had contact with someone in the family who had similar cold-like symptoms, including parents or grandparents, and siblings, from 24 to 38 days before KD symptoms developed.
Five of the contacts in the family were highly suspicious for COVID-19 due to symptoms like ageusia, anosmia, and chest CT findings suggestive of this illness. One tested positive for COVID-19.
Among the 17 patients, seven tested positive for the virus by RT-PCR, though only one had any symptoms of the illness – anosmia, which started 24 hours before the KD symptoms began. IgG antibodies were found in almost 90% of the patients, however. The two patients with negative IgG testing also had negative RT-PCR tests.
Multiplex PCR testing for a variety of respiratory viruses was negative in all 16 patients tested.
The timeline of KD symptoms suggests that they occurred as part of a post-viral immunological reaction. Earlier studies have shown that KD could be linked with viral infections of the respiratory tract, in particular the rhinoviruses, enteroviruses, and human coronaviruses. No study has yet shown any difference in the way KD manifests in infected vs. uninfected patients. However, the current studies showed a more significant number of incomplete KD cases, with many patients showing myocarditis and KDSS.
Severe myocarditis is uncommon in KD; most patients showing only mild myocarditis, which improves rapidly with the resolution of inflammation. It mostly occurs along with KDSS, which is itself extremely uncommon, affecting 1.5% to 7% of patients. It is thought to be due to the high level of circulating inflammatory factors.
In the present study, the PCT levels were much higher than those described in another recent study of 27 KDSS patients – eightfold greater – along with high CRP and IL-6. The researchers remark, “This major proinflammatory state may reflect a particularly strong post-viral immunological reaction to SARS-CoV-2 as compared with other viral agents.” This may also be linked to the cytokine storm in adult patients that is often the precursor to death, and which is due to high inflammatory markers like IL-6.
Other clinical and demographic characteristics also agree with the KDSS diagnosis, like the older age, anemia, high D-dimer level, and low albumin and sodium.
Most patients were above the 75th percentile for weight, and all presented with gut symptoms, which is unusual. These are thought to be caused by vasculitis of the bowel vessels, causing ischemia of the bowel wall. In support of this theory, IVIG rapidly relieved these symptoms.
The high lipase shows pancreatitis, perhaps due to vasculitis. The marked decrease in albumin may have been caused by exudative enteropathy.
The link to African and Caribbean ancestry may reflect poverty or genetic predisposition. The higher rate of COVID-19 among Black Americans could also reflect a higher vulnerability to severe infection with the virus.
The implications include a possible increase in KD among African children in countries with a high number of COVID-19 cases, leading to scarcity of IVIG. However, as the researchers note, “The absence of reported cases of KD associated with SARS-CoV2 infection in Asian countries where the SARS-CoV-2 epidemic started, and where the incidence of KD is the highest, is quite remarkable.” However, KDSS is known to be far less common in Asia compared to Western countries, though the underlying mechanisms are unclear.
The variations in the KD phenotype reported in the current study “should prompt a high degree of vigilance among primary care and emergency physicians, and preparedness in countries with a high proportion of children of African ancestry, while facing SARS-CoV-2 epidemic.”
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.