We describe 2 patients with coronavirus disease who had multiple clinical features suggestive of Kawasaki disease (KD). Both patients presented with fever lasting >5 days and were found to have rash, conjunctival injection, and swollen lips. One patient also had extremity swelling, whereas the other developed desquamation of the fingers. In both cases, laboratory results were similar to those seen in KD. These patients had highly unusual but similar features, and both appeared to respond favorably to treatment. It remains unclear whether these patients had true KD or manifestations of coronavirus disease that resembled KD.
As of May 18, 2020, nearly 1.5 million cases of coronavirus disease (COVID-19) have been confirmed in the United States.1 Although ∼2% of cases are among individuals aged <18 years, and the majority of those patients experience mild symptoms, our understanding of the wide spectrum of clinical patterns and spectrum of disease in this population remains limited.2,3 We report 2 cases of children who tested positive for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) on reverse transcription polymerase chain reaction (RT-PCR) but presented with a multisystem inflammatory syndrome that also may fulfill diagnostic criteria for Kawasaki disease (KD).
Patient I
An 11-year-old boy with a history of febrile seizures had been well until 5 days before his presentation, when fever, sore throat, rhinorrhea, dry cough, emesis, diarrhea, myalgia, bilateral conjunctivitis, and swelling of his hands and feet developed. Two days later, he developed a polymorphous rash on his torso, extremities, and face. The patient contacted his primary care physician, who recommended supportive care. Three days later, symptoms increased in severity, and the patient began to have confusion, severe headache, and neck stiffness, prompting presentation to the emergency department.
On admission, his temperature was 39.3°C, with a heart rate of 145 beats per minute and blood pressure of 78/42. He had a respiratory rate of 24 breaths per minute and oxygen saturation of 97% while he was breathing ambient air. On examination, he appeared unwell and in moderate distress with grunting. He had nuchal rigidity, bilateral bulbar conjunctival injection without exudate, and diffuse maculopapular rash (Fig 1). His lips were initially noted to be swollen and red, although this was not subsequently observed. He did not have cervical adenopathy, and extremity edema was not noted.
Pertinent laboratory results on presentation of the patient (patient 1) to the emergency department are summarized in Table 1. They included a normal white blood cell count with neutrophilic predominance and lymphopenia, hyponatremia, mild transaminitis and hypoalbuminemia, elevated inflammatory markers, and elevated N-terminal B-type natriuretic peptide (NT-ProBNP). Urinalysis and cerebrospinal fluid (CSF) cell counts, which were obtained because of his nuchal rigidity and severe headache, were unremarkable. Urine, blood, and CSF cultures were negative for infection. COVID-19 was diagnosed on the basis of RT-PCR for SARS-CoV-2. Electrocardiogram showed normal sinus rhythm with nonspecific ST- and T-wave changes. Chest radiography revealed patchy perihilar opacities in both lungs. He was started on antimicrobial therapy, including coverage for herpes simplex virus, toxic shock syndrome, and rickettsial disease, with vancomycin, ceftriaxone, acyclovir, clindamycin, and doxycycline. He was fluid resuscitated, and a dopamine infusion was added for hypotension. Supplemental oxygen was provided through a nasal cannula at a rate of 2 L per minute.
Demographic and Clinical Characteristics and Laboratory Findings
Characteristic . | Patient 1 . | Patient 2 . |
---|---|---|
Age, y | 11 | 7 |
Sex | Male | Female |
Medical history | Febrile seizures | None |
Kawasaki criteria | Fever, mucosal changes, conjunctivitis, rash, extremity swelling | Fever, mucosal changes, conjunctivitis, rash |
Imaging features | Patchy perihilar opacities | Patchy perihilar opacities |
Echocardiogram features | Low-normal function, moderate mitral regurgitation, small pericardial effusion | Low-normal function, mild mitral regurgitation |
Findings on presentation | ||
Days since disease onset | 5 | 5 |
Disease severity | Critical | Stable |
Laboratory findings | ||
White blood cell count, per mm3 | 8100 | 3990 (↓) |
Differential count, per mm3 | ||
Total neutrophils | 6642 (↑) | 2872 (↑) |
Total lymphocytes | 543 (↓) | 678 |
Total monocytes | 454 | 80 (↓) |
Platelet count, per mm3 | 116 000 (↓) | 69 000 (↓) |
Hemoglobin, g/L | 117 | 128 |
Albumin, g/L | 27 (↓) | 37 (↓) |
Alanine aminotransferase, U/L | 118.0 (↑) | 52.0 (↑) |
Aspartate aminotransferase, U/L | 84.0 (↑) | 47.0 (↑) |
Sodium, mmol/L | 126 (↓) | 129 (↓) |
Potassium, mmol/L | 3.6 | 3.9 |
Creatinine, mg/L | 80 | 74 |
Creatine kinase, U/L | 223.0 (↑) | 45.0 |
High-sensitivity troponin T, ng/L | 18.0 | 61.0 (↑) |
NT-ProBNP, pg/L | 14 587 000 (↑) | 6 985 000 (↑) |
High-sensitivity C-reactive protein, mg/L | 293.0 (↑) | 185.0 (↑) |
Sedimentation rate, mm/h | 76.0 (↑) | 56.0 (↑) |
Procalcitonin, ng/mL | 24.0 (↑) | N/A |
Ferritin, ng/mL | 1121.0 (↑) | 695.5 (↑) |
Urinalysis | Normal | Normal |
Urine culture | No growth | N/A |
Blood culture | No growth | No growth |
CSF culture | No growth | N/A |
Characteristic . | Patient 1 . | Patient 2 . |
---|---|---|
Age, y | 11 | 7 |
Sex | Male | Female |
Medical history | Febrile seizures | None |
Kawasaki criteria | Fever, mucosal changes, conjunctivitis, rash, extremity swelling | Fever, mucosal changes, conjunctivitis, rash |
Imaging features | Patchy perihilar opacities | Patchy perihilar opacities |
Echocardiogram features | Low-normal function, moderate mitral regurgitation, small pericardial effusion | Low-normal function, mild mitral regurgitation |
Findings on presentation | ||
Days since disease onset | 5 | 5 |
Disease severity | Critical | Stable |
Laboratory findings | ||
White blood cell count, per mm3 | 8100 | 3990 (↓) |
Differential count, per mm3 | ||
Total neutrophils | 6642 (↑) | 2872 (↑) |
Total lymphocytes | 543 (↓) | 678 |
Total monocytes | 454 | 80 (↓) |
Platelet count, per mm3 | 116 000 (↓) | 69 000 (↓) |
Hemoglobin, g/L | 117 | 128 |
Albumin, g/L | 27 (↓) | 37 (↓) |
Alanine aminotransferase, U/L | 118.0 (↑) | 52.0 (↑) |
Aspartate aminotransferase, U/L | 84.0 (↑) | 47.0 (↑) |
Sodium, mmol/L | 126 (↓) | 129 (↓) |
Potassium, mmol/L | 3.6 | 3.9 |
Creatinine, mg/L | 80 | 74 |
Creatine kinase, U/L | 223.0 (↑) | 45.0 |
High-sensitivity troponin T, ng/L | 18.0 | 61.0 (↑) |
NT-ProBNP, pg/L | 14 587 000 (↑) | 6 985 000 (↑) |
High-sensitivity C-reactive protein, mg/L | 293.0 (↑) | 185.0 (↑) |
Sedimentation rate, mm/h | 76.0 (↑) | 56.0 (↑) |
Procalcitonin, ng/mL | 24.0 (↑) | N/A |
Ferritin, ng/mL | 1121.0 (↑) | 695.5 (↑) |
Urinalysis | Normal | Normal |
Urine culture | No growth | N/A |
Blood culture | No growth | No growth |
CSF culture | No growth | N/A |
↑, increased; ↓, decreased; N/A, not applicable.
The patient continued to have fever and developed desquamation of his fingertips. He soon developed leukocytosis (white blood cell count, 19 440 per cubic millimeter) and continued to have thrombocytopenia. On hospital day 2, his high-sensitivity troponin T increased to 136 ng/L (normal value, <22), whereas NT-ProBNP rose to 35 137 000 pg/L (normal range, 10 000–242 000). He continued to have worsening hypotension and was transitioned from dopamine to high-dose epinephrine, norepinephrine, and vasopressin infusions. Cefepime was added for increased Gram-negative coverage, and he was given stress-dose hydrocortisone. A central venous line and an arterial line were placed. Although preparations were made to intubate for impending respiratory failure in the setting of shock, he improved, and his vasoactive infusions were weaned within hours. Transthoracic echocardiography revealed low-normal systolic function of the left ventricle and a shortening fraction of 29% (normal range, 25–45) on vasoactive support, with moderate mitral regurgitation and a small pericardial effusion. The right coronary artery was noted to be prominent with lack of tapering but had a z score of 1.4 (normal value, <2.0), whereas the proximal left coronary artery appeared normal; no aneurysms were evident.
Given the patient’s persistent fevers and mucocutaneous findings, his supportive laboratory data (leukocytosis with neutrophilic predominance, hyponatremia, hypoalbuminemia, elevated inflammatory markers) and his echocardiographic findings (mitral regurgitation and pericardial effusion), there was concern for possible KD. He was treated with intravenous immunoglobulin (IVIg) with a 5-day course of intravenous corticosteroids for cytokine storm resulting from COVID-19. The patient’s fever resolved within 24 hours of receiving IVIg, and his noncardiac symptoms have continued to improve. A repeat echocardiogram did not reveal any further coronary abnormalities.
Patent II
The second patient was a previously healthy 7-year-old girl who presented with 5 days of fever, sore throat, severe abdominal pain, emesis and diarrhea, and an erythema multiforme-like rash on the face, trunk, and back. Two days earlier, her parents took her to a local emergency department, where she tested positive for SARS-CoV-2 on RT-PCR. On examination, in addition to her rash (Fig 2), she was found to have conjunctival erythema, cracked lips, and strawberry tongue. Laboratory results included leukopenia, thrombocytopenia, hyponatremia, mild transaminitis and hypoalbuminemia, elevated troponin, and elevated inflammatory markers. Because of severe abdominal pain concerning for acute appendicitis, abdominal ultrasound and MRI were both obtained and demonstrated acute, aggressive colitis, and ileitis. Shortly after admission, the patient developed agitation and confusion concerning for encephalopathy. On telemetry, she was noted to have frequent premature ventricular contractions. Transthoracic echocardiography revealed low-normal systolic function of the left ventricle and a shortening fraction of 27%, mild mitral regurgitation, and normal coronary arteries. In the context of the previous patient with hyperinflammatory syndrome and evidence of significant multisystem involvement in the setting of COVID-19, treatment with intravenous corticosteroids was initiated, with prompt fever defervescence and improvement in her clinical symptoms. Although the diagnosis of KD was considered, IVIg was not given, because she did not meet diagnostic criteria for incomplete KD, and the consensus at that time was that her symptoms were likely due to COVID-19. After a week course of oral glucocorticoids was completed, she had recurrence of rash and, at present, is once again on glucocorticoids.
Discussion
Despite the ongoing global spread of COVID-19, the broad spectrum of clinical presentations and outcomes associated with the virus remains poorly understood, particularly in the pediatric age group.4,5 Most children appear to experience predictable and milder symptoms, such as diarrhea, fever, and upper respiratory symptoms, and recover within 1 to 2 weeks.6–9 However, uncommon clinical presentations have also been reported.2,4,6 For instance, investigators have recently described cases of patients with confirmed or highly suspected COVID-19 who presented with clinical features of KD.10,11
KD is an acute, systemic inflammatory vasculitis that is predominantly diagnosed in children <5 years of age. Affecting medium-sized arteries, with a predilection for coronary arteries, it is a leading cause of acquired heart disease in the United States.12 The etiology of the disease remains elusive; however, clinical, immunologic, and epidemiological evidence suggests an infectious trigger.13 Although an association of KD with non–COVID-19 subtypes of coronavirus has been suggested, causality has not been established, and this association was contradicted by subsequent investigations.14–17 Although there is no specific test to diagnose KD, the diagnosis is made on the basis of the presence of a constellation of signs and symptoms.12
These patients presented with ≥5 days of fever and were found to have mucosal changes, rash, and conjunctival injection, which are diagnostic criteria for KD (Supplemental Fig 3). They also both had gastrointestinal involvement, which has occasionally been described in patients with KD.18 One patient also had extremity swelling, whereas the other developed desquamation of the fingers. In both cases, the patients exhibited laboratory results consistent with KD, such as hyponatremia, hypoalbuminemia, transaminitis, and elevated inflammatory markers. Given these findings, concern for KD was reasonable. However, some of the patients’ laboratory and clinical findings were not characteristic of KD. Notably, our first patient presented with severe headache, confusion, and nuchal rigidity, prompting a lumbar puncture for CSF fluid collection. This is atypical for patients with KD. The second patient had acute and severe ileitis and colitis and altered mental status. These findings are also not classically reported in KD. Although one patient developed leukocytosis, the other presented with leukopenia, which is uncharacteristic of the disease.12 Both patients had thrombocytopenia throughout their hospital courses, which is rare beyond the acute phase of KD and could suggest increased disease severity (including shock) and higher risk of coronary artery aneurysms, or which could suggest an alternative diagnosis.12,19–22 Additionally, thrombocytopenia is associated with acute SARS-CoV-2 infection and inflammatory disease.23 Although NT-ProBNP can be elevated in patients with KD, particularly those who are resistant to IVIg, both patients had elevations outside of the normal range typically observed with the disease.24 Moreover, neither patient had sterile pyuria, which is a common finding in patients with KD.12
Our patients were older than the majority of patients diagnosed with KD. Studies have shown that children >5 years of age are more likely to have delayed diagnosis and to develop coronary artery aneurysms, which are the major cause of morbidity and mortality in KD.25,26 Both patients had low-normal left ventricle contractility and valvar regurgitation. These findings were consistent with myocarditis, which histologic evidence suggests is universal in KD.12 The presence of a pericardial effusion in one of our patients also indicated pericarditis resulting from his disease. Neither patient’s echocardiogram revealed coronary artery ectasia or aneurysm. Although these findings may be absent at presentation, they can be observed within the first 2 to 3 weeks after fever onset in KD, after which their development is unlikely.27 It is also worth noting that ∼25% of patients with KD develop coronary artery abnormalities.28,29
Selecting the most appropriate treatment strategies for these patients was not straightforward. For the first patient, initial primary focus was to manage his shock syndrome, which is observed in 7% of patients with KD and can be misdiagnosed as bacterial sepsis or toxic shock syndrome.30,31 Given his constellation of symptoms consistent with KD and the severity of his illness, he was treated with IVIg and corticosteroids. This treatment led to immediate resolution of his fever and improvement of his clinical status. Notably, IVIg has been administered with success for acute myocarditis during the acute pandemic period of COVID-19 in China.32 Our second patient was given a course of corticosteroids that similarly led to significant improvement in her symptoms, although she later had recurrence of her rash, which could be due to immunologic phenomena related to SARS-CoV-2 infection or could suggest persistent inflammation from an acute inflammatory illness. Importantly, this patient was not noted to have any coronary artery abnormality on her follow-up echocardiogram 2 weeks after discharge.
It remains unclear if these patients had true KD with coincidental SARS-CoV-2 infection, SARS-CoV-2 infection triggering KD, or an independent multisystem inflammatory syndrome related to SARS-CoV-2 infection. It is possible that indeed these patients demonstrate a novel infectious or postinfectious syndrome of COVID-19 with a similar underlying mechanism of disease to KD, especially given the lack of pulmonary symptoms during disease development. As a result, their risk of developing coronary artery aneurysms remains unknown. Thus, it is important to consider the potential for cardiac sequelae in these patients, and to manage them accordingly. At the same time, recognizing differences in clinical and laboratory presentation will allow us to quickly and appropriately treat patients. Because patients with COVID-19 have a wide variety of clinical presentations overlapping with multiple disease processes, including KD and toxic shock syndrome, treatment must be tailored to patients on an individual level on the basis of discussions between the infectious disease, rheumatology, cardiology, and, if applicable, critical care teams.
Finally, it is also noteworthy to recall that patients with KD can develop a Kawasaki shock syndrome.30 The existence of this syndrome, and the overlap of cytokine expression in KD shock syndrome with COVID-19 and inflammatory response, could add to our understanding of both KD and COVID-19 inflammatory features on a more fundamental, mechanistic level.33,34
Conclusions
We describe 2 young patients with COVID-19 who had features suggestive of KD. Our patients had highly unusual but similar features, and both appeared to respond to treatment. We expect there are more patients with both classic KD features and coincident COVID-19 as well as a potential novel COVID-19 inflammatory syndrome that resembles KD. Better characterization of these 2 syndromes will be important for formulating appropriate diagnostic and therapeutic guidelines. Further research is needed to better understand this unique clinical presentation of SARS-CoV-2 infection.
Acknowledgments
We are grateful to the patients and their families for agreeing to be included in this article; to Drs Joshua Milner, Eva Cheung, and Kara Margolis for their contributions to the care of these patients and for their review of this article; as well as to Dr Julie Glickstein for her limitless energy and commitment to her pediatric cardiology fellows.
Dr Spencer was the consulting cardiology fellow who evaluated 1 of the 2 patients, and he was the primary author of this case report; Dr Closson was the consulting cardiology fellow who evaluated the other patient, and he contributed references and revisions to the manuscript; Dr Gorelik was the consulting rheumatology attending who helped manage one of these patients, and he contributed references and revisions to the manuscript; Dr Boneparth was the consulting rheumatology attending who helped manage these patients, and he contributed references and revisions to the manuscript; Dr Hough was the critical care attending who primarily managed these patients in the ICU, and she contributed references and revisions to the manuscript; Dr Acker was the consulting infectious disease attending who helped manage these patients, and she contributed revisions to the manuscript; Dr Krishnan was the precepting attending for Dr Spencer and senior author, and she conceptualized and contributed revisions to the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
References
Competing Interests
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
Comments