Kawasaki Disease Workup

Updated: Aug 14, 2017
  • Author: Noah S Scheinfeld, JD, MD, FAAD; Chief Editor: Russell W Steele, MD  more...
  • Print

Approach Considerations

No specific laboratory test is used to diagnose Kawasaki disease; however, certain abnormalities coincide with various stages. Acute-phase reactants (ie, erythrocyte sedimentation rate [ESR], C-reactive protein levels [CRP], and alpha1-antitrypsin levels) are almost universally elevated at first; they usually return to baseline 6-10 weeks after the onset of the illness.

More recently, 2 urine proteins hold promise as biomarkers of Kawasaki disease: meprin A or filamin C. These 2 biomarkers were diagnostically superior to ESR or CRP. [6, 7] Investigators identified more than 190 proteins that were present only in children with Kawasaki disease, including the proteins associated with endothelial and myocardial cell injury (filamin C) and immune regulators (meprin A). [6, 7]

Elevated macrophage migration factor (MIF) and Interleukin-6 (IL-6) may be useful markers in the acute stages of Kawasaki disease. The serum complement level is normal or elevated.

On complete blood counts (CBCs), mild-to-moderate normochromic anemia is observed in the acute stage. The white blood cell count (WBC) is moderate to high (50% of patients have a WBC greater than 15,000/µL), with a left shift, which is a predominant sign of immature and mature granulocytes.

During the subacute stage, thrombocytosis is the outstanding marker. The platelet count begins to rise in the second week and continues to rise during the third week. Platelet counts average 700,000/μL, but levels as high as 2 million have been observed. Thrombocytopenia is associated with severe coronary artery disease and myocardial infarction (MI); rarely, it may be associated with disseminated intravascular coagulation.

Serum cholesterol, high-density lipoprotein, and apolipoprotein A levels are decreased; these values tend to persist beyond clinical resolution of the disease. Hypoalbuminemia may be present and is often associated with more severe and prolonged illness.

In the convalescent stage, the levels of platelets and other markers begin to return to values within the reference range. Laboratory values may require 6-8 weeks to normalize.

Normal results on some studies can help narrow the differential diagnosis. levels of antineutrophil cytoplasmic antibodies, antiendothelial cell antibodies, antinuclear antibody, and rheumatoid factors are all within the reference range. Culture results are all negative. On rapid antigen testing, results for adenovirus are negative. Urinalysis may show mild-to-moderate sterile pyuria of urethral origin and proteinuria.

Cardiac imaging, electrocardiography, and cardiac biomarkers

Echocardiography is the study of choice to evaluate for coronary artery aneurysms (CAAs), in both fully manifested and suspected incomplete cases of Kawasaki disease. Serial echocardiograms should be obtained, preferably at the time of Kawasaki disease diagnosis, at 2 weeks, and at 6-8 weeks after the onset of the illness. These may need to be performed more frequently in high-risk patients. [66, 71, 8]

Magnetic resonance imaging (MRI), magnetic resonance angiography (MRA), and ultrafast computed tomography (CT) scanning are other noninvasive tests that can be used to evaluate coronary artery abnormalities. However, larger studies are required to evaluate their reliability.

On electrocardiography (ECG), tachycardia, prolonged PR interval, ST-T wave changes, and decreased voltage of R waves may indicate myocarditis. Q waves or ST-T wave changes may indicate myocardial infarction. Cardiac enzyme levels (eg, creatine kinase [CK], creatine kinase myocardial band [CK-MB], cardiac troponin, lactate dehydrogenase [LD-1 >LD-2]) are elevated during a myocardial infarction.

A select group of patients may require cardiac catheterization and angiography. Coronary CT angiography and magnetic resonance angiography may also prove beneficial in the evaluation and follow-up of the coronary arteries. [72, 73] Cardiac angiography provides a more detailed study of the arteries, but it is associated with greater risks of rupture, especially when performed in the acute phase of the illness; it should be limited to select cases.

Studies for selected patients

Liver function studies and serum lipase measurement may be indicated in selected cases. On liver function testing, serum transaminase values are mildly elevated in 40% of affected patients. Elevated alanine aminotransferase (ALT) levels can indicate a more serious course. Bilirubin values are elevated in 10% of affected patients.

A chest radiograph should be obtained to assess baseline findings, exclude cardiomegaly or subclinical pneumonitis, or confirm clinical suspicion of chronic heart failure.

Gallbladder ultrasonography may be necessary if liver or gallbladder dysfunction is suspected. Acute distention of the gallbladder (hydrops) is identified on abdominal ultrasonography in 15% of patients. Gallbladder hydrops usually resolves without surgical intervention.

Obtain a scrotal sonogram in males to evaluate for epididymitis. Although epididymitis is generally an inflammatory process that affects boys aged 9-14 years, it can be observed in younger boys with Henoch-Schönlein purpura and Kawasaki disease.

Arthrocentesis may be indicated in patients with joint involvement. Joint fluid analysis in affected patients typically shows numerous white blood cells, ranging from 125,000-300,000/µL, with normal glucose levels and negative culture results.

Lumbar puncture may be indicated in patients with clinical signs suggesting meningitis. In children who undergo lumbar puncture, 50% show evidence of aseptic meningitis with a predominance of mononuclear cells, along with normal glucose and protein levels.


Diagnosis of Incomplete Kawasaki Disease

In some cases, patients have many of the typical clinical features of Kawasaki disease but not as many as are required to meet standard diagnostic criteria. Hence, the term "incomplete" rather than "atypical" is used to describe these cases. Incomplete cases usually occur in children younger than 6 months. In this setting, fever plus only 3 features can establish the diagnosis. The rationale is that treatment is safe and effective and that failure to diagnose Kawasaki disease may have a significant negative impact on outcome.

For the diagnosis of incomplete Kawasaki disease, the American Academy of Pediatrics (AAP)/American Heart Association (AHA) recommend that when fever plus 2 or 3 of the typical features are present for 5 days or more and when patient characteristics suggest possible Kawasaki disease, a C-reactive protein (CRP) level and erythrocyte sedimentation rate (ESR) should be obtained. [66] If the CRP level is less than 3 mg/dL and the ESR is more than 40 mm/hr, the child is monitored and actions taken as appropriate.

If the CRP is 3 mg/dL or higher and the ESR is 40 mm/hr or more, the next step is to measure albumin, alanine aminotransferase (ALT), platelets, and WBC count and test the urine for pyuria. Abnormal limits include the following:

  • Albumin < 3 g
  • Anemia for age
  • Elevated ALT level
  • Platelets >450,000 (after 7 d)
  • WBC count >12,000
  • Presence of pyuria

If 3 or more supplemental laboratory criteria are positive, a diagnosis of Kawasaki disease is made. The child should have an echocardiogram and be treated.

If fewer than 3 supplemental laboratory criteria are positive, a cardiac echocardiogram should be performed. If the echocardiogram is negative but fever persists, a repeat echocardiogram may be performed. If the echocardiogram is negative and the fever abates, Kawasaki disease is unlikely. If the echocardiogram is positive, the child is treated for Kawasaki disease.

Newberger questioned the suitability of laboratory findings as evidence of inflammation, noting that in some cases, the clinical criteria are not all present on any given day. Conversely, some patients with an inflammatory disorder did not meet the clinical case definition but developed coronary artery abnormalities consistent with Kawasaki disease. [74]

The results of one study noted that patients with incomplete Kawasaki disease realized a longer median interval from symptom onset to diagnosis and were less likely to be treated with IV immunoglobulin than those patients with complete Kawasaki disease. However, no significant differences between the two groups were noted in demographics, clinical and laboratory characteristics, and coronary artery abnormalities. [75]

A French group has suggested adding another diagnostic category, "uncertain Kawasaki disease”, for children with 5 days of fever, fewer than 4 classic signs, normal echocardiographic findings, and an inflammatory syndrome that does not meet AHA criteria. These researchers found that children in this category did well when treated with IVIG and aspirin. [76]

Hinze et al reported a case of Kawasaki disease in a 3-month-old boy manifested by typical signs and CAAs but without fever. They commented on the difficulty in making the diagnosis in young infants. [77] Case reports of other unusual presentations (eg, GI bleeding, lupuslike illness in a recurrent case, arthritis, rhabdomyolysis) have been published. Such presentations appear to be very uncommon. [33]



Echocardiography is the study of choice to evaluate for CAAs during the acute stage. In order of highest to lowest frequency, the involvement of the coronary arteries is as follows:

  1. Proximal left anterior descending and right coronary artery
  2. Left main coronary artery
  3. Left circumflex artery
  4. Distal right coronary artery
  5. Posterior descending artery

In addition to evaluating the coronary arteries for dilation and thrombosis, the baseline echocardiogram is also performed to evaluate for other cardiac involvement. This includes aortic root dilation, depressed contractility, ventricular and valvular function, and pericardial effusion.

Diffuse dilatation of coronary lumina can be observed in 50% of patients by the 10th day of illness. In children, ensure that pediatric cardiologists perform this study because they are familiar with coronary artery diameters. Coronary artery dimensions must be adjusted for body surface area to accurately identify dilation. A basic rule is that if the internal diameter of a segment is greater than 1.5 times that of an adjacent segment, then dilation probably exists.

The echocardiogram should be repeated in the second or third week and again 1 month after all other laboratory results have normalized. Echocardiograms may need to be performed more frequently in high-risk patients. [66, 71, 8] If by 8 weeks the patient exhibits no coronary involvement, follow-up echocardiography is recommended at 1 year. If the echocardiographic findings are abnormal at any point, refer the child to a pediatric cardiologist for a complete cardiac workup and follow-up care.

In a study that examined echocardiograms obtained at diagnosis and 1 and 5 weeks after diagnosis from 198 patients, Printz et al concluded that noncoronary cardiac abnormalities were associated with coronary artery dilation and laboratory evidence of inflammation within the first 5 weeks after the diagnosis of Kawasaki disease. [78] Left ventricular (LV) systolic dysfunction was noted in 20% of patients and coronary artery dilation in 29%. Mitral regurgitation was present in 27% of patients and aortic root dilation in 8%.

One group of authors recommended that the initial echocardiogram be performed in the emergency department. The same group identified a pulmonary artery aneurysm in one case and noted that this had never been found before. [76]


MRA and CT

Free-breathing 3-dimensional (3D) coronary MRA accurately defines CAA in patients with Kawasaki disease. This technique may provide a noninvasive alternative when the image quality of transthoracic echocardiography is insufficient, thereby reducing the need for serial radiographic coronary angiography in this patient group.

In a small study that used multislice spiral CT in 16 adolescents and young adults with Kawasaki disease, CT was 100% sensitive in the detection of coronary artery aneurysms but only 87.5% sensitive for the detection of significant stenosis or occlusion. False-positive results occurred secondary to severe calcification in 5 arteries and cardiac motion artifact in 2. Specificity was therefore 92.5%. [79]

In another small study, electron beam CT (EBCT) was used to determine whether coronary artery calcifications could be used as a marker of future coronary artery events. The authors felt that EBCT may be useful for risk stratification in long-term management of patients with Kawasaki disease. [80]


Histologic Findings

Biopsy is rarely performed to make the diagnosis; therefore, most specimens are obtained from autopsies or from patients who have had diseased arterial segments removed during bypass operations. Early findings show acute destruction of the media of the vessels by neutrophils, with loss of elastic fibers. Later, the infiltrate is replaced by lymphocytes, monocytes, and fibroblasts involved in arterial remodeling. Chronic lesions show intimal proliferation, neoangiogenesis, and vascular occlusion.

Weedon summarized the reported findings of Kawasaki disease as follows [81] :

  • Features are nonspecific
  • Lymphocytes and mononuclear cells manifest in a perivascular fashion
  • Edema of the papillary dermis may be present
  • Pustules with small intraepidermal and subcorneal abscesses unrelated to eccrine ducts may be present
  • One report noted subtle vascular alterations, subendothelial edema, focal endothelial cell necrosis, and vascular deposition of minute quantities of fibrinoid material

More definite reports have noted the internal vascular changes of Kawasaki disease. Destruction of vascular layers and infiltration of inflammatory cells in blood vessels are observed.

Upon ultrastructural examination, myocardial changes reveal hypertrophy, various degrees of degeneration, proliferation and abnormality of mitochondria, infiltration of a small number of lymphocytes, and fibrosis. Coronary microvascular lesions are characterized by the following:

  • Microvascular dilatation
  • Endothelial cell injury
  • Platelet aggregation with thrombosis
  • Stenotic lumen with thickened walls in the small arterioles

Laboratory Studies

 Elevated blood triglycerides and hypercoagulation characterize Kawasaki disease. [82] .  Soluble LR11 is a new biomarker helps define vascular lesions following Kawasaki disease. [83]