Summary

Kawasaki disease (KD) has replaced acute rheumatic fever as the most common cause of acquired heart disease in children in developed nations.[1] KD can lead to coronary artery aneurysms in about 25% of untreated children,[2] and can result in sudden death from coronary aneurysm rupture or thrombosis leading to myocardial infarction. The existence of effective therapy for a disease with potentially fatal consequences makes accurate diagnosis critical. In this review, we highlight aspects of the diagnosis, epidemiology, etiology, therapy, and cardiology of KD that are of particular interest to the primary care pediatrician.

Educational objectives

At the conclusion of this activity, participants will be able to:

• Evaluate febrile children for a possible diagnosis of acute Kawasaki disease
• Recognize laboratory findings compatible with a diagnosis of acute Kawasaki disease
• List the cardiovascular manifestations of Kawasaki disease sequelae

CME credit

This is an article from The Child's Doctor, Fall 2004 issue. You may take the quiz for learning purposes, but credits are no longer valid.

Author disclosures

Dr. Rowley has no industry relationship to disclose and does not refer to products that are still investigational or not labeled for the use in discussion.

Dr. Shulman has no industry relationship to disclose and does not refer to products that are still investigational or not labeled for the use in discussion.

Printable
version

Currently, no diagnostic test exists for KD, and diagnosis is based upon the presence of at least 5 of 6 classic diagnostic criteria developed by Tomisaku Kawasaki, MD, a general pediatrician in Tokyo who first described the clinical symptoms of this disorder.[3] Unfortunately, the clinical features of KD may mimic many infectious and non-infectious disorders. Moreover, the recognition that incomplete or atypical cases occur in patients who manifest fever and fewer than 4 other criteria, but nonetheless develop serious coronary disease, poses a significant diagnostic dilemma.

Classic Kawasaki disease

All physicians who evaluate children with fever should be familiar with the clinical features of KD:

1. Fever, which is generally high-spiking and remittent

2. Bulbar conjunctival injection, generally without exudates

3. Redness, dryness, and cracking of the lips, with redness of the throat and mouth, and "strawberry tongue"

4. Erythema of the palms and soles, and swelling of the hands and feet

5. Maculopapular, scarlatiniform, or erythema multiforme rash, primarily truncal and often with perineal accentuation

6. Cervical adenopathy

Classic diagnostic criteria for KD require the presence of fever and at least 4 of the other 5 clinical signs. Cervical adenopathy is the least common diagnostic finding among KD patients, but when it is present, it may be striking, and may be misdiagnosed as bacterial lymphadenitis, directing attention away from the other diagnostic features of KD.

Laboratory evaluation in KD patients reveals an elevated peripheral white blood cell count or a normal white blood cell count with a left shift, and very elevated acute phase reactants, such as the sedimentation rate or C-reactive protein level. Laboratory features that are associated with an increased risk of coronary disease include thrombocytopenia, anemia, and hypoalbuminemia. It is important to remember that the more common feature of thrombocytosis does not occur until the second to third week after onset and is therefore not useful in establishing a diagnosis of acute KD.

Incomplete (atypical) Kawasaki disease

Around the world, physicians have reported their experiences with the so-called "incomplete" or "atypical" KD, or children with fever and fewer than 4 of the other diagnostic criteria for KD, who developed coronary artery abnormalities following the illness.[4] Incomplete KD is the preferred term because these children do not present with features that are not normally seen in KD.

Pathologic findings in children who die from classic and incomplete forms of KD are identical, and it is generally accepted that classic and incomplete KD represent the same disease process. Laboratory features of KD in classic and incomplete cases are similar, and supportive laboratory evidence as discussed above, such as an elevated sedimentation rate and a left shift on the complete blood count, is often sought when considering a diagnosis of incomplete KD.

Differential diagnosis

Several infectious and non-infectious diseases may mimic acute KD. Scarlet fever, toxic shock syndrome, measles, drug hypersensitivity reactions, including Stevens Johnson syndrome, and juvenile rheumatoid arthritis (JRA) are illnesses that can sometimes be confused with KD. In toxic shock syndrome, hypotension and an elevated creatine phosphokinase (CPK) level are common, while both of these findings are uncommon in KD. Drug reactions are generally not associated with a markedly elevated sedimentation rate. Patients with JRA may appear to respond to therapy for KD, but generally worsen when high dose aspirin is reduced to low dose treatment. In uncomplicated measles, the white blood cell count and sedimentation rate are usually low, and a measles IgM titer can confirm the diagnosis.

Particularly confusing is the diagnosis of KD in a group A streptococcal (GAS) carrier. Failure to respond within 24 to 48 hours of appropriate antibiotic therapy generally indicates that acute GAS infection is unlikely and that the diagnosis of KD should be reconsidered.

Which patients are at risk of developing Kawasaki disease?

KD is predominantly an illness of young children, with 80% of patients presenting between the ages of 6 months to 5 years.[5] Males are affected more commonly than females.

All racial and ethnic groups are affected, although Asians appear to be at highest risk. Asian children in the U.S. have as high a risk of developing KD as Asian children in Japan, or about 1 in 200 by age 5.[5] However, in the U.S., most KD cases are diagnosed in Caucasian, African American, and Hispanic children because these groups make up most of the population. Therefore, physicians must be alert for clinical features of KD in all febrile children. Asian and non-Asian children have the same risk of developing coronary artery disease following KD.

Age is a stronger predictor of risk than ethnicity; a 12-month-old Caucasian child is at greater risk than a 14-year-old Japanese child. However, KD occasionally occurs in younger and older age groups, and at these ages the diagnosis may be missed or delayed, sometimes leading to severe coronary disease.

Epidemics of illness have been reported in a number of countries throughout the world over the last 30 years, but no recent epidemic has occurred in Japan or other nations.

Etiology

The etiology of KD remains unknown, but clinical and epidemiologic findings strongly support an infectious cause. An attractive hypothesis is that KD is caused by a ubiquitous infectious agent that generally results in asymptomatic infection, but that disease develops in a small subset of genetically predisposed individuals. This hypothesis predicts that the youngest infants are protected by passive maternal antibody and that adults have widespread immunity.

In our research into etiology of KD, we examined the immune response in the arterial wall in acute fatal cases of KD and have made the following significant discoveries:

 • CD8 (cytotoxic/suppressor) T lymphocytes predominate in the arterial wall in acute KD, suggesting the presence of an intracellular pathogen (such as a virus). [6]

 • IgA plasma cells are prevalent in the arterial wall in acute KD, suggesting an immune response to a pathogen with a mucosal portal of entry. [7]

 • IgA plasma cells are increased in the upper respiratory tract in acute KD, in a pattern similar to that seen in patients who died of viral respiratory illnesses, indicating a potential respiratory portal of entry of the etiologic agent(s) of KD. [8]

 • IgA in the arterial wall in acute KD is oligoclonal, indicating that it is antigen-driven. [9]

 • Synthetic antibodies corresponding to the prevalent IgA antibodies in the KD arterial wall bind to ciliated bronchial epithelium and to a subset of macrophages in inflamed acute KD tissues. [10]

Taken together, these studies indicate that KD is likely caused by a single intracellular respiratory infectious agent, probably a virus, which infects bronchial epithelium and macrophages, and travels to coronary artery and other tissues in macrophages via the bloodstream. The development of synthetic KD antibodies provides tools for the identification of these antigens, which is critical to the development of a diagnostic test, improved therapy, and ultimate prevention of KD.

Therapy

It is remarkable that effective therapy exists for KD, despite the elusive nature of the etiologic agent. Children's Memorial researchers participated in multicenter studies in the U.S. and Japan, which confirmed that intravenous gamma globulin (IVGG) at 2 g/kg as a single dose and aspirin at 80-100 mg/kg/day in 4 divided doses, when given within the first 10 days of onset of fever, reduce the risk of coronary artery disease from 25% in untreated patients to 4% in those who received the therapy. [11] KD patients who receive IVGG with aspirin generally experience rapid defervescence and resolution of clinical symptoms. However, up to 10% of KD patients do not respond to this therapy and manifest continued fever and other clinical symptoms. [12]

Prolonged fever is a strong indicator of increased risk of coronary disease in KD. In these patients, a second dose of IVGG is generally recommended and often results in defervescence. However, a small number of patients do not respond to a second dose of IVGG, and the most appropriate treatment for these patients is unknown. Steroids or other immunosuppressive and immunomodulating agents have been administered to such patients, but have unproven efficacy. It is strongly recommended that these patients be referred to a center with considerable experience in caring for KD patients, such as Children's Memorial.

Cardiac complications

Coronary artery aneurysms are the feared complication of acute KD, potentially leading to coronary aneurysm rupture or thrombosis, myocardial infarction, and sudden death in the first few months following the onset of illness. During the acute phase, myocarditis occurs in more than 50% of KD patients, [13] often clinically apparent as tachycardia out of proportion to the degree of fever. Pericarditis is not clinically evident, but a pericardial effusion on echocardiogram is common. KD is an acute vasculitis, and coronary abnormalities are usually evident on echocardiogram toward the end of the first month after the onset of illness, if they have occurred.

Echocardiograms are generally performed at diagnosis, at follow-up 2 to 3 weeks after the onset of illness, and again at 6 to 8 weeks after the onset. It is important that echocardiograms be performed by an individual trained in visualizing coronary arteries in young children. Adult cardiologists and adult echocardiography technicians generally are not trained to perform these types of exams.

If no coronary artery disease is evident by 8 weeks after the onset, it is highly unlikely that coronary disease will develop. There is no convincing evidence of long-term coronary disease in patients who did not have coronary disease during the acute phase.

However, coronary artery thrombosis and/or stenosis leading to myocardial infarction and sudden death may occur months to years following acute KD in patients who developed coronary disease. Patients who develop significant coronary disease should be followed by a pediatric cardiologist.

Summary

Given the potentially fatal consequences of KD, pediatricians should consider the possibility of this disorder in any child with prolonged fever of unexplained cause, especially if any other clinical features of KD are present. Prompt therapy can significantly reduce the prevalence of coronary artery abnormalities following the illness. 

REFERENCES

[1.] Taubert KA, Rowley AH, Shulman ST. Seven-year national survey of Kawasaki disease and acute rheumatic fever. Pediatric Infectious Diseases Journal 1994;13:704-708.

[2.] Newburger JW, Takahashi M, Burns JC, et al. The treatment of Kawasaki syndrome with intravenous gammaglobulin. N Engl J Med 1986;315:341-347.

[3.] Kawasaki T. Acute febrile mucocutaneous lymph node syndrome with lymphoid involvement with specific desquamation of the fingers and toes in children (Japanese). Jpn J Allergol 1967;16:178-222.

[4.] Rowley AH. Incomplete Kawasaki disease. Pediatr Infect Dis J 2002;21:563-564.

[5.] Yanagawa H, Nakamura Y, Yashiro M, et al. Incidence survey of Kawasaki disease in 1997 and 1998 in Japan. Pediatrics 2001;107:e33.

[6.] Brown TJ, Crawford SE, Cornwall M, Garcia F, Shulman ST, Rowley AH. CD8 T cells and macrophages infiltrate coronary artery aneurysms in acute Kawasaki disease. J Infect Dis 2001;184:940-943.

[7.] Rowley AH, Eckerely CA, Jack HM, Shulman ST, Baker SC. IgA plasma cells in vascular tissue of patients with Kawasaki syndrome. J Immunol 1997;159:5946-5955.

[8.] Rowley AH, Shulman ST, Mask CA, et al. IgA plasma cell infiltration of proximal respiratory tract, pancreas, kidney, and coronary artery in acute Kawasaki disease. J Infect Dis 2000;182:1183-1191.

[9.] Rowley AH, Shulman ST, Spike BT, et al. Oligoclonal IgA response in the vascular wall in acute Kawasaki disease. J Immunol 2001;166:1334-1343.

[10.] . Rowley AH, Baker SC, Shulman ST, et al. Detection of antigen in bronchial epithelium and macrophages in acute Kawasaki disease using synthetic antibody. J Infect Dis 2004 (in press).

[11.] Newburger JW, Takahashi M, Beiser AS, et al. A single infusion of gamma globulin as compared with 4 infusions in the treatment of acute Kawasaki syndrome. N Engl J Med 1991;24:1633-1639.

[12.] Freeman AF, Shulman ST. Refractory Kawasaki disease. The Pediatric Infectious Diseases Journal 2004;23:463-464.

[13.] Takahashi M. Myocarditis in Kawasaki syndrome. Circulation 1989;79:1398-1400.