Kawasaki disease was first described in 1967 by Tomisako Kawasaki, a Japanese pediatrician. He characterized the illness, then termed mucocutaneous lymph node syndrome , as including high fever, nonexudative conjunctivitis, inflammation of the oral mucosa, rash, cervical adenopathy, and findings in the limbs, including swollen hands and feet, red palms and soles, and, later, subungual peeling. Initially thought to be self-limited, the disease was later shown to cause coronary artery aneurysms with risk of myocardial infarction and sudden death. Although Kawasaki disease has the highest relative risk in Japanese children, it has been described worldwide in children of all races and ethnicities. In developed countries Kawasaki disease has surpassed rheumatic fever as the most frequent cause of acquired cardiac disease in children. This chapter summarizes current knowledge about the etiology and pathogenesis, diagnosis, treatment, and natural history of Kawasaki disease.
Epidemiology
The incidence of Kawasaki disease differs according to race and ethnicity. In Japan, Kawasaki disease has an incidence of approximately 265 cases for every 100,000 children under the age of 4 years. The incidence of Kawasaki disease has continued to increase in Japan, either because of a greater number of cases or increased awareness and diagnosis of patients with incomplete criteria. In the United States, the incidence is approximately 21 cases per 100,000 children, higher among Asians and Pacific Islanders (30 per 100,000), followed by non-Hispanic African-Americans (17 per 100,000), Hispanics (16 per 100,000), and whites (12 per 100,000). The incidence rate in the United States has been stable.
Kawasaki disease is most common in children younger than 5 years, but one-quarter of cases occur in older children. Young infants have the highest rate of formation of coronary artery aneurysms and often present with incomplete features. Children older than 8 years also have a higher rate of coronary artery involvement. In contrast to young infants, older children generally present with typical findings but are more likely to have a delayed diagnosis, in part because of the belief that Kawasaki disease occurs only in very young children. Males are affected more than females by a ratio of 1.5 to 1, and the illness is most common during the winter and early spring months. Data linking Kawasaki disease to antecedent exposure to carpet cleaning or to viral infections of the respiratory tract have been inconsistent. Other studies have suggested that Kawasaki disease is associated with residence near a standing body of water, antecedent eczema, use of humidifiers, and perinatal exposure.
In Japan the rate of recurrence of the disease is approximately 3.5%, with half of the cases occurring within a year of the index episode. The overall risk of recurrence in the United States has been estimated at 1.7% but is higher (3.5%) in Asians and Pacific Islanders. Patients with coronary complications were more likely to have a recurrence. Siblings have a relative risk that is 10-fold that of the normal Japanese population, with half developing the disease within 10 days of the first case. The risk of occurrence in twins may be as much as 100-fold higher than that in the general population, with a 14% risk of concordance. Of note, individuals of Japanese ancestry in the United States are thought to have similar risk to those in Japan.
Coronary artery disease is responsible for almost all deaths in patients with Kawasaki disease. The case fatality rate is 0.015% in Japan. In the United States, reported in-hospital mortality for Kawasaki disease has varied from 0% to 0.17%. Although the highest risk of myocardial infarction and death occurs in the first months after illness onset, sudden death from ischemic heart disease may occur many years later in patients with coronary artery aneurysms and stenoses.
Etiology and Pathogenesis
The cause of Kawasaki disease remains unknown despite decades of investigation and controversy. An infectious cause or trigger in genetically susceptible individuals is the most widely accepted theory. The self-limited and usually nonrecurring nature of the disease supports the infectious etiology. Moreover, the clinical signs and symptoms overlap with those in known toxin-mediated or viral infections. Predilection of the disease for young children—with rare occurrence in neonates and adults—suggests that immunity is acquired. An infectious agent is further suggested by winter-spring seasonality, outbreaks in the community, and occasional epidemics. Person-to-person transmission does not occur, but infection with a common agent could produce asymptomatic disease in most children and recognizable signs and symptoms of the disease in a subset of susceptible individuals.
Many candidate pathogens have been tested and discarded. More recent studies suggest possible infection with a novel RNA virus that enters through the upper respiratory tract. This concept is supported by intracytoplasmic inclusion bodies in bronchial epithelial cells and coronary arteries of the majority of postmortem specimens from children with Kawasaki disease but not in control patients. The nature of the antigen is still undetermined, and efforts to characterize the molecular details have been limited by the paucity of autopsy tissue available.
Other studies propose an idiosyncratic immune response, influenced by host genetics, that is triggered by an environmental exposure carried by winds. This theory is suggested by links between the seasonality of Kawasaki disease to tropospheric wind patterns.
The importance of genetic factors in susceptibility is supported by the influence of race and family history on its incidence (see Epidemiology section). In addition, an increasing literature has explored the association of genetic polymorphisms to disease susceptibility or to the development of aneurysms.
Pathology
Kawasaki disease is characterized by systemic inflammation of all medium-sized, muscular, intraparenchymal arteries, with the most serious complications related to inflammation of the coronary arteries. Peripheral arterial aneurysms—especially of the celiac, mesenteric, femoral, iliac, renal, axillary and brachial arteries—can occur, but they have been described only in patients with coronary artery aneurysms. A recent pathologic study described three distinct vasculopathic processes. The first process is necrotizing arteritis, which is an acute self-limited neutrophilic process that progressively destroys the arterial wall. This is followed by subacute/chronic vasculitis, an asynchronous nonneutrophilic inflammatory process of small lymphocytes. It starts in the first 2 weeks of the disease but persists for months to years after diagnosis. Luminal myofibroblastic proliferation is a progressive asynchronous intraluminal stenosing process of smooth muscle cell–derived myofibroblasts and their matrix products.
The outcome of coronary aneurysms depends on their severity. Mildly dilated and inflamed arteries may be able to remodel to normal. However, large aneurysms lose their intima, media, and elasticity, which cannot be regenerated. Aneurysms may develop thromboses, which may be completely occlusive but also may organize, recanalize and calcify ( Fig. 53.1 ). Arteries with partially preserved media can develop thrombosis or progressive stenosis from luminal myofibroblastic proliferation.
Myocarditis is found in almost all patients with Kawasaki disease and is the earliest cause of death. Small histopathologic series have demonstrated predominant myocardial edema and a variable cellular inflammatory infiltrate with neutrophils, monocytes, macrophages, and/or eosinophils. After the first 10 days, death was most often caused by myocardial infarction due to thrombosis. Mortality peaks between 15 and 45 days after the onset of fever, when patients, who are in a hypercoagulable state, have thrombocytosis and disrupted vascular endothelium. Beyond the first year after onset of the illness, mortality declines significantly, but myocardial infarction may occur many years later because of progressive coronary artery stenosis.
Clinical Diagnosis
General Aspects of Initial Diagnosis
Diagnostic criteria are summarized in Box 53.1 . The epidemiologic definition for diagnosis includes fever for 4 days with at least four principal clinical criteria ( Fig. 53.2 ) or fever and fewer than four principal criteria in the presence of coronary artery abnormalities. By convention, the first day of the disease is considered to be the first day on which fever occurs. All clinical features are rarely present at the same time, so the diagnosis requires sequential evaluation. In the absence of treatment, the mean duration of fever is 11 days, and fever rarely persists beyond 4 weeks.
Classic KD is diagnosed in the presence of fever for at least 5 days (the day of fever onset is taken to be the first day of fever) together with at least four of the five following principal clinical features. In the presence of four or more principal clinical features, particularly when redness and swelling of the hands and feet are present, the diagnosis of KD can be made with 4 days of fever, although experienced clinicians who have treated many patients with KD may establish the diagnosis with 3 days of fever in rare cases (see Fig. 53.2 ):
- 1.
Erythema and cracking of lips, strawberry tongue, and/or erythema of oral and pharyngeal mucosa.
- 2.
Bilateral bulbar conjunctival injection without exudate.
- 3.
Rash: maculopapular diffuse erythroderma, or like erythema multiforme.
- 4.
Erythema and edema of the hands and feet in acute phase and/or periungual desquamation in subacute phase.
- 5.
Cervical lymphadenopathy (≥1.5 cm diameter), usually unilateral.
A careful history may reveal that one or more principal clinical features were present during the illness but resolved by the time of presentation.
Patients who lack full clinical features of classic KD are often evaluated for incomplete KD ( Fig. 53.3 ). If coronary artery abnormalities are detected, the diagnosis of KD is considered confirmed in most cases.
Laboratory tests typically reveal a normal or elevated white blood cell count with neutrophil predominance and elevated acute-phase reactants such as C-reactive protein and erythrocyte sedimentation rate during the acute phase. Low serum sodium and albumin levels, elevated serum liver enzymes, and sterile pyuria can be present. In the second week after fever onset, thrombocytosis is common.
Other clinical findings may include the following:
- ▪
Cardiovascular
- ▪
Myocarditis, pericarditis, valvar regurgitation, shock
- ▪
Coronary artery abnormalities
- ▪
Aneurysms of medium-sized noncoronary arteries
- ▪
Peripheral gangrene
- ▪
Aortic root enlargement
- ▪
- ▪
Respiratory
- ▪
Peribronchial and interstitial infiltrates on CXR
- ▪
Pulmonary nodules
- ▪
- ▪
Musculoskeletal
- ▪
Arthritis, arthralgia (pleocytosis of synovial fluid)
- ▪
- ▪
Gastrointestinal
- ▪
Diarrhea, vomiting, abdominal pain
- ▪
Hepatitis, jaundice
- ▪
Gallbladder hydrops
- ▪
Pancreatitis
- ▪
- ▪
Nervous system
- ▪
Extreme irritability
- ▪
Aseptic meningitis (pleocytosis of cerebrospinal fluid)
- ▪
Facial nerve palsy
- ▪
Sensorineural hearing loss
- ▪
- ▪
Genitourinary
- ▪
Urethritis/meatitis, hydrocele
- ▪
- ▪
Other
- ▪
Desquamating rash in groin
- ▪
Retropharyngeal phlegmon
- ▪
Anterior uveitis by slit-lamp examination
- ▪
Erythema and induration at BCG inoculation site
- ▪
The differential diagnosis includes other infectious and noninfectious conditions, such as the following:
- ▪
Measles
- ▪
Other viral infections (e.g., adenovirus, enterovirus)
- ▪
Staphylococcal and streptococcal toxin–mediated diseases (e.g., scarlet fever and toxic shock syndrome)
- ▪
Drug hypersensitivity reactions, including Stevens-Johnson syndrome
- ▪
Systemic onset juvenile idiopathic arthritis
- ▪
With epidemiologic risk factors:
- ▪
Rocky Mountain spotted fever or other rickettsial infections
- ▪
Leptospirosis
- ▪
BCG , Bacillus Calmette-Guérin; CXR , chest x-ray; KD , Kawasaki disease.
The clinical findings are nonspecific, so other diseases with similar presentations should be excluded. However, the detection of respiratory viruses should not exclude the diagnosis of Kawasaki disease in a child with clinical findings of Kawasaki disease. The differentiation between adenovirus infection and Kawasaki disease can be more confusing owing to similar clinical presentations. Patients with adenovirus infections are less likely to have nonexudative pharyngitis, erythema and swelling of the hands and feet, strawberry tongue, and desquamating groin rash. Bacterial infections have also been described in patients with Kawasaki disease, most frequently group A streptococcal tonsillitis, pneumonia, and urinary tract infection. Kawasaki disease should be considered in patients with some features of Kawasaki disease and no improvement after 24 to 48 hours of effective antibiotic therapy.
A diagnosis of incomplete disease may be made in patients who do not fulfill the classic criteria outlined in Box 53.1 , whereas the term atypical disease refers to patients who have features not generally seen, for example those presenting with hemophagocytic syndrome or with renal failure. Because coronary artery disease with incomplete or atypical features is well recognized to occur in children, treatment with immunoglobulins must be started early to be effective. An algorithm for evaluation and treatment of the child with suspected disease has been developed and is shown in Fig. 53.3 . Retrospective evaluation of this diagnostic algorithm showed that it was effective in identifying and referring for treatment patients with incomplete clinical criteria. A high level of suspicion is necessary in young infants who are most likely to have incomplete criteria and also have the highest risk of developing aneurysms. Echocardiography is recommended for infants less than 6 months of age with a fever lasting for 7 or more days without other explanation and elevation of either C-reactive protein or erythrocyte sedimentation rate.
Laboratory features are reflective of an acute inflammatory response. Although nonspecific, they provide support for diagnosis in patients with a nonclassic but suggestive presentation. Virtually all patients at presentation have elevation of the erythrocyte sedimentation rate or C-reactive protein, which can, however, be discrepant. The average level of hemoglobin at the time of presentation is two standard deviations below the mean for age, with the anemia being normocytic and normochromic. The number of white blood cells is generally increased, at a median of 15,000/mm with a leftward shift. The platelet count is usually normal in the first week of the illness, peaking in the third week of the illness to values sometimes higher than 1,000,000/mm. Thrombocytopenia is rare but may occur during the acute phase in a subset of patients with more severe disease. Plasma γ-glutamyl transpeptidase, transaminases, and bilirubin are frequently elevated. Synthesis of albumin is decreased in the acute phase, and hypoalbuminemia is common. Microscopic evaluation of the urine may reveal an elevated count of white blood cells with no identified infectious agent, so-called sterile pyuria. Cerebrospinal fluid contains an increased number of white blood cells, predominantly mononuclear cells, with normal levels of glucose and protein.
No biomarkers have been demonstrated superior to clinical criteria and inflammatory markers. N-terminal moiety of B-type natriuretic peptide (NT-proBNP), likely indicative of myocardial involvement, has shown good specificity in identifying patients with Kawasaki disease versus other febrile illnesses but does not have sufficient sensitivity and discriminative ability. Moreover, cutoff values for a positive result have not been clearly defined.
Cardiac Findings
Cardiovascular manifestations and complications are the major cause of morbidity and mortality in Kawasaki disease, both in the acute phase and on long-term follow-up.
The acute phase of the disease may be associated with inflammation of the myocardium, pericardium, or endocardium including valves and coronary arteries. Cardiac auscultation typically reveals a hyperdynamic precordium, or tachycardia. Almost all children have an innocent flow murmur related to anemia and fever. Murmurs of valvar dysfunction can be heard in approximately 25% of patients, most commonly a pansystolic regurgitant murmur of mitral regurgitation. A gallop rhythm can also sometimes be heard, suggesting decreased compliance and diastolic dysfunction secondary to myocardial inflammation. The disease may occasionally present with low cardiac output syndrome or shock.
Electrocardiography may show arrhythmia, a prolonged PR interval, or nonspecific ST- and T-wave changes. Increased QT dispersion, abnormalities of ventricular repolarization, and electrocardiographic signs suggestive of left ventricular dilation and myocardial involvement have also been described.
Based on myocardial biopsies during the acute stage, myocarditis is a nearly universal feature of Kawasaki disease. Evidence of myocardial inflammation is present in 50% to 75% of patients on nuclear imaging. Only a subset of patients presents with overt ventricular dysfunction (25% to 50%). Fortunately, myocardial function usually improves rapidly after administration of intravenous immunoglobulin (IVIG), and long-term abnormalities of systolic function are uncommon in the absence of ischemic heart disease secondary to coronary artery aneurysms.
In the acute phase, mitral regurgitation is the most common valve dysfunction; it may result from valvitis or from transient dysfunction of the papillary muscles. Approximately one-quarter of children have mitral regurgitation at the time of presentation. Late mitral regurgitation is usually the result of ischemic disease. Aortic regurgitation is much less frequent and found in less than 1% of patients. It may be associated with aortic root dilation, which occurs in 8% of patients.
Coronary artery abnormalities during the acute illness range from dilation to aneurysm. The prevalence of coronary artery aneurysms may be as high as 25% when untreated, and decreases to 4% with administration of IVIG in the first 10 days of illness.
The most frequent coronary anomaly is dilation, which is defined as a coronary artery z- score between 2 and 2.5. Some patients will have coronary artery z- scores always within the normal range (below 2) but with a significant reduction in luminal dimensions upon follow-up. It is still unclear if these changes, which occur in 30% to 50% of patients, represent resolution of inflammatory changes or hemodynamic factors related to fever.
The identification of patients with mild coronary artery dilation in the acute phase can be confounded by the fact that normal measurements are based on a population of healthy afebrile children. Coronary artery enlargement has been reported in febrile patients to a lesser extent than in patients with Kawasaki disease. This is important because, especially in patients with incomplete clinical criteria, the presence of coronary artery dilation is often used to confirm the diagnosis. Studies have shown that a z- score of 2.5 identified patients with Kawasaki disease with 98% specificity.
Coronary artery aneurysms are the most serious long-term complication of the disease. Long-term outcomes are related to the extent of dilation of the coronary arteries in the first month of illness. The aneurysms may be detected by echocardiography beginning 7 days after the first appearance of fever, with their diameter usually peaking around 4 weeks after onset of the illness. Independent predictors of development have included age less than 1 year, male gender, Asian/Pacific Islander race, Hispanic ethnicity, delayed treatment with IVIG, persistent or recrudescent fever after IVIG, and laboratory measures suggesting worse inflammation. After 4 to 6 weeks, myointimal proliferation results in regression of approximately half to two-thirds of aneurysms ( Fig. 53.4 ). The likelihood of regression is primarily determined by the peak luminal diameter, with larger aneurysms being less likely to regress. Other factors that predict a greater likelihood of regression are younger age, distal location, and fusiform shape. Despite normalization of the internal luminal diameter, the coronary arteries are not “normal.” This is supported by pathology specimens and intravascular imaging. Intravascular ultrasound has shown that myointimal thickening is present many years after diagnosis. Similarly, optical coherence tomography shows intimal hyperplasia, disappearance of the media, fibrosis, calcifications, macrophage accumulation, and neovascularization. In addition to persistent structural abnormalities, functional abnormalities have also been shown. Pharmacologic testing has demonstrated decreased vasodilation, resulting in impaired coronary artery flow reserve.
Patients with persistent aneurysms are at risk of developing significant stenosis secondary to myointimal proliferation at either end of the aneurysm, calcification, tortuosity, or thrombotic occlusion. Aneurysm rupture is rare and usually occurs as an early complication. The incidence of coronary stenosis increases over time and is highest in patients with giant aneurysms.
The principal cause of death in Kawasaki disease is myocardial infarction as a result of thrombotic occlusion and/or stenosis. Myocardial infarction in children can be clinically silent or present with atypical symptoms. The highest risk is in patients with giant aneurysms, where sluggish flow increases the risk of thrombosis, often combined with stenosis at the distal end. Myocardial infarction is most common in the first year following diagnosis, although patients have a lifelong increased risk of ischemia. Indeed, previously undiagnosed Kawasaki disease may become apparent in adulthood at time of myocardial infarction. Fatalities due to myocardial infarctions are more likely with occlusion of the left main coronary artery or proximal segments of the right and anterior descending coronaries. Thrombosis of the right coronary is more likely to be silent and to recanalize. Patients with prior myocardial infarction are at risk for further ischemic events, although the risk can be decreased with effective revascularization.
Echocardiography
Echocardiographic imaging of the coronary arteries is essential in the evaluation of all patients with definite or suspected disease ( Fig. 53.5 ). Its sensitivity and specificity for the detection of dilation of the proximal coronary artery segments are high when this procedure is performed with appropriate transducers by experienced sonographers in cooperative or sedated children. Because aneurysms also occur in children who do not meet classic criteria, echocardiography has an important role in the evaluation of children with protracted fever and some findings consistent with features of Kawasaki disease. Additional echocardiographic features in the acute phase include coronary artery ectasia and lack of tapering. Left ventricular contractility may also be diminished, and mild valvar regurgitation, particularly mitral regurgitation, is relatively common. Although pericardial effusions may occur, they are rarely significant in size.
Echocardiography should be performed as soon as the diagnosis is suspected but without delaying treatment to provide a baseline examination of coronary dimensions, left ventricular function, valvar regurgitation, and pericardial effusion. The initial echocardiogram in the first week of illness can be normal and does not rule out the diagnosis. In children younger than 3 years or in uncooperative children, a sedated study may be necessary to obtain a good assessment of the coronary arteries. In children whose fever resolves after initial treatment with intravenous immune globulin and who remain afebrile, echocardiography should be repeated within 1 to 2 weeks and again at 4 to 6 weeks after treatment. Children at higher risk—that is, those with recrudescent fever, baseline coronary artery abnormalities, diminished left ventricular function, or pericardial effusion—should undergo more frequent echocardiographic evaluation to guide the need for additional therapies. Failure to escalate thromboprophylaxis at time of rapid aneurysmal expansion is a primary cause of morbidity and mortality.
Two-dimensional echocardiographic imaging should be performed with a transducer of the highest frequency available and recorded in a dynamic video or digital cine format. The imaging planes and transducer positions required for optimal visualization of the coronary artery segments are depicted in Box 53.2 . Whenever possible all major coronary artery segments should be visualized. Multiple imaging planes and transducer position are required for optimal visualization of all major coronary artery segments. The most common site for formation of aneurysms are the proximal segments of the proximal left anterior descending and right coronary arteries, followed in descending order by the left main coronary artery, the circumflex artery, the distal part of the right coronary artery, and the junction between the right coronary artery and inferior interventricular artery. Measurements are made from inner edge to inner edge excluding points of branching. Aneurysms are considered to be saccular when their axial and lateral diameters are nearly equal. Fusiform aneurysms occur when there is symmetric dilation with gradual proximal and distal tapering. Coronary arteries that are dilated without a segment aneurysm are considered ectatic.
- ▪
LMCA: Precordial short axis at level of aortic valve; precordial long axis of left ventricle (superior tangential); subcostal ventricular long axis
- ▪
LAD coronary artery: Precordial short axis at level of aortic valve; precordial superior tangential long axis of left ventricle; precordial short axis of left ventricle
- ▪
Left circumflex branch: Precordial short axis at level of aortic valve; apical four chamber
- ▪
RCA, proximal segment: Precordial short axis at level of aortic valve; precordial long axis (inferior tangential) of left ventricle; subcostal coronal projection of right ventricular outflow tract; subcostal short axis at level of atrioventricular groove
- ▪
RCA, middle segment: Precordial long axis of left ventricle (inferior tangential); apical four chamber; subcostal left ventricular long axis; subcostal short axis at level of atrioventricular groove; RCA proximal (1) and middle (2) observed in the atrioventricular groove from the third intercostal space at the left and right sternal border
- ▪
RCA, distal segment: Apical four chamber (inferior); subcostal atrial long axis (inferior)
- ▪
Posterior descending coronary artery: Apical four chamber (inferior); subcostal atrial long axis (inferior); precordial long axis (inferior tangential) imaging; posterior interventricular groove
LAD , Left anterior descending; LMCA , left main coronary artery; RCA , right coronary artery.
In the most recent American Heart Association guidelines on the management of patients with Kawasaki disease, coronary artery aneurysms are classified using the z- score to account for body surface area. The use of such z- scores has suggested that the previously used Japanese Ministry of Health criteria may underdiagnose and underestimate the true prevalence of coronary artery dilation. A small aneurysm is defined as a z- score of 2.5 to less than 5, medium aneurysm as a z- score of 5 to less than 10 and absolute dimension less than 8 mm, and a large or giant aneurysm as z- score 10 or greater or absolute dimension 8 mm or greater ( Table 53.1 ). However, with most calculators, z- scores are available only for the left main coronary artery and the proximal segments of the left anterior descending and right coronary arteries. For distal coronaries, the Japanese Ministry of Health criteria should still be applied (small, <4 mm; medium, 4 to 8 mm; giant, ≥8 mm; in children 5 years of age and younger, small if <1.5 times the adjacent segment, medium if 1.5 to 4 times, and giant if >4 times).
| Classification | Description |
|---|---|
| 1 | No involvement at any time point ( z- score always <2) |
| 2 | Dilation only ( z- score 2 to <2.5) |
| 3 | Small aneurysm ( z- score ≥2.5 to <5) |
| 3.1 | Current or persistent |
| 3.2 | Decreased to dilation only or normal luminal dimension |
| 4 | Medium aneurysm ( z- score ≥5 to <10 and absolute dimension <8 mm) |
| 4.1 | Current or persistent |
| 4.2 | Decreased to small aneurysm |
| 4.3 | Decreased to dilation only or normal luminal dimension |
| 5 | Large and giant aneurysm ( z- score ≥10, or absolute dimension ≥8 mm) |
| 5.1 | Current or persistent |
| 5.2 | Decreased to medium aneurysm |
| 5.3 | Decreased to small aneurysm |
| 5.4 | Decreased to dilation only or normal luminal dimension |
The number and localization of aneurysms should be assessed, as well as the presence or absence of intraluminal thrombi and stenotic lesions. Evaluation for freshly formed thrombus can be performed using a wider gray scale ( Fig. 53.6 ). However, echocardiography has limited sensitivity for the detection of thrombi and stenotic lesions. Other limitations of echocardiography include difficulty in visualizing coronary arteries in older children as well as evaluating more distal segments. Additional imaging modalities (computed tomography [CT] angiography, cardiac magnetic resonance imaging [MRI], or invasive angiography) should be considered when needed.
Assessment for Inducible Ischemia
Children with aneurysms should undergo periodic stress testing, with assessment of myocardial perfusion or function. Most methods of stress testing used in adult cardiology have been reported in small series of children with Kawasaki disease. Because the sensitivity and specificity of tests to provoke myocardial ischemia have been exhaustively studied in adults with coronary artery disease, adult guidelines should be followed to choose the best test based on patients’ specific characteristics. Additional factors in the choice of modality for testing include institutional expertise with particular techniques and the age and ability of the child to cooperate with exercise. False-positive tests are more likely in patients with a low prior probability of coronary artery disease. Stress testing is therefore not recommended in patients without a history of coronary artery enlargement.
Other Noninvasive Methods for Imaging the Coronary Arteries
In selected patients, noninvasive imaging methods other than echocardiography may be needed to assess the coronary artery anatomy. In particular, ultrafast CT and MRI and angiography may provide valuable data in patients whose coronary arteries cannot adequately be imaged by echocardiography. MRI can assess the coronary arteries as well as ventricular function, myocardial perfusion, and scarring ( Fig. 53.7 , ). In addition to imaging, cardiac resonance tests can be performed together with pharmacologic stress and may allow assessment of myocardial infarction using delayed enhancement.
Ultrafast CT has been shown to have excellent sensitivity for the detection of coronary artery stenosis in patients with the disease ( Fig. 53.8 ). It is more sensitive than MRI for assessment of distal vessels and the presence of thrombus. Because the technique is associated with relatively high doses of ionizing radiation, it should be reserved for circumstances where other noninvasive tests are inadequate. However, newer systems use much lower doses of radiation, which can increase the utility and safety of this modality.
Coronary Angiography
Cardiac catheterization and coronary angiography are the gold standard for imaging against which other methods are assessed ( Fig. 53.9 , ). In addition to providing better delineation of the distal coronary vasculature, angiography is the most reliable method to assess coronary artery stenosis ( Fig. 53.10 , ) or thrombotic occlusion ( Fig. 53.11 , ) and the presence of collateral vessels. Fractional flow reserve can also be performed; it helps to assess the potential for ischemia and need for intervention. To evaluate whether peripheral artery aneurysms are present, abdominal aortography and subclavian arteriography should be performed in patients undergoing coronary arteriography for the first time. Because selective coronary angiography poses greater risks than noninvasive imaging, its use should be restricted to selected patients with inducible myocardial ischemia on noninvasive testing or those in whom noninvasive techniques fail to provide adequate data. It is also useful after surgical or percutaneous revascularization to evaluate efficacy and monitor for complications ( Fig. 53.12 , ).
Treatment
Treatment of the disease in the first weeks after onset is aimed at lowering fever, reducing inflammation and shear stress in the arterial wall, and preventing thrombosis. To reduce shear stress, children who are profoundly anemic and in whom coronary aneurysms are developing should undergo transfusion of red blood cells, ideally to achieve a hematocrit of at least 30%, and β-blockers should be administered to reduce myocardial consumption of oxygen. Among patients with aneurysms, the prevention and, if needed, treatment of coronary thrombosis are key components of therapy. Patients with coronary artery stenosis or occlusion and evidence of reversible ischemia are candidates for interventional catheterization and surgical procedures. Specific therapies are detailed later in this chapter.
Aspirin
Aspirin is used in the acute treatment of Kawasaki disease for its antiinflammatory and antiplatelet effects. There is no consensus on the optimal dose of aspirin as well as significant variation in practice. Recent American Heart Association guidelines suggest that it is reasonable to administer moderate- (30 to 50 mg/kg per day) to high-dose (80 to 100 mg/kg per day) until the patient is afebrile and then to lower the dose to 3 to 5 mg/kg per day for its antiplatelet effects. A multicenter retrospective study showed that low-dose aspirin in not inferior to high-dose aspirin at the time of initial diagnosis for reducing the risk of coronary artery abnormalities; this is in line with prior studies showing no role in the prevention of coronary artery complications. Aspirin in low doses is continued for approximately 6 weeks and then discontinued in patients without coronary artery aneurysms. In children with coronary artery abnormalities, aspirin is continued indefinitely at low doses and may be used together with other antithrombotic therapies such as clopidogrel or warfarin. Because ibuprofen antagonizes the inhibitory effect of aspirin on platelets, sustained therapy with ibuprofen should be avoided in children who are taking aspirin for prophylaxis of coronary artery thrombosis.
Reye syndrome has been reported in children with the disease who are taking aspirin in high doses. Although this syndrome has not been associated with use of aspirin in low doses, annual vaccination against influenza is recommended for all children on chronic treatment with aspirin. When a child medicated on a chronic basis with aspirin develops a flu-like illness, aspirin should be withheld transiently and, if necessary, another antiplatelet medication, such as clopidogrel, should be substituted until resolution of the illness.
Intravenous Immunoglobulin
Control of inflammation decreases the likelihood of aneurysm formation and is the most important aim of therapy in the acute phase of illness. Among the armamentarium of antiinflammatory agents that have been used, only IVIG in high doses has been demonstrated to be effective in multiple randomized multicenter trials with blinded echo interpretation. When administered in the first 10 days, IVIG reduces the prevalence of aneurysms approximately fivefold, to less than 5%. Treatment with IVIG is also beneficial for children beyond the 10th day of illness in whom fever persists or who have coronary artery abnormalities together with persistent clinical and laboratory evidence of inflammation. Patients with recurrent Kawasaki disease, defined as a repeat episode after complete resolution of the previous episode, should also receive standard therapy with aspirin and IVIG. The standard dosage is 2 g/kg administered over 8 to 12 hours. Studies comparing the efficacy of different immune globulin products have been conflicting.
In patients who present with diminished left ventricular function, the agent should be administered more slowly because of the considerable volume load. Coombs-positive hemolytic anemia is a complication of IVIG, especially in patients with the A, B, or AB blood group type. Aseptic meningitis can also occur and resolves quickly with no neurologic sequelae. Immunization for measles, mumps, and varicella should be deferred for 11 months after IVIG administration.
Other Therapies
Whereas the main treatment in the acute phase includes aspirin and IVIG, some patients at higher risk may benefit from primary adjunctive treatment. There are multiple Japanese risk-scoring systems to identify patients at higher risks of IVIG resistance and coronary artery complications, However, their low sensitivity in North American populations limits their use.
Corticosteroids are the mainstay of therapy for many childhood vasculitides, although their use in the treatment of patients with Kawasaki disease has been more controversial. The use of pulsed-dose intravenous methylprednisolone did not improve outcome in a randomized multicenter placebo-controlled trial performed in North America. However, studies in Japan have shown improved coronary outcomes when single-dose methylprednisolone was administered to high-risk patients. Moreover, the use of oral steroids with a slow taper was shown to be beneficial in the randomized controlled trial to assess immunoglobulin plus steroid efficacy for Kawasaki disease (RAISE study) in high-risk patients, with a decreased incidence of coronary artery abnormalities and treatment resistance, lower coronary artery z- scores, and more rapid resolution of fever as well as a decline in inflammatory markers. Thus the challenge in non-Japanese populations is to identify patients who might benefit from more aggressive therapy.
Infliximab, a chimeric monoclonal antibody to TNF-α, was studied for intensification of initial treatment. Although it shortened the number of days of fever and inflammatory parameters normalized more rapidly, it failed to decrease the rate of IVIG resistance, which was the primary outcome of the study. Trials using etanercept for treatment intensification are ongoing.
Additional Therapy for Patients With Intravenous Immunoglobulin Resistance
Approximately 10% to 20% of patients with Kawasaki disease are resistant to intravenous immunoglobulins, defined as persistent or recurrent fever at least 36 hours after the end of their IVIG infusion without other explanation. Patients with resistance to IVIG are at increased risk of coronary artery complications and thus usually warrant retreatment. Meta-analyses have demonstrated a dose-response effect of IVIG ; thus experts recommend a second infusion of IVIG.
Corticosteroids have also been used to treat patients who failed to respond to initial therapy. Retrospective studies and case series suggest that treatment with steroids improves fever and the inflammatory response ; however, not all studies have shown improvement in coronary artery outcome. Both pulse intravenous methylprednisone (30 mg/kg per day) and intravenous prednisolone (2 mg/kg per day) were used, with no clinical trials comparing the efficacy of different regimens. It is hypothesized that longer steroid courses lead to improved outcome due to the suppression of persistent vascular inflammation.
Small nonrandomized studies of the use of infliximab in patients resistant to IVIG suggest a clinical response with resolution of inflammation in the majority of patients but without impact on coronary artery outcomes.
The use of Anakinra, an IL-1 receptor antagonist, has been described in case reports, and clinical trials are ongoing.
Cyclosporine has been used in highly refractory patients with resolution of fever; however, more studies are needed to determine the impact on coronary artery outcome.
Rarely, cytotoxic agents have been used to treat refractory patients with acute disease, but the risks of these therapies are such that they should be used only in patients with expanding aneurysms who are resistant to other agents.
Plasma exchange has been reported to lower the incidence of aneurysms in uncontrolled studies. Because this therapy is technically complex to administer, its use should be reserved for children in whom all other medical therapies have failed.
Prevention of Coronary Artery Thrombosis
Thrombotic occlusion of a coronary artery is the most serious complication in patients with Kawasaki disease, precipitating myocardial infarction or sudden death. Multiple factors contribute to the risk of thrombosis, including thrombocytosis and increased platelet adhesion, inflammation, endothelial dysfunction, and abnormal flow conditions through areas of severe dilation or stenosis. Because randomized trials of antithrombotic regimens have not been performed, the choice of agents is derived primarily from experience in adults with atherosclerotic disease as well as case series and consensus of experts. Anticoagulation is based on the size of the aneurysm at the time of treatment as well as on prior history ( Table 53.2 ). Patients with small aneurysms are treated with aspirin at an antiplatelet dose (3 to 5 mg/kg per day). In patients with moderate aneurysms, clopidogrel may be added to aspirin, but anticoagulation is generally not indicated. Patients with large or giant aneurysms are at highest risk for thrombosis and are usually treated with aspirin and anticoagulation (warfarin or low-molecular-weight heparin). Patients with a recent history of thrombosis are at especially at high risk and may benefit from dual antiplatelet therapy in addition to anticoagulation. In patients on warfarin, an international normalized ratio (INR) of 2 to 3 is usually maintained. Low-molecular-weight heparin is usually preferred in infants or older children who cannot maintain their INRs within goals. Trials of direct oral anticoagulants in children are ongoing.
