Controversies in the management of Kawasaki disease

Best Practice & Research Clinical Rheumatology Vol. 16, No. 3, pp. 427±442, 2002

doi:10.1053/berh.2002.0238, available online at http://www.idealibrary.com on

7 Controversies in the management of Kawasaki disease Bianca Lang

MD, FRCPC

Associate Professor of Paediatrics, Dalhousie University Director, Division of Rheumatology, IWK Health Centre, 5850 University Ave, Halifax, Nova Scotia Canada

Ciaran M. Du€y

MB BCh, MSc, FRCPC

Associate Professor of Paediatrics, McGill University Director, Division of Rheumatology, Montreal Children's Hospital and the McGill University Health Centre, Montreal, Quebec Canada

While there is a generally accepted standard approach to the management of Kawasaki disease (KD) in North America, controversy still exists regarding certain aspects of treatment. Do all patients require treatment with intravenous immunoglobulin (IVIG)? What is the appropriate dose of aspirin (ASA) during the acute phase of the disease? Is there a role for corticosteroids in those who fail IVIG? How should patients with atypical, incomplete or late presentations of KD be managed? What is the appropriate long-term management and follow-up, particularly for those without coronary artery abnormalities (CAA)? Is there a role for surgical intervention, particularly transplantation? These questions, among others, are explored with reference to the pertinent literature. IVIG has been well studied and shown to be ecacious in a number of studies and in two meta-analyses, with clear evidence to support the use of 2 g/kg in a single dose. The appropriate dose of ASA during the acute phase is less clear but, increasingly, data suggest that lower doses of ASA are adequate and perhaps more appropriate. Corticosteroids appear to have a role in those who have failed IVIG but this requires further study before being embraced as accepted treatment. The management of less typical presentations of KD remains controversial, with inadequate data to direct us, although there is a general trend towards treating such patients with IVIG. Careful follow-up of all patients is recommended and, while there are guidelines for this, there is no clear consensus on the most appropriate monitoring investigations for those with and without CAA. There is an expanding role for transplantation, with clearly de®ned indications for this intervention. Key words: Kawasaki disease; treatment; intravenous immunoglobulin; aspirin; cardiac followup; prognosis.

Since the ®rst description of Kawasaki disease (KD) in 1967 and the ®rst Englishlanguage report of 50 patients with KD in 1974,1 there has been considerable Address for correspondence: Division of Rheumatology, Montreal Children's Hospital, 2300 Tupper St., Room C503, Montreal, Quebec, Canada H3H IP3 1521±6942/02/$ - see front matter

c 2002 Elsevier Science Ltd. All rights reserved. *

428 B. Lang and C. M. Du€y

published work describing the clinical features, epidemiology, aetiopathogenesis and management of this interesting entity. In fact, an international symposium devoted to this disorder is held every 2 to 3 years. Despite the profusion of publications on all facets of this disease, controversy abounds, particularly regarding its aetiopathogenesis and management. Herein, we discuss some of the controversies that exist in relation to management and attempt to address them with reference to the pertinent literature, as well as from our collective clinical experience. MANAGEMENT IN THE ACUTE PHASE OF KD Standard management of the acute phase of Kawasaki disease in North America A child with de®nite or suspected KD should be admitted to hospital for prompt evaluation and treatment (Table 1). A careful general assessment, with emphasis on the cardiovascular system, should be carried out and should include an electrocardiogram and an echocardiogram. These are essential to detect evidence of the early cardiac manifestations of the disease, and to document the baseline calibre of the coronary arteries. Unfortunately, there is no speci®c test for establishing the diagnosis. Nonetheless, it is imperative to conduct baseline laboratory tests that might assist in making the diagnosis and in excluding other possible diagnoses, particularly infectious diseases. Once the diagnosis of KD is established, or considered to be likely, the child should receive treatment with intravenous immunoglobulin (IVIG) (2 g/kg over 8±12 hours) as a single infusion and high-dose aspirin (ASA) (80±100 mg/kg per day) in four divided doses. Prompt institution of such treatment results in rapid resolution of the clinical and many of the laboratory features of the acute phase of KD and signi®cantly reduces the risk of subsequently developing coronary artery abnormalities (CAA).2 Although these are standard guidelines for the therapy of KD in North America, controversy exists worldwide regarding the optimal dose and regimen of IVIG and ASA. Furthermore, a number of situations may arise in KD in which the optimal approach to therapy is less clear. What is the optimal dose and regimen of IVIG in KD? Although there is general consensus that treatment with IVIG is ecacious there is still controversy, particularly in Japan, regarding the optimal dose and regimen. Bene®t from IVIG in reducing the incidence of CAA in KD was ®rst demonstrated in 1984.3 In this multicentre controlled trial, IVIG was given at a dose of 400 mg/kg per

Table 1. Standard approach to the management of Kawasaki disease in the acute phase. Hospitalize to conduct appropriate evaluation and to monitor cardiac status Complete cardiovascular evaluation: examination, electrocardiogram, 2D echocardiogram IVIG 2 g/kg in one dose administered over 8±12 hours Acetylsalicylic acid 80±100 mg/kg/day in four doses up to day 14 of illness or until afebrile for 24±48 hours Treat congestive cardiac failure, myocardial dysfunction and other complications

Management of Kawasaki disease 429

day for 5 days. Several subsequent randomized, controlled multicentre studies using various doses and regimens of IVIG have con®rmed the bene®cial e€ects of IVIG in reducing the incidence of CAA compared with treatment with ASA alone.2,4±6 These studies include two randomized controlled trials from the US Multicenter Study Group on KD.2,4 In the initial US multicentre trial, a regimen using IVIG 400 mg/kg per day for 4 days plus ASA was compared with ASA alone. At the 7-week evaluation, CAA were present in 18% of patients in the ASA group but in only 4% in the IVIG group, a di€erence that was statistically highly signi®cant.4 In a subsequent large US randomized controlled trial, the same 4-day regimen of IVIG plus ASA was compared with a single 2 g/kg dose of IVIG given over 8±12 hours, plus ASA. The single-dose regimen resulted in more rapid defervescence, shorter hospital stays and a trend toward fewer CAA. This trend was statistically signi®cant at 2 weeks, but not at 7 weeks after enrolment, when the incidence of CAA was 7.2% in the 4-day infusion group and 3.9% in the single-infusion group (P ˆ 0.098).2 In both trials the dose of ASA used was 100 mg/kg/day. In addition, treatment with high-dose IVIG has also been found to prevent giant CAA (48 mm) which are associated with most of the morbidity and mortality in KD.2 Thus, the single-dose regimen of 2 g/kg IVIG has been accepted as the standard of treatment in North America since 1991. Two recent meta-analyses have demonstrated that the incidence of CAA is signi®cantly lower with higher doses of IVIG (41 g/kg) than with lower doses (51 g/kg)7,8 and that there is a progressive dose±response e€ect, up to 2 g/kg.8 Whether doses of IVIG higher than 2 g/kg may be even more ecacious is not known. Uncontrolled studies from Japan and the USA have reported higher frequencies of cardiac sequelae (21±40%) in those who received 42 g/kg total IVIG dose.9,10 However, these results probably re¯ect selection bias and must be interpreted cautiously. In contrast to the standard North American IVIG regimen for KD, using a single dose of 2 g/kg, in Japan, IVIG doses and regimens have varied signi®cantly and not all patients have received IVIG. According to the 1991±92 Nationwide Survey of KD in Japan, of those patients who did receive IVIG, most (64%) received doses 51.2 g/kg.9,11 Thus, despite the signi®cant increase in IVIG use in Japan between 1983 and 1992 (from 8.2 to 81.8%), the incidence of CAA changed very little during this time (from 16.7 to 13.8%).12 Subsequent to this, a multicentre randomized controlled trial in Japan found a lower incidence of CAA with an IVIG regimen of 400 mg/kg per day for 5 days compared with 200 mg/kg per day for 5 days.6 Therefore, in 1995 the Japanese Medical Insurance System increased the authorized regimen of IVIG to 400 mg/kg per day for 5 days. The most recently published Nationwide Survey of KD for the 2-year period 1997 and 1998 found an incidence of CAA of 10%13, providing further evidence in favour of the use of high-dose IVIG. In summary, IVIG is highly ecacious in KD, particularly when used early in the course of disease, and a dose of 2 g/kg is associated with a better early disease outcome than are lower doses. Is `high-dose' ASA necessary in the acute phase of KD? Although ASA is part of standard therapy in KD because of its anti-in¯ammatory, antipyretic and antiplatelet e€ects, no prospective study has shown that ASA at any dose reduces the incidence of CAA. In the two large randomized controlled US trials on IVIG discussed above, all patients received ASA at a dose of 80±120 mg/kg per day up to day 14 of illness. This was followed by a dose of 3±5 mg/kg per day for an

430 B. Lang and C. M. Du€y

additional 2 months or until the erythrocyte sedimentation rate (ESR) and platelet count returned to normal.2,4 As a result, current recommendations for treatment during the acute phase of KD in North America, although not strictly adhered to, include the use of high-dose ASA (80±100 mg/kg per day).14 In contrast, in Japan, most patients in the acute phase of KD are treated with moderate doses of ASA (30±50 mg/kg per day).15 Two recent meta-analyses have found no signi®cant di€erence in the incidence of CAA between groups of patients treated with high (480 mg/kg/dose) versus lower (480 mg/kg/dose) doses of ASA with IVIG during the acute phase of KD.7,8 A US trial by Melish also found no signi®cant di€erence in incidence of CAA between patients treated with IVIG (2 g/kg) and either high-dose ASA (80±100 mg/kg per day) or lowdose ASA (3±8 mg/kg per day). However, the duration of fever was signi®cantly shorter in the high-dose ASA group.16 It should be noted that the `low-dose' regimen in this study, which provided only an antiplatelet e€ect, was considerably lower than the `low-dose' regimens studied in the meta-analyses which provided both antipyretic and antiplatelet e€ects.7,8 A North American study, published recently in abstract form,17 found an incidence of CAA of 7% in 230 patients treated with IVIG and lowdose ASA (3±5 mg/kg/day). Although ASA is generally well tolerated in patients with acute KD, concern about hepatic toxicity has led to the use of moderate-dose ASA (30±50 mg/kg per day) rather than high-dose ASA (100 mg/kg per day) in Japan.15 Gastritis and gastrointestinal bleeding are also concerns with high-dose ASA in children with KD.18 During the acute phase of KD, because of reduced absorption and increased renal clearance, therapeutic salicylate levels are rarely achieved, even with very high doses of ASA (120 mg/kg per day).19 Despite this, toxic free salicylate levels may occur as a result of the low albumin levels in the acute phase of KD.20 In the subacute phase, salicylate levels typically rise suddenly.20 Thus, in view of the risks of salicylate toxicity, the lack of evidence for signi®cant bene®t from anti-in¯ammatory doses, and the documentation of the early onset of thrombocytosis,21 many physicians prescribe anti-platelet doses prior to day 14 of illness. The earlier use of antiplatelet doses of ASA seems appropriate given the US trial ®ndings that acute symptoms of KD, for which one might otherwise use highdose ASA, usually resolve within 24±48 hours of receiving a single dose of IVIG (2 g/kg).2 As a result, many centres in North America reduce the dose of ASA to lowdose (3±5 mg/kg per day) once the patient has remained afebrile for 24±48 hours, and many other centres use low-dose ASA from the time of diagnosis. Should patients who do not meet diagnostic criteria for Kawasaki disease receive IVIG and ASA? It is well recognized that some patients who do not ful®l criteria for the diagnosis of KD (Table 2) will develop CAA. Patients with fever but fewer than four other diagnostic criteria are classi®ed as having `atypical' disease if they have CAA detected by echocardiography.22 In our view, these patients should receive standard treatment for KD. Patients with fever who ful®l fewer than four other criteria and who have no coronary artery abnormalities are classi®ed as `incomplete KD' if the diagnosis is suspected. The diagnosis of KD may be con®rmed subsequently if the patient ful®ls additional criteria or develops CAA. The decision to treat a patient with incomplete KD is more dicult and has not been studied prospectively. It is more problematic during winter months when viral illnesses with features similar to KD are common

Management of Kawasaki disease 431 Table 2. Diagnostic criteria for Kawasaki disease. Fever 5 5 days Plus four of the following: Bilateral non-purulent conjunctival injection Oropharyngeal changes Erythema and cracking of the lips Strawberry tongue Di€use oropharyngeal erythema Peripheral extremity changes Erythema of palms and soles Swelling and induration of hands and feet Desquamation of ®ngertips and toes in the subacute phase Polymorphous rash Enlarged cervical lymph node mass 5 1.5 cm in diameter

and treatment with IVIG might expose them to the unnecessary risks associated with receiving a blood product.23 In a study to identify factors that might be more predictive of the presence of KD, Burns et al24 found that the presence of anaemia, elevated transaminases and leukocyturia tended to be more frequent in KD but could also be associated with a viral illness. The clinician should have a very low threshold to treat `incomplete KD' if the patient is an infant under 6 months of age, an adolescent, or a child with prolonged fever, all of whom appear to have a higher risk for the development of CAA. How should a child diagnosed with KD after day 10 of illness be managed? In the published controlled trials of IVIG in KD, all patients received IVIG within the ®rst 10 days of their illness.2,4±6 There are no controlled studies to provide speci®c recommendations for treatment of patients who present for the ®rst time after day 10 of illness. An uncontrolled study that analysed data from the 1991±1992 Nationwide Survey of patients with KD in Japan reported a signi®cant increase in CAA in patients in whom IVIG was administered on or after day 9 of the illness. CAA were reported in 25% of patients treated between days 10 and 14 of illness, and 55% of those treated on or after day 15.9 The prevalence of giant CAA was also higher in patients treated on day 15 or later (6.1%) compared with patients treated earlier in the course of their illness (52%).9 In a small, retrospective study of 16 patients with established CAA who were ®rst treated after day 10 of their illness, all seven patients treated with IVIG plus ASA had resolution of CAA, compared with only four of nine patients treated with ASA alone.25 Prospective controlled studies are needed to con®rm the bene®t of IVIG treatment beyond day 10 of the illness, but to conduct such trials will be extremely dicult because numbers will be con®ned to patients who present late in the illness. Prolonged fever is a risk for the development of CAA,26,27 and if a patient with KD has ongoing fever, or other symptoms or laboratory features suggesting ongoing active in¯ammation, treatment with IVIG is standard practice in North America, regardless of the duration of untreated illness.25,28 In addition, if there is evidence of progressive coronary artery dilatation, IVIG treatment should be considered.25,28 There is no

432 B. Lang and C. M. Du€y

evidence that IVIG is of bene®t in patients who have already developed CAA after in¯ammation has resolved.28 Such patients should receive antiplatelet doses of ASA and echocardiographic follow-up. What is the recommended approach to the patient who fails to respond to IVIG? Any child with suspected KD who has persistent fever despite initial treatment with IVIG (2 g/kg) should be re-assessed to exclude the possibility of a bacterial infection. Patients with staphylococcal and streptococcal disease may meet criteria for KD, and particularly during a `mini-outbreak' of KD may incorrectly be given this diagnosis. Documenting bacteraemia and searching for a focus of infection may lead to the correct diagnosis and appropriate treatment. Other less common disorders which can mimic KD, such as other forms of systemic vasculitis, should also be excluded in the child who fails to respond to IVIG. Although the American Heart Association's current diagnostic guidelines for KD require exclusion of infectious illnesses prior to making a de®nite diagnosis of KD, there are several reports of patients with typical features of KD, including CAA, who also have had a documented infection.29±31 Approximately 10±30% of patients with KD fail to respond to the ®rst dose of IVIG.2,10,22,32 They either remain febrile 448 hours after IVIG or have a recurrence of fever 24±48 hours after IVIG.33 Three uncontrolled retrospective studies suggested that most of these patients become afebrile after a second infusion of IVIG.10,22,33 The study of Burns et al suggested a better outcome when a dose of 2 g/kg IVIG was used for re-treatment compared with 1 g/kg. However, the optimal re-treatment dose of IVIG has not yet been determined. In two retrospective studies,22,33 approximately one-third of patients receiving a second dose of IVIG had persistent fever. There are no generally accepted treatment guidelines for this small group of patients. Although a few respond to a subsequent infusion of IVIG, other immunomodulatory or immunosuppressive therapies have also been used.22,32,34,35 Three uncontrolled retrospective studies have reported response to intravenous methylprednisolone in patients refractory to repeated IVIG treatment.22,32,34 In two patients sustained improvement occurred only after the subsequent addition of intravenous cyclophosphamide.32 In a small uncontrolled study the use of pentoxiphylline with IVIG therapy was reported to have additional bene®t;36 however, controlled studies are needed. In the future there may be a role for competitive inhibitors of tumour necrosis factor-alpha (TNF-a), particularly in patients refractory to IVIG, but this has not yet been studied in humans with KD. What is the role for corticosteroids in KD? Although corticosteroids are standard treatment for most forms of systemic vasculitis, their role in KD remains controversial. Steroids have traditionally been contraindicated in KD, based primarily on early negative experience in Japan15 and an uncontrolled clinical trial by Kato et al.37 This study reported an increased incidence of CAA in patients treated with oral prednisolone (65%) compared with patients treated with ASA alone (11%). Because the number of patients treated with steroids was small (n ˆ 17), and because this study was uncontrolled and treatment groups were not strati®ed according to risk factors known to increase the frequency of CAA, the results of this study cannot be considered conclusive. Nevertheless, this study led to general acceptance that corticosteroids should not be used in KD.

Management of Kawasaki disease 433

Since then, several other uncontrolled studies have been published suggesting possible bene®t from corticosteroids in patients with KD and demonstrating no worse outcome in patients treated with corticosteroids compared with ASA. In a series of multicentre prospective randomized but uncontrolled studies carried out by the Kawasaki Disease Research Committee in Japan in 1981, similar incidences of CAA were found at 1 month follow-up in patients treated with ASA (22%) compared with prednisolone and dipyridamole (27%).15 There were no signi®cant di€erences in age and sex in the two treatment groups, and all the 101 patients in each group were treated within 7 days of onset of their illness.15 A small randomized controlled trial of 100 patients treated with either prednisolone (2 mg/kg per day) and ASA, or IVIG (400 mg/kg per day for 3 days) and ASA, failed to detect any signi®cant di€erence in incidence of CAA at 3-month follow-up. However, this may well have been due to the small number of patients studied.38 More recently corticosteroid use in three small retrospective series of patients with KD who failed to respond to IVIG resulted in prompt defervescence in most of these patients and no worsening of CAA.32,34,39 Intravenous methylprednisolone may also have a role in patients with life-threatening complications of KD. Successful treatment with corticosteroids has been reported in patients with severe myocarditis/pancarditis39±41, as well as in a patient with massive cervical adenopathy causing airway obstruction.42 Although corticosteroids are not recommended as standard therapy in patients with KD, and concerns linger regarding their procoagulant e€ects and possible interference with vascular healing, their role in KD deserves further study. LONG-TERM MANAGEMENT OF KD What is the optimal cardiac follow-up in patients who have had KD? Cardiac follow-up is essential in patients with KD because the long-term morbidity and the mortality associated with this disease are related to cardiac complications. Untreated KD can lead to myocardial infarction in early adulthood.43,44 In 1996, Burns et al reviewed the published literature on adult coronary artery disease attributed to prior KD and identi®ed 74 patients with long-term sequelae of KD.44 These patients presented between 12 and 39 years of age, most commonly with angina and myocardial infarction, and less often with sudden death or arrhythmia. Autopsy ®ndings included CAA in all patients and coronary artery occlusion in 72%. Although the long-term natural history of KD remains to be de®ned, many children (50±67%) with CAA will have angiographic evidence of regression within 6 months to 2 years after onset of their illness.45,46 A study by Kato et al, in which 594 consecutive children with KD between 1973 and 1983 were followed for 10±21 years, has helped to de®ne the natural history of this disease.46 Ischaemic heart disease developed in 4.7% of patients, myocardial infarction in 1.9% and death in 0.8%.46 Of 448 patients with normal early ®ndings on angiography, 190 were lost to follow-up; however, no cardiac symptoms or abnormalities were found in the 258 patients in whom 10±21-year follow-up was obtained. Of the 146 patients (25%) with CAA documented in the acute phase of KD, 55% showed regression. Small CAA were most likely to regress, while giant CAA were unlikely to regress. Stenosis developed in 28 patients, including almost half of the patients with giant CAA.46 Poor prognosis in the latter group has been previously reported; 71% of patients with giant CAA progressed to stenosis or obstruction over an 11-month follow-up period in one study.47 Most of the mortality associated with KD is seen in this group of patients.

434 B. Lang and C. M. Du€y

The American Heart Association (AHA) recommendations for cardiac follow-up in KD are shown in Table 3.48 These depend on the risk level assigned to a patient which is determined by the presence or absence of CAA, the size and number of CAA and the presence of coronary artery obstruction. An initial echocardiogram is recommended at the time of diagnosis of KD in all patients, followed by a second echocardiogram 6±8 weeks after disease onset. This allows detection of most CAA, which usually appear between the second and sixth week of illness. If no CAA are detected on these two echocardiograms, or if there is transient ectasia that resolves during the acute phase (risk levels I or II), the AHA recommends a subsequent echocardiogram 6±12 months after onset of the illness. However, the need for this follow-up echo has been debated as it is very unlikely that a CAA will be detected at 1 year if coronary arteries are normal at 2 months after onset of the illness.49 If a single CAA is detected (risk level III) annual echocardiograms are recommended for the ®rst decade of life, and if multiple small to medium, or one or more giant, CAA are detected (risk level IV), annual echocardiograms are recommended inde®nitely. As well, electrocardiograms and possibly pharmacological stress testing are recommended for patients in risk-group IV in the second decade of life. A minimum of six-monthly follow-up is recommended if a patient has evidence of coronary artery obstruction. Angiography is indicated if electrocardiograms or stress testing suggest ischaemia.

Table 3. Recommended long-term follow-up and management of KD (AHA Guidelines, 1994)a. No coronary artery lesions or transient ectasia ASA 3±5 mg/kg/day at least up to 6±8 weeks from disease onset No restrictions on physical activity beyond 8 weeks No invasive testing No follow-up necessary beyond ®rst year Individual physicians may choose to see patients at intervals of 3±5 years Solitary small-to-medium coronary artery lesions ASA 3±5 mg/kg/day at least until resolution of the lesion If 510 years of age, no restrictions on physical activity beyond 8 weeks If 510 years of age, physical activities guided by stress testing in alternate years, contact sports discouraged, non-contact sports permitted Annual follow-up with echocardiogram + ECG Angiography as indicated Giant, or multiple small coronary artery lesions, without obstruction ASA 3±5 mg/kg/day, inde®nitely, +coumadin If 510 years of age, no restrictions on physical activity beyond 8 weeks If 510 years of age, physical activities guided by stress testing annually, contact sports discouraged, noncontact sports permitted if stress test normal Annual follow-up with echocardiogram +ECG +chest radiograph Consider pharmacological stress testing for patients 510 years of age Angiography as indicated Coronary artery obstruction ASA 3±5 mg/kg/day, inde®nitely, +coumadin +calcium channel blockers Physical activities guided by stress testing or myocardial perfusion scans but contact sports and weight training to be avoided Six-monthly follow-up with echocardiogram ‡ ECG; Holter monitoring annually Angiography as indicated Surgery (bypass, transplantation) as indicated aReproduced

from Dajani AS et al (1994, Circulation 89: 916±922) with permission.

Management of Kawasaki disease 435

The best methods for long-term cardiac follow-up are currently a matter of opinion. Exercise stress testing has been reported to be insensitive in detecting evidence of CAA in KD.50 Dobutamine stress echocardiography is easily performed in young children and has increased sensitivity compared with exercise testing.51 Radionuclide techniques using thallium 201 or 99 m Tc-sestamibi have also been used to assess coronary perfusion after exercise or after coronary dilatation with dipyridamole.52,53 Positron emission tomography (PET) scanning provides a non-invasive method of assessment of myocardial blood ¯ow and has been reported to demonstrate lower estimates of myocardial ¯ow reserve in patients with KD.54,55 Further studies are clearly needed to determine optimal techniques for long-term assessment of patients with KD. This is particularly true in patients with regressed CAA, in whom echocardiography and even angiography are limited in their ability to detect abnormalities that may indicate increased risk of myocardial ischaemia.

Can long-term cardiac follow-up be discontinued in patients in whom CAA have resolved? According to the AHA guidelines, the risk level for a given patient with documented CAA may change over time as the morphology of the coronary arteries changes. Regression of aneurysms is assumed to reduce the risk of myocardial ischaemia,48 and one could interpret the AHA guidelines as suggesting that no long-term cardiac follow-up is necessary once CAA have resolved on echocardiography. However, pathological studies, and more recently, studies assessing morphological and functional arterial abnormalities, suggest that it may be prudent to follow patients long-term who have had CAA at any stage during their illness, regardless of regression.56±58 In a pathological study of regressed CAA, Tanaka et al found ®brous intimal thickening of arteries despite normal internal lumen diameters.56 Pathological ®ndings of marked intimal thickening and calci®cation in the intima similar to atherosclerotic lesions have also been reported by Kato et al.59 It appears that intimal thickening is the major mechanism of remodelling of a coronary artery aneurysm that leads to `resolution' of the aneurysm.46 Using intravascular ultrasound, Sugimura and Kato demonstrated intimal thickening of coronary arteries in 38 patients with KD who had persistent or regressed CAA.57 Reassuringly, the intima was normal in the two patients studied who had never had CAA. Suzuki et al found not only intimal thickening on intravascular ultrasound of 23 patients with documented CAA, but also reduced vasodilatory response to nitroglycerine. These abnormalities were found not only in areas of regressed aneurysms but also in some segments of coronary arteries that had not been previously identi®ed as abnormal.58 Furthermore, Paridon et al60 demonstrated inability of vessels to dilate normally in response to increased myocardial oxygen demand. In a study of 46 patients with previous KD, SPECT imaging with Tc-sestamibi at rest and at peak exercise demonstrated stress-induced myocardial perfusion defects in all patients with persistent CAA, two-thirds of patients with regressed aneurysms, and one-third of patients who had had no prior evidence of CAA.60 Long-term studies are needed to determine whether the intimal thickening which occurs during the healing process of CAA predisposes patients to late cardiac sequelae. Until large prospective studies are completed over the next several decades, longterm cardiac follow-up including patients with regressed CAA would seem to be appropriate.

436 B. Lang and C. M. Du€y

What drug therapy is recommended in patients with KD beyond the acute phase of disease? In the subacute phase of KD, low-dose ASA is used for its antiplatelet e€ect. In North America a dose of 3±5 mg/kg given once daily is usually started on day 14 of illness or once the patient has been afebrile for 24±48 hours.2 A dose of 5 mg/kg per day in the subacute phase is also standard therapy in Japan.46 According to the AHA guidelines, patients in whom CAA have never been detected and patients with transient ectasia which resolved during the acute illness should have ASA discontinued after 6±8 weeks. Other recommendations specify that the platelet count and erythrocyte sedimentation rate return to normal prior to discontinuing ASA.28 In patients with a single small-tomedium coronary artery aneurysm, low-dose ASA is continued at least until the coronary artery aneurysm resolves. In patients with multiple small-to-medium or one or more giant aneurysms, and in patients who develop coronary artery obstruction, long-term low-dose ASA is recommended with or without warfarin anticoagulation. In the latter group, the use of calcium channel blockers to reduce myocardial oxygen consumption should also be considered.48 Some investigators advocate the use of lowdose ASA inde®nitely for all patients in whom coronary artery dilatation or aneurysm formation were documented, even after resolution of the abnormalities on echocardiography.49 Although there are no prospective studies to support this approach, data demonstrating that the vessel wall is not normal in areas of regressed aneurysms suggest that this is a reasonable approach.46,57,58 What is the recommended level of activity restriction in patients who have had KD? According to the AHA guidelines, no restrictions for physical activity beyond the initial 6±8 weeks are recommended for patients who have never had CAA or have had only transient coronary artery ectasia in the acute phase. For patients with documented CAA no restrictions are recommended beyond the initial 6±8 weeks during the ®rst decade of life, unless there is evidence of obstruction. In the second decade, however, activity should be guided by stress tests. Competitive contact sports with endurance training are discouraged, and in patients with multiple aneurysms, or one or more giant aneurysms, all strenuous sports are strongly discouraged. In the presence of coronary artery obstruction, contact sports, isometrics and weight training should be avoided and physical activity recommendations should be guided by the outcome of stress testing or myocardial perfusion scans.48 What is the role for surgical intervention in patients with KD? Coronary artery bypass grafting has been successfully performed in children with KD, including infants less than 1 year of age.61 A large series of 170 Japanese patients with prior KD who underwent coronary artery bypass grafting has been reported by Kitamura et al.62 Approximately 75% of these patients received arterial grafts which have been found to have higher patency rates than saphenous vein grafts.62 Percutaneous transluminal coronary angioplasty and stenting are alternatives to bypass grafting in some patients. In 1997, Checchia and Shulman reviewed 13 patients worldwide who underwent cardiac transplantation for severe KD.63 Of the 10 patients with known follow-up 2 years later, nine were alive and well up to 6 years post-transplant and one had died of

Management of Kawasaki disease 437

rejection. Indications for transplantation included severe left ventricular dysfunction resulting from prior myocardial infarction, intractable arrhythmias and extensive distal CAA. Transplantations were performed within 6 months of onset of KD in four of the reported cases, 1 to 5 years post KD in three patients, and many years after KD in another three patients.63 Rarely, KD has resulted in the need for aortic or mitral valve replacement in patients with severe valvulitis and secondary insuciency. What long-term counselling is recommended for patients with KD who have CAA? Speci®c counselling for patients with known CAA following KD depends on the type, number and location of the lesions. Parents of children with ectasia or single small aneurysms can be reassured that most of these abnormalities will resolve spontaneously. Whether or not any residual changes in the healed arteries will predispose to later atherosclerosis is unknown. For patients with multiple CAA, or one or more giant CAA, the importance of careful cardiac follow-up and long-term antiplatelet therapy with or without anticoagulation should be emphasized. The importance of complying with restrictions in physical activities, especially related to sports, must be addressed particularly in children and adolescents in the second decade of life. All of these children and their families should be counselled regarding avoidance of other risk factors for future coronary artery disease, including smoking, diet and appropriate exercise. This should be enforced more strongly in those with any other additional risk factors for the development of ischaemic heart disease. What counselling is recommended for patients with KD in whom CAA have never been detected? There are limited data on which to base counselling recommendations for patients with KD in whom CAA have never been detected. Patients and families can be reassured that, in a large long-term outcome study from Japan, none of the 258 patients with normal coronary arteries after the acute stage of KD developed abnormalities during a follow-up period of 10±21 years. Furthermore, in a prospective study of a cohort of 6500 patients with KD by the Japanese Ministry of Health, no excess mortality has been detected after 10 years. However, data will need to be collected prospectively over the next several decades in this cohort in order to reach de®nite conclusions on longterm prognosis. Concerns regarding the possibility of functional abnormalities of coronary arteries in patients who have had no CAA detected on routine echocardiography have raised concerns that cardiac involvement in KD may be more common than is currently recognized by standard methods of cardiac evaluation.60 Although one recent Japanese study found that vascular function and morphology were normal in patients with previously normal coronary arteries on echocardiography,64 another study has suggested that KD may have not only long-term e€ects on the coronary arteries but also a generalized e€ect on arterial endothelial function. Dhillon et al found that endothelial-dependent dilatation of the brachial arteries, as assessed by high-resolution ultrasound, was markedly reduced in 29 patients 5±17 years post-KD compared with 20 age-matched controls.65 This study requires con®rmation, but suggests that long-term outcome studies in KD should be carried out not only in Japan but also in North America where other risk factors for cardiovascular and peripheral arterial disease may contribute to the ultimate prognosis in patients with KD.

438 B. Lang and C. M. Du€y

CONCLUSION Despite the di€erences between North America and Japan in the management of the acute phase of the disease, an increasing body of data supports the use of IVIG in a single dose of 2 g/kg, in addition to the use of ASA. Whether the latter should be 3±5 mg/kg/day, 30±50/kg/day or 80±100 mg/kg/day continues to be debated, although more recent data support the use of lower doses, with antiplatelet doses most appropriate for the period following defervescence. While there are no con®rmatory studies on the need to treat patients with atypical, incomplete or late presentations of KD, there is an increasing tendency among most practitioners to do so. Re-treatment with a second dose of IVIG (2 g/kg) is indicated in those who have failed to respond to a ®rst dose. There are few data in support of the use of a third dose of IVIG, and although a few patients may respond, a switch to pulse intravenous methyl prednisolone should be considered. Whether there is a role for corticosteroids in the initial management of the disease remains to be determined although this approach might bene®t those with signi®cant myocarditis. It is still not clear for how long patients with and without CAA should be followed. Nor is it clear what investigative tool should be used to determine risk of development of ischaemic heart disease. Long-term studies are needed to determine whether the intimal thickening which occurs during the healing process of CAA predisposes patients to late cardiac sequelae. Until large prospective studies are completed, long-term cardiac follow-up, including patients with regressed CAA, is appropriate. There are limited data on which to base follow-up recommendations for patients with KD in whom CAA have never been detected. For those with severe cardiac sequelae there are encouraging reports on the success of cardiac transplantation. Practice points . once the diagnosis of KD is established, or considered to be likely, treatment should be started with IVIG (2 g/kg over 8±12 hours) as a single infusion and highdose aspirin (ASA) (80±100 mg/kg per day) in four divided doses. Prompt institution of such treatment results in rapid resolution of the clinical and many of the laboratory features of the acute phase of KD, as well as a signi®cant reduction in the risk of subsequent development of coronary artery abnormalities (CAA) . in the subacute phase of KD, low-dose ASA is used for its antiplatelet e€ect. In North America a dose of 3±5 mg/kg given once daily is usually started on day 14 of illness or once the patient has been afebrile for 24±48 hours . according to the American Heart Association guidelines, patients in whom CAA have never been detected and patients with transient ectasia which resolves during the acute illness, should have their ASA discontinued after 6±8 weeks once their platelets and ESR return to normal . in patients with a single small-to-medium coronary artery aneurysm, low-dose ASA is continued at least until the coronary artery aneurysm resolves. In patients with multiple small-to-medium or one or more giant aneurysms, and in patients who develop coronary artery obstruction, long-term low-dose ASA is recommended with or without warfarin anticoagulation . while there are no con®rmatory studies on the need to treat patients with atypical, incomplete or late presentations of KD, there is an increasing tendency among most practitioners to do so

Management of Kawasaki disease 439

. the clinician should have a very low threshold to treat a patient with `incomplete KD' if the patient is an infant under 6 months of age, an adolescent, or a child with prolonged fever, all of whom appear to have a higher risk for the development of CAA . any child with suspected KD who has persistent fever despite initial treatment with IVIG (2 g/kg) should be re-assessed to exclude the possibility of a bacterial infection . approximately 10±30% of patients with KD fail to respond to the ®rst dose of IVIG. Re-treatment with a second dose of IVIG (2 g/kg) is indicated in these patients. There are few data in support of the use of a third dose of IVIG, and although a few patients may respond, a switch to pulse intravenous methyl prednisolone should be considered . cardiac follow-up is essential in patients with KD because the long-term morbidity and the mortality associated with this disease are related to cardiac complications. Untreated KD can lead to myocardial infarction in early adulthood . the American Heart Association has published recommendations for cardiac follow-up in KD which depend on the presence or absence of CAA, the size and number of CAA and the presence of coronary artery obstruction . parents of children with ectasia or single small aneurysms can be reassured that most of these abnormalities will resolve spontaneously. Whether or not any residual changes in the healed arteries will predispose to later atherosclerosis is unknown

Research agenda . what is the optimal dose of ASA during the acute phase of KD? . what is the role of corticosteroids in the treatment of KD? . does the intimal thickening which occurs during the healing process of CAA predispose to late cardiac sequelae in patients with regressed aneurysms? . does KD have a generalized e€ect on arterial endothelial function? . what are the optimal techniques for long-term assessment of patients with KD, particularly patients with regressed CAA in whom echocardiography and even angiography are limited in their ability to detect abnormalities that may indicate an increased risk of myocardial ischaemia? . will there be a signi®cant di€erence in the long-term outcome of children with KD in Japan compared with North America where other risk factors for cardiovascular and peripheral arterial disease may contribute to the ultimate prognosis in KD?

REFERENCES 1. Kawasaki T, Kosaki F, Okawa S et al. A new infantile acute febrile mucocutaneous lymph node syndrome (MLNS) prevailing in Japan. Pediatrics 1974; 54: 271±276. 2. Newburger JW, Takahashi M, Beiser AS et al. A single intravenous infusion of gamma globulin as compared with four infusions in the treatment of acute Kawasaki syndrome. New England Journal of Medicine 1991; 324: 1633±1639. 3. Furusho K, Kamiya T, Nakano H et al. High-dose intravenous gammaglobulin for Kawasaki disease. Lancet 1984; ii: 1055±1058.

440 B. Lang and C. M. Du€y 4. Newburger JW, Takahashi M, Burns JC et al. The treatment of Kawasaki syndrome with intravenous gamma globulin. New England Journal of Medicine 1986; 315: 341±347. 5. Harada K. Intravenous gamma-globulin treatment in Kawasaki disease. Acta Paediatrica Japonica 1991; 33: 805±810. 6. Morikawa Y, Ohashi Y, Harada K et al. A multicenter, randomized, controlled trial of intravenous gamma globulin therapy in children with acute Kawasaki disease. Acta Paediatrica Japonica 1994; 36: 347±354. 7. Durongpisitkul K, Gururaj VJ, Park JM et al. The prevention of coronary artery aneurysm in Kawasaki disease: a meta-analysis on the ecacy of aspirin and immunoglobulin treatment. Pediatrics 1995; 96: 1057±1061. 8. Terai M & Shulman ST. Prevalence of coronary artery abnormalities in Kawasaki disease is highly dependent on gamma globulin dose but independent of salicylate dose. Journal of Pediatrics 1997; 131: 888±893. 9. Yanagawa H, Nakamura Y, Sakata K et al. Use of intravenous gamma-globulin for Kawasaki disease: e€ects on cardiac sequelae. Pediatric Cardiology 1997; 18: 19±23. 10. Burns JC, Capparelli EV, Brown JA et al. Intravenous gamma-globulin treatment and retreatment in Kawasaki disease. US/Canadian Kawasaki Syndrome Study Group. Pediatric Infectious Disease Journal 1998; 17: 1144±1148. 11. Yanagawa H, Yashiro M, Nakamura Y et al. IV gamma globulin treatment of Kawasaki disease in Japan: results of a nationwide survey. Acta Paediatrica 1995; 84: 765±768. 12. Yanagawa H, Yashiro M, Nakamura Y et al. Results of 12 nationwide epidemiological incidence surveys of Kawasaki disease in Japan. Archives of Pediatric Adolescent Medicine 1995; 149: 779±783. 13. Yanagawa H, Nakamura Y, Yashiro M et al. Incidence survey of Kawasaki disease in 1997 and 1998 in Japan. Pediatrics 2001; 107: E33. 14. Dajani AS, Taubert KA, Gerber MA et al. Diagnosis and therapy of Kawasaki disease in children. Circulation 1993; 87: 1776±1780. 15. Kusakawa S & Tatara K. Ecacies and risks of aspirin in the treatment of the Kawasaki disease. Progress in Clinical and Biological Research 1987; 250: 401±413. 16. Melish ME. Comparison of low dose aspirin (LDA) vs. high dose aspirin (HDA) as an adjunct to intravenous gamma globulin (IVIG) in the treatment of Kawasaki syndrome (KS). Pediatric Research 1992; 31: 170A. 17. Scuccimarri R. Incidence of early cardiac abnormalities in children with Kawasaki disease treated with low dose aspirin and intravenous immunoglobulin in the acute phase. Arthritis & Rheumatism 2001; 44 (supplement 9): S169. 18. Matsubara T, Mason W, Kashani IA et al. Gastrointestinal hemorrhage complicating aspirin therapy in acute Kawasaki disease. Journal of Pediatrics 1996; 128: 701±703. 19. Koren G, Scha€er F, Silverman E et al. Determinants of low serum concentrations of salicylates in patients with Kawasaki disease. Journal of Pediatrics 1988; 112: 663±667. 20. Koren G, Silverman E, Sundel R et al. Decreased protein binding of salicylates in Kawasaki disease. Journal of Pediatrics 1991; 118: 456±459. 21. Scuccimarri R. Thrombocytosis occurs early in Kawasaki disease. Arthritis & Rheumatism 2001; 44 (supplement 9): S169. 22. Han RK, Sinclair B, Newman A et al. Recognition and management of Kawasaki disease. CMAJ 2000; 162: 807±812. 23. Bjoro K, Froland SS, Yun Z et al. Hepatitis C infection in patients with primary hypogammaglobulinemia after treatment with contaminated immune globulin. New England Journal of Medicine 1994; 331: 1607±1611. 24. Burns JC, Mason WH, Glode MP et al. Clinical and epidemiologic characteristics of patients referred for evaluation of possible Kawasaki disease. United States Multicenter Kawasaki Disease Study Group. Journal of Pediatrics 1991; 118: 680±686. 25. Marasini M, Pongiglione G, Gazzolo D et al. Late intravenous gamma globulin treatment in infants and children with Kawasaki disease and coronary artery abnormalities. American Journal of Cardiology 1991; 68: 796±797. 26. Koren G, Lavi S, Rose V et al. Kawasaki disease: review of risk factors for coronary aneurysms. Journal of Pediatrics 1986; 108: 388±392. 27. Ichida F, Fatica NS, Engle MA et al. Coronary artery involvement in Kawasaki syndrome in Manhattan, New York: risk factors and role of aspirin. Pediatrics 1987; 80: 828±835. 28. Melish ME. Kawasaki syndrome. Pediatric Reviews 1996; 17: 153±162. 29. Reller M, DeCristofaro J & Schwartz DC. Coronary aneurysms in a patient with atypical Kawasaki syndrome and a streptococcal infection. Pediatric Cardiology 1984; 5: 205±207. 30. Holm JM, Hansen LK & Oxhoj H. Kawasaki disease associated with parvovirus B19 infection. European Journal of Pediatrics 1995; 154: 633±634.

Management of Kawasaki disease 441 31. Johnson D & Azimi P. Kawasaki disease associated with Klebsiella pneumoniae bacteremia and parain¯uenza type 3 virus infection. Pediatric Infectious Disease 1985; 4: 100. 32. Wallace CA, French JW, Kahn SJ et al. Initial intravenous gammaglobulin treatment failure in Kawasaki disease. Pediatrics 2000; 105: E78. 33. Sundel RP, Burns JC, Baker A et al. Gamma globulin re-treatment in Kawasaki disease. Journal of Pediatrics 1993; 123: 657±659. 34. Wright DA, Newburger JW, Baker A et al. Treatment of immune globulin-resistant Kawasaki disease with pulsed doses of corticosteroids. Journal of Pediatrics 1996; 128: 146±149. 35. Raman V, Kim J , Sharkey A et al. Response of refractory Kawasaki disease to pulse steroid and cyclosporin A therapy. Pediatric Infectious Disease Journal 2001; 20: 635±637. 36. Furukawa S, Matsubara T, Umezawa Y et al. Pentoxifylline and intravenous gamma globulin combination therapy for acute Kawasaki disease. European Journal of Pediatrics 1994; 153: 663±667. 37. Kato H, Koike S & Yokoyama T. Kawasaki disease: e€ect of treatment on coronary artery involvement. Pediatrics 1979; 63: 175±179. 38. Nonaka. Randomized controlled study of intravenous prednisolone and gammaglobulin treatment in 100 cases with Kawasaki disease. Elsevier Science B.V, 1995: p 328±331. 39. Dale RC, Saleem MA, Daw S et al. Treatment of severe complicated Kawasaki disease with oral prednisolone and aspirin. Journal of Pediatrics 2000; 137: 723±726. 40. Bierman FZ & Gersony WM. Kawasaki disease: clinical perspective. Journal of Pediatrics 1987; 111: 789±793. 41. Haney I, Beghetti M, McCrindle BW et al. Ventricular arrhythmia complicating Kawasaki disease. Canadian Journal of Cardiology 1995; 11: 931±933. 42. Shetty AK, Homsi O, Ward K et al. Massive lymphadenopathy and airway obstruction in a child with Kawasaki disease: success with pulse steroid therapy. Journal of Rheumatology 1998; 25: 1215±1217. 43. Kato H, Inoue O, Kawasaki T et al. Adult coronary artery disease probably due to childhood Kawasaki disease. Lancet 1992; 340: 1127±1129. 44. Burns JC, Shike H , Gordon JB et al. Sequelae of Kawasaki disease in adolescents and young adults. Journal of the American College of Cardiology 1996; 28: 253±257. 45. Takahashi M, Mason W & Lewis AB. Regression of coronary aneurysms in patients with Kawasaki syndrome. Circulation 1987; 75: 387±394. 46. Kato H, Sugimura T, Akagi T et al. Long-term consequences of Kawasaki disease. A 10- to 21-year followup study of 594 patients. Circulation 1996; 94: 1379±1385. 47. Nakano H, Ueda K, Saito A et al. Repeated quantitative angiograms in coronary arterial aneurysm in Kawasaki disease. American Journal of Cardiology 1985; 56: 846±851. 48. Dajani AS, Taubert KA, Takahashi M et al. Guidelines for long-term management of patients with Kawasaki disease. Report from the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease, Council on Cardiovascular Disease in the Young American Heart Association. Circulation 1994; 89: 916±922. 49. Rowley AH & Shulman ST. Kawasaki syndrome. Pediatric Clinics of North America 1999; 46: 313±329. 50. Rhodes J, Hijazi ZM, Marx GR et al. Aerobic exercise function of patients with persistent coronary artery aneurysms secondary to Kawasaki disease. Pediatric Cardiology 1996; 17: 226±230. 51. Noto N, Ayusawa M, Karasawa K et al. Dobutamine stress echocardiography for detection of coronary artery stenosis in children with Kawasaki disease. Journal of the American College of Cardiology 1996; 27: 1251±1256. 52. Kondo C, Nakanishi T, Sonobe T et al. Scintigraphic monitoring of coronary artery occlusion due to Kawasaki disease. American Journal of Cardiology 1993; 71: 681±685. 53. Fukazawa M, Fukushige J, Takeuchi T et al. Discordance between thallium-201 scintigraphy and coronary angiography in patients with Kawasaki disease: myocardial ischemia with normal coronary angiogram. Pediatric Cardiology 1993; 14: 67±74. 54. Muzik O, Paridon SM, Singh TP et al. Quanti®cation of myocardial blood ¯ow and ¯ow reserve in children with a history of Kawasaki disease and normal coronary arteries using positron emission tomography. Journal of the American College of Cardiology 1996; 28: 757±762. 55. Ohmochi Y, Onouchi Z, Oda Y et al. Assessment of e€ects of intravenous dipyridamole on regional myocardial perfusion in children with Kawasaki disease without angiographic evidence of coronary stenosis using positron emission tomography and H2(15)O. Coronary Artery Disease 1995; 6: 555±559. 56. Tanaka N, Naoe S, Masuda H et al. Pathological study of sequelae of Kawasaki disease (MCLS). With special reference to the heart and coronary arterial lesions. Acta Pathologica Japonica 1986; 36: 1513±1527. 57. Sugimura T, Kato H, Inoue O et al. Intravascular ultrasound of coronary arteries in children. Assessment of the wall morphology and the lumen after Kawasaki disease. Circulation 1994; 89: 258±265. 58. Suzuki A, Yamagishi M, Kimura K et al. Functional behavior and morphology of the coronary artery wall in patients with Kawasaki disease assessed by intravascular ultrasound. Journal of the American College of Cardiology 1996; 27: 291±296.

442 B. Lang and C. M. Du€y 59. Kato H, Akagi T , Sugimura T et al. Kawasaki disease. Coronary Artery Disease 1995; 6: 194±206. 60. Paridon SM, Galioto FM, Vincent JA et al. Exercise capacity and incidence of myocardial perfusion defects after Kawasaki disease in children and adolescents. Journal of the American College of Cardiology 1995; 25: 1420±1424. 61. Dalzell V. Coronary artery bypass grafting for Kawasaki disease at Children's Memorial Hospital, Chicago. Pediatric Research 1997; 41: 118A. 62. Kitamura S, Kameda Y, Seki T et al. Long-term outcome of myocardial revascularization in patients with Kawasaki coronary artery disease. A multicenter coopereative study. Journal of Thoracic and Cardiovascular Surgery 1994; 107: 663±673, discussion 673±674. 63. Checchia PA, Pahl E , Shaddy RE et al. Cardiac transplantation for Kawasaki disease. Pediatrics 1997; 100: 695±699. 64. Iemura M, Ishii M , Sugimura T et al. Long term consequences of regressed coronary aneurysms after Kawasaki disease: vascular wall morphology and function. Heart 2000; 83: 307±311. 65. Dhillon R, Clarkson P , Donald AE et al. Endothelial dysfunction late after Kawasaki disease. Circulation 1996; 94: 2103±2106.