Treatment of Kawasaki syndrome: A comparison of two dosage regimens of intravenously administered immune globulin Karyl S. Barren, MD, Daniel J. Murphy, Jr., MD, Earl D. Silverman, MD, Herbert D. R u t t e n b e r g , MD, G r e g o r y B. Wright, MD, W a y n e Franklin, MD, Stanley J. G o l d b e r g , MD, Stanley M. Higashino, MD, D i a n n e G. Cox, BS, a n d Martin Lee, Phb From Texas Children's Hospital and the Department of Pediatrics, Baylor College of Medicine, Houston, Texas;the Hospital for Sick Children, Toronto, Ontario, Canada; the Primary Children's Medical Center and the Department of Pediatrics, University of Utah School of Medicine, Salt Lake City; the Minneapolis Children's Medical Center, Minneapolis, Minnesota; the Children's Hospital of Philadelphia, Philadelphia, Pennsylvania; the Department of Pediatrics, University of Arizona, Tucson; Children's Hospital Medical Center, Oakland, California; and Baxter Healthcare Corporation, Hyland Division, Glendale, California Because intravenously administered immune globulin (IVIG) is effective in reducing the i n c i d e n c e of coronary artery aneurysms in Kawasaki syndrome when given at a dose of 400 m g / k g daily for 4 days, we undertook a multicenter clinical trial c o m p a r i n g two d o s a g e regimens of IVIG. Patients were randomly assigned to receive IVIG at either 400 m g / k g daily for 4 days (22 patients) or I g m / k g as a single dose (22 patients). All patients received aspirin therapy, and all were enrolled within 7 days of onset of fever. The presence of coronary artery aneurysms was e v a l u a t e d by means of two-dimensional e c h o c a r d i o g r a p h y before infusion; at days 4 to 6, 14 to 21, and 42 to 49 after infusion; and at I year. Coronary artery aneurysms were d e t e c t e d in 3 of the 44 patients, including one patient receiving 400 m g / k g and two patients receiving 1 g m / k g (p value not significant). No giant aneurysms were d e t e c t e d . No major side effects occurred with either d o s a g e regimen. Patients receiving the I g m / k g dose had a faster resolution of fever and were discharged from the hospital a p p r o x i m a t e l y I d a y sooner than the 400 m g / k g group (p = 0.01). Although the relatively small sample size in this trial does not allow for a more definitive statement regarding the o c c u r r e n c e of coronary artery aneurysms, it appears that the 1 g m / k g dose is associated with a more rapid clinical improvement and a shorter hospital stay. (J PEDIATR1990;117:638-44)
High-dose aspirin therapy has until recently been accepted as standard therapy in the treatment of Kawasaki syndrome. Although aspirin may be effective in reducing fever Supported by a grant from Baxter Healthcare Corporation, Hyland Division, Glendale, Calif., and the American Red Cross. Submitted for publication July 18, 1989; accepted April 11, 1990. Reprint requests: Karyl S. Barron, MD, Pediatric Rheumatology, 6621 Fannin St., Houston, TX 77030. 9/25/21525 638
and inflammation in the acute phase of the disease, there is 2 15% to 25% incidence of coronary artery abnormalities. Clinical studies have shown that intravenously administered immune globulin, used concomitantly with aspirin, is effective in reducing the incidence of coronary artery abnormalities. 1-7 However, the optimal dose and treatment schedule of IVIG remain unknown. Our study was designed as a preliminary trial to investigate the efficacy and safety of two different treatment regimens of IVIG in preventing the development of coronary artery abnormalities in chil-
Volume 117 Number 4
ALT AST IVIG KS
IVIG for Kawasaki syndrome
Alanine aminotransferase Aspartate aminotransferase Intravenously administered immune globulin Kawasaki gyndrome
dren with KS. The primary goal was to show no clinically significant difference between these two regimens. METHODS
Patient selection. Patients were enrolled from January 1987 through April 1988 at seven centers in the United States and Canada. Kawasaki syndrome was defined by the presence of fever for at least 3 consecutive days and at least four of the five following features: (1) changes in the oropharynx, including reddening and fissuring of the lips, hyperemia of the buccal mucosa, or "strawberry" tongue, (2) nonexudative conjunctivitis, (3) erythema of the palms and soles, edema of the hands and feet, or periungual desquamation in the subacute phase of the disease, (4) rash, and (5) cervical lymphadenopathy, defined as one or more nodes at least 1.5 cm in diameter. All patients were enrolled in the study within 7 days of the onset of KS, which was defined as the first day of fever. The patient was excluded if there was clinical or laboratory evidence of any other disease known to mimic KS. Other exclusion criteria included known heart disease (e.g., congenital malformations or rheumatic carditis), recurrent KS, known immunologic defects, or prior treatment with an immune globulin product or any immunomodulating agents. Informed consent was obtained from parents. The protocol was approved by the human subjects committee of each institution. Study design. Patients meeting the entry criteria were randomly assigned in blocks of four to receive one of the two dosage regimens of IVIG. Each institution was provided with a randomization schedule, and patients were assigned in sequential order to the proper treatment group. Patients received either IVIG at 1 gm/kg, administered as a single dose for 5 to 8 hours, or at 400 mg/kg, administered once daily for 2 to 4 hours for 4 consecutive days. The IVIG used was Gammagard (Baxter Healtheare Corp., Hyland Division, Glendale, Calif.), a highly purified preparation of human IgG derived from the cold ethanol fractionation process and further purified by ion exchange adsorption. 8 The manufacturing process does not involve chemical or enzymatic modification procedures; the Fc portion remains intact. Patients in both treatment groups received aspirin at an initial dosage of 80 to 100 m g / k g / d a y divided every 6 hours. Salicylate levels were obtained on days 4 to 6 of the study. An attempt was made to adjust the dose to achieve a minimal salicylate blood level of 15 mg/dl. The high dose of aspirin was continued until 24 hours after resolution of fe-
639
ver, and then reduced to 3 to 5 m g / k g / d a y and continued for a minimum of 2 months. Vital signs were monitored before, during, and after infusion. Any adverse reactions were noted. Maximum temperature was recorded daily for the duration of the hospitalization. A requirement for discharge was that the patient be afebrile for at least 24 hours. Laboratory evaluation. Laboratory evaluations were performed at entry into the study and at days 4 to 6, 14 to 21, and 42 to 49 after enrollment. Complete blood cell and differential cell counts, platelet count, sedimentation rate, alantitrypsin level, and quantitative immunoglobulin determinations were obtained at each visit. In addition, A L T and AST levels were measured at entry, days 14 to 21, and days 42 to 49. Total serum bilirubin concentration was determined at entry and at days 42 to 49. Urinalysis was performed and blood urea nitrogen value obtained at entry. Cardiac evaluation. Cardiac evaluation was performed at entry into the study (before infusion) and at 4 to 6, 14 to 21, and 42 to 49 days after infusion. Echocardiographic evaluation of the coronary arteries included morphologic evaluation of the left main, left anterior descending, circumflex, right, and posterior descending coronary arteries. The diameter of each coronary artery segment was measured from inner surface to inner surface. Coronary artery segments were graded as follows: normal (uniform appearance, ~4 mm diameter); or aneurysmal (localized increase > 150% of adjacent vessel diameter). Further evaluation included measurement of left ventricular dimension and shortening fraction, examination of the pericardial space, and Doppler assessment of the presence and degree of any mitral or aortic insufficiency. Each echocardiogram was recorded on videotape and reviewed by two observers who were unaware of patient identity. If there was a discrepancy in assessment between the two reviewers, the tape was assigned to a third reader for review in a blind fashion. Criteria for determining safety and efficacy. Any change in vital signs (respiration, pulse, blood pressure, and temperature) was noted, as well as chills, pain, nausea, emesis, headache, anxiety, flushing, chest tightness, abdominal cramps, wheezing, dizziness, or rash. The primary criterion for assessing the efficacy of IVIG administered at 1 gm/kg as a single dose, in comparison with 400 mg/kg administered daily for 4 days, was the difference in the incidence of coronary artery aneurysms. Differences between the two treatment groups in laboratory values, duration of fever, and reduction of clinical symptoms associated with the disease were also compared. Data analysis. The incidence of aneurysms in the two treatment groups was compared by chi-square analysis. Some quantitative variables, such as laboratory values,
640
Barron et al.
The Journal o f Pediatrics October 1990
T a b l e I. Patient characteristics on entry into the study
Gender (M/F) Age (mo) Mean _+ SD Range Children 30,000/mm 3 Hgb 500,000/ram 3 ESR >100 mm/hr arAntitrypsin (mg/dl)*
Group 4: I gm/kg (n=25)
Group 2: 400 mg/kg • 4 days (n=22)
P
17/8
9/13
0.06
38.2 _+ 22.2 10-111 3 5.2 _+ 0.8 0 3 4 1 368.9 _+ 72.6
30.1 _+ 19.1 9-98 1 5.0 _+ 1.3 2 4 8 2 390.7 _+ 97.8
0.48 0.48 0.53 0.12 0.55 0.06 0.48 0.39
WBC. Whitebloodcell count;Hgb, hemoglobin;ESR, erythrocytesedimentationrate. *Valuesare expressedas mean _+SD.
T a b l e II. Echocardiographic findings Aneurysm Treatment group
No.
Location
Size (mm)
Study day when first detected
1 gm/kg 1 gm/kg 400 mg/kg • 4 days
1 1 4
Left main CA Proximal right CA Left main CA Circumflex CA LAD CA Proximal right CA
4 5 5 4 4 5
4 21 15
CA. Coronaryartery; LAD. left anteriordescendingartery.
were tested for statistical significance by the independent t test (two-tailed comparison). Others, such as febrile days and duration of hospital stay, were examined by the MannWhitney (Wilcoxon) test. The p values --30,000/ram 3, number of patients with a hemoglobin level 8 mm) aneurysms. Once visualized, the aneurysms persisted at least until days 42 to 49 of the study. There was no statistical difference in rate of aneurysm formation between the two groups of patients. The approximate 95% confidence interval for this difference is -0.10 to 0.19. At 1-year follow-up, we could no longer visualize aneurysms by echocardiography in the two patients with solitary aneurysms who had received IVIG in a single 1 g m / kg dose. The patient with multiple aneurysms who had received IVIG in a dosage regimen of 400 mg/kg daily for 4 days had a normal-appearing right coronary artery shown by echocardiogram at 1-year follow-up; however, the other aneurysms were still visualized. Thus at 1-year follow-up, only 1 of the 44 patients had persistent aneurysms detected by echocardiogram. Nine patients had a pericardial effusion detected on the first (before infusion) echocardiogram, including six patients in the 1 gm/kg group and three in the 400 mg/kg group (p = NS). In most instances the pericardial effusion was not detected after days 4 to 6 of the study. None of the patients with aneurysms had evidence of pericardial effusion during the study. Six patients had evidence of mitral insufficiency at entry into the study, including three patients receiving 1 gm/kg and three on the 400 mg/kg regimen. In three patients, the mitral insufl%iency resolved by days 4 to 6 of the study; in two patients (one from each group) the insufficiency resolved before the third echocardiogram, and in one patient (1 gm/kg) the insufficiency persisted until the third echocardiogram. The one patient with multiple aneurysms had evidence of transient mitral insufficiency on entry into the study. No patients had evidence of aortic insufficiency. There was no significant difference in left ventricular shortening fraction between the two groups at study entry (1 gm/kg -- 34.3 + 4.6% vs 400 mg/kg = 35.1 +__5.3%; p = 0.63). There was also no difference in shortening fraction between the groups at the other 3 study days, and shortening fraction did not change significantly from one study day to the next in either group. Clinical course. Both groups of patients had a fairly rapid disappearance of fever after intravenous infusion of im-
IVIG for Kawasaki syndrome
64l
800
,,,, . ~ I ~
700
a. Platelets
600 cq E E
500 400 300 b=.08
200
-- m
400 mg/Kg +SEM 9 lg/Kg++SEM
100
i
i
0
I
8O i
M,,
70
601
.....
~
~
~
b, ESR
~
50 =.72
4O E
30 2O 10 i
0
i
40o] - ~ . . . .
~.. ~~ ~" p =~- 4 2l ~ " _' ~ " ,
i
C. Alpha-1 Antitrypsin
-.,
~176176
-
100 ]-
'~
.
Day 0
.
.
Day 4-6
.
Day 14-21
Day 42-49
Day of Study
Fig. 1. Laboratory parametersobtained at day 0, clays4 to 6, days 14 to 21, and days 42 to 49 of study for patients on IVIG dosage regimens of 400 mg/kg daily for 4 days (----) and 1 gm/kg ( _ _ ) . a, Platelet count; b, erythrocyte sedimentation rate; e, cqantitrypsin level. Note p values.
mune globulin, with a more rapid decline in the 1 gm/kg group. Between the first and second study days, maximum temperature fell an average of 1.41 ~ C. in the 1 gm/kg group, compared with 0.78 b C. in the 400 mg/kg group (p = 0.013). Patients were discharged once they were afebrile for at least 24 hours, so there was a significant difference in the number of days of hospitalization, with the 1 gm/kg group discharged at a median of 4 days after entry into the study and the 400 mg/kg group at 5 days (p = 0.01).This difference at least partially reflects the fact that patients receiving the four doses required a minimum of 4 days of hospitalization. Laboratory evaluation. Platelet count increased in both groups on days 4 to 6 of the study. Thereafter the platelet count gradually decreased. There was no significant difference in platelet count at these various periods for patients receiving either dosage regimen of IVIG (Fig. 1, a). There was a slight increase in erythrocyte sedimentation rate at days 4 to 6 of the study in the 400 mg/kg group of patients,
64 2
Barron et al.
2,000
}~
1,8oo
~
/
a. IgG ~" ~.
1,600
, 1,4oo
- -
The Journal o f Pediatrics October 1990
/
1,000
800 / r 600 ~
p=.03
400 mO/Kg+_SEM I g/K~_ SEM
--
I
I
I
160
14 2o ~
b. lgA
100
60 40
, "~" ~=.03
p=.34
~ ~,~"~ p=.12
20 0
I
I
I
I
I
200 180 160 140
_== 120if/ 100
p=.l I 80 6Day 0
p=.10 Day 4-6
Day 14-21
I
Day 42-49
Day of Study Fig. 2. Mean serum immunoglobulin level obtained at day 0, days 4 to 6, days 14 to 21, and days 42 to 49 for patients on IVIG dosage regimens of 400 mg/kg daily for 4 days (..... ) and 1 gm/kg ( ). a, Mean serum IgG level; b, mean serum IgA level, e, mean serum IgM level. Note p values.
with a subsequent fall to normal values (Fig. 1, b). The erythrocyte sedimentation rate for the 1 gm/kg group did not statistically differ from that in the 400 mg/kg group at any point in the study. There was a gradual decrease in alantitrypsin concentration after infusion in both groups (Fig. 1, c), and no statistical difference between the two groups. There was a transient mild increase in ALT and AST values on days 4 to 6 of the study in two patients (one receiving 1 gm/kg and one receiving 400 mg/kg), with subsequent resolution by days 14 to 21 of the study. No increase in serum bilirubin concentration followed infusion of either dose of 1VIG. Both groups had a significant rise in serum IgG values. Despite a significantly lower mean IgG level in the 400 mg/kg group at entry into the study, the mean level at days 4 to 6 of the study was significantly higher in the 400 mg/ kg group (Fig. 2, a). Thereafter the serum IgG level for both groups gradually declined but remained significantly greater than pretreatment levels at days 42 to 49. The mean serum IgA levels increased at days 4 to 6 of the study in both
groups, despite negligible amounts of IgA in the IVIG preparation (Fig. 2, b). Thereafter, there was a gradual decline in mean serum IgA levels. At days 42 to 49 of the study, the mean serum IgA level was significantly lower than the pretreatment level in the 1 gm/kg group but not in the 400 mg/kg group; in four patients (one receiving 1 gm/kg and three on the 400 mg/kg regimen), serum IgA levels became markedly less than age-related normal values (with development of an absolute deficiency in one child). Follow-up of two of these children revealed normalization of serum IgA levels during the subsequent few months. Mean serum IgM levels were significantly increased at days 4 to 6 and days 14 to 21 of the study, compared with day 0 for both groups of patients. This transient increase in serum IgM was not due to passive transfer of IgM, because there were negligible amounts of IgM detectable in the IVIG preparation. By days 42 to 49 of the study, mean serum IgM levels did not differ significantly from pretreatment levels. Adverse reactions. None of the patients experienced serious adverse reactions to IVIG. Four children receiving 1 gm/kg had mild adverse reactions. One child had mild flushing that required no change in the infusion rate. Another patient had mild flushing, chills, and nausea and vomiting, which required temporarily slowing the infusion rate until symptoms resolved. One child had mild hypotension, which responded to stopping of the infusion temporarily and administration of 0.9% saline solution (10 ml/kg). In the fourth child a morbilliform rash developed 3 days after the infusion but resolved in 2 days without therapy. Three children on the 400 mg/kg regimen had mild adverse reactions. Two hours into the fourth infusion, one child had shaking chills, which resolved spontaneously. A second child had chills and noisy breathing 11/2 hours into the first infusion and responded to stopping of the infusion temporarily; this problem did not recur with subsequent infusions. A third child had headache, flushing, and abdominal cramping 1V2hours into the second infusion. The infusion had infiltrated; it was stopped and then restarted at a slower rate. No patient had clinical signs of congestive heart failure. DISCUSSION A current theory is that KS is an immunologically mediated generalized vasculitis that is triggered in a susceptible host by a variety of common infectious agents prevalent in the community. 9 It may be postulated that there is a predisposition to abnormal immune responsiveness after an asyet-unrecognized challenge to the immune system. The list of suspected infectious causes of KS is long, but no etiologic agent has been clearly identified. 1~ Currently, most therapies are based on decreasing the inflammatory response. Most standard treatment protocols involve the use of as-
Volume 117 Number 4
pirin. Most of our patients became afebrile soon after the start of IVIG therapy, regardless of dosage regimen, so the duration of treatment with high doses of aspirin in most children was just a few days. As a result, the maximum salicylate level obtained in most children was less than 10 mg/dl. Furusho et al. 1' 2 first reported the efficacy of IVIG in KS. Newburger et al. 4 described a multieenter collaborative controlled study demonstrating the efficacy of IVIG plus aspirin, compared with aspirin alone, in reducing the frequency of coronary artery abnormalities. They also reported more rapid resolution of fever and more rapid return of laboratory measurements to normal in the acutephase response. Engle et al. 7 recently reported an uncontrolled, one-arm study in which all patients received a 1 gm/kg dose of IVIG. Non e had subsequent aneurysms, and fever abated in 27 of the 32 patients within the first day after treatment; however, five children required additional IVIG. It is difficult to compare the various IVIG studies because of different patient populations, differing definitions of coronary artery abnormalities, and differing doses of IVIG, aspirin, or both. It has become apparent, however, that the dose of IVIG may be an important determinant of efficacy. In various other studies, 6' 17, 18 smaller doses of IVIG were ineffective in significantly reducing the incidence of coronary artery abnormalities. In addition, the efficacy of IVIG may be related to the number of days of illness at the time of infusion. 19 The incidence of aneurysms in our study compares favorably with that in other IVIG studies. Our findings suggest that IVIG given as a single high dose, 1 gm/kg, may be as efficacious as a regimen of 400 mg/kg daily for 4 days. The 95% confidence interval for the difference in the incidence rates of coronary aneurysms in the two groups (-0.10 to 0.19) suggests that the rate for the 1 gm/kg regimen may be lower by as much as 10% or higher by as much as 19% compared with the 400 mg/kg regimen; the wide range is a result of the relatively small sample size. Perhaps most important is the fact that aneurysms were detected in only one child at 1-year follow-up. The duration of hospitalization may also be shortened with the single dose, thereby increasing the cost-effectiveness of this regimen, which requires less total IVIG. ,Both regimens of our therapeutic protocol were well tolerated; only minor adverse reactions were noted in a small number of children. The effects on erythrocyte sedimentation rate, platelet count, and al-antitrypsin level were similar with the two dosage regimens. There was an expected increase in serum IgG levels after infusion of the IVlG in both treatment groups. The serum IgG levels were still elevated at days 42 to 49 of the study, an expected finding because the half-life of IgG is approximately 25 days. An unexpected finding was the increase of both IgA and IgM af-
IVIG for Kawasaki syndrome
643
ter the infusion of an IVIG preparation that is almost entirely IgG. More surprising was the marked fall in serum IgA levels in four patients by days 42 to 49 of the study. Whether these changes are a reflection of the action of IVIG or an immunologic change intrinsic to KS needs further evaluation. Significant morbidity and mortality rates in KS have been limited primarily to patients with coronary artery aneurysms. The use of IVIG appears to reduce the overall incidence of coronary artery aneurysms and the occurrence of giant aneurysms, which are associated with increased morbidity and mortality rates. In this study, two of the three patients in whom small solitary coronary artery aneurysms developed had ech0cardiographic resolution of their lesions at 1-year follow-up, but such resolution does not eliminate the risk of eventual development of coronary stenosis. Therefore the first goal of therapy should be to prevent the formation of coronary artery aneurysms. The population that should receive IVIG therapy remains unclear. Aneurysms do not develop in approximately 80% of patients. Therefore the use IVIG for every patient with KS, as recommended by some authors, 4' 2o may be unnecessarily costly. It would be ideal if, before undertaking therapy of this type, clinicians could identify prospectively the patients in whom aneurysms or long-term sequelae would ultimately develop. The substrate for aneurysm formation occurs in the first days of the illness, when perivaseular inflammation gives way to a panvasculitis, so it is critical t h a t high-risk patients be differentiated early in their clinical course. Unfortunately, none of the methods now available is sufficiently helpful to allow early identification of all patients at risk. The dose of IVIG for use in KS also remains a subject of investigation and controversy. The earlier reduction in fever in the 1 gm/kg group in this study raises the possibility that the general antiinflammatory effect of this regimen is at least as potent as that of the regimen of 400 mg/kg daily for 4 days. Although the relatively small sampl e size in this trial does not allow for a definitive statement regarding the occurrence of coronary artery aneurysms in patients with KS, it appears that the use of IVIG in a dose Of 1 gm/kg is associated with a more rapid clinical improvement and a shorter hospital stay. Definitive therapy for patients with KS and, ultimately, prevention of this disorder will not be possible until the cause and pathogenesis can be explicitly defined. Documentation of the effectiveness of IVIG in ameliorating the course of the disease and its complications in selected patients may suggest new approaches for research into the pathophysiologic nature of this disease. We acknowledge and thank the following persons who participated in the clinical trials: Ivonne Arroyo, MD, and Martha Curry, RN, Houston; Norman N. Musewe, MD, Toronto; Frederic Stone,
644
Barron et al.
MD, and Donald L. Uden, PharmD, Minneapolis; Victoria Vetter, MD, Gregg HeRon, MD, and Jennifer Puck, MD, Philadelphia; and Christian E. Hardy, MD, Oakland, Calif. We thank Masato Takahashi, MD, for his continued assistance and interest in this project; Fred Bierman, MD, for technical contributions in the field of echocardi01ogy; Theodore Colton, PhD, and Edward Giannini, DrPH, for statistical services; Marian Melish, MD for assistance in study design; and Robert Warren, MD, PhD, for his critical review of the manuscript.
The Journal of Pediatrics October 1990
9.
10. 11. 12.
REFERENCES
1. Furusho K, Sato K, Soeda T, et al. High-dose intravenous gammaglobulin for Kawasaki disease [Letter]. Lancet 1983;2:1359. 2. Furusho K, Hiroyuki N, Shinomiya K, et al. High-dose intravenous gammaglobulin for Kawasaki disease. Lancet 1984; 2:1055-8. 3. Nagashima M, Matsushima M, Matsuoka H, Ogawa A, Okumura N. High-dose gammaglobulin therapy for Kawasaki disease. J PEDIATR 1987;110:710-2. 4. Newburger JW, Takahashi M, Burns JC, et al. The treatment of Kawasaki syndrome with intravenous gamma globulin. N Engl J Med 1986;315:341-7. 5. Isawa M, Sugiyama K, Kawase A, et al. Prevention of coronary artery involvement in Kawasaki disease by early intravenous high-dose gamma globulin. In: Doyle EF, Engle MA, Gersony WM, Rashkind W J, Talner NS, eds. Pediatric Cardiology. New York: Springer-Verlag, 1986:1083-5. 6. Okuni M, Harada K, Yamaguchi H, et al. Intravenous gamma globulin therapy in Kawasaki disease: trial of low-dose gamma globulin. In: Shulman ST, ed. Kawasaki disease. New York: Alan R Liss, 1987:433-9. 7. Engle MA, Fatica NS, Bussel JB, O'Loughlin JE, Snyder MS, Lesser ML. Clinical trial of single-dose intravenous gamma globulin in acute Kawasaki disease. Am J Dis Child 1989; 143:1300-4. 8. Ochs HD, Lee ML, Fischer SH, Kingdon HS, Wedgwood RJ.
13.
14. 15.
16.
17.
18.
19.
20.
EfficacY of a new intravenous immunoglobulin in primary immunodeficient patients. Clin Ther 1987;9:512-22. Barron KS, DeCunto CL, Montalvo JM, Orson FM, Lewis DE. Abnormalities of immunoregulation in Kawasaki syndrome. J. Rheumatol 1988;15:1243-9. Feigin RD, Barron KS. Treatment of Kawasaki syndrome. N Engl J Med 1986;315:388-90. Carter RF, Haynes ME, Morton J. Rickettsia-like bodies and splenitis in Kawasaki disease. Lancet 1976;2:1254-5. Hamashima Y, Tasaki K, Fujiwara H, Hoshino T, Matsuda S, Kao T. Kawasaki disease: its pathological features and possible pathogenesis. Aeta Paediatr Jpn 1983;25:108- t 7. Burns JC, Geha RS, Schneeberger EE, et al. Polymerase activity in lymphocyte culture supernatants from patients with Kawasaki disease. Nature 1986;2:323-814-6. Shulman ST, Rowley AH. Does Kawasaki disease have a retroviral etiology? Lancet 1986;2:545-6. Melish ME, Marchette N J, Kaplan JC, Kihara S, Ching D, Ho DD. Absence of significant RNA-dependent DNA polymerase activity in lymphocytes from patients with Kawasaki syndrome. Nature 1989;337:268-90. Burns JC, Huang AS, Newburger JW, et al. Characterization of the polymerase activity associated with cultured peripheral blood mononuclear cells from patients with Kawasaki disease. Pediatr Res 1990;27:109-12. Furusho K, Kamiya T, Nakano H, et al. Japanese gamma globulin trials for Kawasaki disease. In: Shulman ST, ed. Kawasaki disease. New York: Alan R Liss, 1987:425-32. Ogino H, Ogawa M, Harima Y, et al. Clinical evaluation of gammaglobulin preparations for the treatment of Kawasaki disease. In: Shulman ST, ed. Kawasaki disease. New York: Alan R. Liss, 1987;555-6. Nakano H, Ueda K, Saito A, et al. Scoring method for identifying patients with Kawasaki disease at high risk of coronary artery aneurysms. Am J Cardiol 1986;58:739-42. Plotkin SA, Daum RS, Giebink GS, et al. Intravenous gamma-globulin use in children with Kawasaki disease. Pediatrics 1988;82:122.
High-dose aspirin therapy has until recently been accepted as standard therapy in the treatment of Kawasaki syndrome. Although aspirin may be effective in reducing fever Supported by a grant from Baxter Healthcare Corporation, Hyland Division, Glendale, Calif., and the American Red Cross. Submitted for publication July 18, 1989; accepted April 11, 1990. Reprint requests: Karyl S. Barron, MD, Pediatric Rheumatology, 6621 Fannin St., Houston, TX 77030. 9/25/21525 638
and inflammation in the acute phase of the disease, there is 2 15% to 25% incidence of coronary artery abnormalities. Clinical studies have shown that intravenously administered immune globulin, used concomitantly with aspirin, is effective in reducing the incidence of coronary artery abnormalities. 1-7 However, the optimal dose and treatment schedule of IVIG remain unknown. Our study was designed as a preliminary trial to investigate the efficacy and safety of two different treatment regimens of IVIG in preventing the development of coronary artery abnormalities in chil-
Volume 117 Number 4
ALT AST IVIG KS
IVIG for Kawasaki syndrome
Alanine aminotransferase Aspartate aminotransferase Intravenously administered immune globulin Kawasaki gyndrome
dren with KS. The primary goal was to show no clinically significant difference between these two regimens. METHODS
Patient selection. Patients were enrolled from January 1987 through April 1988 at seven centers in the United States and Canada. Kawasaki syndrome was defined by the presence of fever for at least 3 consecutive days and at least four of the five following features: (1) changes in the oropharynx, including reddening and fissuring of the lips, hyperemia of the buccal mucosa, or "strawberry" tongue, (2) nonexudative conjunctivitis, (3) erythema of the palms and soles, edema of the hands and feet, or periungual desquamation in the subacute phase of the disease, (4) rash, and (5) cervical lymphadenopathy, defined as one or more nodes at least 1.5 cm in diameter. All patients were enrolled in the study within 7 days of the onset of KS, which was defined as the first day of fever. The patient was excluded if there was clinical or laboratory evidence of any other disease known to mimic KS. Other exclusion criteria included known heart disease (e.g., congenital malformations or rheumatic carditis), recurrent KS, known immunologic defects, or prior treatment with an immune globulin product or any immunomodulating agents. Informed consent was obtained from parents. The protocol was approved by the human subjects committee of each institution. Study design. Patients meeting the entry criteria were randomly assigned in blocks of four to receive one of the two dosage regimens of IVIG. Each institution was provided with a randomization schedule, and patients were assigned in sequential order to the proper treatment group. Patients received either IVIG at 1 gm/kg, administered as a single dose for 5 to 8 hours, or at 400 mg/kg, administered once daily for 2 to 4 hours for 4 consecutive days. The IVIG used was Gammagard (Baxter Healtheare Corp., Hyland Division, Glendale, Calif.), a highly purified preparation of human IgG derived from the cold ethanol fractionation process and further purified by ion exchange adsorption. 8 The manufacturing process does not involve chemical or enzymatic modification procedures; the Fc portion remains intact. Patients in both treatment groups received aspirin at an initial dosage of 80 to 100 m g / k g / d a y divided every 6 hours. Salicylate levels were obtained on days 4 to 6 of the study. An attempt was made to adjust the dose to achieve a minimal salicylate blood level of 15 mg/dl. The high dose of aspirin was continued until 24 hours after resolution of fe-
639
ver, and then reduced to 3 to 5 m g / k g / d a y and continued for a minimum of 2 months. Vital signs were monitored before, during, and after infusion. Any adverse reactions were noted. Maximum temperature was recorded daily for the duration of the hospitalization. A requirement for discharge was that the patient be afebrile for at least 24 hours. Laboratory evaluation. Laboratory evaluations were performed at entry into the study and at days 4 to 6, 14 to 21, and 42 to 49 after enrollment. Complete blood cell and differential cell counts, platelet count, sedimentation rate, alantitrypsin level, and quantitative immunoglobulin determinations were obtained at each visit. In addition, A L T and AST levels were measured at entry, days 14 to 21, and days 42 to 49. Total serum bilirubin concentration was determined at entry and at days 42 to 49. Urinalysis was performed and blood urea nitrogen value obtained at entry. Cardiac evaluation. Cardiac evaluation was performed at entry into the study (before infusion) and at 4 to 6, 14 to 21, and 42 to 49 days after infusion. Echocardiographic evaluation of the coronary arteries included morphologic evaluation of the left main, left anterior descending, circumflex, right, and posterior descending coronary arteries. The diameter of each coronary artery segment was measured from inner surface to inner surface. Coronary artery segments were graded as follows: normal (uniform appearance, ~4 mm diameter); or aneurysmal (localized increase > 150% of adjacent vessel diameter). Further evaluation included measurement of left ventricular dimension and shortening fraction, examination of the pericardial space, and Doppler assessment of the presence and degree of any mitral or aortic insufficiency. Each echocardiogram was recorded on videotape and reviewed by two observers who were unaware of patient identity. If there was a discrepancy in assessment between the two reviewers, the tape was assigned to a third reader for review in a blind fashion. Criteria for determining safety and efficacy. Any change in vital signs (respiration, pulse, blood pressure, and temperature) was noted, as well as chills, pain, nausea, emesis, headache, anxiety, flushing, chest tightness, abdominal cramps, wheezing, dizziness, or rash. The primary criterion for assessing the efficacy of IVIG administered at 1 gm/kg as a single dose, in comparison with 400 mg/kg administered daily for 4 days, was the difference in the incidence of coronary artery aneurysms. Differences between the two treatment groups in laboratory values, duration of fever, and reduction of clinical symptoms associated with the disease were also compared. Data analysis. The incidence of aneurysms in the two treatment groups was compared by chi-square analysis. Some quantitative variables, such as laboratory values,
640
Barron et al.
The Journal o f Pediatrics October 1990
T a b l e I. Patient characteristics on entry into the study
Gender (M/F) Age (mo) Mean _+ SD Range Children 30,000/mm 3 Hgb 500,000/ram 3 ESR >100 mm/hr arAntitrypsin (mg/dl)*
Group 4: I gm/kg (n=25)
Group 2: 400 mg/kg • 4 days (n=22)
P
17/8
9/13
0.06
38.2 _+ 22.2 10-111 3 5.2 _+ 0.8 0 3 4 1 368.9 _+ 72.6
30.1 _+ 19.1 9-98 1 5.0 _+ 1.3 2 4 8 2 390.7 _+ 97.8
0.48 0.48 0.53 0.12 0.55 0.06 0.48 0.39
WBC. Whitebloodcell count;Hgb, hemoglobin;ESR, erythrocytesedimentationrate. *Valuesare expressedas mean _+SD.
T a b l e II. Echocardiographic findings Aneurysm Treatment group
No.
Location
Size (mm)
Study day when first detected
1 gm/kg 1 gm/kg 400 mg/kg • 4 days
1 1 4
Left main CA Proximal right CA Left main CA Circumflex CA LAD CA Proximal right CA
4 5 5 4 4 5
4 21 15
CA. Coronaryartery; LAD. left anteriordescendingartery.
were tested for statistical significance by the independent t test (two-tailed comparison). Others, such as febrile days and duration of hospital stay, were examined by the MannWhitney (Wilcoxon) test. The p values --30,000/ram 3, number of patients with a hemoglobin level 8 mm) aneurysms. Once visualized, the aneurysms persisted at least until days 42 to 49 of the study. There was no statistical difference in rate of aneurysm formation between the two groups of patients. The approximate 95% confidence interval for this difference is -0.10 to 0.19. At 1-year follow-up, we could no longer visualize aneurysms by echocardiography in the two patients with solitary aneurysms who had received IVIG in a single 1 g m / kg dose. The patient with multiple aneurysms who had received IVIG in a dosage regimen of 400 mg/kg daily for 4 days had a normal-appearing right coronary artery shown by echocardiogram at 1-year follow-up; however, the other aneurysms were still visualized. Thus at 1-year follow-up, only 1 of the 44 patients had persistent aneurysms detected by echocardiogram. Nine patients had a pericardial effusion detected on the first (before infusion) echocardiogram, including six patients in the 1 gm/kg group and three in the 400 mg/kg group (p = NS). In most instances the pericardial effusion was not detected after days 4 to 6 of the study. None of the patients with aneurysms had evidence of pericardial effusion during the study. Six patients had evidence of mitral insufficiency at entry into the study, including three patients receiving 1 gm/kg and three on the 400 mg/kg regimen. In three patients, the mitral insufl%iency resolved by days 4 to 6 of the study; in two patients (one from each group) the insufficiency resolved before the third echocardiogram, and in one patient (1 gm/kg) the insufficiency persisted until the third echocardiogram. The one patient with multiple aneurysms had evidence of transient mitral insufficiency on entry into the study. No patients had evidence of aortic insufficiency. There was no significant difference in left ventricular shortening fraction between the two groups at study entry (1 gm/kg -- 34.3 + 4.6% vs 400 mg/kg = 35.1 +__5.3%; p = 0.63). There was also no difference in shortening fraction between the groups at the other 3 study days, and shortening fraction did not change significantly from one study day to the next in either group. Clinical course. Both groups of patients had a fairly rapid disappearance of fever after intravenous infusion of im-
IVIG for Kawasaki syndrome
64l
800
,,,, . ~ I ~
700
a. Platelets
600 cq E E
500 400 300 b=.08
200
-- m
400 mg/Kg +SEM 9 lg/Kg++SEM
100
i
i
0
I
8O i
M,,
70
601
.....
~
~
~
b, ESR
~
50 =.72
4O E
30 2O 10 i
0
i
40o] - ~ . . . .
~.. ~~ ~" p =~- 4 2l ~ " _' ~ " ,
i
C. Alpha-1 Antitrypsin
-.,
~176176
-
100 ]-
'~
.
Day 0
.
.
Day 4-6
.
Day 14-21
Day 42-49
Day of Study
Fig. 1. Laboratory parametersobtained at day 0, clays4 to 6, days 14 to 21, and days 42 to 49 of study for patients on IVIG dosage regimens of 400 mg/kg daily for 4 days (----) and 1 gm/kg ( _ _ ) . a, Platelet count; b, erythrocyte sedimentation rate; e, cqantitrypsin level. Note p values.
mune globulin, with a more rapid decline in the 1 gm/kg group. Between the first and second study days, maximum temperature fell an average of 1.41 ~ C. in the 1 gm/kg group, compared with 0.78 b C. in the 400 mg/kg group (p = 0.013). Patients were discharged once they were afebrile for at least 24 hours, so there was a significant difference in the number of days of hospitalization, with the 1 gm/kg group discharged at a median of 4 days after entry into the study and the 400 mg/kg group at 5 days (p = 0.01).This difference at least partially reflects the fact that patients receiving the four doses required a minimum of 4 days of hospitalization. Laboratory evaluation. Platelet count increased in both groups on days 4 to 6 of the study. Thereafter the platelet count gradually decreased. There was no significant difference in platelet count at these various periods for patients receiving either dosage regimen of IVIG (Fig. 1, a). There was a slight increase in erythrocyte sedimentation rate at days 4 to 6 of the study in the 400 mg/kg group of patients,
64 2
Barron et al.
2,000
}~
1,8oo
~
/
a. IgG ~" ~.
1,600
, 1,4oo
- -
The Journal o f Pediatrics October 1990
/
1,000
800 / r 600 ~
p=.03
400 mO/Kg+_SEM I g/K~_ SEM
--
I
I
I
160
14 2o ~
b. lgA
100
60 40
, "~" ~=.03
p=.34
~ ~,~"~ p=.12
20 0
I
I
I
I
I
200 180 160 140
_== 120if/ 100
p=.l I 80 6Day 0
p=.10 Day 4-6
Day 14-21
I
Day 42-49
Day of Study Fig. 2. Mean serum immunoglobulin level obtained at day 0, days 4 to 6, days 14 to 21, and days 42 to 49 for patients on IVIG dosage regimens of 400 mg/kg daily for 4 days (..... ) and 1 gm/kg ( ). a, Mean serum IgG level; b, mean serum IgA level, e, mean serum IgM level. Note p values.
with a subsequent fall to normal values (Fig. 1, b). The erythrocyte sedimentation rate for the 1 gm/kg group did not statistically differ from that in the 400 mg/kg group at any point in the study. There was a gradual decrease in alantitrypsin concentration after infusion in both groups (Fig. 1, c), and no statistical difference between the two groups. There was a transient mild increase in ALT and AST values on days 4 to 6 of the study in two patients (one receiving 1 gm/kg and one receiving 400 mg/kg), with subsequent resolution by days 14 to 21 of the study. No increase in serum bilirubin concentration followed infusion of either dose of 1VIG. Both groups had a significant rise in serum IgG values. Despite a significantly lower mean IgG level in the 400 mg/kg group at entry into the study, the mean level at days 4 to 6 of the study was significantly higher in the 400 mg/ kg group (Fig. 2, a). Thereafter the serum IgG level for both groups gradually declined but remained significantly greater than pretreatment levels at days 42 to 49. The mean serum IgA levels increased at days 4 to 6 of the study in both
groups, despite negligible amounts of IgA in the IVIG preparation (Fig. 2, b). Thereafter, there was a gradual decline in mean serum IgA levels. At days 42 to 49 of the study, the mean serum IgA level was significantly lower than the pretreatment level in the 1 gm/kg group but not in the 400 mg/kg group; in four patients (one receiving 1 gm/kg and three on the 400 mg/kg regimen), serum IgA levels became markedly less than age-related normal values (with development of an absolute deficiency in one child). Follow-up of two of these children revealed normalization of serum IgA levels during the subsequent few months. Mean serum IgM levels were significantly increased at days 4 to 6 and days 14 to 21 of the study, compared with day 0 for both groups of patients. This transient increase in serum IgM was not due to passive transfer of IgM, because there were negligible amounts of IgM detectable in the IVIG preparation. By days 42 to 49 of the study, mean serum IgM levels did not differ significantly from pretreatment levels. Adverse reactions. None of the patients experienced serious adverse reactions to IVIG. Four children receiving 1 gm/kg had mild adverse reactions. One child had mild flushing that required no change in the infusion rate. Another patient had mild flushing, chills, and nausea and vomiting, which required temporarily slowing the infusion rate until symptoms resolved. One child had mild hypotension, which responded to stopping of the infusion temporarily and administration of 0.9% saline solution (10 ml/kg). In the fourth child a morbilliform rash developed 3 days after the infusion but resolved in 2 days without therapy. Three children on the 400 mg/kg regimen had mild adverse reactions. Two hours into the fourth infusion, one child had shaking chills, which resolved spontaneously. A second child had chills and noisy breathing 11/2 hours into the first infusion and responded to stopping of the infusion temporarily; this problem did not recur with subsequent infusions. A third child had headache, flushing, and abdominal cramping 1V2hours into the second infusion. The infusion had infiltrated; it was stopped and then restarted at a slower rate. No patient had clinical signs of congestive heart failure. DISCUSSION A current theory is that KS is an immunologically mediated generalized vasculitis that is triggered in a susceptible host by a variety of common infectious agents prevalent in the community. 9 It may be postulated that there is a predisposition to abnormal immune responsiveness after an asyet-unrecognized challenge to the immune system. The list of suspected infectious causes of KS is long, but no etiologic agent has been clearly identified. 1~ Currently, most therapies are based on decreasing the inflammatory response. Most standard treatment protocols involve the use of as-
Volume 117 Number 4
pirin. Most of our patients became afebrile soon after the start of IVIG therapy, regardless of dosage regimen, so the duration of treatment with high doses of aspirin in most children was just a few days. As a result, the maximum salicylate level obtained in most children was less than 10 mg/dl. Furusho et al. 1' 2 first reported the efficacy of IVIG in KS. Newburger et al. 4 described a multieenter collaborative controlled study demonstrating the efficacy of IVIG plus aspirin, compared with aspirin alone, in reducing the frequency of coronary artery abnormalities. They also reported more rapid resolution of fever and more rapid return of laboratory measurements to normal in the acutephase response. Engle et al. 7 recently reported an uncontrolled, one-arm study in which all patients received a 1 gm/kg dose of IVIG. Non e had subsequent aneurysms, and fever abated in 27 of the 32 patients within the first day after treatment; however, five children required additional IVIG. It is difficult to compare the various IVIG studies because of different patient populations, differing definitions of coronary artery abnormalities, and differing doses of IVIG, aspirin, or both. It has become apparent, however, that the dose of IVIG may be an important determinant of efficacy. In various other studies, 6' 17, 18 smaller doses of IVIG were ineffective in significantly reducing the incidence of coronary artery abnormalities. In addition, the efficacy of IVIG may be related to the number of days of illness at the time of infusion. 19 The incidence of aneurysms in our study compares favorably with that in other IVIG studies. Our findings suggest that IVIG given as a single high dose, 1 gm/kg, may be as efficacious as a regimen of 400 mg/kg daily for 4 days. The 95% confidence interval for the difference in the incidence rates of coronary aneurysms in the two groups (-0.10 to 0.19) suggests that the rate for the 1 gm/kg regimen may be lower by as much as 10% or higher by as much as 19% compared with the 400 mg/kg regimen; the wide range is a result of the relatively small sample size. Perhaps most important is the fact that aneurysms were detected in only one child at 1-year follow-up. The duration of hospitalization may also be shortened with the single dose, thereby increasing the cost-effectiveness of this regimen, which requires less total IVIG. ,Both regimens of our therapeutic protocol were well tolerated; only minor adverse reactions were noted in a small number of children. The effects on erythrocyte sedimentation rate, platelet count, and al-antitrypsin level were similar with the two dosage regimens. There was an expected increase in serum IgG levels after infusion of the IVlG in both treatment groups. The serum IgG levels were still elevated at days 42 to 49 of the study, an expected finding because the half-life of IgG is approximately 25 days. An unexpected finding was the increase of both IgA and IgM af-
IVIG for Kawasaki syndrome
643
ter the infusion of an IVIG preparation that is almost entirely IgG. More surprising was the marked fall in serum IgA levels in four patients by days 42 to 49 of the study. Whether these changes are a reflection of the action of IVIG or an immunologic change intrinsic to KS needs further evaluation. Significant morbidity and mortality rates in KS have been limited primarily to patients with coronary artery aneurysms. The use of IVIG appears to reduce the overall incidence of coronary artery aneurysms and the occurrence of giant aneurysms, which are associated with increased morbidity and mortality rates. In this study, two of the three patients in whom small solitary coronary artery aneurysms developed had ech0cardiographic resolution of their lesions at 1-year follow-up, but such resolution does not eliminate the risk of eventual development of coronary stenosis. Therefore the first goal of therapy should be to prevent the formation of coronary artery aneurysms. The population that should receive IVIG therapy remains unclear. Aneurysms do not develop in approximately 80% of patients. Therefore the use IVIG for every patient with KS, as recommended by some authors, 4' 2o may be unnecessarily costly. It would be ideal if, before undertaking therapy of this type, clinicians could identify prospectively the patients in whom aneurysms or long-term sequelae would ultimately develop. The substrate for aneurysm formation occurs in the first days of the illness, when perivaseular inflammation gives way to a panvasculitis, so it is critical t h a t high-risk patients be differentiated early in their clinical course. Unfortunately, none of the methods now available is sufficiently helpful to allow early identification of all patients at risk. The dose of IVIG for use in KS also remains a subject of investigation and controversy. The earlier reduction in fever in the 1 gm/kg group in this study raises the possibility that the general antiinflammatory effect of this regimen is at least as potent as that of the regimen of 400 mg/kg daily for 4 days. Although the relatively small sampl e size in this trial does not allow for a definitive statement regarding the occurrence of coronary artery aneurysms in patients with KS, it appears that the use of IVIG in a dose Of 1 gm/kg is associated with a more rapid clinical improvement and a shorter hospital stay. Definitive therapy for patients with KS and, ultimately, prevention of this disorder will not be possible until the cause and pathogenesis can be explicitly defined. Documentation of the effectiveness of IVIG in ameliorating the course of the disease and its complications in selected patients may suggest new approaches for research into the pathophysiologic nature of this disease. We acknowledge and thank the following persons who participated in the clinical trials: Ivonne Arroyo, MD, and Martha Curry, RN, Houston; Norman N. Musewe, MD, Toronto; Frederic Stone,
644
Barron et al.
MD, and Donald L. Uden, PharmD, Minneapolis; Victoria Vetter, MD, Gregg HeRon, MD, and Jennifer Puck, MD, Philadelphia; and Christian E. Hardy, MD, Oakland, Calif. We thank Masato Takahashi, MD, for his continued assistance and interest in this project; Fred Bierman, MD, for technical contributions in the field of echocardi01ogy; Theodore Colton, PhD, and Edward Giannini, DrPH, for statistical services; Marian Melish, MD for assistance in study design; and Robert Warren, MD, PhD, for his critical review of the manuscript.
The Journal of Pediatrics October 1990
9.
10. 11. 12.
REFERENCES
1. Furusho K, Sato K, Soeda T, et al. High-dose intravenous gammaglobulin for Kawasaki disease [Letter]. Lancet 1983;2:1359. 2. Furusho K, Hiroyuki N, Shinomiya K, et al. High-dose intravenous gammaglobulin for Kawasaki disease. Lancet 1984; 2:1055-8. 3. Nagashima M, Matsushima M, Matsuoka H, Ogawa A, Okumura N. High-dose gammaglobulin therapy for Kawasaki disease. J PEDIATR 1987;110:710-2. 4. Newburger JW, Takahashi M, Burns JC, et al. The treatment of Kawasaki syndrome with intravenous gamma globulin. N Engl J Med 1986;315:341-7. 5. Isawa M, Sugiyama K, Kawase A, et al. Prevention of coronary artery involvement in Kawasaki disease by early intravenous high-dose gamma globulin. In: Doyle EF, Engle MA, Gersony WM, Rashkind W J, Talner NS, eds. Pediatric Cardiology. New York: Springer-Verlag, 1986:1083-5. 6. Okuni M, Harada K, Yamaguchi H, et al. Intravenous gamma globulin therapy in Kawasaki disease: trial of low-dose gamma globulin. In: Shulman ST, ed. Kawasaki disease. New York: Alan R Liss, 1987:433-9. 7. Engle MA, Fatica NS, Bussel JB, O'Loughlin JE, Snyder MS, Lesser ML. Clinical trial of single-dose intravenous gamma globulin in acute Kawasaki disease. Am J Dis Child 1989; 143:1300-4. 8. Ochs HD, Lee ML, Fischer SH, Kingdon HS, Wedgwood RJ.
13.
14. 15.
16.
17.
18.
19.
20.
EfficacY of a new intravenous immunoglobulin in primary immunodeficient patients. Clin Ther 1987;9:512-22. Barron KS, DeCunto CL, Montalvo JM, Orson FM, Lewis DE. Abnormalities of immunoregulation in Kawasaki syndrome. J. Rheumatol 1988;15:1243-9. Feigin RD, Barron KS. Treatment of Kawasaki syndrome. N Engl J Med 1986;315:388-90. Carter RF, Haynes ME, Morton J. Rickettsia-like bodies and splenitis in Kawasaki disease. Lancet 1976;2:1254-5. Hamashima Y, Tasaki K, Fujiwara H, Hoshino T, Matsuda S, Kao T. Kawasaki disease: its pathological features and possible pathogenesis. Aeta Paediatr Jpn 1983;25:108- t 7. Burns JC, Geha RS, Schneeberger EE, et al. Polymerase activity in lymphocyte culture supernatants from patients with Kawasaki disease. Nature 1986;2:323-814-6. Shulman ST, Rowley AH. Does Kawasaki disease have a retroviral etiology? Lancet 1986;2:545-6. Melish ME, Marchette N J, Kaplan JC, Kihara S, Ching D, Ho DD. Absence of significant RNA-dependent DNA polymerase activity in lymphocytes from patients with Kawasaki syndrome. Nature 1989;337:268-90. Burns JC, Huang AS, Newburger JW, et al. Characterization of the polymerase activity associated with cultured peripheral blood mononuclear cells from patients with Kawasaki disease. Pediatr Res 1990;27:109-12. Furusho K, Kamiya T, Nakano H, et al. Japanese gamma globulin trials for Kawasaki disease. In: Shulman ST, ed. Kawasaki disease. New York: Alan R Liss, 1987:425-32. Ogino H, Ogawa M, Harima Y, et al. Clinical evaluation of gammaglobulin preparations for the treatment of Kawasaki disease. In: Shulman ST, ed. Kawasaki disease. New York: Alan R. Liss, 1987;555-6. Nakano H, Ueda K, Saito A, et al. Scoring method for identifying patients with Kawasaki disease at high risk of coronary artery aneurysms. Am J Cardiol 1986;58:739-42. Plotkin SA, Daum RS, Giebink GS, et al. Intravenous gamma-globulin use in children with Kawasaki disease. Pediatrics 1988;82:122.
