Op

Sm

RyhT(t

CSd

Ksarpao

c(n

Dect

4a3

ACEG

Resistance to Intravenous Immunoglobulin in Children withKawasaki Disease

ADRIANA H. TREMOULET, MD, BROOKIE M. BEST, PHARMD, SUNGCHAN SONG, BS, SUSAN WANG, ELENA CORINALDESI, MD,JULIA R. EICHENFIELD, DANIELLE D. MARTIN, BS, JANE W. NEWBURGER, MD, MPH, AND JANE C. BURNS, MD

bjectives To explore the increased incidence of intravenous immunoglobulin- (IVIG) resistance among San Diego Countyatients with Kawasaki disease (KD) in 2006 and to evaluate a scoring system to predict IVIG-resistant patients with KD.

tudy design We performed a retrospective review of patients with KD treated within 10 days of fever onset. Withultivariate analysis, independent predictors of IVIG-resistance were combined into a scoring system.

esults In 2006, 38.3% of patients with KD in San Diego County were IVIG-resistant, a significant increase over previousears. IVIG-resistance was not associated with a particular brand or lot of IVIG. Resistant patients were diagnosed earlier, hadigher percent bands, and higher concentrations of C-reactive protein, alanine aminotransferase, and �-glutamyl transferase.hey also had lower platelet counts and age-adjusted hemoglobin concentrations and were more likely to have aneurysmsP � .0008). A scoring system developed to predict IVIG-resistant patients using illness day, percent bands, �-glutamylransferase, and age-adjusted hemoglobin had a sensitivity of 73.3% and specificity of 61.9%.

onclusions An unexplained increase in IVIG-resistance was noted among patients with KD in San Diego County in 2006.coring systems based on demographic and laboratory data were insufficiently accurate to be clinically useful in our ethnicallyiverse population. (J Pediatr 2008;153:117-21)

awasaki disease (KD), the leading cause of pediatric acquired heart disease in the United States and Japan, is an acute,systemic vasculitis. Treatment with a single dose of intravenous immunoglobulin (IVIG) and high-dose aspirin resultsin resolution of fever in most patients and significantly reduces the rate of coronary artery aneurysms.1 Despite this

uccess, 10% to 20% of children will have persistent or recrudescent fever after their first infusion of IVIG.2-5 These patients aret increased risk for development of coronary artery abnormalities.5,6 Additional therapies used in these patients includeetreatment with IVIG, immunomodulatory agents such as infliximab, high-dose methylprednisolone, cyclophosphamide, andlasmapheresis.7-11 Identification of patients who are likely to be IVIG-resistant wouldllow the use of additional therapies early in the course of their illness when preventionf coronary artery damage might still be possible.

A number of recent studies from Asia have identified demographic and laboratoryharacteristics, including age, illness day, platelet count, erythrocyte sedimentation rateESR), and concentrations of hemoglobin, C-reactive protein (CRP), lactate dehydroge-ase, and alanine aminotransferase (ALT) as predictors of IVIG-resistance.2,12-14

We noted an increase in the numbers of IVIG-resistant patients with KD in Saniego County in 2006 as compared with previous years. This prompted us to examine the

pidemiology of IVIG-resistance in our population. We tested the hypothesis thatomparison of characteristics of patients with KD and their response to IVIG could leado a scoring system that predicted IVIG-resistance.

METHODSAll patients met the standard case definition for KD and had fever and at least

of the 5 standard KD clinical criteria (rash, conjunctival injection, cervical lymph-denopathy, changes in the oral mucosa, and changes in the extremities) or fever andcriteria plus coronary artery abnormalities (dilatation or aneurysm) documented by

LT Alanine aminotransferaseRP C-reactive proteinSR Erythrocyte sedimentation rate

IVIG Intravenous immunoglobulinKD Kawasaki diseasezHgb Age-adjusted hemoglobin

From the Department of Pediatrics (A.T.,B.B., S.S., S.W., E.C., J.E., J.B.), University ofCalifornia San Diego School of Medicineand Rady Children’s Hospital San Diego, SanDiego, the Skaggs School of Pharmacy andPharmaceutical Sciences (B.B.), University ofCalifornia San Diego, La Jolla, CA, and theDepartment of Cardiology, Children’s Hospi-tal Boston (D.M., J.N.), Boston, MA.

Supported by grants from the NationalHeart, Lung, and Blood Institute (HL69413and K24 HL074864), from the NationalInstitute for Child Health and Human De-velopment, Pediatric Pharmacology Re-search Unit (5U10 HD031318), and theCiaranello Family Fund at Children’s Hos-pital Boston.

Submitted for publication Jul 4, 2007; lastrevision received Oct 22, 2007; acceptedDec 10, 2007.

No reprints available.

0022-3476/$ - see front matter

Copyright © 2008 Mosby Inc. All rightsreserved.

GT �-Glutamyl transferase

10.1016/j.jpeds.2007.12.021

117

efieDoapsirrawhtrwC

cbi(n�abaco

oaaSJdDtpdPat2Rerppit

bKt

pipv.ww[TragppscvEvicpVa(Gbftpm�1Tsrnl

cts2

ctRIfrid

1

chocardiography.15 Only patients diagnosed within therst 10 days after fever onset were analyzed. Patients werenrolled at 2 clinical sites (Rady Children’s Hospital, Saniego and the University of California, San Diego) after

btaining parental informed consent and subject assent, asppropriate. The study protocol was reviewed and ap-roved by the Institutional Review Boards at both studyites. Demographic and clinical data including sex, ethnic-ty, race, age, illness day (first day of fever � illness day 1),esponse to IVIG therapy, and coronary artery status wereecorded for all subjects. Laboratory testing at hospitaldmission and prior to IVIG administration includedhite blood cell count with differential, platelet count,emoglobin, hematocrit, ESR, CRP, ALT, and �-glu-amyl transferase (GGT). IVIG brand and lot data andesponse to IVIG therapy were also analyzed for patientsith KD treated within the first 10 days after fever onset athildren’s Hospital Boston.

IVIG-resistance was defined as persistent or recrudes-ent fever (T � 100.4° F rectally or orally) at least 48 hoursut not longer than 7 days after completion of the first IVIGnfusion (2 g/kg). Patients were classified as having normal�2.5 standard deviation units [z score] from the mean,ormalized for body surface area), dilated (2.5 � z score4.0), or aneurysmal (focal or diffuse dilatation of a coronary

rtery segment with z score �4.0) coronary arteries on theasis of the maximal internal diameters of the right coronaryrtery and left anterior descending artery measured by echo-ardiography at the time of diagnosis and at 2 to 4 weeks afternset of fever.15

The analysis of the epidemiologic data and the devel-pment of a scoring system were performed with the data setsvailable at the time each component of the study was initi-ted. For the epidemiologic evaluation of IVIG-resistance inan Diego County, data from subjects admitted betweenanuary 1, 1998, and December 31, 2006, were evaluated toetermine the annual incidence of KD in San Diego County.emographic and pre-IVIG laboratory data collected be-

ween October 1998 and September 2006 from subjects whoresented within the first 10 days of illness were used toevelop the scoring system to predict IVIG-resistant subjects.harmacy records from Rady Children’s Hospital, San Diegond Children’s Hospital Boston were retrospectively reviewedo retrieve IVIG brand and lot data for patients treated in006. These data were not available for previous years forady Children’s Hospital. To test whether there were differ-

nces in clinical presentation in years with high or low IVIG-esistance rates, we compared clinical criteria for KD betweenatients diagnosed in 2005 and 2006. Greater than 90% of theatients during these 2 years were evaluated during their acutellness by a single physician (J.C.B.), thus reducing the po-ential for interobserver variability.

The scoring system to identify IVIG-resistant patientsy Egami et al12 was applied to our cohort of patients withD in San Diego County (October 1998 to September 2006)

o determine its predictive value in an ethnically diverse o

18 Tremoulet et al

opulation. To create a new scoring system that might betterdentify children at risk for IVIG-resistance in a multiethnicopulation, we first performed univariate analysis to identifyariables significantly associated with IVIG-resistance (P �05). Because hemoglobin concentrations are age dependent,e analyzed age-adjusted hemoglobin concentrations (zHgb)ith the following formula: ([Observed hemoglobin] �

Mean hemoglobin for age])/Standard deviation for age.16

he standard deviations were estimated as one quarter of theeported range of normal hemoglobin concentrations for eachge interval. We then performed a multivariate logistic re-ression analysis with backward elimination to identify inde-endent predictors of IVIG-resistance. The model was sim-lified by retaining those variables with a P � .2 in the scoringystem. Results were expressed as an odds ratio with a 95%onfidence interval. Analyses were performed with NCSSersion 2007 (NCSS, Kaysville, Utah) and SAS Learningdition 2.0 (SAS Institute, Inc., Cary, NC). Continuous

ariables were converted to dichotomous variables by choos-ng a break point based on receiver-operator characteristicurves and the upper or lower quartile for each independentredictor identified by the multivariate logistic regression.ariables and break points that were tested included age (�6

nd �12 months), illness day (�4 days), platelet count�30,000 and �150,000 cells/mm3), ALT (�80 IU/L),GT (�60 and � 100 IU/L), CRP (�8 mg/dL), percent

ands (�20), and zHgb (��1.0, ��1.5, and ��2.0). Dif-erent scoring systems were developed by use of the odds ratioo determine the weight of each variable, giving a weight of 2oints for those with the highest odds ratios. Even thoughost variables were given 1 point, illness day �4 days, GGT60 IU/L, and percent bands �20 were weighted with bothand 2 points to evaluate the effects on the scoring system.he scores were calculated for each patient, with a final risk

core consisting of 2 risk strata, low-risk (score 0-1) or high-isk (score 2-5). The sensitivity, specificity, and positive andegative predictive values for each scoring system were calcu-

ated.The t test and Wilcoxon rank-sum test were used to

ompare normally distributed and skewed continuous labora-ory values, respectively, between IVIG-resistant and -re-ponsive patients. Categorical variables were compared by a-sided �2 test and Fisher exact test as appropriate.

RESULTSFrom January 1, 1998, to December 31, 2006, 362

hildren with a median age of 2.3 years were admitted andreated for KD at the University of California, San Diego andady Children’s Hospital, San Diego. The percentage of

VIG-resistant patients with KD from 1998 to 2005 rangedrom 9.8% to 20% (Figure). In 2006 the incidence of IVIG-esistance increased to 38.3% (P � .009). To explain thisncrease in IVIG-resistance, we postulated that either theisease or the IVIG had changed.

To test the first hypothesis, we evaluated the frequency

f the classical clinical signs associated with KD between

The Journal of Pediatrics • July 2008

2ant

irMBtHbtFdtwtpItsd(oraWroI2aGwfi

c1((T

tcwE�cCprts

cactt(taca

difif�(auttctw

sr(saae

a

FCIf

R

005, when the incidence of IVIG-resistance was only 11.1%,nd 2006, when the incidence was 38.3%. No difference wasoted in the frequency of the classical clinical signs betweenhe 2 years (data not shown).

Because of our concern about the possibility of a changen the IVIG product, IVIG- resistant patients from 2006 wereeported to the Food and Drug Administration through the

edWatch system and the Pharmacovigilance Division ofaxter Pharmaceuticals was contacted. The lots of IVIG used

o treat these resistant patients were investigated by Baxterealthcare Corporation. The investigation included review of

atch records, release data, manufacturing changes, tempera-ure deviations during shipment, and relevant stability data.urthermore, the immunoglobulin G content (monomer/imer/polymer/fragments), anticomplementary activity, andrypsin/chymotrypsin exposure time of the IVIG batches usedere within the historical trend for these measures. In addi-

ion, Baxter verified that no significant changes in the donorool or collection sites had occurred between 2005 and 2006.n our investigation in San Diego, we found no difference inhe distribution of brands and lots used to treat IVIG-respon-ive and -resistant patients in 2006, although our power toetect a difference was limited by the small sample size.Supplemental Table I; available at www.jpeds.com). We alsobtained data from Children’s Hospital Boston on the IVIGesponse of 25 patients with KD treated with the same brandsnd lots as the San Diego patients (Supplemental Table I).

hereas 13 of 29 San Diego patients (45%) were IVIG-esistant when treated with 3 specific IVEEGAM lots, only 1f 23 Boston patients (4%) treated with these same lots wasVIG-resistant. The rate of IVIG-resistance in Boston in006 was 8%. To confirm that IVIG-resistant patients actu-lly received the prescribed dose, the serum immunoglobulin

levels were analyzed in a subset of patients (n � 20). Levelsere not significantly different from those published in the

igure. IVIG-responsive and -resistant KD patients in San Diegoounty (1998-2006) Number of patients and percentages of

VIG-responsive and -resistant patients with KD for San Diego Countyrom 1998 to 2006 (P � 0.009 for 2006 versus other years, overall �2).

IVIG-responsive, � IVIG-resistant.

rst trial of single-infusion (2 g/kg) IVIG.1 2

esistance to Intravenous Immunoglobulin in Children with Kawasaki Di

IVIG-resistance was strongly associated with an in-reased aneurysm rate. Analysis in 362 patients with KD from998 to 2006 revealed that aneurysms developed in 9 of 6015%) IVIG-resistant patients as compared with 9 of 3023%) IVIG-responsive patients (P � 0.0008) (Supplementalable II; available at www.jpeds.com).

Multiple scoring systems for predicting IVIG-resis-ance in patients with KD have been published from otherountries.2,12-14 However, the only scoring system for whiche had all of the variables in our database was published bygami et al12 from Japan. This scoring system includes age6 months (1 point), illness day �4 day (1 point), platelet

ount �30,000 (1 point), ALT �80 IU/L (2 points), andRP �8 mg/dL (1 point). When applied to our patientopulation, this model missed more than 60% of IVIG-esistant patients (Table III). When applied only to Asian pa-ients in the San Diego County cohort, the Egami score had apecificity of 89.3% and a sensitivity of 33.3%.

A new scoring system was developed with data from ourohort of patients with KD diagnosed between October 1998nd September 2006. Comparison of the demographic andlinical characteristics of IVIG-resistant and -responsive pa-ients with KD revealed that IVIG-resistant patients tendedo be younger (P � .06) and presented earlier (P � 0.002)Supplemental Table III). Univariate analysis of the labora-ory data revealed that percent bands, absolute band count,nd concentrations of CRP, ALT, and GGT were signifi-antly higher in IVIG-resistant patients (Table I). Plateletsnd zHgb were significantly lower in IVIG-resistant patients.

Multivariate logistic regression analysis found illnessay at diagnosis, percent bands, GGT, and zHgb to bendependent predictors of IVIG-resistance (Table II). In thenal scoring system, dichotomous variables were weighted asollows: Illness day at diagnosis �4 days (1 point), % bands

20 (2 points), GGT �60 IU/L (1 point), and zHgb � �2.01 point). This scoring system had a sensitivity of 73.3% andspecificity of 61.9% (Table III). If this scoring system weresed prospectively to identify patients with KD who are likelyo be resistant to IVIG and therefore candidates for additionalherapy, 31.8% of patients (false-positive results) would re-eive additional therapy when not needed and 26.7% of pa-ients (false-negative results) would be IVIG-resistant butould not have received additional treatment.

When our cohort was stratified by ethnicity, the scoringystem showed decreasing sensitivity for detecting IVIG-esistance as follows: Caucasians � Hispanics � Asians81.3%, 68.2%, and 66.7%, respectively). When patients weretratified by coronary artery status (normal or dilated versusneurysm), our scoring system had a sensitivity of 72.2% andspecificity of 57.6% for predicting the development of an-

urysms.

DISCUSSIONThe cause of the unusual increase in IVIG-resistance

mong patients with KD treated in San Diego County in

006 was investigated. No change in the clinical presentation

sease 119

oblaoiD

GcpIcip

Utpo

cpaatpE(b

Tw

W%A%APzECAG

WNa

b

c

d al va

T

I%Gz

Ca

Trp

SSPNM

Pa

c

1

f KD or an association of resistance with particular IVIGrands or lots was found, although the small sample sizeimited our ability to detect a difference. The same lotsssociated with IVIG-resistance in San Diego were used atther sites across the United States without apparent changesn their IVIG-resistance rates (personal communication,

able I. Univariate analysis comparing laboratory vaith KD

No.a

BC, �103/mm3 [6-17] 60:301 12Polys 58:295

NC, cells/mm3 58:295 96Bands 58:295

BC cells/mm3 58:295 28latelets, �103/mm3 [150-350] 60:299 3Hemoglobind 60:300 �1SR, mm/hr [4-20] 58:298RP, mg/dL [0-1] 50:227 11LT, IU/L [13-45] 51:271GT, IU/L [0-23] 52:286

BC, White blood cell; Polys, polymorphonuclear leukocytes; ANC, absolute neutrophumbers in brackets are normal values (The Harriet Lane: a manual for pediatric hou

Number of subjects available for comparison, IVIG-resistant: IVIG-responsive.Expressed as median (25th-75th percentiles).Wilcoxon rank sum test.zHemoglobin, standard deviation units from the mean based on age-adjusted norm

able II. Multivariate predictors of IVIG-resistance in

Variables Logistic coefficient (�) Standard

llness day �4 0.61 0.35Bands �20 1.02 0.34GT �60 IU/L 0.80 0.34

Hgb ��2 0.71 0.33

I, Confidence interval limits.�2 test.

able III. Ability of scoring systems to predict IVIG-esistance and coronary artery aneurysms inatients with KD in the San Diego County cohort

IVIG-resistance Aneurysms

Egami score(%)

San Diegoscore (%)

San Diegoscore (%)

ensitivity 38.3 73.3 72.2pecificity 83.8 61.9 57.6PV 31.9 27.7 8.2PV 87.2 92.1 97.5isseda 61.7 26.7 27.8

PV, Positive predictive value; NPV, negative predictive value.Percentage of patients with IVIG-resistance (first two columns) or aneurysms (lastolumn).

onald Baker, Baxter Healthcare Corporation, January 2007). o

20 Tremoulet et al

iven that the mechanism of action of IVIG, as well as theause of KD remain unknown, constructing testable hy-otheses to explain what might have led to the increase inVIG-resistance is difficult. We publish this report to alertlinicians to examine the rate of IVIG-resistance at theirnstitutions and to track IVIG brand and lot use in theseatients.

Rates of IVIG-resistance at institutions across thenited States, including Hawaii, varied by year and by insti-

ution, but rates as high as 30% were sporadically observed. Inublished series of IVIG-resistance, the rates vary from a lowf 9.4% in Iran to a high of 23% in the United States.3,4,6,17,18

Because IVIG-resistant patients are at higher risk fororonary artery aneurysms, it is important to identify theseatients who might benefit from more aggressive initial ther-py. In our study, because of differences in the type of clinicalnd laboratory data recorded for our cohort, we were only ableo test 1 of the 4 previously published scoring systems. Theoor performance, particularly the low sensitivity of thegami score in our population of ethnically diverse children

18% Asian, predominantly Korean and Filipino), may beecause the Egami scoring system was created and validated

between IVIG-resistant and -responsive patients

istantb Responsiveb P valuec

0.0-17.3) 14.0 (11.2-18.0) �.054-59) 51 (38-60) �.05606-12730) 9027 (6716-12000) �.053-41) 14 (5-23) .00001670-5291) 1852 (691-3356) .000288-444) 404 (309-491) .0503.0, �0.8) �1.3 (�2, �0.6) .014

7-80) 65 (46-86) �.05.0-18.7) 7.4 (4.5-14.1) .0415-137) 31 (19-87) .0027-161) 32 (16-92) .0001

t; ABC, absolute band count.ers. 16th ed. Philadelphia: Mosby; 2002).

lues.16

tients with KD

r Odds ratio (95% CI)a P valuea Points

1.84 (0.91-3.72) .09 12.76 (1.42-5.37) .003 22.22 (1.15-4.31) .02 12.04 (1.06-3.93) .03 1

lues

Res

.8 (150 (300 (624 (116 (162 (2.9 (�

59 (4.9 (6

56 (384 (3

il counse offic

pa

erro

nly in cohorts of Japanese children. Despite robust statistical

The Journal of Pediatrics • July 2008

edqpopsa

tttsmwp

oatwcpuwitKctaac

s2Itlushf

Wac

1st2KC3oK4i5gw6o27og8ga9r1EP1I1rJ1ea1Pd1efKA1p1Sc2

R

valuation of available demographic, clinical, and laboratoryata, our own scoring system would miss more than oneuarter of IVIG-resistant patients. Because genetics likelylay an important role in determining disease severity andutcome, genetic differences between cohorts may affect theredictive value of different scoring systems. In the future,coring systems might incorporate polymorphic alleles to cre-te a more robust predictor of disease outcome.

Limitations of this study include the small sample size,he availability of IVIG brand and lot data only for 2006, andhe collection of limited laboratory data that precluded theesting of other scoring systems. We were unable to test ourcoring system in an independent cohort because measure-ent of GGT concentrations and reporting of percent bandsere not routine at other US institutions with large cohorts ofatients with KD.

Surveillance for IVIG-resistance by a broader networkf institutions would be helpful in defining the epidemiologynd providing clues to the cause of IVIG-resistance. As newherapies become available, identification of patients with KDho would be candidates for more aggressive therapy be-

omes increasingly important. Infliximab, which binds theroinflammatory cytokine tumor necrosis factor–�, was eval-ated in a retrospective case review of IVIG-resistant patientsith KD. No infusion reactions or complications attributed to

nfliximab administration were seen in this patient popula-ion.11 A randomized trial in IVIG-resistant patients withD has recently been completed (J.C.B., personal communi-

ation, January 2007). To improve the identification of pa-ients with KD who are candidates for more aggressive ther-py, we need a better understanding of the mechanism ofction of IVIG, the pathogenesis of aneurysms, and theontribution of host genetics to disease outcome.

In summary, although we were unable to explain theudden increase in IVIG-resistance in San Diego County in006, it is critical that clinicians be aware that increases inVIG-resistant KD may occur in their community. In addi-ion, scoring systems that were based on demographic andaboratory data were insufficiently sensitive to be clinicallyseful in our ethnically diverse population. A better under-tanding of the pathogenesis and host genetics in KD willelp identify key factors that may be important to include in

uture scoring systems.

1t

esistance to Intravenous Immunoglobulin in Children with Kawasaki Di

e thank Joan Pancheri, RN, and Annette Baker, RN, forssistance with data collection. We are indebted to the families andhildren who agreed to participate in this study.

REFERENCES. Newburger JW, Takahashi M, Beiser AS, Burns JC, Bastian J, Chung KJ, et al. Aingle intravenous infusion of gamma globulin as compared with four infusions in thereatment of acute Kawasaki syndrome. N Engl J Med 1991;324:1633-9.. Durongpisitkul K, Soongswang J, Laohaprasitiporn D, Nana A, Prachuabmoh C,angkagate C. Immunoglobulin failure and retreatment in Kawasaki disease. Pediatrardiol 2003;24:145-8.

. Han RK, Silverman ED, Newman A, McCrindle BW. Management and outcomef persistent or recurrent fever after initial intravenous gamma globulin therapy in acuteawasaki disease. Arch Pediatr Adolesc Med 2000;154:694-9.

. Kashef S, Safari M, Amin R. Initial intravenous gamma-globulin treatment failuren Iranian children with Kawasaki disease. Kaohsiung J Med Sci 2005;21:401-4.. Burns JC, Capparelli EV, Brown JA, Newburger JW, Glode MP. Intravenousamma-globulin treatment and retreatment in Kawasaki disease. US/Canadian Ka-asaki Syndrome Study Group. Pediatr Infect Dis J 1998;17:1144-8.. Sittiwangkul R, Pongprot Y, Silvilairat S, Phornphutkul C. Management andutcome of intravenous gammaglobulin-resistant Kawasaki disease. Singapore Med J006;47:780-4.. Mori M, Imagawa T, Katakura S, Miyamae T, Okuyama K, Ito S, et al. Efficacyf plasma exchange therapy for Kawasaki disease intractable to intravenous gamma-lobulin. Mod Rheumatol 2004;14:43-7.. Hashino K, Ishii M, Iemura M, Akagi T, Kato H. Re-treatment for immunelobulin-resistant Kawasaki disease: a comparative study of additional immune globulinnd steroid pulse therapy. Pediatr Int 2001;43:211-7.. Sundel RP, Burns JC, Baker A, Beiser AS, Newburger JW. Gamma globuline-treatment in Kawasaki disease. J Pediatr 1993;123:657-9.0. Chiyonobu T, Yoshihara T, Mori K, Ishida H, Nishimura Y, Yamamoto Y, et al.arly intravenous gamma globulin retreatment for refractory Kawasaki disease. Clinediatr (Phila) 2003;42:269-72.1. Burns JC, Mason WH, Hauger SB, Janai H, Bastian JF, Wohrley JD, et al.nfliximab treatment for refractory Kawasaki syndrome. J Pediatr 2005;146:662-7.2. Egami K, Muta H, Ishii M, Suda K, Sugahara Y, Iemura M, et al. Prediction ofesistance to intravenous immunoglobulin treatment in patients with Kawasaki disease.Pediatr 2006;149:237-40.3. Fukunishi M, Kikkawa M, Hamana K, Onodera T, Matsuzaki K, Matsumoto Y,t al. Prediction of non-responsiveness to intravenous high-dose gamma-globulin ther-py in patients with Kawasaki disease at onset. J Pediatr 2000;137:172-6.4. Kobayashi T, Inoue Y, Takeuchi K, Okada Y, Tamura K, Tomomasa T, et al.rediction of intravenous immunoglobulin unresponsiveness in patients with Kawasakiisease. Circulation 2006;113:2606-12.5. Newburger JW, Takahashi M, Gerber MA, Gewitz MH, Tani LY, Burns JC,t al. Diagnosis, treatment, and long-term management of Kawasaki disease: a statementor health professionals from the Committee on Rheumatic Fever, Endocarditis andawasaki Disease, Council on Cardiovascular Disease in the Young, American Heartssociation. Circulation 2004;110:2747-71.

6. Gunn L, Nechyba C, eds. The Harriet Lane Handbook: 16th edition. Philadel-hia: Mosby; 2002.7. Durongpisitkul K, Sangtawesin C, Khongphatthanayopthin A, Panamonta M,opontammarak S, Sittiwangkul R, et al. Epidemiologic study of Kawasaki disease andases resistant to IVIG therapy in Thailand. Asian Pac J Allergy Immunol 2006;4:27-32.

8. Wallace CA, French JW, Kahn SJ, Sherry DD. Initial intravenous gammaglobulinreatment failure in Kawasaki disease. Pediatrics 2000;105:E78.

sease 121

St

I

I

G

G

Na ponsive patients receiving Iveegam vs Gammagard or Gamunex with a Fisher’s exact test(

1

upplemental Table I. Comparison of IVIG brands breated in 2006a

Brand (Lot), no. (% of total)

San D

NIVIG-resist

(n � 17)

VEEGAM 36 (77%)VNE1A001 4 (50%)VNE1A003A 5 (50%)VNE1F003 4 (36%)Lot unknown 0 (0%)

VEEGAM & Gammagard liquid 1 (2%)VNE1F003 & LE12F021AC 1 (100%)ammagard liquid 8 (17%)LE12CA10AD1 1 (50%)LE08F005AB 0 (0%)LE12F003AA/LE12F017AA 0 (0%)LE12F003AA N/ALE12F068AB 1 (100%)amunex liquid 2 (4%)26N6P71, 26N7H2, 26N7JK1, 26N7981 0 (0%)26N6P71, 26N6R01 1 (100%)

/A, Not applicable.No statistical difference was found between the number of IVIG-resistant and -resP � .05).

etween IVIG-resistant and -responsive patients with KD

iego Boston

ant IVIG-responsive(n � 30) N

IVIG-resistant(n � 1)

IVIG-responsive(n � 24)

23 (92%)4 (50%) 0 (0%) 8 (100%)5 (50%) 1 (14%) 6 (86%)7 (64%) 0 (0%) 8 (100%)7 (100%) N/A N/A

N/A0 (0%) N/A N/A

2 (8%)1 (50%) N/A N/A4 (100%) N/A N/A1 (100%) N/A N/A

N/A 0 (0%) 2 (100%)0 (0%) N/A N/A

N/A1 (100%) N/A N/A0 (0%) N/A N/A

21.e1 Tremoulet et al The Journal of Pediatrics • July 2008

Sea

NDA

Na

b

pNfms

R

upplemental Table II. Comparison ofchocardiogram results between IVIG-resistantnd -responsive patients with KD (1998-2006)

Classification ofcoronary arteries

Resistant(n � 60)

Responsive(n � 302) P valuea

ormal, n (%)b 31 (52%) 193 (64%) NSilated 20 (33%) 100 (33%) NSneurysm/ectasia 9 (15%) 9 (3%) .0008

S, Not significant.Fisher exact test.Classification was based on the most abnormal of at least 2 echocardiograms peratient. The right and left anterior descending coronary arteries were classified as:ormal � z score �2.5 standard deviations from the mean internal diameter normalized

or body surface area; Dilated � z score �2.5 but �4.0 and returns to �2.5 within 2onth follow-up period; Aneurysm � focal or diffuse dilation of coronary artery

egment with z score �4.0; Ectasia � z � 4.0 and persists �2 months.

esistance to Intravenous Immunoglobulin in Children with Kawasaki Disease 121.e2

Sa

AIME

Na

b

c

d

e

1

upplemental Table III. Comparison of demographic and clinical characteristics between IVIG-resistantnd -responsive patients with KD

Resistant (n � 60) Responsive (n � 302) P value

ge in yearsa 2 (0.8-3.4) (range 2 mo-9 yrs) 2.4 (1.2-4.2) (range 2 mo-14 yrs) .061b

llness day at diagnosisa,c 5 (4-6) 6 (5-7) .002b

ales, n (%) 38 (63) 184 (61) NSd

thnicity/Race, n (% of total) NSd

Asian, n � 65 (18.0%) 9 (13.8%)e 56 (86.2%)Black, n � 16 (4.4%) 4 (25%) 12 (75%)Caucasian, n � 99 (27.3%) 16 (16.2%) 83 (83.8%)Hispanic, n � 113 (31.2%) 22 (19.5%) 91 (80.5%)AI/Alaskan, n � 1 (0.3%) 0 (0%) 1 (100%)Hawaiian/PI, n � 1 (0.3%) 0 (0%) 1 (100%)Mixed, n � 59 (16.3%) 9 (15.3%) 50 (84.7%)Unknown, n � 8 (2.2%) 0 (0%) 8 (100%)

S, Not significant; AI, American Indian; PI, Pacific Islander.Expressed as median (25th-75th percentiles).Wilcoxon rank sum test.Illness day at diagnosis: Illness Day 1 � the first day of fever.�2 test.

Number of patients with specified ethnicity/race who were IVIG-resistant or -responsive divided by total number of patients with that ethnicity/race.

21.e3 Tremoulet et al The Journal of Pediatrics • July 2008