Introduction

Currently, the standard therapy for pediatric patients with Kawasaki disease is the combination of intravenous immunoglobulin (IVIG) and aspirin (A). Platelets are activated during the acute phase of Kawasaki disease, which provides biological plausibility for antiplatelet therapy in these patients [13]. However, the role and impact of anti-inflammatory drugs (ADs), including high- or medium-dose A on IVIG therapy during the acute phase of Kawasaki disease remain unclear.

Two studies have shown that ADs may be unnecessary in the acute phase of Kawasaki disease [4, 9]. Another two studies disclosed that the prevalence of coronary artery lesions (CAL) differed between patients who received IVIG therapy without ADs and those who received concomitant ADs with IVIG [7, 10]. In one of these studies [7], the regimen of IVIG was 400 mg/kg day over 5 days, which is not standard at present. Therefore, it is important to clarify the role of ADs in pediatric patients initially treated with a 1–2-g/kg/day IVIG regimen for one day.

The hypothesis of this study was that ADs are unnecessary in the acute phase of Kawasaki disease and negatively impact the effects of IVIG on CAL suppression. Accordingly, this study aimed to investigate the necessity of ADs and their impact on initial IVIG therapy in the acute phase of Kawasaki disease.

Methods

This retrospective study included 182 consecutive patients (94 boys, 88 girls; mean age, 2 years and 9 months; range, 2 months to 13 years 3 months) who received IVIG therapy for Kawasaki disease between January 1999 and September 2013 at the Department of Pediatrics, Aomori Prefectural Central Hospital. Diagnosis of Kawasaki disease was based on the Japanese criteria (fifth edition) [12]. Patients with either disease relapse or CAL before therapy were excluded. All patients were divided into S and T groups. The S group comprised 111 patients who received initial single IVIG therapy with delayed administration of ADs, while the T group comprised 71 patients who received concomitant ADs with IVIG. In the S group, ADs were initiated within 24 h after the end of initial IVIG infusion.

AID Therapy

The choice between A and flurbiprofen (F) was made by each doctor after consideration of the patient’s liver function, and the timing of the influenza pandemic. Flurbiprofen was preferred before 2009. Aspirin was initiated at a dose of 30 mg/kg/day and decreased to 5–10 mg/kg/day when patients became afebrile. Flurbiprofen was initiated at a dose of 3–5 mg/kg/day and decreased to 3 mg/kg/day when the patient became afebrile. The regimen of the S group was used after 2004. A part of patients received this therapy using the regimen of the S group between 2004 and 2008. After 2009, this regimen was used for all patients.

IVIG Therapy

During the study period, an initial 2-g/kg/day IVIG regimen on the fifth day of illness was used as the first-line therapy when possible. The indication for additional therapy in resistant patients was determined between 48 and 72 h after the end of the initial IVIG infusion. Diagnosis was made according to clinical parameters, including body temperature, major signs of Kawasaki disease, general condition, and laboratory data. The second-line therapy was additional IVIG therapy, and the third-line therapy was urinastatin infusion. Written informed consent was obtained from the parents or guardians of all the patients before initial therapy.

Patients who responded to IVIG therapy were defined as those who became afebrile (temperature <37.5 °C for 24 h) within 24 h after completion of initial IVIG infusion. IVIG-resistant patients were defined as those not meeting these criteria.

Diagnosis of CAL

CAL was diagnosed by echocardiography according to the Japan Ministry of Health and Welfare criteria [11]. Coronary artery dilation was defined as an artery diameter exceeding 3 mm in a child aged below 5 years or exceeding 4 mm in a child aged 5 years or older. Transient dilatation was defined as the disappearance of CAL within 30 days of illness.

Statistical Analysis

Statistical analyses were performed with StatFlex® Ver. 5 and 6 for Windows (Artech Co., Ltd. Osaka, Japan). The Mann–Whitney U test, χ 2test, Fisher’s exact test, and logistic regression analysis were used as appropriately. A P value of <0.05 was considered statistically significant.

Results

Steroids were not administered to any patient. The initial IVIG regimens of 2 and 1 g/kg/day were used in 176 patients and 6 patients, respectively. The median start time of initial IVIG therapy was the fifth day of illness (range; day 3–16 of illness). Initial IVIG therapy resistance occurred in 45 of 182 patients (25 %), and 13 patients (7 %) received additional IVIG. Three patients received urinastatin.

The prevalence of CAL up to day 30 of illness was 7 % (12/182); after 30 days, it was 2 % (4/182). The maximal internal CAL diameters were within 4 mm in all patients. The 4 patients who developed CAL after 30 days of illness were evaluated by selective coronary arteriography at a median of 12.5 months (range, 6–85 months) after disease onset. The coronary arteriograms of all 4 patients revealed that all CALs regressed without leaving stenotic lesions.

The comparison of clinical findings between the S and T groups is shown in Table 1. The severity of Kawasaki disease before initial IVIG therapy, as evaluated using the Egami score [3], was not significantly different between groups. However, serum albumin levels before initial IVIG therapy were significantly lower in the S group than in the T group (P < 0.001). The following clinical findings were significantly different between the S and T groups: disease onset before 2003 (0 vs. 59 %, P < 0.001) and after 2009 (70 vs. 0 %, P < 0.001), use of 2-g/kg/day IVIG therapy (100 vs. 93 %, P = 0.034), AD type (A/F: 62/49 vs. 17/54, P < 0.001), and the prevalence of CAL up to (1/111 vs. 11/71, P < 0.001) and after 30 days (0/111 vs. 4/71, P = 0.022) of illness.

Table 1 Comparison of clinical findings between the S and T groups
Full size table

Logistic regression performed for all 182 patients, including IVIG therapy only (S group), 2-g/kg/day IVIG therapy, AD type (A/F), and disease onset after 2009 as variables revealed that IVIG therapy only (S group; P = 0.009) and 2-g/kg/day IVIG therapy (P = 0.015) were significant factors for CAL suppression (Table 2), while AD type (A/F) and disease onset after 2009 were not significant.

Table 2 The results of logistic regression analysis for coronary artery lesions
Full size table

Discussion

This study identified a possible negative impact of ADs on initial IVIG therapy in the acute phase of Kawasaki disease. The establishment of a safe and effective regimen for initial IVIG therapy and to suppress CAL development in the acute phase is clinically important.

The prevalence of CAL did not increase after the administration of IVIG therapy without ADs; this is consistent with the results of a recent study [9]. Low serum albumin levels suggest severe arteritis in the acute phase of Kawasaki disease. The lower serum albumin levels in the S group compared with those in the T group suggested more severe arteritis in the former. However, CAL was less prevalent in the S group and no patient in this group developed CAL after 30 days of illness. In addition, IVIG therapy only (S group) was a significant factor for CAL suppression according to logistic regression analysis. These findings suggest that initial IVIG therapy without ADs may be useful for the suppression of CAL due to Kawasaki disease.

It was previously reported that ADs, including A, affected the immunological function of T-cells [1, 2, 5]. A recent study suggested that the pathway comprising both T-cells and Fcγ receptors may play a role in the mechanism of action of IVIG [6, 14]. Furthermore, a recent immunological study highlighted that T cell activation in the early and middle stages was involved in the mechanism underlying cardiovascular injury in Kawasaki disease [15]. These findings suggest that ADs can alter the effects of IVIG on Kawasaki disease. Results of this study suggest that the use of ADs may negatively affect the ability of IVIG to suppress CAL development in patients with Kawasaki disease.

The previous study used the Kawasaki disease model to show that high-dose A enhances tumor necrosis factor (TNF)-alpha, which promotes CAL development. These findings from that study suggested that further research on the biological effects of A in patients with acute Kawasaki disease was required to facilitate rational therapy [8]. The present study also revealed the necessity of determining the optimal time for adding ADs to initial IVIG therapy in patients with Kawasaki disease.

The limitation to this study is the small number of patients. In addition, this is a retrospective study. Finally, the use of Japanese Ministry of Health and Welfare criteria may underestimate the true incidence of CAL due to Kawasaki disease [16].

Conclusions

Aspirin and flurbiprofen appeared to have a negative impact on the suppressive effects of initial IVIG therapy on CAL development in the acute phase of Kawasaki disease; these results may also apply to other ADs. The prevalence of CAL did not increase after the administration of initial single IVIG therapy with delayed administration of ADs; therefore, this approach may be useful for the suppression of CAL due to Kawasaki disease.