SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Objective

To investigate the roles of serum Th1 and Th2 cytokines in Kawasaki disease (KD) and determine whether the Th1/Th2 cytokine profiles in children with KD may be involved in intravenous immunoglobulin (IVIG) resistance and development of coronary artery lesions (CALs).

Methods

Serum Th1 and Th2 cytokines, including interferon-γ (IFNγ), tumor necrosis factor α (TNFα), interleukin-10 (IL-10), IL-6, IL-4, and IL-2, were measured using a cytometric bead array in the serum of 143 patients with KD before and after treatment with IVIG (pre-IVIG, at 3 days after temperature normalization following IVIG treatment [post-IVIG], and 1 month posttreatment).

Results

Levels of IL-6, IL-10, TNFα, and IFNγ were significantly increased in KD patients pre-IVIG. Post-IVIG, the levels of IL-6, IL-10, and IFNγ quickly decreased. The levels of TNFα decreased significantly after IVIG treatment in KD patients without CALs post-IVIG and in KD patients who were IVIG responders, but increased slightly in KD patients with CALs post-IVIG and in KD patients who were IVIG nonresponders. Before IVIG treatment, the levels of IL-4, IL-6, IL-10, and IFNγ were significantly higher in KD patients with CALs than in those without CALs. The post-IVIG levels of IL-6 and IL-10 were significantly higher in IVIG nonresponders than in IVIG responders. Pre-IVIG, an IL-10 level >8 pg/ml had a sensitivity of 75.0% and a specificity of 64.4% for predicting CALs, while a TNFα level <2 pg/ml had a sensitivity of 66.7% and a specificity of 74.2% for predicting IVIG resistance. Post-IVIG, an IL-6 level >10 pg/ml had a sensitivity of 67.9% and a specificity of 81.7% for predicting CALs, while an IL-10 level >6 pg/ml had a sensitivity of 53.6% and a specificity of 86% for predicting CALs.

Conclusion

Determination of the serum Th1/Th2 cytokine profile may be helpful for predicting the disease prognosis and targeting treatment strategies in patients with KD.

Kawasaki disease (KD) is an acute childhood vasculitis of unknown etiology that most frequently affects infants and children younger than 5 years of age. It was first described by Dr. Tomasaku Kawasaki in 1967 (1). KD is characterized by high fever, rashes, cervical lymphadenopathy, conjunctivitis, oral enanthema, and erythematous induration of the hands and feet (2). Coronary vascular lesions would previously occur in ∼20% of KD patients, but with the advent of treatment with intravenous immunoglobulin (IVIG), the aneurysm rate has been reduced to 3–5%. Although the etiology of KD remains a mystery, important progress has been made in characterizing the features of the arterial wall, myocardial pathology, and long-term clinical consequences of KD (3).

Both the clinical and the epidemiologic features of KD strongly support the notion that an infectious agent may be the inducing factor, but it may not necessarily be the proximate cause of the disease. Some studies have proposed the idea that KD is related to a bacterial superantigenic toxin, or superantigen. To date, most investigators believe that derangement of the immune system and functional disorder of T cells are the primary pathophysiologic features in patients with KD. However, the definitive cause of KD is as yet unknown, and the mode of action of IVIG remains largely unclear.

There are still a number of important unanswered questions regarding the pathogenesis of KD. Accurate diagnosis and early therapeutic interventions, such as a high dose of IVIG, can decrease the risk of developing coronary artery abnormalities. Approximately 10% of patients are unresponsive to IVIG, and these patients may, unfortunately, experience higher incidence rates of coronary artery abnormalities (4).

Recent studies have shown that stimulation of the cytokine cascade and activation of endothelial cells are key events that occur in the acute phase of KD. The concentrations of many proinflammatory cytokines and chemokines are higher than normal during the acute phase of the disease (5). These include tumor necrosis factor α (TNFα), interleukins such as interleukin-4 (IL-4), and interferon-γ (IFNγ).

TNFα (also known as cachectin) is a monocyte-derived cytotoxin that has been implicated in tumor regression, septic shock, and cachexia (6). IL-4 is a cytokine that induces differentiation of naive T helper cells to Th2 cells. IL-6 is an interleukin that acts as both a proinflammatory and antiinflammatory cytokine. It can be secreted by T cells and macrophages to stimulate an immune response to specific microbial molecules, referred to as pathogen-associated molecular patterns (7, 8). IL-10 is a powerful Th2 cell cytokine produced by lymphoid cells that exerts its functions by inhibiting the replication of monocyte/macrophages and T cell lymphocytes and the secretion of inflammatory cytokines (9). IFNγ is produced mainly by natural killer cells and T cells, including CD8+ T cells and Th1 T cells. Those cytokines play an important role in the progression from systemic activation of the immune system to local in- inflammation in coronary vessels. The effects of cytokines on the coronary artery are wide-ranging, from the activation of endothelial cells, up-regulation of adhesion molecules, and induction of inflammatory histopathologic features (10). In addition, a gene-dose effect of cytokines on the risk of KD has been found (11).

Although the cytokine profiles are related to the pathogenesis of KD, a clear profile of Th1/Th2 cytokines and its predictive significance in the treatment and prognosis of KD have not been reported. In this study, we investigated the Th1/Th2 cytokine profiles, including the levels of IL-2, IL-4, IL-6, IL-10, TNFα, and IFNγ, in the serum of patients with KD. The levels of these cytokines were determined rapidly by flow cytometry in patients with KD at several time points, at the acute stage of the disease before IVIG treatment (pre-IVIG), at 3 days after the patients' temperature had returned to normal following IVIG treatment (post-IVIG), and at 1 month post–IVIG treatment.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Patients and healthy controls.

This study was approved by the ethics committee of the Children's Hospital, Zhejiang University School of Medicine. The study protocol conforms to the ethics guidelines of the 1975 Declaration of Helsinki, as was reflected in the a priori approval provided by the institution's human research committee. Informed consent was obtained from each patient's parent or guardian.

For this study, 143 patients with KD (82 boys and 61 girls), whose median age was 2.7 years (range 2 months to 12.5 years), were enrolled in this study from January 2011 to August 2012. The duration of fever was 5–20 days (mean ± SD 7.5 ± 3.4 days). During the acute stage of the disease, the mean ± SD erythrocyte sedimentation rate was 68 ± 33 mm/hour, and the C-reactive protein level was 90 ± 48 mg/liter. All of the patients with KD fulfilled the diagnostic criteria established by the Kawasaki Disease Research Committee (12).

All patients were treated with IVIG at a dosage of 1 gm/kg/day for 2 days and oral aspirin at 30–50 mg/kg/day. After 3–5 days of treatment, when the patients' temperature had returned to values within the normal range, the dosage of aspirin was reduced to 3–5 mg/kg/day for 12 weeks. There were 15 KD patients (9 boys and 6 girls) who were deemed to be IVIG resistant (designated nonresponders), whose median age was 2.9 years (range 2 months to 8.1 years) and whose temperatures were still higher than 38°C after 48 hours of standard treatment. These patients continued to receive IVIG treatment at a dosage of 1 gm/kg/day for another 2 days (13).

Twenty-eight patients (18 boys and 10 girls), whose median age was 3.1 years (range 13 months to 8.5 years), were found to have coronary artery lesions (CALs) by echocardiography during the disease process (within 3 months of diagnosis). The CALs in the majority of these KD patients were temporary and disappeared within 3 months of onset, but the CALs persisted for more than 3 months in 10 patients, persisted for more than 6 months in 5 patients, and persisted for more than 12 months in 2 patients. CALs were defined according to the following criteria: 1) internal lumen diameter of >2.5 mm in children <3 years of age, >3 mm in children 3–9 years of age, and >3.5 mm in children 9–14 years of age; 2) internal diameter of a segment measuring ≥1.5 times that of an adjacent segment; and 3) lumen that was clearly irregular (14). For all patients with CALs, treatment with either dipyridamole or warfarin was added, depending on the severity of the CALs.

From January 2011 to August 2012, control samples were obtained from 80 healthy children (46 boys and 34 girls, median age 5.4 years [range 8 months to 13.7 years]) without any evidence of infection at the time of a routine health visit. For the sepsis group, 70 patients with fever symptoms in whom bacterial sepsis was diagnosed based on positive blood cultures (38 boys and 32 girls, median age 2.4 years [range 1 month to 12.3 years]) and who were treated in our hospital were enrolled. For the Epstein-Barr virus (EBV) infection group, 20 patients who had a definitive diagnosis of EBV infection (10 boys and 10 girls, median age 4.1 years [range 9 months to 10.9 years]) and who were treated in our hospital were enrolled.

Analysis of blood samples.

Venous blood samples were collected from the KD patients at the acute stage of disease before IVIG treatment, at 3 days after the patients' temperature had returned to values in the normal range post–IVIG treatment, and at 1 month posttreatment. The blood samples from the healthy control group were collected at the time of the routine health visit. The blood samples from patients in the sepsis and EBV groups were collected at the time of admission to the hospital.

One milliliter of blood was transferred to a serum-separating tube and centrifuged at 1,000g at 20°C for 20 minutes after clotting. The sera were carefully harvested, and the aliquots were stored at 2–8°C until analyzed (within 12 hours). The serum concentrations of IL-2, IL-4, IL-6, IL-10, TNFα, and IFNγ were quantitatively determined using a cytometric bead array (CBA) kit (CBA Human Th1/Th2 Cytokine Kit II; BD Biosciences) as previously described (15). Briefly, the CBA technique was based on 6 bead populations, each with a distinct fluorescence intensity, that had been coated with capture antibodies specific for IL-2, IL-4, IL-6, IL-10, TNFα, and IFNγ proteins. The fluorescent dye had a maximal emission wavelength of ∼650 nm (FL-3), which was detectable by flow cytometry. The cytokine capture beads were mixed with phycoerythrin-conjugated detection antibodies, and then incubated with recombinant standards or test samples to form sandwich complexes. Following the acquisition of sample data on a FACSCalibur flow cytometer (Becton Dickinson), the sample results were generated in graphic and tabular format using BD Biosciences CBA software. Six standard curves (ranging from 0 to 5,000 pg/ml) were obtained from one set of calibrators, and 6 results were obtained from each test sample.

Statistical analysis.

Analysis of the difference between the groups of patients was accomplished using analysis of variance followed by the least significant difference test for multiple comparisons of normal data, while the Mann-Whitney test was applied to analyze nonparametric data. Receiver operating characteristic (ROC) curves were derived from the cytokine levels in all KD patients. In an ROC curve, the sensitivity, specificity, positive predictive value, and negative predictive value for the prediction of CALs and IVIG resistance in the KD patients were calculated by combining the optimal cutoff values for each cytokine. The difference was considered significant at P values less than 0.05. All data analyses were performed using SPSS version 16.0 software.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Serum Th1 and Th2 cytokine levels in patients with KD.

Prior to IVIG treatment, the levels of IL-6 were significantly increased in KD patients. Among these patients, 75 (52.4%) had IL-6 levels >27.4 pg/ml, 55 (38.5%) had IL-6 levels >50 pg/ml, 38 (26.6%) had IL-6 levels >100 pg/ml, and 7 (4.9%) had IL-6 levels >1,000 pg/ml. The IL-6 level decreased quickly after IVIG treatment, with a return to normal levels in the majority of patients (Table 1 and Figure 1), but 15 patients (10.5%) still had an IL-6 level >50 pg/ml and 10 patients (7%) still had an IL-6 level >100 pg/ml.

Table 1. Serum Th1 and Th2 cytokine levels pre- and post-IVIG in patients with Kawasaki disease, compared to healthy control subjects, patients with sepsis, and patients with EBV infection*
Th1/Th2 cytokineHealthy controls (n = 80)Patients with Kawasaki disease  
Pre-IVIG (n = 143)Post-IVIG (n = 143)One month post-IVIG (n = 33)Patients with sepsis (n = 70)Patients with EBV (n = 20)
  • *

    Values are the median (range) pg/ml. IVIG = intravenous immunoglobulin; EBV = Epstein-Barr virus; IL-2 = interleukin-2; TNFα = tumor necrosis factor α; IFNγ = interferon-γ.

  • P < 0.01 versus patients with Kawasaki disease pre-IVIG.

  • P < 0.05 versus patients with Kawasaki disease pre-IVIG.

IL-22.9 (1.4–5.8)2.0 (1.0–12.2)2.0 (1.0–17.0)1.5 (0.9–2.1)2.5 (1.0–10.0)2.3 (1.6–6.6)
IL-43.0 (1.2–4.5)1.9 (0.7–23.3)1.9 (0.8–17.3)2.1 (0.8–3.3)2.1 (0.4–17.8)3.1 (1.3–6.1)
IL-65.4 (3.0–27.4)30.9 (2.6–5,000)4.5 (1.0–567.2)4.0 (1.9–26.8)226.7 (2.1–5,000)22.2 (5.1–70.2)
IL-106.7 (2.9–19.8)6.7 (1.5–152.3)3.9 (1.2–21.8)3.1 (1.5–17.3)37.5 (1.2–5,000)34.4 (9.1–190.7)
TNFα3.8 (1.5–7.0)2.5 (1.0–226.9)2.4 (1.0–72.3)2.5 (1.5–17.1)3.6 (1.0–1,194)4.0 (2.2–13.0)
IFNγ5.4 (1.9–16.4)3.4 (1.0–204.8)2.6 (1.0–146.5)2.8 (1.1–14.1)7.8 (1.5–232.1)12.1 (3.5–67.1)
thumbnail image

Figure 1. Serum Th1 and Th2 cytokine levels in patients with Kawasaki disease before and after treatment with intravenous immunoglobulin (IVIG) (pre-IVIG, at 3 days after temperature normalization following IVIG treatment [post-IVIG], and at 1 month after IVIG treatment). Healthy children were used as the control group. Solid circles represent individual subjects; horizontal bars show the median. IL-2 = interleukin-2; TNFα = tumor necrosis factor α; IFNγ = interferon-γ.

Download figure to PowerPoint

The levels of IL-10 were also significantly increased in KD patients pre-IVIG. Of these patients, 31 (21.7%) had an IL-10 level >19.8 pg/ml and 6 (4.2%) had an IL-10 level >50 pg/ml. The cytokine level decreased quickly after IVIG treatment, and only 1 patient continued to have an IL-10 level >19.8 pg/ml.

The median TNFα level in KD patients pre-IVIG was 2.5 pg/ml (range 1–226.9 pg/ml); in 15 patients (10.5%), it was >7.0 pg/ml. Post-IVIG, the TNFα level did not decrease significantly, and 13 patients (9.1%) had a TNFα level >7.0 pg/ml. A significantly elevated TNFα level (>20 pg/ml) was found in 5 patients pre-IVIG, but in 7 patients post-IVIG.

The median IFNγ level in KD patients pre-IVIG was 3.4 pg/ml (range 1–204.8 pg/ml). In 9 patients, the IFNγ level was >16.4 pg/ml pre-IVIG, but this decreased quickly post-IVIG, to a median level of 2.6 pg/ml (range 1–146.5 pg/ml), and the level remained high in only 1 patient.

The median IL-2 levels in KD patients pre- and post-IVIG were 2.0 pg/ml (range 1–12.2 pg/ml) and 2.0 pg/ml (range 1–17 pg/ml), respectively. Only 2 patients had a level of IL-2 that was >5.8 pg/ml.

The median IL-4 levels in KD patients pre- and post-IVIG were 1.9 pg/ml (range 0.7–23.3 pg/ml) and 1.9 pg/ml (range 0.8–17.3 pg/ml), respectively. Only 3 patients had an IL-4 level >4.5 pg/ml.

In all KD patients, the levels of IL-2, IL-4, IL-6, IL-10, and IFNγ were within the normal range at 1 month posttreatment. The TNFα levels were within the normal range at 1 month posttreatment in the majority of patients, although 2 patients continued to have levels that were higher than normal.

Serum Th1 and Th2 cytokine levels in patients with sepsis and patients with EBV infection, compared to patients with KD.

The levels of IL-6, IL-10, TNFα, and IFNγ were significantly higher in patients with sepsis compared to KD patients pre-IVIG, whereas there was no difference in the levels of IL-2 and IL-4 between the sepsis and KD pre-IVIG groups (Table 1). Among patients with EBV infection, the levels of IL-6 and TNFα were significantly lower than those in KD patients pre-IVIG, but the levels of IL-10 and IFNγ were significantly higher in patients with EBV infection compared to KD patients pre-IVIG. Levels of IL-2 and IL-4 were not significantly different between the EBV and KD pre-IVIG groups.

Serum Th1 and Th2 cytokine levels in KD patients with CALs and those without CALs.

In KD patients in whom CALs were present during the disease process, the pre-IVIG levels of IL-4, IL-6, IL-10, and IFNγ were significantly higher than in KD patients who did not have CALs during the disease process (Table 2). The post-IVIG levels of IL-2, IL-4, IL-6, IL-10, TNFα, and IFNγ were significantly higher than in those without CALs. Although the levels of IL-10 and IFNγ in KD patients with CALs were slightly higher than in those without CALs after 1 month of IVIG treatment, these cytokines had levels within the normal range in both groups after 1 month. The levels of IL-2, IL-4, IL-6, and TNFα at 1 month posttreatment were not significantly different between KD patients with CALs and those without CALs.

Table 2. Serum Th1 and Th2 cytokine levels pre- and post-IVIG in patients with Kawasaki disease in whom CALs were absent and those in whom CALS were present during the disease process*
Th1/Th2 cytokineWithout CALs (n = 115)Pre-IVIGPost-IVIGOne month post-IVIG
With CALs (n = 28)Without CALs (n = 115)With CALs (n = 28)Without CALs (n = 27)With CALs (n = 6)
  • *

    Values are the median (range) pg/ml. IVIG = intravenous immunoglobulin; CALs = coronary artery lesions; IL-2 = interleukin-2; TNFα = tumor necrosis factor α; IFNγ = interferon-γ.

  • P < 0.05 versus patients without CALs at the same time point.

  • P < 0.05 versus patients pre-IVIG.

  • §

    P < 0.05 versus patients post-IVIG.

  • P < 0.01 versus patients without CALs at the same time point.

  • #

    P < 0.01 versus patients pre-IVIG.

  • **

    P < 0.01 versus patients post-IVIG.

IL-22.0 (1.0–11.5)2.2 (1.0–12.2)1.9 (1.0–17.0)2.8 (1.7–15.5)1.4 (0.9–2.1)§1.6 (1.3–2.0)§
IL-41.7 (0.7–5.1)2.1 (1.1–23.3)1.8 (0.8–5.9)2.7 (0.8–17.3)2.0 (0.9–3.3)2.0 (0.8–2.7)§
IL-629.2 (2.6–3,236.5)63.4 (2.8–5,000)4.1 (1.0–234.1)#24.0 (2.0–567.2)#3.8 (1.9–26.8)§5.3 (2.2–8.6)**
IL-106.0 (1.5–152.3)11.5 (3.0–71.0)3.5 (1.2–19.1)#7.5 (2.8–21.8)#3.0 (1.8–5.5)§3.8 (1.5–17.3)§
TNFα2.5 (1.0–226.9)2.7 (1.0–29.0)2.3 (1.0–72.3)3.3 (1.0–42.5)2.3 (1.5–17.1)§2.9 (1.7–3.7)§
IFNγ3.2 (1.0–204.8)4.0 (1.6–97.1)2.4 (1.0–8.1)#3.4 (1.2–146.5)2.8 (1.1–6.0)3.0 (2.0–14.1)§

In patients without CALs during the disease process, the levels of IL-2 and IL-4 did not change from pre- to post-IVIG, but in patients with CALs during the disease process, the IL-2 and IL-4 levels slightly increased from pre- to post-IVIG. The levels of both cytokines then decreased from the post-IVIG period to 1 month after treatment.

Levels of IL-6 in KD patients decreased significantly from pre- to post-IVIG, but the decrease was significantly greater in KD patients without CALs than in those with CALs; the median IL-6 levels decreased from 29.2 pg/ml to 4.1 pg/ml in KD patients without CALs and decreased from 63.4 pg/ml to 24 pg/ml in KD patients with CALs. After 1 month of IVIG treatment, the IL-6 levels in all KD patients recovered to normal levels.

Levels of IL-10 in KD patients decreased significantly from pre- to post-IVIG. In KD patients without CALs, the IL-10 levels recovered to normal levels after IVIG treatment. However, in KD patients with CALs, the IL-10 levels were slightly higher than normal after treatment, and then recovered to normal levels at 1 month after treatment.

The TNFα levels decreased significantly from pre- to post-IVIG in KD patients without CALs, but increased slightly from pre- to post-IVIG in patients in whom CALs were present during the disease process. Two of these patients with CALs had a level of TNFα >20 pg/ml pre-IVIG, while 4 of these patients had a TNFα level >20 pg/ml post-IVIG.

Among KD patients in whom no CALS were present during the disease process, the IFNγ levels decreased significantly after IVIG treatment, and recovered to normal levels post-IVIG. In KD patients in whom CALs were present during the disease process, the IFNγ levels showed no significant change from pre- to post-IVIG, but the levels decreased significantly and recovered to normal levels after 1 month of treatment.

Serum Th1 and Th2 cytokine levels in KD patients who were IVIG responders compared to those who were IVIG nonresponders.

Prior to treatment with IVIG, the levels of IL-4, IL-6, and TNFα were slightly lower in IVIG nonresponders compared to IVIG responders, but the IL-10 level was slightly higher in IVIG nonresponders compared to IVIG responders (Table 3). Post-IVIG, the IL-6 and IL-10 levels were significantly higher in IVIG nonresponders than in IVIG responders. Although the IL-10 level was significantly higher in IVIG nonresponders than in IVIG responders after 1 month of treatment, the level was within the normal range in both groups after 1 month.

Table 3. Serum Th1 and Th2 cytokine levels pre- and post-IVIG in patients with Kawasaki disease who were IVIG responders and those who were nonresponders*
Th1/Th2 cytokineResponders (n = 128)Pre-IVIGPost-IVIGOne month post-IVIG
Nonresponders (n = 15)Responders (n = 128)Nonresponders (n = 15)Responders (n = 30)Nonresponders (n = 3)
  • *

    Values are the median (range) pg/ml. IVIG = intravenous immunoglobulin; IL-2 = interleukin-2; TNFα = tumor necrosis factor α; IFNγ = interferon-γ.

  • P < 0.05 versus responders or nonresponders post-IVIG.

  • P < 0.05 versus responders at same time point.

  • §

    P < 0.05 versus responders or nonresponders pre-IVIG.

  • P < 0.01 versus responders or nonresponders pre-IVIG.

  • #

    P < 0.01 versus responders at same time point.

  • **

    P < 0.01 versus responders or nonresponders post-IVIG.

IL-21.9 (1.0–12.2)2.4 (1.2–11.5)1.9 (1.0–17.0)2.8 (1.1–5.2)1.5 (0.9–2.1)1.3 (1.0–1.9)
IL-41.9 (0.8–23.3)1.4 (0.7–3.8)1.9 (0.8–17.3)2.0 (1.1–5.4)§2.0 (0.8–3.3)2.4 (1.9–2.7)
IL-631.8 (2.6–5,000)19.8 (4.7–377.5)4.2 (1.0–567.2)15.6 (1.7–456.7)#3.9 (1.9–26.8)**4.9 (3.3–6.8)**
IL-106.7 (1.5–152.3)9.5 (2.6–87.4)3.6 (1.2–21.8)7.3 (4.0–12.1)3.0 (1.5–5.5)7.6 (2.2–17.3)
TNFα2.6 (1.0–226.9)1.8 (1.0–3.7)2.4 (1.0–72.3)§2.9 (1.0–9.1)§2.5 (1.5–17.1)2.4 (1.8–3.3)
IFNγ3.3 (1.0–204.8)4.0 (1.4–38.1)2.6 (1.0–146.5)§3.0 (1.8–10.2)§2.8 (1.1–14.1)4.0 (1.9–7.6)

From pre- to post-IVIG, the IL-6 and IL-10 levels decreased significantly in IVIG responders (from a median 31.8 pg/ml to 4.2 pg/ml for IL-6 and from 6.7 pg/ml to 3.6 pg/ml for IL-10), but showed no significant change in IVIG nonresponders (from 19.8 pg/ml to 15.6 pg/ml for IL-6 and from 9.5 pg/ml to 7.3 pg/ml for IL-10). After 1 month of IVIG treatment, the IL-6 and IL-10 levels in all KD patients recovered to normal levels.

From pre- to post-IVIG, the TNFα level decreased significantly in IVIG responders, but increased slightly in IVIG nonresponders. After 1 month of IVIG treatment, the TNFα levels in the majority of KD patients recovered to normal levels.

In both IVIG responders and nonresponders, the IFNγ levels decreased progressively from pre- to post-IVIG. Furthermore, in both groups, the IFNγ levels recovered to normal levels after 1 month of treatment.

Predictive sensitivity and specificity of the Th1 and Th2 cytokine levels for CALs and IVIG resistance.

Pre-IVIG, IL-4 levels had a sensitivity of 17.9% and a specificity of 93.0% for predicting CALs when the cutoff point was >3 pg/ml (Table 4). IL-10 levels had a sensitivity of 75.0% and a specificity of 64.4% for predicting CALs when a cutoff point of >8 pg/ml was applied, and had a sensitivity of 39.3% and a specificity of 77.4% for predicting CALs at a cutoff point of >15 pg/ml. TNFα levels had a sensitivity of 66.7% and a specificity of 74.2% for predicting IVIG resistance when the cutoff point was <2 pg/ml (Table 5).

Table 4. Sensitivity and specificity of Th1 and Th2 cytokine levels pre- and post-IVIG for predicting coronary artery lesions in patients with Kawasaki disease*
Time point, Th1/ Th2 cytokine, cutoffSensitivity, % (95% CI)Specificity, % (95% CI)PPV, % (95% CI)NPV, % (95% CI)
  • *

    IVIG = intravenous immunoglobulin; 95% CI = 95% confidence interval; PPV = positive predictive value; NPV = negative predictive value; IL-4 = interleukin-4.

Pre-IVIG    
 IL-4    
  >3 pg/ml17.86 (7.88–35.59)93.04 (86.87–96.43)38.46 (17.71–64.48)82.31 (74.85–87.91)
 IL-10    
  >3 pg/ml96.43 (82.29–99.37)13.91 (8.75–21.41)27.43 (15.17–29.38)94.12 (73.02–98.95)
  >8 pg/ml75 (56.64–87.32)64.35 (55.26–72.51)33.87 (23.34–46.28)91.36 (83.22–95.75)
  >15 pg/ml39.29 (23.57–57.59)77.39 (68.93–84.08)29.73 (17.49–45.78)83.96 (75.81–89.74)
Post-IVIG    
 IL-2    
  >1.8 pg/ml96.43 (82.29–99.37)48.70 (39.75–57.72)31.40 (22.56–41.82)98.25 (90.71–99.69)
  >3 pg/ml21.43 (10.21–39.54)81.74 (73.69–87.73)22.22 (10.61–40.76)81.03 (72.95–87.13)
 IL-4    
  >1.5 pg/ml92.86 (77.35–98.02)42.61 (33.95–51.74)28.26 (20.08–38.19)96.08 (86.78–98.92)
  >2.1 pg/ml78.57 (60.46–89.79)65.25 (56.3–73.24)34.92 (24.33–47.25)90.59 (82.51–95.15)
  >3 pg/ml28.57 (15.25–47.06)91.30 (84.73–95.21)44.44 (24.56–66.28)77.78 (70.05–83.97)
 IL-6    
  >3 pg/ml92.86 (77.35–98.02)20.87 (14.44–29.18)22.22 (15.64–30.57)92.31 (75.86–97.86)
  >4.2 pg/ml80.77 (62.12–91.49)53.04 (43.97–61.92)28 (19.10–39.04)92.42 (83.46–96.72)
  >10 pg/ml67.86 (49.34–82.07)81.74 (73.69–87.73)47.50 (32.94–62.50)91.26 (84.22–95.33)
  >100 pg/ml28.57 (15.25–47.06)98.26 (93.88–99.52)80 (49.02–94.33)84.96 (77.91–90.05)
 IL-10    
  >2.7 pg/ml100 (87.94–100)23.48 (16.67–32)24.14 (17.26–32.67)100 (87.54–100)
  >3.9 pg/ml82.14 (64.41–92.12)58.26 (49.12–66.86)32.39 (22.66–43.94)93.06 (84.75–97)
  >6 pg/ml53.57 (35.81–70.47)85.96 (78.41–91.17)48.39 (31.97–65.16)88.29 (80.99–93.03)
Table 5. Sensitivity and specificity of Th1 and Th2 cytokine levels pre- and post-IVIG for predicting IVIG resistance in patients with Kawasaki disease*
Time point, Th1/Th2 cytokine, cutoffSensitivity, % (95% CI)Specificity, % (95% CI)PPV, % (95% CI)NPV, % (95% CI)
  • *

    IVIG = intravenous immunoglobulin; 95% CI = 95% confidence interval; PPV = positive predictive value; NPV = negative predictive value; TNFα = tumor necrosis factor α; IL-6 = interleukin-6.

Pre-IVIG    
 TNFα    
  <4 pg/ml100 (79.61–100)24.22 (17.62–32.32)13.39 (8.29–20.93)98.63 (92.64–99.76)
  <2 pg/ml66.67 (41.71–84.82)74.22 (66.01–81.01)23.26 (13.15–37.74)95 (88.82–97.85)
Post-IVIG    
 IL-6    
  >4.55 pg/ml93.33 (70.18–98.81)61.02 (52–69.34)23.33 (14.44–35.44)98.63 (92.64–99.76)
  >10 pg/ml66.67 (41.71–84.82)75.78 (67.68–82.38)24.39 (13.83–39.34)95.10 (89.03–97.89)
  >200 pg/ml20 (7.05–45.19)98.44 (94.48–99.57)60 (23.07–88.24)91.30 (85.42–94.96)
 IL-10    
  >3.95 pg/ml100 (79.61–100)61.02 (52–69.34)24.59 (15.51–36.68)100 (94.93–100)
  >5 pg/ml73.33 (48.05–89.1)71.18 (63.54–78.94)23.40 (13.60–37.22)95.83 (89.77–98.37)
  >10 pg/ml33.33 (15.18–58.29)89.84 (83.4–93.97)27.78 (12.50–50.87)92 (85.9–95.6)

Post-IVIG, IL-2 levels had a sensitivity of 96.4% and a specificity of 48.7% for predicting CALs when a cutoff point of >1.8 pg/ml was applied. IL-4 levels had a sensitivity of 92.9% and a specificity of 42.6% for predicting CALs at a cutoff point of >1.5 pg/ml, and had a sensitivity of 28.6% and a specificity of 91.3% for predicting CALs at a cutoff point of >3 pg/ml. IL-6 levels had a sensitivity of 67.9% and a specificity of 81.7% for predicting CALs when the cutoff point was >10 pg/ml. IL-10 levels had a sensitivity of 53.6% and a specificity of 86% for predicting CALs at a cutoff point of >6 pg/ml.

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Th1 cells play an important role in cellular immunity by secreting IL-2 and IFNγ, while the helper function of Th2 cells involves the development of antibody-producing B cells via the secretion of IL-4, IL-5, IL-6, and IL-10. TNFα is produced by both T cells and macrophages after superantigenic stimulation. Although activation of the immune system and production of various cytokines have both been reported in patients with KD, the role of T cells in KD and the functional state of Th1 and Th2 cells in KD are still not fully understood.

In our research, the increase in IL-6 levels was the most obvious of the changes observed among all cytokines tested. Pre-IVIG, the median level of IL-6 in KD patients was more than 5.7-fold higher than that in healthy controls. Although the IL-6 level in KD patients decreased significantly post-IVIG, it still remained higher than that in healthy children. The decrease in the IL-6 level was more significant in KD patients without CALs than in those with CALs. Pre-IVIG and post-IVIG, the percentages of patients whose IL-6 level was higher than that in healthy controls were 50.9% and 4.3%, respectively, in KD patients without CALs and 57.1% and 46.4%, respectively, in KD patients with CALs. The IL-6 level in KD patients with CALs was significantly higher than that in KD patients without CALs at both the pre-IVIG and post-IVIG time points (median 2.17-fold higher and 5.85-fold higher, respectively).

Post-IVIG, an IL-6 level >10 pg/ml had a high sensitivity (67.9%) and high specificity (81.7%) for predicting CALs. Pre-IVIG, the IL-6 level in IVIG nonresponders was slightly lower than that in IVIG responders. From pre- to post-IVIG, the IL-6 level decreased significantly in IVIG responders; the percentages of patients whose levels were higher than those in healthy controls decreased from 53.1% pre-IVIG to 10.1% post-IVIG. However, no significant change in the IL-6 level occurred in IVIG nonresponders from pre- to post-IVIG; the percentages of patients whose levels were higher than those in healthy controls decreased from 46.7% pre-IVIG to 33.3% post-IVIG. After IVIG treatment, the IL-6 level was significantly higher in IVIG nonresponders than in IVIG responders (median 3.71-fold higher). After 1 month of treatment, the IL-6 levels in all KD patients recovered to normal levels.

These results suggest that a high level of IL-6 may be related to the occurrence of CALs in KD. The slower decrease in IL-6 levels after treatment with IVIG may be related to IVIG resistance and the occurrence of CALs in these KD patients. Thus, the attenuation of proinflammatory cytokine responses, especially IL-6, following infusions of IVIG may play an integral role in the rapid resolution of symptoms and in reduced levels of acute-phase proteins in children with KD (16).

A number of studies have demonstrated interindividual variability in IL-10 production in stimulated peripheral blood mononuclear cells, leading to the hypothesis that variation in IL-10 levels may be due to genetic polymorphisms (8). The effects of IL-10 genetic polymorphisms on CALs are important, and this has been supported by the results of studies assessing IL-10 polymorphisms and CALs in patients with KD (17, 18). In this study, we found that the IL-10 level was markedly elevated pre-IVIG and decreased substantially after IVIG treatment in IVIG responders, but there was no significant change in IVIG nonresponders from pre- to post-IVIG. Pre-IVIG and post-IVIG, the IL-10 level was significantly higher in KD patients with CALs than in those without CALs (median 1.92-fold higher and 2.14-fold higher, respectively). Moreover, the IL-10 levels pre-IVIG and post-IVIG were significantly higher in IVIG nonresponders than in IVIG responders (median 1.42-fold higher and 2.03-fold higher, respectively).

Pre-IVIG, an IL-10 level >8 pg/ml had a high sensitivity (75.0%) and high specificity (64.4%) for predicting CALs. Post-IVIG, an IL-10 level >6 pg/ml had a high sensitivity (53.6%) and high specificity (86%) for predicting CALs. This finding suggests that there is a significant relationship between the IL-10 level and CALs in patients with KD, and that KD patients may develop CALs when their IL-10 levels significantly increase from pre- and post-IVIG. This finding also suggests that KD patients whose IL-10 levels do not decrease after IVIG treatment may be IVIG resistant.

In this study, the IFNγ levels in the majority of KD patients appeared to stay within the normal range. This finding is consistent with that from a study by Matsubara et al (19), in which the authors observed a decrease in the levels of IFNγ-producing CD3+ T cells during the acute stage of KD. This may also reflect the hypofunction of a proportion of peripheral blood T cells during the acute stage of KD (20). However, in our study, the IFNγ levels pre-IVIG and post-IVIG were higher in KD patients with CALs than in those without CALs (median 1.25-fold higher and 1.42-fold higher, respectively). The IFNγ level decreased progressively from pre- to post-IVIG, and recovered to normal levels after 1 month of treatment. There was no difference in the IFNγ levels between IVIG responders and nonresponders. This finding indicates that KD patients with significantly increased IFNγ levels may develop CALs. IFNγ is now generally considered to be a cytokine with immune-suppressive activities. It is also known to be secreted by many cell types, including B cells, macrophages, dendritic cells, and CD4+ T cells. Although some recent studies have addressed the effects of IFNγ in patients with KD (19–22), the role of IFNγ in KD patients and the relationship between IFNγ and CALs in KD patients are largely unknown, and therefore further study is warranted.

TNFα is a key inflammatory cytokine that is initially produced by T lymphocytes, followed by a secondary release of TNFα from monocyte/macrophages within 24 hours. The pleiotropic effects of TNFα work in synergy with IFNγ to activate and recruit other immune cell populations to sites of inflammation. TNFα mediates endothelial cell activation through the increased expression of adhesion molecules, such as intercellular adhesion molecule 1, vascular cell adhesion molecule 1, and E-selectin. TNFα can also up-regulate the expression of chemokines, such as macrophage inflammatory protein 1α and RANTES, which are important in the orchestration of leukocyte–endothelial cell interactions, leading to activation of the vascular endothelium. However, no information is as yet available with regard to the relationship between peripheral blood TNFα levels and CALs or the role of TNFα in the pathogenesis of coronary vessel damage in humans.

In this study, we found that the TNFα level was not significantly different between KD patients with CALs and those without CALs pre-IVIG. The TNFα level decreased in KD patients without CALs post-IVIG or in IVIG responders, but increased in the patients with CALs post-IVIG or in IVIG nonresponders. Two patients with CALs had a TNFα level >20 pg/ml pre-IVIG, but 4 patients with CALs had a TNFα level >20 pg/ml post-IVIG. This finding suggests that KD patients may be nonresponsive to IVIG or might develop CALs when the TNFα levels increase during the process after treatment with IVIG.

Up until now, no systematic study has addressed whether there is a relationship between TNFα and KD, although it has been reported that the TNFα level in KD patients unresponsive to an initial course of IVIG was high and could be rapidly reduced by treatment with intravenous pulse methylprednisolone (23). Furthermore, in an animal model of KD, it was found that TNFα was necessary for the induction of coronary artery inflammation and aneurysm formation (24).

With regard to a possible relationship between IL-4 levels and KD, the results of studies have been contradictory. Hirao et al (25) reported that the IL-4 level was increased in patients with KD and, as the key cytokine, might regulate the cytokine network, but Matsubara et al (19) reported that there was no significant difference in the percentages of IL-4–producing T cells in the peripheral blood between patients with KD and healthy controls. Kimura et al (26) demonstrated that IL-4 messenger RNA levels in peripheral blood mononuclear cells were significantly decreased in patients with KD compared to healthy controls. In this study, IL-4 levels in KD patients with CALs were higher than those in patients without CALs both pre-IVIG and post-IVIG, but the levels were within the normal range in most patients, which suggests that the IL-4 level may be related to CALs. However, among the recent studies, the relationship between IL-4 and KD/CALs is rarely illustrated. Further studies are needed to explore the role of IL-4 in the pathophysiologic mechanisms of KD.

In this study, we measured and analyzed the serum Th1 and Th2 cytokine levels in patients with KD. We found that the levels of IL-6, IL-10, TNFα, and IFNγ were significantly increased pre-IVIG, and that the levels of IL-6, IL-10, and IFNγ decreased quickly post-IVIG. After IVIG treatment, the TNFα level decreased significantly in KD patients without CALs and in KD patients who were IVIG responders, but increased slightly in KD patients with CALs and in KD patients who were IVIG nonresponders. The levels of IL-4, IL-6, IL-10, and IFNγ were significantly higher in KD patients with CALs than in those without CALs pre-IVIG. Furthermore, the post-IVIG levels of IL-6 and IL-10 were significantly higher in IVIG nonresponders compared to IVIG responders.

We calculated the sensitivity and specificity of these cytokine levels for predicting the development of CALs and IVIG resistance in the KD patients. Pre-IVIG, an IL-10 level >8 pg/ml had a sensitivity of 75.0% and a specificity of 64.4% for predicting CALs, while a TNFα level <2 pg/ml had a sensitivity of 66.7% and a specificity of 74.2% for predicting IVIG resistance. Post-IVIG, an IL-6 level >10 pg/ml had a sensitivity of 67.9% and a specificity of 81.7% for predicting CALs, while an IL-10 level >6 pg/ml had a sensitivity of 53.6% and a specificity of 86% for predicting CALs.

These results suggest that KD patients may be nonresponsive to IVIG when the serum levels of IL-6 and IL-10 decrease slowly and the levels of IL-4 and TNFα increase after treatment with IVIG. It may be a warning sign for the development of CALs when the levels of IL-6, IL-10, and IFNγ are significantly increased in the serum of KD patients prior to IVIG treatment and then decrease slowly post-IVIG, and when TNFα levels increase post-IVIG. Thus, the findings in this study may be helpful in making a preliminary estimate for the prognosis and treatment of KD.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Gong had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Y. Wang, W. Wang, Gong.

Acquisition of data. Y. Wang, W. Wang, Fu, Q. Zhang, Hu.

Analysis and interpretation of data. Gong, Qi, Xie, Y. Zhang.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES
  • 1
    Kawasaki T. Acute febrile mucocutaneous syndrome with lymphoid involvement with specific desquamation of the fingers and toes in children. Arerugi 1967; 16: 178222. In Japanese.
  • 2
    Galeotti C, Bayry J, Kone-Paut I, Kaveri SV. Kawasaki disease: aetiopathogenesis and therapeutic utility of intravenous immunoglobulin. Autoimmun Rev 2010; 9: 4418.
  • 3
    Burns JC. Kawasaki disease update. Indian J Pediatr 2009; 76: 716.
  • 4
    Kim DS. Kawasaki disease. Yonsei Med J 2006; 47: 75972.
  • 5
    Furukawa S, Matsubara T, Jujoh K, Yone K, Sugawara T, Sasai K, et al. Peripheral blood monocyte/macrophages and serum tumor necrosis factor in Kawasaki disease. Clin Immunol Immunopathol 1988; 48: 24751.
  • 6
    Kriegler M, Perez C, DeFay K, Albert I, Lu SD. A novel form of TNF/cachectin is a cell surface cytotoxic transmembrane protein: ramifications for the complex physiology of TNF. Cell 1988; 53: 4553.
  • 7
    Heinrich PC, Behrmann I, Haan S, Hermanns HM, Muller-Newen G, Schaper F. Principles of interleukin (IL)-6-type cytokine signalling and its regulation. Biochem J 2003; 374: 120.
  • 8
    Smith AJ, Humphries SE. Cytokine and cytokine receptor gene polymorphisms and their functionality. Cytokine Growth Factor Rev 2009; 20: 4359.
  • 9
    Weng KP, Hsieh KS, Hwang YT, Huang SH, Lai TJ, Yuh YS, et al. IL-10 polymorphisms are associated with coronary artery lesions in acute stage of Kawasaki disease. Circ J 2010; 74: 9839.
  • 10
    Furukawa S, Imai K, Matsubara T, Yone K, Yachi A, Okumura K, et al. Increased levels of circulating intercellular adhesion molecule 1 in Kawasaki disease. Arthritis Rheum 1992; 35: 6727.
  • 11
    Weng KP, Ho TY, Chiao YH, Cheng JT, Hsieh KS, Huang SH, et al. Cytokine genetic polymorphisms and susceptibility to Kawasaki disease in Taiwanese children. Circ J 2010; 74: 272633.
  • 12
    Ayusawa M, Sonobe T, Uemura S, Ogawa S, Nakamura Y, Kiyosawa N, et al. Revision of diagnostic guidelines for Kawasaki disease (the 5th revised edition). Pediatr Int 2005; 47: 2324.
  • 13
    Iwashima S, Kimura M, Ishikawa T, Ohzeki T. Importance of C-reactive protein level in predicting non-response to additional intravenous immunoglobulin treatment in children with Kawasaki disease: a retrospective study. Clin Drug Investig 2011; 31: 1919.
  • 14
    Fu S, Gong F, Xie C, Zhu W, Wang W, Shen H, et al. S100A12 on circulating endothelial cells surface in children with Kawasaki disease. Pediatr Res 2010; 68: 1658.
  • 15
    Chen R, Lowe L, Wilson JD, Crowther E, Tzeggai K, Bishop JE, et al. Simultaneous quantification of six human cytokines in a single sample using microparticle-based flow cytometric technology. Clin Chem 1999; 45: 16934.
  • 16
    Gupta M, Noel GJ, Schaefer M, Friedman D, Bussel J, Johann-Liang R. Cytokine modulation with immune gamma-globulin in peripheral blood of normal children and its implications in Kawasaki disease treatment. J Clin Immunol 2001; 21: 1939.
  • 17
    Hsueh KC, Lin YJ, Chang JS, Wan L, Tsai YH, Tsai CH, et al. Association of interleukin-10 A-592C polymorphism in Taiwanese children with Kawasaki disease. J Korean Med Sci 2009; 24: 43842.
  • 18
    Jin HS, Kim HB, Kim BS, Lee JK, Seo EJ, Yoo HW, et al. The IL-10 (−627 A/C) promoter polymorphism may be associated with coronary aneurysms and low serum albumin in Korean children with Kawasaki disease. Pediatr Res 2007; 61: 5847.
  • 19
    Matsubara T, Katayama K, Matsuoka T, Fujiwara M, Koga M, Furukawa S. Decreased interferon-γ (IFN-γ)-producing T cells in patients with acute Kawasaki disease. Clin Exp Immunol 1999; 116: 5547.
    Direct Link:
  • 20
    Chen J, Liu XS. Development and function of IL-10 IFN-γ-secreting CD4+ T cells. J Leukoc Biol 2009; 86: 130510.
  • 21
    Hahn Y, Kim Y, Jo S, Han H. Reduced frequencies of peripheral interferon-γ-producing CD4+ and CD4- cells during acute Kawasaki disease. Int Arch Allergy Immunol 2000; 122: 2938.
  • 22
    Chan WC, Duong TT, Yeung RS. Presence of IFN-γ does not indicate its necessity for induction of coronary arteritis in an animal model of Kawasaki disease. J Immunol 2004; 173: 3492503.
  • 23
    Miura M, Kohno K, Ohki H, Yoshiba S, Sugaya A, Satoh M. Effects of methylprednisolone pulse on cytokine levels in Kawasaki disease patients unresponsive to intravenous immunoglobulin. Eur J Pediatr 2008; 167: 111923.
  • 24
    Hui-Yuen JS, Duong TT, Yeung RS. TNF-α is necessary for induction of coronary artery inflammation and aneurysm formation in an animal model of Kawasaki disease. J Immunol 2006; 176: 6294301.
  • 25
    Hirao J, Hibi S, Andoh T, Ichimura T. High levels of circulating interleukin-4 and interleukin-10 in Kawasaki disease. Int Arch Allergy Immunol 1997; 112: 1526.
  • 26
    Kimura J, Takada H, Nomura A, Ohno T, Mizuno Y, Saito M, et al. Th1 and Th2 cytokine production is suppressed at the level of transcriptional regulation in Kawasaki disease. Clin Exp Immunol 2004; 137: 4449.