Importance of C-reactive protein level in predicting non-response to additional intravenous immunoglobulin treatment in children with Kawasaki disease: a retrospective study.

BACKGROUND: Intravenous immunoglobulin (IVIG) therapy in the acute stage of Kawasaki disease (KD; mucocutaneous lymph node syndrome) is the treatment of first choice for preventing the development of coronary artery lesions (CALs). Failure of initial treatment with IVIG remains the most consistent risk factor for CALs. However, there are few reports on nonresponders to additional IVIG therapy in KD.

OBJECTIVE: The goal of the present study was to predict non-responders to additional IVIG therapy in children with KD.

METHODS: This was a retrospective study aimed at predicting non-responders to additional IVIG therapy for KD in a cohort of 446 patients. The IVIG response group ('responders') was defined as those patients who were afebrile 48 hours after administration of initial IVIG. The IVIG non-response group ('non-responders') was defined as those patients who remained febrile 48 hours after administration of initial IVIG and was divided into two subgroups: (i) those patients who remained febrile 48 hours after administration of additional IVIG (non-responders 1), and (ii) those patients who were afebrile 48 hours after additional IVIG (non-responders 2).

RESULTS: Ninety-one patients received additional IVIG; of these, 25 patients (non-responders 1) received additional rescue therapy because no improvement was observed and 66 patients (non-responders 2) were afebrile. Mean – SD C-reactive protein (CRP) levels were higher in non-responders 1 than in non-responders 2 (12.05 – 5.14 vs 7.67 – 4.99 mg/dL; p = 0.002). The optimal cutoff point of sensitivity and specificity for predicted non-responders was ≥8 mg/dL. The sensitivity and specificity for prediction of IVIG response was 76.0% and 63.6%, respectively. Forty-three patients had a CRP level of ≥8 mg/dL after initial IVIG, 18 of whom developed CALs (eight persistent lesions and ten transient lesions). Forty-eight patients had a CRP level of <8 mg/dL after initial IVIG, of whom only eight developed CALs (all transient).

CONCLUSION: We have discovered a biomarker able to identify KD patients at high risk of complications who require additional IVIG treatment, thus avoiding overtreatment of low-risk individuals. We suggest that patients who have a CRP level of ≥8mg/dL after initial IVIG are likely to fail additional IVIG and may require further IVIG plus rescue therapy.