Variability in Response to Intravenous Immunoglobulin in the Treatment of Kawasaki Disease.

OBJECTIVES

To characterize the pattern of temperature response to intravenous immunoglobulin (IVIG) infusion in patients with Kawasaki disease (KD).

STUDY DESIGN

Patients nonresponsive to IVIG (axillary temperature ≥37.5°C >24 hours after end of IVIG) were identified. Each patient with IVIG-nonresponsive KD was matched to a control patient with IVIG-responsive KD of the same age, sex, and duration of fever before IVIG. Hourly temperature profiles were obtained from immediately before the start of IVIG infusion until complete defervescence.

RESULTS

A total of 182 patients nonresponsive to IVIG were matched (total n = 364). Nonresponders were further classified as partial nonresponders (68%) (axillary temperature decreased to <37.5°C but fever recurred) and complete nonresponders (32%) (axillary temperature consistently ≥37.5°C throughout IVIG treatment). The temperature profile during IVIG infusion was similar between responders and partial nonresponders (EST: -0.061 [0.007]°C/h, P < .001 for responders vs EST: -0.027 (0.012)°C/h, P = .03 for partial nonresponders [responders vs partial nonresponders, P = .65]), where EST is the parameter estimate from the regression model, representing the change in degrees Celsius for each hour since start of IVIG. In complete nonresponders, IVIG was not associated with significant decreases in temperature (EST: -0.008 [0.010]°C, P = .42). Factors associated with complete (vs partial) nonresponse included laboratory-confirmed infection, greater C-reactive protein, and IVIG brand. Defervescence in partial nonresponders was achieved with a second IVIG dose for 72% of patients compared with 58% of complete nonresponders (P = .001). Complete nonresponders were more likely to develop coronary artery aneurysms vs partial nonresponders (OR: 2.4 [1.1-5.4], P = .03) or responders (OR: 3.2 [1.5-6.9], P = .002).

CONCLUSIONS

Nonresponse to initial IVIG can be further characterized by temperature profile, and complete nonresponders may require more aggressive second-line therapy.