TABLE 4

Univariate and Multivariable Logistic Regression for Increase in Coronary Artery Z Score During Follow-up

PredictorsUnivariateMultivariable
No Increase in CA Z Scores >1, n = 81 Increase in CA Z Scores >1, n = 40POR95% CIP
Male, n (%)59 (73)29 (73).231.720.60 to 4.88.31
Age at time of diagnosis in y, median (IQR)1.5 (0.6 to 3.4)2.2 (0.7 to 4.1).161.080.94 to 1.24.30
Length of fever in d at time of diagnosis, median (IQR)6 (4.5 to 7)7 (4 to 8).371.040.83 to 1.30.73
Primary treatment, n (%).003
 IVIg only, n = 3314 (42)19 (58)1
 IVIg and corticosteroids, n = 3023 (77)7 (23)0.210.06 to 0.77.02
 IVIg and infliximab, n = 5844 (76)14 (24)0.170.06 to 0.51.002
IVIg resistance, n (%)10 (12)6 (15).784.310.70 to 26.37.11
Second dose of IVIg, n (%)24 (30)14 (35).680.630.18 to 2.26.48
Other immunosuppressive therapies, n (%)15 (19)11 (28).350.980.32 to 3.01.97
Baseline bilateral CAAs, n (%)26 (32)16 (40).421.900.62 to 5.81.26
Baseline coronary z score, median (IQR)3.4 (2.9, 4.2)3.5 (2.9, 4.6).901.250.91 to 1.71.17
  • CA, coronary artery; CI, confidence interval; OR, odds ratio; —, not applicable.