Risk factors for persistent CMA
In a comparison of baseline characteristics between the tolerant and persistent groups, statistically significant differences were observed for the presence of coexisting atopic dermatitis, anaphylaxis to CM before the age of 6 years, complete elimination of CM at the age of 6 years, and sIgE to CM and casein (Table 2). In both groups, the CM-sIgE level was the highest at ages 6–7 years and then decreased annually, except for ages 8 and 9 years in the tolerant group (Supplementary figure 2). At all ages between 6 and 12 years, the CM-sIgE levels were significantly higher in the persistent group than in the tolerant group.
The results of the Cox hazard model analysis are presented in Table 3. Using the significant risk factors identified by univariate analysis, the multivariate Cox regression analysis showed that previous anaphylaxis to CM (adjusted HR [95% CI]: 2.42 [1.24–4.69],p = 0.01), complete CM elimination at the age of 6 years (adjusted HR [95% CI]: 5.18 [2.45–10.99], p < 0.001), and higher CM-sIgE levels at the age of 6 years (adjusted HR [95% CI]: 2.58 per ten-fold increase [1.62–4.12], p< 0.001) were independent risk factors for persistent CMA.
The area under the curve for CM-sIgE levels was 0.83 (95% CI: 0.76–0.90), and the optimal cutoff level of CM-sIgE was 34.4 kUA/L. Using this cutoff value, the sensitivity and specificity were 60% and 91%, respectively.
Although there was no statistical significance, these baseline characteristics tended to be higher in the persistent group with enrollment in OIT than in the persistent group without enrollment in OIT (Supplementary table 1).