Purpose: To understand the association between biomarkers and exacerbations of severe asthma in adult patients in Taiwan. Materials and Methods: Demographic, clinical characteristics and biomarkers were retrospectively collected from the medical charts of severe asthma patients in six hospitals in Taiwan. Exacerbations were defined as those requiring asthma-specific emergency department visits/hospitalizations, or systemic steroids. Enrolled patients were divided into: (1) those with no exacerbations (non-exacerbators) and (2) those with one or more exacerbations (exacerbators). Receiver operating characteristic curves were used to determine the optimal cut-off value for biomarkers. Generalized linear models evaluated the association between exacerbation and biomarkers. Results: 132 patients were enrolled in the study with 80 non-exacerbators and 52 exacerbators. There was no significant difference in demographic and clinical characteristics between the two groups. Exacerbators had significantly higher eosinophils (EOS) counts (367.8 ± 357.18 vs. 210.05 ± 175.24, p = 0.0043) compared to non-exacerbators. The optimal cut-off values were 292 for EOS counts and 19 for the Fractional exhaled Nitric Oxide (FeNO) measure. Patients with an EOS count ≥ 300 (RR = 1.88; 95% CI, 1.26–2.81; p = 0.002) or FeNO measure ≥ 20 (RR = 2.10; 95% CI, 1.05–4.18; p = 0.0356) had a significantly higher risk of exacerbation. Moreover, patients with both an EOS count ≥ 300 and FeNO measure ≥ 20 had a significantly higher risk of exacerbation than those with lower EOS count or lower FeNO measure (RR = 2.16; 95% CI, 1.47–3.18; p = < 0.0001). Conclusions: Higher EOS counts and FeNO measures were associated with increased risk of exacerbation. These biomarkers may help physicians identify patients at risk of exacerbations and personalize treatment for asthma patients.