Response to letter regarding article, ‘impact of beta blockers on long-term mortality in Takotsubo syndrome: a real-world analysis of the TriNetX global collaborative network database’

We thank Rizk et al. for their interest in our work evaluating the impact of beta-blocker use on mortality in Takotsubo syndrome (TTS).1 They bring up important considerations regarding the potential influence of cardiogenic shock (CS) and vasoactive medication use on the observed mortality benefit associated with beta-blocker therapy. To address these concerns, we performed additional analyses incorporating propensity-score matching for CS and vasoactive medication use to further mitigate confounding related to acute illness severity. In the unmatched cohort, CS prevalence was 6% in TTS, which is consistent with our previously reported data.2 As noted by Rizk et al., early initiation of beta-blockers may be clinically challenging in patients presenting with CS and may result in delayed treatment. Accordingly, time to beta-blocker initiation differed by CS status, with a mean of 8 days among patients without CS compared with 15 days among those with CS. In the revised matched cohort, the additional baseline characteristics were as follows (beta-blocker vs. no beta-blocker cohort): CS (3.1% vs. 3.0%), epinephrine (2.7% vs. 2.5%), norepinephrine (2.7% vs. 2.7%), phenylephrine (2.1% vs. 2.0%), ephedrine (0.7% vs. 0.7%), dobutamine (0.5% vs. 0.5%), and dopamine (0.3% vs. 0.3%). Beta-blocker use in patients with TTS, despite matching for CS and vasoactive medication use, was associated with a significantly lower risk of short-term (1-year) (hazard ratio 0.55; 95% confidence interval 0.52, 0.59; P < 0.001) and long-term (5-year) mortality (hazard ratio 0.70; 95% confidence interval 0.66, 0.74; P < 0.001).