Abstract Aims Takotsubo syndrome (TTS) is an acute heart failure (AHF) syndrome mimicking the symptoms of acute myocardial infarction. Impaired outcome has been shown, making risk stratification and novel therapeutic concepts a necessity. We hypothesized insulin resistance with elevated plasma glucose and potentially myocardial glucose deprivation to contribute to the pathogenesis of TTS and investigated the therapeutic benefit of insulin in vivo. Methods and results First, we retrospectively analysed patient data of n = 265 TTS cases (85.7% female, mean age 71.1 ± 14.1 years) with documented initial plasma glucose from the Department of Cardiology of the University Hospital Heidelberg in Germany (May 2011 to May 2021). Median split of the study population according to glucose levels (≤123 mg/dL vs. >123 mg/dL) yielded significantly elevated mean heart rate (80.75 ± 18.96 vs. 90.01 ± 22.19 b.p.m., P < 0.001), left ventricular end-diastolic pressure (LVEDP, 18.51 ± 8.35 vs. 23.09 ± 7.97 mmHg, P < 0.001), C-reactive protein (26.14 ± 43.30 vs. 46.4 ± 68.6 mg/L, P = 0.006), leukocyte count (10.12 ± 4.29 vs. 15.05 ± 9.83/nL, P < 0.001), peak high-sensitive Troponin T (hs-TnT, 515.44 ± 672.15 vs. 711.40 ± 736.37 pg/mL, P = 0.005), reduced left ventricular ejection fraction (EF, 34.92 ± 8.94 vs. 31.35 ± 8.06%, P < 0.001), and elevated intrahospital mortality (2.3% vs. 12.1%, P = 0.002) in the high-glucose group (Student's t-test, Mann–Whitney U test, or chi-squared test). Linear regression indicated a significant association of glucose with HR (P < 0.001), LVEDP (P = 0.014), hs-TnT kinetics from admission to the next day (P < 0.001), hs-TnT the day after admission (P < 0.001), as well as peak hsTnT (P < 0.001). Logistic regression revealed significant association of glucose with a composite intrahospital outcome including catecholamine use, respiratory support, and resuscitation [OR 1.010 (1.004–1.015), P = 0.001]. To further investigate the potential role of glucose in TTS pathophysiology experimentally, we utilized an in vivo murine model of epinephrine (EPI)-driven reversible AHF. For this, male mice underwent therapeutic injection of insulin (INS, 1 IU/kg) or/and glucose (GLU, 0.5 g/kg) after EPI (2.5 mg/kg), both of which markedly improved mean EF (EPI 34.3% vs. EPI + INS + GLU 43.7%, P = 0.025) and significantly blunted mean hs-TnT (EPI 14 393 pg/mL vs. EPI + INS 6864 pg/mL at 24 h, P = 0.039). Particularly, insulin additionally ameliorated myocardial pro-inflammatory gene expression, suggesting an anti-inflammatory effect of acute insulin therapy. Conclusions Elevated initial plasma glucose was associated with adverse outcome-relevant parameters in TTS and may present a surrogate parameter of heightened catecholaminergic drive. In mice, insulin- and glucose injection both improved EPI-induced AHF and myocardial damage, indicating insulin resistance rather than detrimental effects of hyperglycaemia itself as the underlying cause. Future studies will investigate the role of HbA1c as a risk stratifier and of insulin-based therapy in TTS.