Impact of Admission Hyperglycemia on Heart Failure Events and Mortality in Patients With Takotsubo Syndrome at Long-term Follow-up: Data From HIGH-GLUCOTAKO Investigators

Takotsubo syndrome (TTS) is a cardiovascular disease characterized by heart failure (HF) with reduced ejection fraction, usually in the absence of significant angiographic coronary artery stenoses (1). To date, TTS meets the diagnostic criteria for acute coronary syndrome and can occur during stressful events (1). Indeed, the excessive release of sympathetic hormones together with the particular distribution of adrenergic receptors in the left ventricular apical segments may determine the acute and long-term clinical features of TTS (2,3). Overall, data showed 1) an excessive increase in serum catecholamines can cause acute hyperadrenergic myocardium stunning (3) and 2) patients with TTS had enhanced myocardial sympathetic nerve activity resulting from an increase in presynaptic vesicle exocytosis and a decrease in terminal nerve axon reuptake of norepinephrine (3). Whether hyperglycemia may serve as a metabolic trigger to unbalance the sympathetic system axis through the alteration of signals and transduction pathways of adrenergic receptors as well as through overinflammation (4–6) is not fully understood. Although the harmful effects of hyperglycemia on infarct size and inflammatory response have been well evaluated in the Myocardial Infarction in Obstructive Coronary Artery Disease (MIOCA) and Myocardial Infarction With Nonobstructive Coronary Arteries (MINOCA) studies, no studies have investigated the sympathetic activity and long-term prognosis in patients with TTS and hyperglycemia versus normoglycemia (7). Therefore, we assessed the role of admission hyperglycemia in patients with TTS, focusing on the interplay between sympathetic activity and inflammatory burden on myocardial contractility and long-term outcome in terms of HF occurrence and mortality.

In this multicenter study, we screened 4,783 patients undergoing coronary angiography within the first 72 h of hospitalization for suspected acute coronary syndrome between January 2015 and January 2018. All enrolled patients met the InterTAK diagnostic criteria proposed in the European Society of Cardiology position statement (8). Exclusion criteria encompassed patients with previous myocardial infarction, TTS events, or chronic kidney or liver disease (1). The authors conducted the study in accordance with the Declaration of Helsinki. The ethics committees of all participating institutions approved the protocol. All patients were informed about the nature of the study and provided written informed consent to participate.

Patients with TTS were divided into those with hyperglycemia versus those with normoglycemia according to a cutoff admission blood glucose value of 140 mg/dL (9). Sympathetic activity was assayed by blood values of norepinephrine and ¹²³I-labeled metaiodobenzylguanidine (MIBG) cardiac scintigraphy with late heart-to-mediastinum ratio (H/M_late) and washout rate (WR), performed in 30 patients who did not present any contraindication to the examination, evaluated at baseline and at follow-up. Similarly, systemic inflammatory markers (CRP, white blood cell count [leukocytes and neutrophils], and tumor necrosis factor-α [TNF-α]) (1,6) and B-type natriuretic peptide (BNP) were assessed. Details on coronary angiography, transthoracic echocardiography, MIBG technique, and outpatient management are reported in the data supplement. Prespecified end points (HF and all-cause deaths) were assessed at long-term follow-up (12 and 24 months).

Our final study population consisted of 76 patients with confirmed TTS, 28 with hyperglycemia and 48 with normoglycemia. Study results were divided into hospital admission and long-term follow-up data (12 and 24 months).

At admission, no significant differences were observed in terms of demographic characteristics and cardiovascular risk factors between patients with hyperglycemia and normoglycemia, except for presence of diabetes, which was more prevalent among those with hyperglycemia (P = 0.02). Likewise, no significant differences in admission cardiovascular medications between study groups were noted. Similar atherosclerotic burden and admission STEMI diagnosis between the two groups were found. As for clinical presentation, compared with patients with normoglycemia, those with admission hyperglycemia presented with lower left ventricular ejection fraction (LVEF) values both at hospital admission (P = 0.011) and discharge (P = 0.021), with no differences in mitral regurgitation degree or Killip class. Moreover, patients with hyperglycemia had significantly higher values of inflammatory markers, norepinephrine, and BNP (P < 0.05 for all) (Supplementary Table 1). Direct proportional correlation between glucose blood values and norepinephrine levels at hospital admission was found (R² = 0.39; P = 0.001). In a subset of 30 patients with TTS (13 with hyperglycemia vs. 17 with normoglycemia), ¹²³I-MIBG cardiac scintigraphy in the acute phase showed lower values of H/M_late among those with hyperglycemia (P < 0.001). At the same time, no significant differences were found for WR parameters (Fig. 1A and B). Therefore, we tested whether the absence of difference in WR between those with hyperglycemia and normoglycemia in the acute state was independent of LVEF between the two groups. No differences in length of hospital stay or intrahospital deaths between the two study groups were observed.

At long-term follow-up, patients with TTS and hyperglycemia had higher inflammatory marker levels (CRP and TNF-α) and BNP serum levels (P < 0.05 for all) both at 12 and 24 months (Supplementary Tables 3 and 4). Among patients undergoing ¹²³I-MIBG cardiac scintigraphy, H/M_late and WR values observed during the 12- and 24-month follow-up were similar to those detected in the acute phase (Fig. 1D and E). Changes in H/M_late after 24 months of follow-up were linked to a significant reduction in glycemia (R² = 0.38; P = 0.021). Patients with hyperglycemia showed more significant mortality (25 vs. 8.3%; P < 0.05) and a higher rate of HF events (21.4 vs. 8.3%; P = 0.001) during follow-up, as depicted in the Kaplan-Meier curves (Fig. 1C and F). Recurrence of TTS was observed in one patient with hyperglycemia at 12 months and in two patients with hyperglycemia and one with normoglycemia at 24 months of follow-up. In multivariate Cox regression analysis, admission hyperglycemia (hazard ratio [HR] 1.01; 95% CI 1.003–1.018; P = 0.008), TNF-α (HR 1.775; 95% CI 1.26–2.51; P = 0.001), and norepinephrine (HR 1.001; 95% CI 1.00–1.001; P = 0.035) were independent predictors of HF events at 24 months. Conversely, only admission hyperglycemia (HR 1.016; 95% CI 1.001–1.031; P = 0.031) was a significant predictor for all-cause deaths (Supplementary Table 2).

Our study investigated the role of admission hyperglycemia in the acute phase and long-term follow-up of patients with TTS. The main findings were 1) hyperglycemia was associated with higher norepinephrine levels; 2) patients with TTS and hyperglycemia had lower LVEF values at hospital admission and discharge compared with patients with normoglycemia; 3) in a subset of patients with TTS who underwent ¹²³I-MIBG cardiac scintigraphy, those with hyperglycemia had lower H/M_late values at the acute phase and follow-up; 4) glycemic compensation was linked to better H/M_late during follow-up; 5) patients with hyperglycemia exhibited worse outcomes in terms of all-cause deaths and HF occurrence; 6) hyperglycemia, TNF-α, and norepinephrine predicted HF events at 2 years of follow-up; and 7) only admission hyperglycemia was a significant predictor of death.

The negative effect of hyperglycemia might induce a chronic alteration of cardiac metabolism that could amplify the sympathetic tone dysfunction and cardiac denervation, negatively influencing hyperglycemic prognosis in patients with TTS. In fact, hyperglycemic status might impair norepinephrine reuptake and prolong hyperadrenergic stimulation with the persisting transduction defect of β-adrenergic receptors in the stunning myocardium, resulting in more severe and protracted myocardial sympathetic denervation (10), as described in diabetic cardiomyopathy (11). Moreover, stressful triggers could worsen both hyperglycemic status and oversympathetic stimulation in patients with TTS, which are potentially interrelated (8). Interestingly, admission hyperglycemia increased by 7.5-fold the risk of HF at the 24-month follow-up. Furthermore, the persistent hyperglycemic environment promotes the production of adipose tissue proinflammatory cytokines, representing a parallel but not secondary pathway of cardiovascular damage via overinflammation.

Thus, persisting/chronic overactivity of the sympathetic system (9,10), cardiac denervation (12), and the hyperglycemia-mediated proinflammatory pathway (13) could affect acute and long-term prognoses regarding HF events and all-cause deaths, often observed among patients with TTS and hyperglycemia.

Acknowledgments. The authors thank the physicians of the participating institutions.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. P.P., L.B., and C.S. were responsible for the conception, funding, and design of the study and wrote the first draft of the manuscript. L.B. and C.S. carried out the statistical analysis. P.R., G.G., A.F., F.A., and M.F. collated data. A.F., G.C., M.B., N.G., R.M., and C.P. corrected and approved the revisions and final version of the manuscript. C.S. is the guarantor of this work and, as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.