Response to: Correspondence on ‘Beta-blockers are associated with better long-term survival in patients with Takotsubo syndrome’ by Chang et al

The Authors’ reply We appreciate the interest of Dr Chang and Dr Liu in our study. Cardiogenic shock (CS) complicates the inhospital course of patients with takotsubo syndrome (TTS) in 6%–20% of cases and is the main cause of inhospital mortality. Treatment of CS in TTS is a clinical challenge and, due to the complexity of investigating this subject in dedicated randomised controlled trials, current evidence is limited to case reports or retrospective observational studies. The use of catecholamines in TTS complicated by CS is still debated since the administration of exogenous catecholamines might reexacerbate the acute phase and increase the risk of inhospital adverse events. In this scenario, previous case series have suggested the use of the Ca sensitiser levosimendan as a safe and feasible nonadrenergic alternative to common inotropic agents in TTS. Owing to the key role of catecholamine overstimulation in TTS pathophysiology, the use of betablockers has been proposed to mitigate the sympathetic drive and the effects of further catecholamine surges during the acute phase. In a monkey model of epinephrineinduced TTS, metoprolol improved left ventricular dysfunction, diminished the catecholamineinduced cardiomyocytolysis and modified the expression of heart failurerelated genes. However, in a recent propensity score matching analysis including 2110 Japanese patients with TTS, the early use of betablockers (started on hospitalisation day 1 or 2) did not influence patient 30day mortality. One interpretation of these conflicting results may be that the expected benefit of betablockers during the acute phase depends on the correct identification of the patient most likely to benefit. In TTS scenario, the early identification of patients at high risk of developing CS or with initial signs of haemodynamic instability constitutes the first step of an individualised patienttailored therapy. In a study from the INTERTAK Registry including 2078 patients with TTS, 198 (9.5%) developed CS. Some parameters easily detectable on admission including apical TTS, physical stress, lower left ventricular ejection fraction, diabetes mellitus and atrial fibrillation were associated with the risk of CS, hence emphasising their early detection for the identification of patients more prone to develop CS. We concur with Dr Chang and Dr Liu that betablockers should be considered in all patients after clinical stabilisation and before discharge, even in those who did not tolerate their early use. This certainly includes patients who developed CS during the acute phase, who have a significantly higher risk of developing adverse events compared with those without CS over the long term. 7 In our study, the prescription of betablockers at discharge was associated with a lower risk of mortality at longterm followup, particularly in patients with TTS who developed CS during the hospitalisation. This finding suggests that these patients may constitute a particular TTS phenotype characterised by higher sympathetic overdrive and greater susceptibility to catecholaminemediated myocardial damage. This result was also consistent with a previous study from RETAKO registry, which showed a significantly lower 1year mortality in patients who received betablockers compared with those who did not in the subset with CS during the hospitalisation. During the acute phase, the choice of the most appropriate treatment demands deep understanding of the multiple mechanisms potentially involved in haemodynamic instability. Although CS in TTS is generally attributable to acute pump failure, the haemodynamic deterioration can be determined by mechanical complication including dynamic left ventricular outflow tract obstruction (LVOTO) and functional transient mitral regurgitation. LVOTO is generally considered haemodynamically relevant if the gradient exceeds 50 mm Hg and represents a dynamic phenomenon closely related to the interactions between loading conditions and myocardial contractility. Rule out of LVOTO is usually achieved by transthoracic echocardiography. In patients undergoing coronary angiography and ventriculography, a slow pullback of pigtail catheter from left ventricle can exclude an intraventricular pressure gradient if TTS is suspected. While positive inotropes may be considered for the treatment of patients with primary pump failure, these agents should be possibly avoided in patients with TTS and LVOTO since they may increase basal myocardial contractility and worsen intraventricular gradient. In contrast to the therapy of acute heart failure with pulmonary oedema, diuretics are also not indicated in presence of LVOTO as they can further deteriorate intraventricular gradient through the preload reduction. In this setting, the use of shortacting betablockers such as esmolol or landiolol along with careful fluid administration have been proposed. In fact, these agents may reduce the hypercontractility of the basal segments and increase the ventricular preload with improvement of intraventricular gradient and of the haemodynamic status. In patients with refractory CS, shortterm mechanical circulatory support may also be considered as bridge to recovery, particularly in the presence of LVOTO. The use of Impella left ventricular assist device might restore haemodynamics by expelling aspirated blood from the left ventricle into the ascending aorta and skipping the left ventricular outflow tract. Conversely, intraaortic balloon pump is contraindicated since it leads to afterload decrease and to a worsening of intraventricular gradient. In case of refractory CS with multiple organ failure, the use of venoarterial extracorporeal membrane oxygenation may be considered depending on centre expertise and availability. Pending the definition of the best treatment for CS in TTS, based inevitably on the pathophysiological mechanism involved, our study suggests that betablockers should be started once adequate haemodynamic conditions have been restored, and maintained over time.