In the present study, the clinical implications of elevated IMR in patients with TTS were investigated. The main results of the investigation are: (I) elevated caIMR could be consistently demonstrated in the TTS cohort; (II) apical and midventricular TTS form showed comparable microvascular dysfunction patterns, only a trend towards higher caIMR in the LAD of the apical types was observed; (III) impairment of the microvascular function in TTS showed a transient nature with complete recovery at a median follow-up of 28 months; and (IV) in-hospital outcomes correlated significantly with the extent of caIMR elevation at the index coronary angiography.

Dysfunctional coronary microcirculation in the acute TTS phase has previously been described. In a small cohort of TTS patients, Kim et al. reported that invasively measured IMR was similarly impaired as compared to STEMI patients after revascularization [22]. In contrast to the microvascular dysfunction following a reperfused STEMI [23, 24], impaired microcirculatory function in TTS is likely due to excessive vasoconstriction and augmented sympathetic responses to acute stress [22, 25] and subsequent catecholamine excess [4, 26,27,28]. High catecholamine levels following activation of the sympathetic nervous system have been associated with the development of endothelial CMD either via direct cardiomyocyte toxicity or by inducing coronary artery spasms (evidence of decreased plasma levels of microRNA 125a-5p and increased levels of its target endothelin-1) resulting in inadequate myocardial blood supply [29,30,31].

Angiography-derived IMR, such as caIMR, has been developed as a non-invasive alternative to wire-based IMR and is nowadays a validated and useful tool to assess coronary microcirculatory function [15, 16]. Since no wire introduction into the coronary arteries is required for this technique, angiography-derived IMR is a safe and feasible alternative to wire-based approaches for the evaluation of the three main epicardial vessels in patients with TTS.

The present study was able to demonstrate higher caIMR in the acute setting of TTS compared to non-culprit vessels of matched ACS patients, which have been proven to function as a control and therefore represent a valid model in wire-based IMR studies [16, 17]. In addition, it was demonstrated that caIMR elevation is not limited to one specific coronary artery but is present in the three main coronary arteries with highest caIMR in the LCX followed by the LAD and the RCA. These findings are in contrast with previous evidence reporting higher caIMR in the LAD of TTS patients [9]. A possible explanation for the discrepancy in the magnitude of the differences between caIMR in the LAD and LCX is the relatively higher number of TTS patients with midventricular types in our cohort.

By comparing TTS patients to a matched control group, the presence of elevated caIMR in the three main epicardial vessels was also endorsed. In a comparison of all vessels (all vessels of TTS patients and all culprit + non-culprit vessels of STEMI patients), these findings were consistent (Suppl. Material Page 1). This could possibly be of help in differentiating ACS from TTS with bystander coronary artery disease when the cause of an apical wall motion abnormality is in doubt, but due to a high prevalence of elevated caIMR in non-culprit ACS vessels (53%) this hypothesis will need evaluation in a prospective all-comer cohort. However, the high percentage of elevated caIMR in non-culprit vessels of ACS patients in this study is in contrast to previously published results but could be a consequence of the unusually high female prevalence in our ACS cohort, which can be mainly attributed to the matched fashion of the study, as TTS cohorts usually have a female prevalence of ~ 85%.

We observed a trend towards higher caIMR values in the LAD of patients with apical TTS types compared to midventricular TTS types. This trend could potentially achieve statistical significance with an increased number of patients. While the elevated caIMR in the LAD could reflect the akinesia in the anterior wall and apical cap, IMR alone does not appear to fully explain the different wall motion patterns. In contrast to our findings, Sans-Rosello et al. demonstrated differences in caIMR between different TTS types. However, in their study, the authors were splitting apical and midventricular TTS types into “limited” and combined groups (“apical limited” vs “apical + midventricular” and “midventricular limited” vs “midventricular + basal”). When considering only the “limited” groups, their results seem comparable to the findings in this study.

This study includes 10 TTS patients with follow-up angiographies for various reasons (no recurrent TTS) at median of 28 months from the TTS event. A transient nature of elevated caIMR in TTS was demonstrated, independent of the extent of caIMR elevation during the acute TTS event. Therefore, the study provides evidence that patients who suffer TTS do not necessarily have a certain degree of chronic CMD at baseline but can resemble the general population apart from their TTS event. These findings support those reported by Rivero and colleagues, who demonstrated an inverse relation between invasively measured IMR and the duration between the time of symptom onset and hospital admission in a small TTS cohort of 15 patients, suggesting a transient nature of caIMR elevation in TTS [32].

In this cohort, no statistically significant correlation between cardiac troponins, LVEF or LVEDP and caIMR was observed (Suppl. Figure 2). The lack of correlation with troponin is challenging to interpret, however, the prognostic role of troponin in TTS has not been fully elucidated and may be influenced by other factors. When considering wall-motion abnormalities, it is important to note that LVEF and, consequently, LVEDP during the acute event are influenced not only by the extent of akinesia but also by the degree of hyperkinesia, as seen in the basal myocardium in the apical TTS variant. Moreover, this correlation might vary significantly between typical and atypical TTS forms. While in another previously published study [9], the caIMR in the LAD correlated with LVEF, in our study, with a substantially higher number of atypical TTS cases, these findings could not be reproduced.

Since from a physiological standpoint, an increased LVEDP might correspond to elevated microcirculatory resistance, a sub-analysis including only patients with normal LVEDP (≤ 12 mmHg) at the time of angiography was performed. In this context, comparable results to the main analysis were obtained (Suppl. Material p. 1).

Similar to the prognostic impact of elevated caIMR in ACS patients [15, 16], this study demonstrated that the extent of caIMR elevation might predict in-hospital outcomes in patients with TTS. While in ACS cohorts CMD most likely correlates with infarct size and myocardial scaring, thereby influencing long-term outcomes, the transient nature of caIMR elevations in TTS patients might mainly influence the in-hospital outcome. In this regard, the TTS-specific major in-hospital adverse cardiac event (MACE) rate of 19.8% defined as death, cardiac arrest, ventricular arrhythmia and cardiogenic shock as a result of cardiogenic shock was associated with higher caIMR values in the LAD. The implications on in-hospital outcomes, together with the evidence for the transient nature of elevated caIMR in TTS suggest that the microvascular function might represent a promising therapeutic target in the acute TTS event.

Sans-Roselló et al. recently published their cohorts’ one-year follow-up MACE rate of 21.2% in 166 TTS patients in which long-term outcome significantly correlated with the extent of caIMR elevation at the time of the acute TTS event [33]. Of note, the end-point definition included all-cause death, cardiovascular death, heart failure event, acute myocardial infarction and symptomatic tachyarrhythmia/bradyarrhythmia, which occurred also after complete recovery and, therefore, might not be a consequence of the TTS event, but rather the old age and high rate of comorbidities in patients with TTS. However, within the cohort examined in this study, increased caIMR did not exhibit a significant association with established TTS-specific adverse long-term outcomes. This finding aligns with our initial expectations, indicating that elevated caIMR may not be a major contributor to established TTS-specific adverse long-term outcomes but rather in-hospital outcomes.

The design of the present study did not allow for comparison of caIMR to invasively measured IMR. However, reports validating the caIMR in comparison to invasive hyperemic IMR have been published [13, 16]. The number of TTS patients with follow-up angiographies in our study was limited and the timing of the follow-up exam varied from 22 to 35 months. Therefore, no conclusions can be drawn on the exact timely dynamics of caIMR elevations and its recovery in the setting of TTS.