Author + information
- Dana K. Dawson, DM, DPhil∗ (, )
- Christopher J. Neil, MBBS,
- Anke Henning, PhD,
- Donnie Cameron, PhD,
- Baljit Jagpal, RDMS,
- Margaret Bruce, RN,
- John Horowitz, MBBS and
- Michael P. Frenneaux
- ↵∗School of Medicine & Dentistry, University of Aberdeen, Polwarth Building, Foresterhill, Aberdeen, AB25 2ZD, United Kingdom
Tako-tsubo cardiomyopathy (TTC) is an acute form of left ventricular (LV) systolic dysfunction often following intense stress, but its pathophysiology remains elusive. Because of its reversible natural course without much apparent myocyte damage, as inferred by the lack of late gadolinium-enhanced (LGE) cardiac magnetic resonance (CMR), it has been a priori assumed that its evolution is benign (1). Whatever the etiology of the acute insult, it is important to establish the nature of this state of intense myocardial contractile dysfunction with preserved viability, as this may point to a potential therapeutic target in those cases with a less favorable clinical course. It is known that impaired cardiac energetic status is directly linked to clinical and subclinical LV dysfunction. We hypothesized that impaired cardiac energetics may occur in acute TTC, further investigating its recovery at 4 months follow-up.
We prospectively recruited 26 patients with a clear-cut diagnosis of TTC (2) and 11 matched healthy controls; all provided informed consent. Patients underwent 31P-CMR spectroscopy (31P-CMRS) (3) and imaging (inclusive of native T1 mapping with a 3(3)3(3)5 scheme) on a 3-T Philips scanner (Best, the Netherlands), short term (days 0 to 3) and at 4 months follow-up. 31P-CMRS was analyzed with JMRUi-3 (University of Lyon, Lyon, France), LV volumes/mass with CMRTools (Cardiovascular Imaging Solutions, London, United Kingdom), wall motion scored (1 = normal, 2 = hypokinetic, 3 = akinetic, 4 = dyskinetic) and T1 maps generated with RelaxMaps (Philips)/Segment (Medviso, Lund, Sweden). C-reactive protein (CRP) and troponin I were recorded on admission. Follow-up 31P-CMRS and imaging were achieved in 20 patients (3 device implantations, 2 in-hospital deaths, 1 claustrophobia were excluded).
Data are shown as mean ± SD unless otherwise stated. Comparisons between groups were performed using independent/paired Student t tests with significance at p < 0.05.
The patients’ mean age was 63 years (range 41 to 87 years), and 92% were women. The majority presented with chest pain (80%) and had a stressful trigger (73%), widespread ST-segment elevation electrocardiogram (70%), and apical ballooning variant (77%). The 12-h troponin level was 3.45 ng/ml (range 0.22 to 11.97 ng/ml), and CRP was 21.5 (range <4 to 75).
Mean LV ejection fraction was reduced and LV mass index increased during the acute study versus healthy controls (54 ± 12% vs. 66 ± 4% and 77 ± 15 g/m2 vs. 66 ± 10 g/m2, respectively, p < 0.05 for both); all normalized at follow-up. LGE was present in only 4 cases, as a transmural band at the hinge points of the wall motion abnormality in 3 cases and as nonconfluent small foci in another case.
A significant increase in native whole myocardium T1 relaxation time was seen during the acute presentation compared with healthy volunteers (1,253 ± 63 ms vs. 1,188 ± 16 ms, p = 0.004), and this improved significantly (1,196 ± 29 ms) at follow-up (p = 0.01) (Figure 1A). The acute changes were driven by both the T1 values measured from acutely dysfunctional segments (T1 = 1,279 ± 86 ms, p = 0.03 vs. healthy controls) but also by the segments with normal wall motion in the acute study (1,226 ± 49 ms, p = 0.02 vs. healthy controls). At follow-up, the T1 relaxation times significantly improved (1,177 ± 42 ms) compared with the acute study in the nondysfunctional segments (p = 0.008), but this improvement only showed a trend in the previously dysfunctional myocardium (p = 0.05), which remained significantly abnormal compared with healthy controls (T1 = 1,224 ± 40 ms, p = 0.013 vs. healthy controls), implying that a certain degree of myocardial edema remains present until at least this stage.
A profound decrease in resting cardiac energetic status was seen during the acute tako-tsubo phase compared with healthy volunteers (phosphocreatine/adenosine triphosphate [PCr/ATP] ratio 1.1 ± 0.39 vs. 1.9 ± 0.43, p << 0.0001). The PCr/ATP ratio increased to 1.5 ± 0.48 at 4 months follow-up (p = 0.018 compared with initial presentation) but remained significantly reduced compared with normal healthy volunteers (p = 0.02) (Figure 1B).
We demonstrate severe global edema associated with profound cardiac energetic impairment with incomplete resolution of both at 4 months. These observations contribute to the understanding of TTC pathophysiology.
The authors thank all their NHS consultant colleagues at Aberdeen Royal Infirmary for their help with prompt recruitment of these patients (Dr. Malcolm Metcalfe, Dr. Andrew Stewart, Dr. Andrew Hannah, Dr. Awsan Noman, Dr. Paul Broadhurst, Dr. Duncan Hogg, Dr. Deepak Garg) and to Dr. Caroline Scally, specialist cardiology trainee, for special effort with recruitment. Thanks also go to Ms. Bernice Kai Ng, medical student, for her initial work with cardiac spectroscopy in healthy volunteers and data entry.
Please note: This work was supported by Tenovus Scotland (to Dr. Dawson) and NHMRC Australia (to Prof. Horowitz). Dr. Dawson has a research agreement with Philips Healthcare. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation