Takotsubo cardiomyopathy (TTC) has generally been regarded as a relatively transient disorder, characterized by reversible regional left ventricular systolic dysfunction. However, most patients with TTC experience prolonged lassitude or dyspnea after acute attacks. Although this might reflect continued emotional stress, myocardial inflammation and accentuated brain-type natriuretic peptide (BNP) release persist for at least 3 months. We therefore tested the hypotheses that this continued inflammation is associated with (1) persistent contractile dysfunction and (2) consequent impairment of quality of life. Echocardiographic parameters (global longitudinal strain [GLS], longitudinal strain rate [LSR], and peak apical twist [AT]) were compared acutely and after 3 months in 36 female patients with TTC and 19 age-matched female controls. Furthermore, correlations were sought between putative functional anomalies, inflammatory markers (T2 score on cardiovascular magnetic resonance, plasma NT-proBNP, and high-sensitivity C-reactive protein levels), and the physical composite component of SF36 score (SF36-PCS). In TTC cases, left ventricular ejection fraction returned to normal within 3 months. GLS, LSR, and AT improved significantly over 3-month recovery, but GLS remained reduced compared to controls even at follow-up (−17.9 ± 3.1% vs −20.0 ± 1.8%, p = 0.003). Impaired GLS at 3 months was associated with both persistent NT-proBNP elevation (p = 0.03) and reduced SF36-PCS at ≥3 months (p = 0.04). In conclusion, despite normalization of left ventricular ejection fraction, GLS remains impaired for at least 3 months, possibly as a result of residual myocardial inflammation. Furthermore, perception of impaired physical exercise capacity ≥3 months after TTC may be explained by persistent myocardial dysfunction.
Takotsubo cardiomyopathy (TTC), also known as acute stress-induced cardiomyopathy, apical ballooning syndrome, and “broken heart syndrome,” is a form of regional left ventricular (LV) systolic dysfunction most commonly seen in aging women after physical or emotional stress, which can be attributed to a dysfunctional myocardial reaction to high concentrations of catecholamines. The classic course of TTC is “spontaneously reversible” LV systolic dysfunction: initially hypokinetic or akinetic (usually apical) regions recover to apparently normal wall motion within days in most cases. However, recent clinical studies report persistence of symptoms well beyond this, whereas myocardial inflammation persists on LV biopsy and myocardial edema on magnetic resonance imaging, 3 months after acute attacks. Similarly, the associated (inflammatory) release of brain-type natriuretic peptide/N-terminalpro BNP (BNP/NT-proBNP) persists for at least that duration. We have now performed an echocardiographic comparison of TTC and matched control subjects, to test 2 hypotheses: (1) LV systolic function remains abnormal 3 months after TTC, despite normal LVEF, reflecting underlying myocardial inflammation and (2) the extent of putative LV dysfunction after 3 months correlates with symptomatic limitation of physical performance.
Methods
Patients with TTC were eligible for inclusion on the basis of an acute diagnosis, according to the Mayo Clinic criteria including (1) a presentation with abnormal findings of ST segments/T waves, with or without chest pain or dyspnea, (2) a sustained elevation of cardiac troponin levels, and (3) demonstration of a characteristic wall motion abnormality, not confined to a single epicardial coronary territory and the exclusion of obstructive coronary disease, through selective coronary angiography, in the territory subserving the regional wall motion abnormality. Exclusion criteria were preexisting LV or valvular disease, pheochromocytoma, and myocarditis, each of which was prospectively considered.
Age- and gender-matched controls were recruited by advertisement and were eligible in the absence of known cardiac disease. The study complies with the Declaration of Helsinki and was approved by the institutional ethics of human research committee. All participants in each group provided written informed consent.
Echocardiography was performed, using a Vivid 7 echocardiography machine with an M5 ultrasound transducer (GE Vingmed, Horton, Norway). In patients with TTC, echocardiography was performed at the time of initial diagnosis, 10 days and 3 months thereafter. Standardized echocardiographic data included pulse wave Doppler, tissue Doppler, and 2-dimensional gray-scale imaging in all conventional views, as recommended by the American Society of Echocardiography. Loops of 3 consecutive cardiac cycles, with a minimum frame rate of 50 seconds, were stored offline, for subsequent analysis with a dedicated software package (EchoPAC version BT11, GE Vingmed). LVEF, LV end-systolic volume index, and LV end-diastolic volume index (corrected to body surface area of individual subjects) were calculated using Simpson’s biplane method. The wall motion score index (WMSI) was determined as previously described.
For speckle tracking analyses, the endocardium was manually traced at end systole to generate a segmented region of interest, the width of which was adjusted to include the thickness of the LV myocardium. Segmental motion was then automatically tracked throughout a cardiac cycle. Strain was defined as the change in the length of the segment studied, as a percentage of its final (end-diastolic) length, with respect to the 3 orthogonal directions of tissue deformation (longitudinal, radial, and circumferential, in relation to the cardiac axis); strain rate (SR) was defined as the rate at which this deformation occurs. Peak values were derived from strain–time and strain rate (SR)–time curves, respectively. Mean global longitudinal strain (GLS)/SR were derived from the 3 individual apical views and thus represent triplanar averages. Radial and circumferential function was studied as strain and SR (averaged from all segments) at the base, midventricle, and apex. Twist mechanics were studied at the LV base and apex, and the average peak rotation and rotation rate were determined from automatically generated curves. Counterclockwise rotation, as observed at the apex, was expressed as a positive value, whereas basal clockwise rotation was expressed as a negative value. LV twist was defined as the net difference between peak apical and peak basal rotations, whereas LV torsion was defined as LV twist, normalized for LV diastolic length, as elsewhere described.
Cardiovascular magnetic resonance (CMR) imaging was performed both acutely and after 3 months to quantitate myocardial edema through integrated estimates of T2-weighted signal intensity within the entire left ventricle, as previously described.
A number of biochemical investigations were performed to quantitate the initial catecholamine stimulus toward development of TTC, together with the extent of inflammatory response and cellular damage. Furthermore, most of these investigations were repeated at 3-month follow-up.
- (1)
Catecholamine release was measured through plasma metanephrine and normetanephrine concentrations.
- (2)
Inflammatory activation was quantitated through high-sensitivity C-reactive protein and NT-proBNP concentrations.
- (3)
Extent of myocardial necrosis was quantitated through release of troponin T and creatine kinase (CK).
We sought to evaluate patients’ quality of life as assessed by the Short-Form health survey (SF36), completed at ≥3 months by 24 patients. The SF36 questionnaire includes a physical composite score (SF36-PCS), which was used for assessment of hypothesis 2.
The data were analyzed using the SPSS version 20 software (SPSS, Chicago, Illinois) and presented as mean ± 1SD unless otherwise specified. To evaluate functional indexes of recovery, 3 indexes were chosen as potential markers of recovery of LV systolic function. These were (1) GLS, (2) longitudinal strain rate (LSR), and (3) peak apical twist (AT). It was prospectively elected that correlations between recovery and ongoing inflammation would use whichever of these parameters displayed the most consistent impairment during recovery from TTC. As regards testing of hypothesis 2, we elected to seek correlations between GLS at 3 months and SF36-PCS using Pearson correlation, and correlations with the total SF36 were also evaluated as a secondary consideration. Comparisons between control subjects and patients with TTC were performed using the 3-month TTC data. Methods included nonpaired t tests or the Wilcoxon signed rank tests for quantitative data and the chi-square tests for categorical data. Correlations within the TTC group were performed through Pearson or Spearman rank correlation as appropriate.
Results
Patients with TTC (n = 36) and control subjects (n = 19) were well matched for age and distribution of coronary risk factors ( Table 1 ). In the TTC cohort, 33% presented with ST elevation, and all had abnormally elevated plasma troponin T (0.46 ± 0.39 mg/dl) and NT-proBNP levels (7557 ± 8626 pg/ml). Of the patients with TTC, 75% were treated with ACE inhibitors for at least the first 3 months after presentation.
| Parameters | TTC patients (n = 36) | Controls (n = 19) |
|---|---|---|
| Age (years) (mean±SD) | 67 ± 11 | 63 ±12 |
| BMI(kg/m 2 ) (mean±SD) | 28 ± 6 | 26 ± 4 |
| Current smoking | 6 (17%) | 2 (11%) |
| Systemic hypertension | 20 (56%) | 6 (32%) |
| Diabetes mellitus | 8 (22%) | 4 (21%) |
| Hyperlipidemia | 18 (50%) | 6 (32%) |
Over the first 3 months of follow-up, all parameters of LV systolic function improved progressively ( Table 2 ). For example, LVEF increased from 52 ± 14% at presentation to 60 ± 7% (p = 0.003), with significant reductions in LV end-systolic and end-diastolic volumes. Furthermore, there was significant improvement in all rotational and longitudinal parameters measured ( Table 2 ). Transmitral E/A ratio decreased from 1.1 ± 0.5 to 0.9 ± 0.4 (p <0.05).
| Echocardiographic parameters | Controls | TTC | |
|---|---|---|---|
| Acute | 3 months | ||
| 1. Non-deformational parameters | |||
| LVEDVi (mL/m 2 ) | 36.1 ± 7 | 45.1 ± 15.9 | 34.8 ± 9.6 ‡ |
| LVESVi (mL/m 2 ) | 13.6 ± 3.8 | 22.7 ± 11.5 | 14.1 ±4.7 # |
| LVEF (%) | 63 ± 6 | 52 ± 14 | 60 ± 7 ‡ |
| Transmitral E/A ratio | 1.03 ± 0.3 | 1.1 ± 0.5 | 0.9 ± 0.4 ≠≠ |
| E/Ea ratio | 10.5 ± 4.8 | 12.7 ± 10 | 10.3 ± 3.5 |
| 2. 2D speckle tracking parameters | |||
| GLS (%) | -20.0 ± 1.8 | -12.7 ± 3.9 | -17.9 ± 3.1 #,≠ |
| Strain rate (s -1 ) ∗ | -1.1 ± 0.1 | -0.9 ± 0.2 | -1.0 ± 0.2 † |
| Basal twist ∗ | -4.5 ± 2.3 | -6.2 ± 4.7 | -6.6 ± 3.3 |
| Apical twist ∗ | 11.3 ± 5 | 4.9 ± 3.9 | 9.3 ± 4 ‡ |
| LV twist ( ° ) | 15.7 ± 5.2 | 11.1 ± 6.8 | 15.9 ± 5.9 † |
| LV torsion ( ° cm -1 ) ∗ | 2.1 ± 0.7 | 1.4 ± 0.9 | 2.1 ± 0.8 † |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
