Takotsubo cardiomyopathy (TC) is generally recognized to have a good prognosis, but it can be rarely aggravated. We sought to investigate the clinical characteristics of TC and to evaluate the effects of clinical parameters on predicting delayed recovery. We enrolled consecutive patients with TC admitted to our hospital from January 1991 to January 2014. We defined delayed recovery as sustained left ventricular (LV) systolic dysfunction requiring ≥10 days for LV contraction to normalize. We screened 9,630 patients suspected of having acute coronary syndrome, and 60 patients (0.6%; men/women: 20/38; mean age: 69.7 ± 11.9 years) were diagnosed as having TC. With the exception of 2 patients who died before LV systolic function improved, all patients recovered from LV systolic dysfunction within 6 months; the mean recovery period was 9.1 ± 11.5 days. Twenty-eight patients met the criteria for delayed recovery. Univariate logistic regression analyses showed that male gender, LV end-diastolic diameter, brain natriuretic peptide (BNP) level, body mass index (BMI), and nonuse of calcium channel blockers (CCBs) at baseline were associated with delayed recovery. Among these factors, multiple logistic regression analysis identified BNP ≥238 pg/ml (relative risk [RR] 11.6, p = 0.002) and nonuse of CCBs (RR 22.2, p = 0.0014) as independent risk factors for delayed recovery and leptosomic build (BMI <20 kg/m 2 ) as an independent predictor of rapid recovery (RR 0.11, p = 0.02). In conclusion, BNP level, BMI, and use of CCBs are associated with recovery speed of LV systolic function in patients with TC.
Takotsubo cardiomyopathy (TC) is a cardiac syndrome characterized by transient left ventricular (LV) systolic dysfunction with apical ballooning and electrocardiographic changes that mimic acute coronary syndrome in the absence of significant obstructive coronary disease. The onset of TC is frequently (but not always) triggered by an acute medical illness or by intense emotional or physical stress. TC is also generally recognized as a disease with good prognosis, but it may be rarely aggravated, causing various complications such as heart failure, arrhythmia, cardiac rupture, and thromboembolism. LV systolic dysfunction varies among patients and may affect prognosis. It remains unclear what clinical parameters are most important in predicting delayed recovery from LV systolic dysfunction in patients with TC. The aim of this study was to investigate the effect of clinical parameters on delayed LV recovery by retrospectively analyzing follow-up data from patients with TC.
Methods
We screened all patients who underwent cardiac catheterizations for suspicion of acute coronary syndrome from January 1999 to January 2014 at the Nippon Medical School Hospital and finally analyzed the patients among them diagnosed as having TC. The clinical data, along with blood examination, 12-lead electrocardiogram, cardiac catheterization, and echocardiography data, were retrospectively retrieved from electronic medical records and the cardiac catheter database. Patients with TC satisfied the Mayo Clinic criteria for the clinical diagnosis of TC as follows: (1) transient akinesia or dyskinesia in the LV apical and midventricular segments with regional wall motion abnormalities extending beyond a single epicardial vascular distribution, (2) the absence of obstructive coronary disease or angiographic evidence of an acute plaque rupture, (3) new electrocardiographic abnormalities (either ST-segment elevation or T-wave inversion), and (4) the absence of intracranial bleeding, recent significant head trauma, pheochromocytoma, obstructive epicardial coronary artery disease, myocarditis, or hypertrophic cardiomyopathy. We divided patients with TC into 2 groups: the early recovery group (patients who recovered from LV systolic dysfunction within 10 days) and the delayed recovery group (patients who recovered from LV systolic dysfunction in longer than 10 days) because the average of admission period in the present study was 9.8 days and the average recovery time from LV systolic dysfunction was 9.1 days. We defined the ejection fraction >50% as the improvement of the cardiac function. Echocardiograms of all cases were independently reviewed by at least 2 echocardiographers, and the LV ejection fraction was calculated using the biplane Simpson’s method from orthogonal 2-chamber and 4-chamber apical views with PHILIPS IE 33 (Amsterdam, The Netherlands) or GE Vivid E 9 (Fairfield, Connecticut) ultrasound systems. Peripheral blood samples were collected for laboratory data, including serum brain natriuretic peptide (BNP) and cardiac troponin-T (cTnT) levels. Serum BNP and cTnT were measured with an AIA-2000ST analyzer (TOSOH, Tokyo, Japan) in accordance with the manufacturer’s instructions. The assay detection limit of cTnT was ≥0.003 ng/ml. The Nippon Medical School Research Ethics Committee approved this study, and written consent was obtained from all patients.
Continuous variables were tested for normality by the Shapiro-Wilk test. Normally distributed continuous variables are shown as means ± SD, and nonparametrically distributed variables are shown as medians (interquartile range). Categorical variables are presented as frequencies (percentages). The correlation between 2 continuous variables was examined by Pearson’s test (if relevant by Spearman’s test). Logistic regression analyses were performed with the early/delayed recovery from LV ejection fraction (the cut-off point of recovery time was defined as 10 days) and different clinical factors as covariables. Multiple regression analyses of the relation between the 2 groups and study variables were performed to detect clinical characteristics related to these markers after adjustment for interrelation among study variables. Potential factors were eligible for entry into the multivariate logistic regression models by univariate analysis (significance level for entry into the model = 0.15). Then, the optimal discriminating thresholds of these factors values were determined. A 2-sided probability value of p <0.05 was considered to be statistically significant. JMP (version 9.0.3; SAS Institute, Cary, North Carolina) was used for analysis.
Results
We screened 9,630 patients who were suspected to have acute coronary syndrome at our hospital from January 1991 to January 2014, and 60 of these (0.6%) were diagnosed as having TC. Two cases were excluded because these patients died from noncardiac causes before the LV dysfunction improved. Finally, 58 patients were enrolled in this study ( Figure 1 ).
The baseline characteristics of the patients are summarized in Table 1 . Of the 58 patients, 65.5% (n = 38) were women (all postmenopausal), 62.7% (n = 37) had hypertension, 16.9% (n = 10) had diabetes mellitus, 32.2% (n = 19) had a history of smoking (current or past), and 16.9% (n = 10) required respirator management because of difficulty in breathing caused by heart failure. The inhospital mortality in patients with TC was 3.0%. Echocardiographic findings showed that LV ejection fraction at admission was 41.0 ± 6.8%, and all patients recovered from LV dysfunction within 6 months, with a mean recovery period of 9.1 ± 11.5 days. Laboratory data at admission showed that the peak serum BNP level was 1,057 ± 2,070 pg/ml, and the peak of sensitive cTnT level was 0.94 ± 1.14 mg/dl. White blood cell count was 10,312 ± 5,697 per microliter, sensitive C-reactive protein was 2.92 ± 4.43 mg/dl, and creatinine kinase MB fraction was 26.8 ± 28.6 IU/L.
| Variable | ALL (N = 58) | Early recovery (N = 38) | Delayed recovery (N = 20) | P value |
|---|---|---|---|---|
| Age (years) | 69.7 ± 12.0 | 70.5 ± 12.6 | 68.2 ± 10.8 | 0.5 |
| Women | 38 (64 %) | 29 (76 %) | 9 (45 %) | 0.02 ∗ |
| Body mass index (kg/m 2 ) | 21.4 ± 4.0 | 20.7 ± 3.6 | 22.5 ± 4.6 | 0.1 |
| Hypertension | 37 (63 %) | 24 (63 %) | 13 (68 %) | 0.7 |
| Diabetes mellitus | 10 (17 %) | 5 (14 %) | 5 (26 %) | 0.2 |
| Dyslipidemia | 21 (36 %) | 17 (47 %) | 4 (22 %) | 0.7 |
| Hyperuricemia | 12 (20 %) | 7 (19 %) | 5 (28 %) | 0.5 |
| Systolic blood pressure (mmHg) | 123.4 ± 30.5 | 124.8 ± 27.7 | 120.8 ± 35.8 | 0.6 |
| Diastolic blood pressure (mmHg) | 72.2 ± 20.7 | 73.0 ± 20.7 | 70.8 ± 21.4 | 0.7 |
| Heart rate (bpm) | 92.1 ± 22.5 | 91.5 ± 20.1 | 27.1 ± 6.2 | 0.8 |
| Types of stress cause TCM | 0.17 | |||
| Emotional stress | 9 (16 %) | 8 (89 %) | 1 (11 %) | |
| Physical stress | 21 (36 %) | 14 (67 %) | 7 (63 %) | |
| Infection | 8 (14 %) | 3 (38 %) | 5 (63 %) | |
| Others | 20 (35 %) | 13 (65 %) | 7 (35 %) | |
| Medications at Admission | ||||
| Beta-blockers | 5 (9 %) | 3 (8 %) | 2 (12 %) | 0.7 |
| Angiotensin-converting enzyme inhibitors or angiotensin receptor blockers | 9 (15 %) | 7 (20 %) | 2 (11 %) | 0.4 |
| Diuretics | 7 (12 %) | 5 (14 %) | 2 (11 %) | 0.8 |
| Nicorandil | 2 (3 %) | 1 (3 %) | 1 (6 %) | 0.6 |
| Statins | 8 (14 %) | 5 (14 %) | 3 (17 %) | 0.8 |
| Calcium channel blockers | 16 (27 %) | 14 (40 %) | 2 (11 %) | 0.02 ∗ |
| Antiplatelet drugs | 12 (20 %) | 9 (25 %) | 3 (17 %) | 0.5 |
Patients were also divided into 2 groups according to their recovery time from LV dysfunction: the early recovery group (recovery time <10 days) and the delayed recovery group (recovery time ≥10 days). The comparison of clinical characteristics between the early recovery group and the delayed recovery group by univariate logistic regression is listed in Tables 1 to 3 . The delayed recovery group contained 34.5% (n = 20) of the patients with TC. LV end-diastolic diameter and body mass index (BMI) at admission were higher in the delayed recovery group than in the early recovery group (LV end-diastolic diameter: 50.7 vs 44.9 mm, p = 0.01; BMI 22.5 vs 20.8 kg/m 2 , p = 0.12). An increased level of serum BNP at admission was observed in the delayed recovery group compared with the early recovery group (log BNP: 2.91 vs 2.50, p = 0.01). Use of calcium channel blockers (CCBs) was higher in the early recovery group compared with the delayed recovery group (40% vs 11.1%, p = 0.03). Of patients using CCBs, 4 (18%) received amlodipine, 7 (31%) benidipine, 2 (9%) nifedipine, 1 (4.5%) nicardipine, and 2 (9%) did diltiazem. Patients using benidipine tended to be associated with delayed recovery, but this did not reach to statistical significance.
| Variable | All (N = 58) | Early recovery (N = 38) | Delayed recovery (N = 20) | P value |
|---|---|---|---|---|
| Max brain natriuretic peptide (pg/mL) | 1057 ± 2070 | 596 ± 635 | 1953 ± 3319 | 0.02 ∗ |
| Max Troponin-T (mg/dL) | 0.94 ± 1.14 | 0.94 ± 1.21 | 1.05 ± 0.26 | 1.0 |
| White blood cell count (/μL) | 10,312 ± 5698 | 10,052 ± 5830 | 10,831 ± 5538 | 0.6 |
| C-reactive protein (mg/dL) | 2.92 ± 4.43 | 2.55 ± 0.83 | 3.50 ± 1.05 | 0.5 |
| Creatinine kinase muscle and brain (IU/L) | 26.8 ± 28.6 | 28.4 ± 31.5 | 23.6 ± 22.0 | 0.5 |
| Variables | All | Early recovery | Delayed recovery | P value |
|---|---|---|---|---|
| Left ventricular ejection fraction at admission (%) | 41.4 ± 16.8 | 42.0 ± 18.6 | 40.2 ± 12.9 | 0.7 |
| Recovery time of left ventricular ejection fraction (days) | 9.1 ± 11.5 | 3.8 ± 2.6 | 19.3 ± 14.7 | <0.001 ∗ |
| E/A ratio | 1.07 ± 1.09 | 1.13 ± .123 | 0.91 ± 0.62 | 0.6 |
| Left ventricular end-diastolic dimension (mm) | 46.7 ± 7.57 | 44.9 ± 6.3 | 50.7 ± 8.9 | 0.01 ∗ |
| Posterior wall thickness (mm) | 7.70 ±.701 | 10.5 ± 2.3 | 11.3 ± 2.0 | 0.3 |
| Inter-ventricular septal thickness (mm) | 10.6 ± 2.60 | 10.5 ± 2.4 | 10.6 ± 3.0 | 0.9 |
| Left atrium diameter (mm) | 34.4 ± 6.57 | 33.8 ± 6.3 | 35.6 ± 7.0 | 0.4 |
| Left ventricular thrombus | 2 (3.4%) | 1 (2.6%) | 1 (5.0%) | 0.6 |
| Left ventricular obstruction | 7 (13.6%) | 6 (15.8%) | 1 (5.0%) | 0.2 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
