Abstract
Background:
Acute coronary syndrome (ACS) from non-ST-segment elevation myocardial infarction (NSTEMI) and Takotsubo (TK) cardiomyopathy present with similar initial clinical features and can result in left ventricular (LV) dysfunction and acute heart failure.
Methods:
This study was a retrospective case-control study that identified patients aged 18 years and older who presented with ACS and underwent cardiac catheterization.
Results:
There were a total of 321 patients in the TK group and 1031 patients in the NSTEMI group. There was significantly worse LV dysfunction in the TK group with average ejection fraction (EF) of 44.35% (±15.11%) versus NSTEMI with an average EF of 47.36% (±13.5%) (P < .001). The presence of TK yielded of an odds ratio (OR) of 2.373 (95% confidence interval [CI]: 1.165-3.618) and presence of peripheral artery disease (PAD) yielded an OR of 2.053 (95% CI: 1.165-3.618).
Conclusions:
The presence of TK cardiomyopathy and PAD were independent predictors of patients who had LVEF of <35% and elevated B-type natriuretic peptide levels.
Keywords
Introduction
Takotsubo (TK) cardiomyopathy, also called apical ballooning syndrome, broken heart syndrome, or stress cardiomyopathy, is generally characterized by transient systolic dysfunction of the apical and/or mid segments of the left ventricle that mimics myocardial infarction, but in the absence of obstructive coronary artery disease (CAD).1–5 In the acute phase, the clinical presentation, echocardiographic findings, and biomarker profiles are often similar to those of an acute coronary syndrome (ACS).2,5 It predominantly affects postmenopausal women after exposure to an unexpected physical or emotional stressor; however, often no stressor is identified.1–5 The exact mechanism of TK cardiomyopathy is unknown although a neurohormonal effect from high circulating levels of norepinephrine and epinephrine on cardiac myocytes leading to hypercontraction and functional basal outflow obstruction has been described.3,6
Although typically viewed as benign and transient, TK is not free of serious complications, mostly occurring during the hospital stay. Acute complications of TK can include heart failure (HF), tachyarrhythmias, bradyarrhythmias, mitral regurgitation (MR), and cardiogenic shock.1,4 Reported in-hospital mortality rates have ranged from 0% to 8%.1,5,7 A previous study found that age >70 years, physical stressor, admission troponin, and low ejection fraction (EF) (defined as EF < 40%) were independent predictors of acute HF in TK cardiomyopathy. 5 It is generally considered that these patients who develop acute HF from TK cardiomyopathy have fewer complications, less severity of signs and symptoms, and overall less mortality from an equal decrease in left ventricular (LV) EF compared with patients who present with acute HF from non-ST-segment elevation myocardial infarction (NSTEMI). The specific aim of this study is to compare demographic, echocardiographic, and clinical data in patients with acute LV systolic dysfunction in TK cardiomyopathy to patients with acute LV systolic dysfunction in NSTEMI and identify independent predictors of LV dysfunction and elevated B-type natriuretic peptide (BNP).
Methods
This study was an institutional review board–approved retrospective case-control study that identified patients aged 18 years and older who underwent cardiac catheterization during their hospital visit to confirm active or nonobstructive CAD. Patients were diagnosed with TK cardiomyopathy by a clinician-based diagnosis or patients who presented as NSTEMI but did not undergo percutaneous coronary intervention (PCI) during their cardiac catheterization. Patients who required PCI or were referred for coronary artery bypass graft (CABG) surgery were identified as having an NSTEMI. Patients also had an echocardiogram to assess left ventricular ejection fraction (LVEF).
The primary outcome of this study was to identify patient variables that predict the presence of reduced LVEF (defined as <35%) and elevated BNP (defined as >1200) in a cohort of patients with either TK cardiomyopathy or NSTEMI. Presence and severity of MR was collected to determine whether this influenced outcomes in either group.
Setting and participants
Patient data were obtained from October 1, 2009 to August 31, 2015. Patients diagnosed with TK were either diagnosed by clinician or diagnosed as NSTEMI who subsequently underwent cardiac catheterization without PCI. All subjects included in the study underwent cardiac catheterization during their hospital visit to confirm active or nonobstructive CAD. Also required was echocardiogram to assess LVEF, presence and severity of MR, aortic stenosis (AS), and aortic regurgitation (AR). Patients were excluded from both NSTEMI and TK groups if they had no cardiac catheterization and/or echocardiogram during their hospitalization or were under 18 years of age. Patients who were clinician diagnosed as TK cardiomyopathy but had cardiac catheterization with PCI were excluded from the study. Ultimately, 321 patients with TK cardiomyopathy and 1031 patients with NSTEMI were included in the study.
Data collection
Patient data was collected using International Classification of Diseases, Ninth Revision (ICD-9) coding for acute myocardial infarction (ICD-9 410.9) and TK cardiomyopathy (ICD-9 429.83). Patient’s charts were reviewed through electronic medical records to obtain demographic, clinical, transthoracic echocardiogram, and cardiac catheterization data.
Patient’s demographic information collected included age, sex, weight, and height to calculate body mass index. Factors regarding patient medical history such as documented history of HF, relevant comorbid conditions listed in the medical record, social history of smoking, and family history of heart disease were obtained. Laboratory values collected also included was average troponin, and also included was average BNP. Variables from transthoracic echocardiograms included presence of AS, AR, presence and severity of MR, and EF as percentage. Any valvular data labeled as trace in the echocardiogram (MR, AS, AR) were included in the group labeled as none.
Data points collected from the in-hospital cardiac catheterization included the following: severity of CAD, any PCI including angioplasty and/or stent, and referral for CABG.
Statistical analysis
Most of the data presented as continuous variables are expressed as mean ± standard deviation or counts with population as appropriate.
Variables with P < .1 on univariate analysis are considered for inclusion in the multiple regression model. Using forward conditional method, variables were entered into the model and the regression analysis was performed. All potential explanatory variables were assessed for collinearity, and only independent variables were included in the regression model. Hosmer-Lemeshow goodness-of-fit test was used to measure how well the model reflects the data on which it was created, and area under the receiver operator characteristic curve was calculated. SPSS 16.0 computer software was used to perform statistical calculations.
Results
Baseline characteristics
There were a total of 321 patients in TK group and 1031 patients in the NSTEMI group that fulfilled study criteria and included in this analysis. The groups were similar in mean age, 64.7 (SD: ±13.9) years in the TK group and 64.3 (SD: ±12.9) years in the NSTEMI group. Baseline demographics are presented in Table 1.
Demographics and characteristics of study population stratified by patients in Takotsubo versus NSTEMI groups.
Abbreviations: BIPAP, bilevel positive airway pressure; CABG, coronary artery bypass graft; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; CVA, cerebrovascular accident; FHx, family history; HTN, hypertension; ICU, intensive care unit; NSTEMI, non-ST-segment elevation myocardial infarction; OSA, obstructive sleep apnea; PAD, peripheral artery disease; PCI, percutaneous coronary intervention; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.
Vasopressor support is defined as an order in the hospital electronic medical records (EMR) for norepinephrine or epinephrine. Inotropic support is an order in the hospital EMR for either dobutamine or milrinone. Chronotropic support is an order in the hospital EMR for the medication dopamine.
There were significantly more women in the TK group versus NSTEMI group: 62.2% and 36.9% (P < .01), respectively. Other baseline demographic data that were statistically significant in the NSTEMI group were presence of type 2 diabetes mellitus (T2DM) with 25.2% (P = .04), current smokers with 29.8% (P = .01), history of CAD with 43.5% (P = .01), history of peripheral artery disease (PAD) with 9.9% (P < .001), and family history of premature CAD with 17.5% (P = .02). In the TK group, 17.8% of patients were significant for depression and 18.8% of patients had chronic obstructive pulmonary disease (COPD).
In-hospital mortality, length of stay, and intensive care unit admission
Inpatient mortality was 4.7% and 3.2% (P = .21) for TK and NSTEMI groups, respectively. For the TK group, the minimum, average, and maximum inpatient length of stay was 0, 5.5, and 77 days, respectively, whereas for the NSTEMI group, the minimum, average, and maximum inpatient length of stay was 0, 6.1, and 113 days, respectively. There was no statistically significant difference between average length of stay for TK group versus the NSTEMI group (P = .14). There were 156 patients in the TK group who were admitted to the intensive care unit (ICU), representing 48.8% of the population, whereas 718 patients in the NSTEMI group were admitted to the ICU, representing 69.7% of the population (P < .001). There were a significant number of patients in the NSTEMI group referred for CABG (21.5%). Significant differences were present regarding the use of a vasopressor (18.6% vs 7.8%, P < .001), chronotropic support (15.7% vs 5.9%, P < .001), and inotropic support (4.5% vs 1.6%, P = .02) for NSTEMI versus TK groups.
Cardiac biomarkers
Table 2 demonstrates that the mean peak troponin (ng/ml) for TK was 3.62 and for NSTEMI was 5.4 (P = .02). The maximum, average, and minimum BNP (pg/ml) value for the TK group was >70 000, 3226.07, and 8, respectively. The average admission or presenting BNP (pg/ml) was 2695.1 for the TK group. The maximum, average, and minimum BNP (pg/ml) value for the NSTEMI group was 63 451, 1974, and 7.0, respectively. The average admission or presenting BNP (pg/ml) was 1652.6 for the NSTEMI group.
Clinical variables of study population stratified by patients in the Takotsubo versus NSTEMI groups.
Abbreviations: BMI, body mass index; BNP, B-type natriuretic peptide; bpm, beats per minute; CABG, coronary artery bypass graft; kg, kilograms; NSTEMI, non-ST-segment elevation myocardial infarction; PCI, percutaneous coronary intervention.
Echocardiographic data
Results are shown in Table 2. Patients in the TK group had an average EF of 44.35% (±15.11%). Meanwhile, for NSTEMI group, the average EF was 47.35% (±13.6%) (P < .001). There were 165 patients in the TK group who had mild, moderate, or severe MR and represented 51.6% of the population, whereas 520 patients in the NSTEMI group had mild, moderate, or severe MR which represented 50.4% of the population. The presence of mild, moderate, or severe AR represented 17.8% of the population in the TK group versus 24.5% of the NSTEMI group (P < .01). Mild, moderate, or severe AS was present in 8.8% of the TK group versus 11.4% of the NSTEMI group (P = .19).
Subgroup analysis
Table 3 demonstrates a comparison of TK cardiomyopathy versus NSTEMI with an EF < 35%. A total number of 262 patients are found to have an EF < 35% in either group. Data in Table 3 are presented as a percentage of the total subset population for EF < 35%. There was a significant difference in history of depression within the TK group representing 25.6% of the population versus 5% in the NSTEMI group (P = .0001). Former smokers represented 50% of the TK population versus 33.8% of the NSTEMI population (P = .018).
Comparison of clinical data, echocardiographic data, and past medical history of Takotsubo cardiomyopathy versus NSTEMI within EF < 35%.
Abbreviations: BNP, B-type natriuretic peptide; bpm, beats per minute; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; CVA, cerebrovascular accident; HTN, hypertension; ICU, intensive care unit; NSTEMI, non-ST-segment elevation myocardial infarction; OSA, obstructive sleep apnea; PAD, peripheral artery disease; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.
A total number of 262 patients are found with an EF < 35% in either TK cardiomyopathy or NSTEMI group. Values are listed as a percentage of the total population of EF < 35%.
Values in Table 3 that demonstrated significance in the NSTEMI versus the TK group include history of HF (26.9% vs 14.6%, P = .035), history of CAD (40.6% vs 18.3%, P = .0001), family history of CAD (15.1% vs 7.3%, P = .06), and obstructive sleep apnea (OSA) (13.8% vs 3.7%, P = .014). There was a significant difference in history of PAD with 16.9% associated with the NSTEMI group versus 6.1% in the TK group (P = .026). The presence of PAD was significantly higher within the EF < 35% subset population; there was a total of 32 patients with PAD within the subset of EF < 35% which represented 13.2% of the subset population versus 82 patients within the subset population of EF ≥ 40% representing 7.4% of the subset population (P = .005).
Table 3 demonstrates the laboratory and clinical data for patients in the TK and NSTEMI groups with an EF < 35%. There was no significant difference between these variables.
Multiple regression model
Variables that achieved statistical significance with P < .10 from the univariate analysis were included in the multivariable logistic regression model. This included presence of TK cardiomyopathy, sex, T2DM, COPD, depression, PAD, and OSA. Using a forward conditional method, variables were entered into the model and the regression analysis was performed. With the forward conditional method, univariate predictors (P < .10) that were not significant on multivariate analysis were eliminated from the final model. As described in Table 4, multivariate predictors of patients having a reduced LVEF <35% and elevated BNP levels included the presence of PAD and presence of TK cardiomyopathy. The goodness-of-fit calculation revealed that the model appropriately fits the data, and area under the receiver operator characteristics curve was 0.61 (0.55-0.66) and the Hosmer-Lemeshow goodness-of-fit test: χ 2 = 0.005, df = 1, P = .941.
Score performance for the prediction of patients who have an elevated B-type natriuretic peptide and left ventricular dysfunction with left ventricular ejection fraction < 35% (N = 767).
Discussion
TK cardiomyopathy cardiomyopathy has been proposed as an acute HF syndrome. 2 Consistent with this concept, there was a significantly higher elevation in BNP in the TK cardiomyopathy group compared with the NSTEMI group. Also, there was a significantly worse degree of LV dysfunction in the TK group compared with the NSTEMI group. HF is a clinical diagnosis based on signs and symptoms as well as support of laboratory and echocardiographic data. 8 This study uses LV dysfunction with LVEF < 35% and BNP > 1200 as surrogates to define patients in acute HF. Many complications can arise from acute HF from TK cardiomyopathy such as cardiogenic shock, tachyarrhythmias, thromboembolism, and death. 2 Thus, recognition of acute HF syndrome may lead to earlier and more directed medical therapy.
There was a significantly higher presence of depression and female prevalence in the TK group. These data are consistent with common demographics seen in previous studies with a female predominance in the TK group. 2 There was also a significant male population in the NSTEMI group which is consistent with men having a higher prevalence of CAD. Many of our patients had no identifiable stressor with our chart review, and most of this is due to lack of documentation. As these patients often present as a traditional acute coronary syndrome (ACS), focused history taking on risk factors and identifiable causes for ACS without consideration for TK cardiomyopathy are also likely attributable to lack of identifiable stressor. The patients with NSTEMI had significantly higher presence of previous CAD, T2DM, current tobacco use, and PAD. All of which are known risk factors for ACS and CAD in general. 9 The study specifically looked at presence and severity of MR as presence of functional MR in HF has been shown to have worse prognosis in LV dysfunction. 10 However, there were no differences in outcomes through multivariate analysis.
TK cardiomyopathy is believed to be transient and benign compared with NSTEMI; however, in-hospital mortality rates were similar in both groups and did not achieve statistical difference. In the NSTEMI group, there was significantly more ICU admissions compared with the TK group, but this may be due to a number of reasons: 21.5% of our patients went for CABG who were likely placed in ICU for perioperative management, complicated PCI required higher level of perioperative care, and perhaps clinician preference for management of post-PCI patients. There was a significantly higher vasopressor support in the NSTEMI group. However, these data represent an order in the electronic medical records but does not reflect actual use of the medication. Thus, the differences between these groups may represent common electronic medical record order sets for CABG rather than requiring actual use of these medications.
The presence of PAD within the subset of EF<30% within both NSTEMI and Takotsubo groups was significantly different, with a higher frequency of PAD in the NSTEMI group. PAD disease is a known risk factor for CAD and perhaps the degree of CAD was worse in this subset of population; further analysis looking into degree of CAD on cardiac catheterization reports and future studies would be warranted to further explain this difference. None of our patients underwent biopsy or cardiac magnetic resonance imaging to rule out the presence of myocarditis. Acute myocarditis also exhibits some degree of cellular infiltration on histopathology and interstitial inflammation on T2 imaging in cardiac magnetic resonance imaging both of which mimic TK cardiomyopathy.11,12 There were no data collected on prior assessment of LVEF; thus, we cannot rule out patients with new onset cardiomyopathy in either NSTEMI or TK group. This may have caused us to underestimate the overall prevalence. Data on renal function including creatinine and glomerular filtration rate were not collected which can affect clearance and levels of BNP.
Conclusions
The presence of TK cardiomyopathy and PAD was independent predictors of patients who had LVEF of <35% and elevated BNP levels.
Footnotes
Acknowledgements
The authors acknowledge the support of Genesys Regional Medical Center’s departments of medical education and clinical research, specifically, Dr Heather Kirkpatrick, PhD; Dr Kimberly Barber, PhD; and Tara Knisely, MPH, in the preparation of this manuscript.
Peer Review:
Three peer reviewers contributed to the peer review report. Reviewers’ reports totaled 523 words, excluding any confidential comments to the academic editor.
Funding:
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests:
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Author Contribution
AH performed conception and research design, data analysis and data interpretation, wrote manuscript and acted as corresponding author. TV and SE supervised concept and development of work, helped in data interpretation, and manuscript evaluation. NE, KG, BM, and SR performed data collection and manuscript evaluation. All the authors reviewed and approved the final manuscript.