Pulmonary Embolism in Patients Admitted With Takotsubo Cardiomyopathy: Prevalence and Associated In-Hospital Adverse Events

Introduction Takotsubo cardiomyopathy (TCM) is a poorly understood condition typically characterized by abnormal left ventricular wall motion without causative coronary artery disease and typically presents in post-menopausal women after the experience of a physical or emotional stressor. The pathophysiology of TCM is complex and multifactorial, resulting in complications with varied severity; one of the most concerning complications is thromboembolism, specifically, pulmonary embolism (PE), which is understudied in its relation to TCM. The purpose of this study was to characterize and evaluate the real-world prevalence and outcomes of PE in TCM. Methods Data were derived from the National Inpatient Sample database from January 2016 to December 2019. The primary outcomes assessed were baseline and hospital admission characteristics and comorbidities for patients with TCM with and without PE. Outcomes for TCM patients with PE and predictors of mortality in TCM were also analyzed. Results PE developed in 788 of 40,120 patients with TCM (1.96%). After multivariate adjustment, PE was found to be independently associated with intracardiac thrombus (adjusted odds ratio (aOR) 2.067; 95% confidence interval (CI): 1.198-3.566; p = 0.009) and right heart catheterization (RHC) (aOR: 1.971; 95% CI: 1.160-3.350; p = 0.012). Mortality in patients with TCM was associated with, among other factors, age in years at admission (aOR: 1.104; 95% CI: 1.010-1.017; p = 0.001), African American race (aOR: 1.191; 95% CI: 1.020-1.391; p = 0.027), Asian or Pacific Islander race (aOR: 1.637; 95% CI: 1.283-2.090; p = 0.001), coagulopathy (aOR: 3.393; 95% CI: 2.889-2.986; p = 0.001), liver disease (aOR: 1.446; 95% CI: 1.147-1.824; p = 0.002), atrial fibrillation (aOR: 1.460; 95% CI: 1.320-1.615; p = 0.001), and pulmonary embolism (aOR: 2.217; 95% CI: 1.781-2.760; p = 0.001). Conclusion In a large cohort of patients admitted with TCM, we found the prevalence of PE to be 1.96%. PE, along with comorbidities such as coagulopathy and atrial fibrillation, was found to be a significant predictor of mortality in this patient cohort.


Introduction
Takotsubo cardiomyopathy (TCM) is an understudied syndrome that is typically characterized by abnormal left ventricular wall motion without evidence of attributable coronary artery disease (CAD) [1].The most commonly described movement is apical ballooning, but other anatomic patterns have also been observed which distinguish subtypes of TCM [1].Of all patients who present with acute coronary syndrome (ACS), TCM accounts for 1-4% [1].Nearly 90% of patients diagnosed with TCM are post-menopausal women, and the incidence is highest in Asian and Caucasian populations [1].Patients classically present with symptoms concerning for ACS, including chest pain, dyspnea, syncope, and palpitations [1].International diagnostic criteria have been developed to identify TCM cases and distinguish them from acute myocardial infarction, as they both present with ECG abnormalities like ST-segment elevation and/or pronounced T-wave inversions [2,3].Left ventriculography during catheterization is frequently used to further confirm left ventricular wall motion abnormalities that extend beyond the domain of a single epicardial artery distribution [4].However, it is important to note that while TCM generally presents in the absence of CAD, patients may also have concomitant CAD, which complicates its diagnosis [1].
Interestingly, the pathophysiology of TCM is disputed [1].The onset of TCM typically follows the experience of a physical or emotional stressor, suggesting sympathetic nervous system overstimulation and catecholamine release.While some studies have shown higher levels of circulating catecholamines in the subacute phase of TCM compared to myocardial infarction, other studies have failed to demonstrate higher circulating levels during the acute phase [1].Alternatively, adrenergic receptor function has been postulated to contribute to TCM pathology via a protective switch to β2 receptor signaling in the presence of elevated epinephrine levels [1].Other proposed mechanisms include endothelial dysfunction, microvascular vasospasm, hormone-mediated changes, genetic predispositions, and pro-inflammatory states [1,3,5].Angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs) have been shown to improve survival, whereas β-blockers have not been shown to benefit recovery despite frequent use [1].Anticoagulation can be prescribed for patients at high risk for thromboembolism, and antiplatelet agents such as aspirin can be used, though there is no evidence to suggest cardiovascular benefit [1].Importantly, no randomized clinical trials have been conducted to date to support treatment recommendations for TCM [5].
In most patients, TCM is transient, and left ventricular ejection fraction (LVEF) is recovered spontaneously [1].One of the most serious complications of TCM, however, is thromboembolism, which has been reported in 2-14% of patients with TCM [6].Thromboembolic events occur most commonly in patients with low LVEF (<30%) and detectable apical ballooning and are treated like other thromboembolic diseases with unfractionated heparin, low molecular weight heparin, and vitamin K epoxide reductase antagonists [7].
Pulmonary embolism (PE), a lethal manifestation of post-TCM thromboembolism, is underreported in the literature [8,9].The first reported case of concomitant TCM and PE was in the setting of COVID-19, which causes a hypercoagulable state and can present with PE [8].Similarly, the first case of TCM in the setting of PE was reported in a postmenopausal woman with thrombotic disease [9].Apart from these stand-alone cases, little has been published describing the association between PE and TCM.Thus, this article aims to explore the prevalence, predictors, and in-hospital incidence of PE in patients with TCM.

Data acquisition
This is a retrospective study that obtained data from The National Inpatient Sample (NIS) database.The NIS is part of the Healthcare Cost and Utilization Project (HCUP) set forth by the Agency for Healthcare Research and Quality.It utilizes the International Classification of Disease, Tenth Edition, Clinical Modification (ICD-10-CM) codes for diagnosis and procedures.The data set was utilized to examine patients admitted between 2016 and 2019.Encounters with a primary diagnosis of TCM were selected using ICD-10 code I51.81.This cohort of patients was further divided into patients who developed PE versus patients without PE.Adult patients ≥ 18 years old were included.We abstracted data from 40,144 charts, excluded 24, and were left with 40120 charts for analysis.IRB approval was not required as the NIS database provides de-identified patient information.

Outcomes and variables
Patient baseline characteristics such as age, sex, race, and insurance status were extracted.Comorbidities, hospital complications, mortality rates, disposition status, length of stay, and total charges were also analyzed.
The primary aim of the study was to assess whether or not there is a difference in outcomes (mortality, inhospital complications, length of stay, total charges) between the cohort of patients with TCM and PE vs. patients with TCM and without PE.We also analyzed the independent association of PE with outcomes after controlling for confounders such as age, race, sex, and comorbidities.

Statistical analysis
Categorical values were analyzed via Pearson's Chi-square analysis and continuous variables were analyzed via independent Student's t-test.Logistic regression was performed to generate odds ratios (ORs) with 95% confidence intervals (CIs) to assess predictors of mortality in patients with TCM.We also used logistic regression to assess the independent association of PE with clinical outcomes after controlling for confounders like age, sex, race, and comorbidities.A p-value of <0.05 was considered statistically significant.All analyses were completed using IBM SPSS Statistics for Windows, Version 29.0 (Released 2023; IBM Corp., Armonk, New York, United States).

Discussion
In the 40,120 patients with TCM studied, the prevalence of PE was 1.96%.Patients with TCM presenting with PE had longer hospital stays, higher total charges, and higher mortality rates.In the cohort of TCM patients studied, a higher incidence of PE was found in males.Previous literature notes that despite higher rates of TCM in females, males are more likely to have adverse clinical outcomes when presenting with TCM [10].A variety of findings including an exaggerated inflammatory response, relatively lower levels of estrogen, which normally provides a cardioprotective effect, and a difference in the type of stressors in males compared to females have been suggested as possible explanations, although the underlying mechanisms for the difference in the severity of clinical outcomes remain to be explored [11,12].
Additionally, in the cohort studied, African American and Hispanic patients were more likely to present with PE.Previous studies have shown a greater prevalence of PE in African American patients without TCM in comparison to White patients [13,14], although the same has not been found for Hispanic patients [13,15].Currently, evidence suggests that racial disparities in venous thromboembolism between African American and White patients are due to greater risk factors such as higher body weight and factor VIII concentrations, as well as lower family income [16].However, limited literature is published regarding Hispanic patients and TCM, and further research is needed to understand the higher incidence of PE in Hispanic patients with TCM.
Patients with PE and TCM were found to have a higher burden of iron-deficiency anemia.Recent case reports have noted an association between iron-deficiency anemia and thromboembolism [17,18].While the relationship between iron-deficiency anemia and thromboembolism is not completely understood, the most probable mechanism is that iron-deficiency anemia results in thrombocytosis via decreased inhibition of thrombopoiesis, which leads to an increased risk of thromboembolism [19].This association may also explain why TCM patients with iron-deficiency anemia have lower odds of mortality; the easily treatable nature of iron-deficiency anemia facilitates the decrease of risk factors for coagulopathy and thus mortality [18].
We also observed that smoking, obesity, coronary artery disease, hypertension, and hyperthyroidism decreased the odds of mortality in TCM patients.It seems contradictory that conditions such as smoking and obesity, which are known to contribute to the pathogenesis of cardiovascular disease, were shown to decrease the odds of mortality.The potential advantage of excess adipose storage during illness is an oftencited explanation [20].However, a recent investigation found that the "obesity paradox" is a product of reverse causation and confounding by smoking; since individuals with obesity are less likely to smoke than individuals of normal weight and are selected into a disease group for which both smoking and obesity are risk factors, selection bias and downward bias may occur, leading to statistical error [20].
The contributions of hypertension and coronary artery disease to decreased mortality in TCM patients are also perplexing.To our knowledge, there is no literature suggesting that hypertension or coronary artery disease decreases mortality for TCM.It is possible that therapies targeting pre-existing hypertension and coronary artery disease in TCM patients improve outcomes in TCM patients, though further studies should be conducted to evaluate this relationship further.
Finally, we found that hyperthyroidism decreases the odds of mortality in TCM patients.A recent metaanalysis found that both overt and subclinical hyperthyroidism are associated with an increased risk of cardiovascular events and cardiovascular mortality [21].Common cardiovascular symptoms in hyperthyroid patients include palpitation, atrial fibrillation, and various arrhythmias [22].Perhaps because of, not despite, this close relationship with cardiovascular function, the management of hyperthyroidism via diuretics and β-blockers may decrease the risk for mortality in TCM patients.Most importantly, hyperthyroidism increases the density of β1-adrenergic receptors on the myocardium [22], which increases sensitivity to circulating catecholamines and counteracts the upregulation of β2 receptors thought to contribute to the pathogenesis of TCM [1,23].
Limitations of our study include that it is observational, therefore our results merely reflect correlation, and caution must be exercised when drawing conclusions.Because the NIS is an administrative database that uses ICD-10 codes, it is prone to human errors in coding.Other confounders may be present that are not controlled for in our study, including medication use that is not provided by the NIS.Results from our study can only be interpreted in the inpatient setting and cannot extend to outpatient practice.Patients are also not followed after discharge; therefore, long-term outcomes for patients cannot be established.We did not study the treatment modality of PE in our patients; therefore, outcomes may differ depending on which treatment (anticoagulation vs. lysis vs. endovascular therapy) the patient received.Further studies could prospectively study the impact of the different modalities of PE treatment on outcomes among patients hospitalized with TCM.

Conclusions
In this large real-world registry of patients with TCM, we found the prevalence of PE in patients with TCM to be 1.96%.PE in patients with TCM was found to be independently associated with increased outcomes of intracardiac thrombus and right heart catheterization.Mortality in patients with TCM was associated with age in years at admission, race, and various cardiac and pulmonary complications including PE.

TABLE 3 : Outcomes for the study population of TCM patients with and without PE
The data has been represented as n (%).p-values are significant at <0.05.PE: pulmonary embolism; PPM: permanent pacemaker; VF: ventricular fibrillation; VT: ventricular tachycardia; RHC: right heart catheterization; LHC: left heart catheterization; IABP: intra-aortic balloon pump; STEMI: ST-segment elevation myocardial infarction; NSTEMI: non-ST elevation myocardial infarction; TCM: takotsubo cardiomyopathyWe used multivariate logistic regression to assess the independent association of pulmonary embolism with outcomes in TCM patients, controlling for confounding variables such as sex, race, age, and comorbidities.