Ventricular arrhythmias not meeting criteria for terminating cardiopulmonary exercise testing stratify prognosis and disease severity in heart failure of preserved, midrange, and reduced ejection fraction

Abstract Background Continued high mortality in heart failure patients indicates the need for additional methods of risk stratification and phenotyping. Hypothesis We hypothesized that ventricular arrhythmias that do not meet test‐termination criteria (non‐terminating ventricular arrhythmias [NTVA]) during cardiopulmonary exercise testing (CPET) may help in phenotyping disease severity and prognosis in heart failure with reduced (HFrEF) and midrange (HFmrEF)/preserved (HFpEF) left ventricular ejection fraction (LVEF). Methods About 319 patients with heart failure (199 HFrEF; 80 HFmrEF; 41 HFpEF) underwent CPET. Tricuspid annular plane systolic excursion (TAPSE) and pulmonary artery systolic pressure (PASP) were measured by echocardiography. B‐type natriuretic peptide (BNP) at rest and peak exercise was also determined. The patients were tracked for primary (cardiac death) and secondary composite outcomes (all‐cause death, heart transplantation/left ventricular assist device implantation, hospitalization for cardiac reasons). Results Forty‐seven (15%) of the patients demonstrated NTVA during CPET, regardless of coronary artery disease prevalence. Patients without arrhythmias had a significantly higher LVEF (P < .05), TAPSE/PASP ratio (P < .001), peak oxygen consumption (P < .01), lower resting and peak BNP (P < .001), and the minute ventilation/carbon dioxide production slope (P < .001) compared to those with NTVA. Seventy‐one patients died during the tracking period, 54 for cardiac reasons. NTVA during CPET was a significant predictor of primary and secondary outcomes in the total heart failure cohort (HR: 5.3, 3.7; 95% CI: 3.1‐9.1, 2.4‐5.5; P < .001, respectively), as well as in subgroups categorized according to reduced and middle‐range/preserved LVEF (P < .001). Conclusion Exercise‐induced ventricular arrhythmias that do not reach test‐termination criteria are nonetheless indicative of an advanced disease severity phenotype and worse prognosis.

Conclusion: Exercise-induced ventricular arrhythmias that do not reach testtermination criteria are nonetheless indicative of an advanced disease severity phenotype and worse prognosis.

K E Y W O R D S
cardiopulmonary exercise testing, prognosis, HFpEF, HFmrEF, HFrEF

| INTRODUCTION
Ventricular arrhythmias (VA) may cause or be a consequence of heart failure (HF). They are common, increase in frequency according to disease severity and portends poor prognosis. 1,2 VA may be specifically associated with an ischemic etiology in HF 1,3 ; however randomized trials do not show a reduction in overall mortality by revascularization therapies. 4 There is evidence that more than 10 premature ventricular beats per hour and nonsustained ventricular tachycardia (NSVT) increase mortality risk in patients with structural heart disease, although providing little discrimination between sudden cardiac death or death due to progressive HF. 5 Some other reports demonstrate that in patients with HF and an ejection fraction (EF) below 35%, premature ventricular beats did not have prognostic value beyond other clinical variables. 6 Exercise may be associated with VA, indicating a higher risk of all-cause mortality, sudden cardiac death or acute coronary syndrome. [7][8][9] In HF patients, cardiopulmonary exercise testing (CPET) is a standard of measure, whose main derived variables also have prognostic independent information for risk of sustained VA in HF. 10,11 The incidence of exercise-induced VA in patients with HF is high, 7 with limited and mixed evidence of its prognostic value. Moreover, a number of HF patients may develop VA during CPET that do not meet testtermination criteria set by guidelines (ie, sustained ventricular tachycardia). 12 The data on specific prognostic value of exercise nonterminating ventricular arrhythmias (NTVA), such as ectopic beats or NSVT, are not conclusive suggesting or excluding an increased risk of death. [7][8][9] In one report of asymptomatic adults, exercise induced NSVT was reported in nearly 4% and no association with mortality was observed. 13 Data inconsistency may point toward a lack of prognostic value of NTVA in the general population, as well as for HF.
Given the important clinical implications and current lack of firm evidence in the area, the purpose of the current investigation was to define the prognostic significance of NTVA during CPET in a HF cohort across left ventricular ejection fraction category. When left ventricular ejection fraction (LVEF) was ≥50%, along with the additional proposed criteria, 1 patients were considered to have a HF with preserved EF (HFpEF); when EF was 40-49% they were classified as midrange EF (HFmrEF), and when EF was <40%, patients were classified as HF with reduced EF (HFrEF). We considered ischemic all patients with documented coronary artery disease (CAD; myocardial infarction, revascularization, ≥ 50% reduction in luminal diameter on coronary angiography). The study was approved by the local Ethical Institutional Review Board and informed consent was obtained from all subjects.

| Event tracking and endpoints
Subjects were followed for primary outcome (cardiac death) and secondary outcome of composite cardiac events (all cause death, heart transplantation and left ventricular assist device [LVAD] implantation, rehospitalization for cardiac reasons), via hospital and outpatient medical chart review for up to maximum 193 months. Subjects were followed by the HF program providing a high likelihood that all events were captured.
Cardiac death was considered to be death due to cardiac reasons, and hospitalization for cardiac reasons and admission to the heart failure unit.

| Echocardiography
A 2D and Doppler echocardiography was performed with a Hewlett-Packard 77 020/A (Andover, MA) and Philips IE33 devices (Andover) by two experienced cardiologists following current guidelines. 14 A prespecified protocol was used to optimize RV imaging. 15 The tricuspid annular plane systolic excursion (TAPSE) was obtained by M-mode analysis in the apical four-chamber view and was measured as the total displacement of the tricuspid annulus (millimeters) from end-diastole to end-systole, on an average of three to five beats. 15 Pulmonary artery systolic pressure (PASP) was estimated by Doppler echocardiography from the systolic right ventricular to right atrial pressure gradient using the modified Bernoulli equation. Right atrial pressure (assessed jugular venous pressure), estimated by size and respiratory variation of vena cava inferior, was added to the gradient to yield PASP. 14,15 TAPSE/PASP ratio, a measure of right ventricular-pulmonary vasculature (RV-PV) coupling, 16 was derived. In cases of HFpEF, care was taken to identify the proper etiology of coexistent pulmonary hypertension excluding idiopathic pulmonary arterial hypertension. Accordingly, we referred to Opotowsky et al 17

| Blood analysis
Blood for NT-pro-BNP analysis (20 mL) was taken at rest and at the peak effort, from intravenous cannula placed into the patient's brachial vein before the test. Samples were kept at −80 C and centrifuged on 4000 Hz. NT-pro-BNP was measured in all samples by immunoassay sandwich technique (pro-BNP II, Cobas, Roche, Burgess Hill, England) with lower sensitivity limit of 5 pg/mL.

| Exercise testing procedures
Symptom-limited CPET was performed on a bicycle ergometer for all subjects, according to established guidelines. 18 Pharmacologic therapy was maintained during CPET. Ventilatory expired gas analysis was performed using a Sensormedics metabolic cart (Vmax, Yorba Linda, CA).
Standard 12-lead electrocardiograms were obtained after adequate skin preparation, at rest, each minute during exercise, and for at least 5 minutes during the recovery phase. Heart rate (HR) was determined at rest, peak exercise and after 1 minute of recovery (HHR-1). Minute ventilation (VE, BTPS), oxygen uptake (VO 2 , STPD), and carbon dioxide output (VCO 2 , STPD) were acquired breath-by-breath and printed using rolling averages every 10 seconds. Peak VO 2 and peak respiratory exchange ratio (RER) were expressed as the highest 10-second averaged sample obtained during the last 20 seconds of testing. VE and VCO 2 values, acquired from the initiation of exercise to peak, were input into spreadsheet software (Microsoft Excel, Microsoft Corp., Bellevue, WA) to calculate the VE/VCO 2 slope via least squares linear regression (y = mx + b, m = slope). Exercise oscillatory ventilation during CPET was defined as previously described in detail. 18 Test termination criteria consisted of symptoms (ie, dyspnea and/or fatigue), sustained ventricular tachycardia (VT) and NSVT that interfered with hemodynamic stability, > 2 mm of horizontal or downsloping ST segment depression, or a drop of systolic blood pressure > 20 mmHg during progressive exercise. VA other than sustained VT, including unifocal or multifocal ectopy, NSVT without hemodynamic stability, ventricular triplets and couplets were considered as nonterminating. Arrhythmias were tracked actively during the testing and registered by ECG tracings.
All subjects were also evaluated by performing the 6 minutes walk test (6MWT) as a measure of submaximal exercise performance.

| Statistical analysis
The results are expressed by classic descriptive parameters-mean and SD for parametric variables and median for variables that were not normally distributed. In order to apply parametric statistics, analysis of distribution was performed by the Kolmogorov-Smirnov test, followed by power transformation of not normally distributed data.
Categorical data are expressed as percentages. The unpaired t test was used to assess differences in key continuous variables between subjects who did and did not demonstrate NTVA during the test. The chi-square test assessed differences in categorical data between these subgroups. Univariate and multivariate Cox regression analysis was used to assess the prognostic value of key CPET and Echo measures.

| RESULTS
Of 319 subjects with HF enrolled, mean age 63.0 ± 9.9 years, 78% were male. LVEF of studied population was 36.0 ± 11.1% and 62% of F I G U R E 1 Kaplan-Meier analysis of NTVA appearance during CPET in distinguishing between HF patients with and without primary ( Figure 1A) and secondary outcome ( Figure 1B) Figure 1A F I G U R E 2 Kaplan-Meier analysis of NTVA appearance during CPET in distinguishing between patients with and without primary outcome in HFrEF during 25.9 ± 24.3 months ( Figure 2A) and HFmrEF/HFpEF during 25.6 ± 29.7 months follow-up period ( Figure 2B). CPET, cardiopulmonary exercise testing; HF, heart failure; HFrEF, HF with reduced EF; HFpEF, HF with preserved EF; HFmrEF, HF midrange EF; NTVA, nonterminating ventricular arrhythmias F I G U R E 3 Kaplan-Meier analysis of NTVA appearance during CPET in distinguishing between patients with and without secondary outcome in HFrEF during 25.9 ± 24.3 months ( Figure 3A) and HFmrEF/HFpEF during 25.6 ± 29.7 months follow-up period ( Figure 3B)

| Pathophysiological insights into VA appearance during exercise in HF
It is well established that the VA rate in HF patients is associated with electrical conduction heterogeneity secondary to myocardial scar, ischemia or QT dispersion due to certain drugs, myocardial late potentials or re-entry phenomena, and sympathetic system activation. 19 Exercise may suppress cardiac arrhythmias detectable at rest by an overdrive suppression of the ectopic Purkinje pacemaker activity through sinus tachycardia favored by increased sympathetic tone and vagal withdrawal. 19 On the other hand, increased sympathetic tone during exercise may induce ectopic impulse formation in the Purkinje tissue by increasing automaticity due to acceleration of the phase 4 of the action potential, provoking spontaneous discharge. 19 Accordingly, in the present study, patients with NTVA achieved higher RER, suggesting the relation of higher metabolic achievements and VA expression. Ectopic ventricular beats are the most common cardiac arrhythmia during exercise, and they are usually associated with cardiac abnormalities, older age and obesity. 7,19 Furthermore, exerciserelated types of VA include catecholamine-triggered polymorphic VT, but also right ventricular outflow tract VT associated usually with arrhythmogenic right ventricular dysplasia, suggesting the importance of the right ventricular pathology in generating VA. 19 Actually, we do not have specific elements to dissect, which trigger mechanism may be predominant in our population. Nonetheless, no association was found between NTVA with older age, obesity, or increased CAD prevalence, suggesting that HF itself promotes some hemodynamic and metabolic derangements responsible for a lower oxygenation of myocardial cells. One explanation of this phenomenon may be the drop in cardiac output in association with the high catecholamine levels, followed by generalized vasodilation in exercising muscles, further affecting cardiac output. 1,13,19,20 This condition may lead to a reduction in coronary perfusion while HR is still elevated.
NTVA during CPET was associated with lower EF and RV to PC coupling, which is indicative for lower cardiac output during CPET.
Accordingly, patients with NTVA demonstrated increased plasma BNP at rest and peak exercise which reflect higher myocardial stress and consequently lower contractility. 1 However, in the group of patients with HFmrEF and HFpEF NTVA appearance during CPET was not related to a lower LV EF suggesting other implicated mechanisms.
A role would have been played by chronotropic incompetence, 21 leading to insufficient cardiac output increase during exercise. In accordance with literature data, a lower HR at peak exercise in NTVA patients was observed regardless of beta receptor blockade. HRR-1, a prognostic indicator in HF, was significantly lower as well. Overall, the constellation of these factors 21-23 reflects the strong role of autonomic dysregulation which may be the basis of NTVA appearance, as well.
The fact that in patients with NTVA during CPET pulmonary vascular pressure was increased, systolic function of the right ventricle diminished, and the TAPSE/PASP ratio, a measure of RV-PV uncoupling, decreased, may point toward a role of the right heart and pulmonary vasculature activity in generating arrhythmias. 24 It seems reasonable to speculate that right ventricle and left ventricular cardiac output decrease, with a likely anticipation for the right ventricle.
Namely, in the condition of right ventricular failure and increased PASP, the blood flow from the right heart to the left heart is usurped, leading to a reduction in cardiac output during exercise that could not be explained by the HR response per se.
It was shown previously that patients prescribed with nitrates show VA less likely. 5,7,25 It is also known that exercise can induce car-  [7][8][9] Some studies have shown that frequent or complex repetitive ventricular activity during exercise, and especially ventricular ectopy in the recovery period after exercise, heralds increased risk of death. 8,9,28 Furthermore, exerciseinduced VA is an independent predictor of cardiovascular mortality and, in combination with resting premature ventricular contractions, carries the highest risk. 9 Moreover, an origin of the ventricular ectopy was suggested as important in a prognostic sense, indicating that ectopy with a right bundle-branch block morphology, common in patients with LV dysfunction, more likely predict adverse events than ectopy originating from the right ventricular outflow tract. 29 The present study proposes the existence of a potential link between VA appearance and right heart function, both significantly determining prognosis, which is in accordance to previous studies which already demonstrated that right heart function is a crucial determinant of outcome in HF patients regardless of LV function or predominance of systolic or diastolic HF. 16 It seems that the worse outcome in patients with HF is diminished right-sided cardiac function and increased pulmonary pressure, with a real-time decrease in left-sided cardiac output, followed by arrhythmogenic presentation.
The finding that NTVA appearance during CPET demonstrated strong predictive value for cardiac events in the subgroups of patients with HFrEF, HFmrEF and HFpEF has to be stressed taking into consideration the poor availability of prognostic criteria for HFmrEF. 1  Another limitation of this study is the lack of specific definition of pathophysiological mechanisms leading to NTVA and no in-depth evaluation of structural cardiac muscle changes; however, we used strict echocardiographic protocol for noninvasive assessment rejecting data without good quality to minimize potential errors. Additionally, although NTVA during exercise can be easily registered and interpreted, technical improvement of ECG recordings is warranted. In our study, we used data derived from CPET, a comprehensive method for evaluation of HF patients, however, its broad applicability is lacking.
Nonetheless, it is reasonable to speculate that our results may be transferable to standard exercise testing procedure and together with 24 hours Holter monitoring may further contribute to more efficient risk stratification in HF patients.
In conclusion, exercise-induced arrhythmias not reaching criteria for test termination seem nonetheless indicative of an advanced HF severity phenotype, and worse prognosis, independently of HF subtype. Marked abnormalities in CPET-derived variables drive the outcome prediction. Among the many potential causes, our data suggest a role of unfavorable combination of RV-PV uncoupling and autonomic dysregulation leading to reduced cardiac output during exercise.

| Clinical perspectives: Competencies in medical knowledge and translational outlook
In HF of any left ventricular ejection fraction, mortality remains high and multiparametric assessment of risk is of basic importance. The