Does Resting Cardiac Power Index Affect Survival Post Transcatheter Aortic Valve Replacement?

Abstract: Objective. Cardiac power index (CPI) is an integrative hemodynamic measure of cardiac pumping capability and is the product of the simultaneously measured mean arterial pressure and the cardiac output. We assessed the association between baseline resting CPI and survival post transcatheter aortic valve replacement (TAVR). Methods and Results. We retrospectively abstracted data of patients who underwent TAVR at the Mayo Clinic Foundation with follow-up data available at 1 year. Baseline demographic, clinical, and echocardiographic data were abstracted. CPI was calculated using the formula, (cardiac output x mean arterial blood pressure) / (451 x body surface area) W/m². Patients were divided into CPI <0.48 W/m² (group 1) and CPI ≥0.48 W/m² (group 2). Survival according to CPI was determined using Kaplan-Meier method. Multivariate Cox regression analysis was performed to adjust for covariates. Nine hundred and seventy-five patients were included in the final analysis. CPI in group 1 vs group 2 was 0.41 ± 0.05 W/m² vs 0.66 ± 0.14 W/m², respectively (P<.001, two-sided t-test). Patients in group 1 were more likely to be male and to have a prior history of myocardial infarction, coronary revascularization, peripheral arterial disease, diabetes mellitus, transient ischemic attack, carotid artery disease, atrial fibrillation, lower left ventricular ejection fraction, and moderate to severe mitral and tricuspid regurgitation. After adjusting for baseline covariates, a lower CPI was associated with higher 1-year mortality among patients undergoing TAVR (24.39% in group 1 vs 8.28% in group 2; P<.001). Conclusion. Low baseline CPI (<0.48 W/m²) confers higher mortality risk among patients undergoing TAVR and provides additional prognostic information, which can help risk-stratify patients.  

J INVASIVE CARDIOL 2020;32(4):129-137. Epub Ahead of Issue 2020 March 20.

Key words: aortic valve replacement, cardiac power, mortality, risk stratification

Cardiac power (CP), the product of the simultaneously measured mean arterial pressure and the cardiac output, is an integrative hemodynamic measure of cardiac pump function. By combining both flow and pressure domains in the left ventricle (LV), CP provides an accurate estimate of the pumping capacity of the LV.¹ CP reflects the rate of energy expenditure by the LV to circulate the blood in the systemic vasculature. Multiple studies have shown CP to be a powerful predictor of mortality in patients with cardiogenic shock and chronic heart failure.^2-6 Transcatheter aortic valve replacement (TAVR) is currently the treatment of choice for severe aortic stenosis in patients with prohibitive and high surgical risk,⁷ and is increasingly performed for intermediate surgical risk patients. The Transcatheter Valve Therapy (TVT) registry has reported a mortality rate of 23.7% at 1 year following TAVR.⁸

Risk prediction is an important part of patient selection criteria. There is a lack of literature describing the impact of resting cardiac power on long-term mortality in patients undergoing TAVR. In this study, we therefore aim to evaluate the relationship between baseline resting cardiac power indexed (CPI) to the body surface area and 1-year mortality in patients with severe aortic stenosis who underwent TAVR. 

Methods

Study population. We performed a multicenter, retrospective cohort study on consecutive patients with severe symptomatic aortic stenosis undergoing TAVR between January 1, 2012 and June 30, 2017 at Mayo Clinic hospitals located in Rochester, Minnesota, Phoenix, Arizona, and Jacksonville, Florida. We assessed all-cause mortality at 1 year post TAVR. The study protocol was approved by the Mayo Clinic institutional review board and all patients included in the study provided research authorization to utilize their medical information for clinical research. Inclusion criteria were: (1) patients age ≥18 years; (2) symptomatic severe aortic stenosis; (3) TAVR with either balloon-expandable or self-expanding aortic valve prostheses; (4) transthoracic echocardiogram (TTE) performed within 2 months of TAVR; and (5) follow-up data available at 1 year. Exclusion criteria were: (1) patients lost to follow-up before 1 year from date of TAVR; and (2) patients with moderate-to-severe aortic regurgitation. Baseline data were collected by retrospective chart review. Follow-up data were obtained by Valve Clinic coordinators at the time of follow-up clinic visits, contacting the patient/family by phone or via a letter from the patient’s medical provider at an outside facility. In this analysis, a total of 1070 patients underwent TAVR from January 2012 to June 2017. Among the 1070 patients, ninety-five had to be excluded (37 patients without follow-up at 1 year, 20 patients without TTE performed within 2 months of the index procedure, and 38 patients who refused to provide research authorization); the remaining 975 patients were included in the final analysis (Figure 1). The primary outcome of the study was 1-year all-cause mortality. Patients lost to follow-up were censored and eliminated from our analysis.

Informed, written consent was obtained from all individuals for whom information is included in this article. The authors have conformed to institutional guidelines and those of the American Physiological Society.

Baseline clinical variables. We obtained preprocedural variables including age, sex, race, history of diabetes mellitus (DM), hypertension (HTN), smoking, myocardial infarction, percutaneous coronary intervention (PCI), coronary artery bypass graft (CABG) surgery, stroke, transient ischemic attack (TIA), infective endocarditis, permanent pacemaker implantation, implantable cardioverter device implantation, carotid artery stenosis, carotid artery stenting/endarterectomy, dialysis, atrial fibrillation/flutter, heart failure within 2 weeks, New York Heart Association (NYHA) class within 2 weeks, cardiogenic shock within 24 hours, cardiac arrest within 24 hours, peripheral arterial disease (PAD), chronic obstructive pulmonary disease (COPD), home oxygen use, left main stenosis ≥50%, porcelain aorta, vascular access site, type of TAVR valve, serum albumin level, elective vs urgent procedure, type of anesthesia, presence of mechanical assist device prior to initiation of TAVR, and Society of Thoracic Surgeons-Predicted Risk of Operative Mortality (STS-PROM) score by chart review using standardized definitions developed by the TVT registry.⁹ Blood pressure (BP) was measured using a manual sphygmomanometer at the time of TTE, as per guidelines.¹⁰ Mean arterial blood pressure (MAP) was calculated using the formula: (systolic BP + [2 x diastolic BP])/3 mm Hg.

Echocardiography. Comprehensive Doppler and two-dimensional (2D) TTE were performed before TAVR using the standard ultrasound scanners (Philips iE33 [Phillips Medical Systems]; GE Vivid E9 [GE Healthcare]). The TTE images were performed and interpreted according to the guidelines set by the European Association of Echocardiography and American Society of Echocardiography by experienced cardiologists.^11-13 Offline measurements of TTE images were performed using ProSolv Cardiovascular Analyzer 3.0 (ProSolv Cardiovascular, Inc). Only patients with TTE performed 2 months before TAVR were included in our study.

Echocardiographic variables. The TTE variables collected for the current study include left ventricular ejection fraction (LVEF), severity of mitral regurgitation (MR), severity of tricuspid regurgitation (TR), aortic valve area (AVA), aortic valve systolic mean gradient (AVSMG), right ventricular systolic pressure (RVSP), and cardiac output.

The AVA was calculated using the continuity equation and also indexed to body surface area (AVAI). The AVSMG was estimated using the modified Bernoulli equation. Severe aortic stenosis was defined as AVAI ≤0.6 cm² or AVA ≤1 cm² and AVSMG ≥40 mm Hg or a peak velocity ≥4.0 m/s.^7,12 The LVEF was derived using Simpson’s biplane method of disks.¹³ Mild, moderate, and severe regurgitation across mitral, tricuspid, and aortic valves was assessed using standard diagnostic criteria.¹¹ Patients with moderate or severe aortic valve regurgitation were excluded from our study. The RVSP was derived using a simplified Bernoulli equation: 4(V²) + right atrial pressure, where V is the peak velocity of the TR jet. Right atrial pressure was estimated from the diameter of inferior vena cava and its respirophasic changes.¹⁴ Left ventricular outflow tract (LVOT) diameter was measured in systole and long parasternal view between the bases of aortic valve cusps. Cardiac output was measured using the formula: (LVOT diameter² x 0.785 x LVOT velocity time integral) in L/min.¹⁵

Cardiac power was derived as (MAP x cardiac output)/ (451 x BSA) Watts² and indexed to BSA (CPI in W/m²) using the baseline pre-TAVR echocardiogram. For the measurement, we used the blood pressure and BSA obtained during TTE.

Statistical analysis. Descriptive statistics were summarized as number and percentage for categorical variables, and mean ± standard deviation for continuous variables. The Chi-square test or Fisher’s exact test was used to compare baseline characteristics between both groups for categorical variables, as appropriate. The t-test was used for continuous variables. Univariate logistic regression analysis was performed to assess the relationship between baseline variables and all-cause mortality at 1 year. Patient survival score was calculated with step up to identify the cut-off value for CPI, where the maximum difference in mortality occurred.¹⁶ The cohort was subsequently divided into group 1 (CPI <0.48 W/m²) and group 2 (CPI ≥0.48 W/m²). Kaplan-Meier methods were used to calculate event rates and multivariate Cox regression was applied for survival analysis to adjust for the baseline covariates. A subgroup survival analysis was performed after excluding patients with moderate-to-severe MR. A two-sided P-value of <.05 was defined as statistically significant. We used R 3.4.1 software (Foundation for Statistical Computing) to run our analysis. 

To investigate the independence of CPI and other covariates, both unsupervised and supervised methods were used. We first investigated the independence between CPI and covariates without considering the mortality. We then calculated the Spearman’s correlation coefficient,¹⁷ which measures the non-parametric rank correlation, between CPI and other continuous covariates (age, STS risk score, total albumin, RVSP, and AVSMG). 

Results

The mean age of patients in the cohort was 81 ± 8.6 years and 59.4% of study participants were men. Mean follow-up time for the cohort was 378 days. CPI itself shows a strong relationship with patient mortality. Using raw CPI values to predict 1-year mortality achieves an area under the curve (AUC) of 0.67 (Figure 2). The CPI in patients who were deceased was significantly lower than patients who were alive at 1 year (0.51 ± 0.15 W/m² vs 0.6 ± 0.16 W/m², respectively; P<.001). Remaining baseline characteristics for the study cohort are described in detail in Table 1. At 1 year after the index procedure, a total of 135 patients (13.84%) were deceased. Patients in the deceased group at baseline had a higher prevalence of moderate-to-severe chronic lung disease, atrial fibrillation, heart failure within 2 weeks of index procedure, higher STS-PROM score, lower serum albumin level, higher prevalence of moderate-to-severe TR, elevated RVSP, and higher prevalence of patients on dialysis. The remaining baseline covariates were similar between both the alive and deceased groups.

Survival score analysis with step up in CPI showed that a cut-off of 0.48 W/m² produced the maximum distance between the survival curves. Subsequently, the study cohort was divided into group 1 (CPI <0.48 W/m²) and group 2 (CPI ≥0.48 W/m²); unadjusted Kaplan-Meier survival curves are shown in Figure 3 (P<.001). Group 1 had a higher proportion of men, and patients more often had a history of myocardial infarction, CABG surgery, DM, permanent pacemaker implantation, TIA, carotid artery stenosis, atrial fibrillation/flutter, proximal left anterior descending artery stenosis ≥70%, STS-PROM score, lower left ventricular ejection fraction, smaller aortic valve area, higher prevalence of moderate-to-severe MR and TR, and higher RVSP compared with patients in group 2 (Table 1). After adjusting for baseline covariates, a lower CPI was associated with higher 1-year mortality among patients undergoing TAVR (24.39% in group 1 vs 8.28% in group 2; P<.001) (Figure 4). 

The results of univariate logistic regression analysis of individual covariates have been described in detail in Table 2. Among patients who underwent TAVR, presence of moderate-to-severe chronic lung disease, requirement of home oxygen, presence of atrial fibrillation, heart failure within 2 weeks of index procedure, higher STS-PROM score, lower serum albumin level, emergent TAVR, moderate-to-severe TR, elevated RVSP, lower CPI, and requirement of dialysis conferred an elevated risk of all-cause mortality at 1 year. Predicted 1-year mortality as a function of CPI with 97.5 confidence bands is displayed in Figure 5.

Multivariate Cox regression analysis was performed, adjusting for baseline covariates including age, gender, race, history of PCI, stroke, CABG, PAD, smoking, HTN, DM, hemodialysis, home oxygen use, atrial fibrillation/flutter, STS-PROM score, serum albumin, MR, TR, RVSP, and AVSMG to determine the independent association of CPI with 1-year mortality post TAVR. CPI <0.48 was shown to be independently associated with elevated 1-year mortality (hazard ratio [HR], 0.033; 95% confidence interval [CI], 0.008-0.142; P<.001) (Figure 6). To exclude the potential impact of MR on underestimating the CPI, which could result in decreased effective cardiac output, a subgroup analysis was performed after excluding patients with moderate-to-severe MR and the association still remained statistically significant (HR, 0.005; 95% CI, 0.001-0.036; P<.001) (Figure 7 and Table 3).

We calculated the Spearman’s correlation coefficient,¹⁷ which measures the non-parametric rank correlation between CPI and other continuous covariates (age, STS risk score, total albumin, RVSP, and AVSMG). The Spearman’s correlation coefficients between CPI and listed covariates are -0.06, -0.13, 0.01, -0.07, and 0.17, respectively. None of them showed a high correlation with CPI. We then utilized supervised methods, logistic regression, and random forest models to examine the variable importance using all variables to predict 1-year mortality (Figure 8). Both methods showed that CPI is the most important variable to predict patient mortality.

Discussion

In this retrospective study of 975 patients who underwent TAVR at Mayo Clinic, lower resting CPI (<0.48 W/m²) at baseline was associated with a significant increase in all-cause mortality at 1 year independent of other covariates. The association remained statistically significant in the subgroup analysis of patients without clinically significant MR. This is the first study to test the utility of CPI for prognostication in patients undergoing TAVR. The prognostic value of resting CPI remained significant after multivariate adjustment of patient demographics, comorbidities, laboratory, preprocedural and echocardiographic characteristics. This highlights the potential benefit of resting CPI as an important metric for risk stratification in a patient undergoing TAVR evaluation. CPI is a measure of the rate of energy input transmitted to the systemic vasculature from the heart, by incorporating the flow and blood pressure domains of the cardiovascular system.² CPI has been shown to predict outcomes in patients with cardiogenic shock, advanced heart failure, critical cardiac illness including cardiac arrhythmias, acute coronary syndromes, and post percutaneous coronary interventions.^5,18-20 Systemic arterial pressure and cardiac output are both measures of cardiac function; however, their values are not always concordant. For instance, a distributive shock is characterized by high cardiac output and low systemic blood pressure, whereas cardiogenic shock has low systemic blood pressure and cardiac output. Therefore, integrating both measures may provide a more accurate assessment of the patient’s physiologic status, especially since it combines separate hemodynamic axes (volume and flow) into a single entity that could potentially be targeted.

The 1-year mortality rate of our study cohort was 13.84%, which is significantly lower than the national average of 23.7% as reported by Holmes et al.²¹ The mean STS-PROM score for the entire cohort was 8.24%, as compared with the 7.1% rate reported by the STS/TVT registry. The baseline covariates associated with 1-year mortality in our study, including chronic lung disease, requirement of home oxygen, presence of atrial fibrillation, heart failure within 2 weeks of index procedure, higher STS-PROM score, lower serum albumin level, emergent TAVR, moderate-to-severe TR, elevated RVSP, lower CPI, and requirement of dialysis are consistent with results reported by STS/TVT registry.²¹ 

The cardiac output measured using echocardiography is not a reflection of the true cardiac output of the LV in patients with clinically significant MR, as part of the stroke volume is pumped back into the left atrium. In a subgroup analysis excluding patients who had clinically significant MR, low CPI remained statistically significant, further supporting CPI as a reliable hemodynamic parameter to predict prognosis in patients undergoing TAVR.

The attrition rate of cardiac myocytes is approximately 1/3 moving from early childhood to old age in the absence of major cardiac insults. The attrition leads to a decrease in cardiac functional reserve, aerobic capacity, and CPI. Patients with a greater cardiac reserve are less likely to deteriorate in comparison to patients with low cardiac reserve. Therefore, CPI provides useful information for prognostication in patients with advanced heart failure.²² In our study cohort, approximately 80% of patients had worsening heart failure 2 weeks prior to the index procedure. Since aortic stenosis and heart failure have a complex interplay with pathogenesis, clinical progression, and prognosis, CPI could help guide clinical management and risk stratify patients. A subsequent study will be performed to test the roles of resting CPI, peak stress CPI, and ΔCPI, especially in patients with low-flow low-gradient aortic stenosis, to assess the relationship with short-term and long-term mortality after undergoing TAVR.

Study limitations. This study has several limitations. First, it is a retrospective analysis, which is subject to inherent limitations such as study design, selection bias, etc. Second, cardiac output was measured by TTE, which is affected by multiple factors including preload, technique of image acquisition, presence of arrhythmias, and patient factors such as obesity. Third, cardiac output was measured at one point in time (median, 45 days) before TAVR; therefore, it is unclear if there was any progressive decline in the cardiac function until the index procedure was performed. Ideally, cardiac output measurements should be obtained on the day of TAVR prior to the procedure. However, this was lacking in most patients. Given that this is a retrospective study, it is hypothesis-generating and future prospective/randomized control trials must be performed to test the hypothesis and evaluate the clinical utility of CPI for the prognostication of patients undergoing TAVR.

Conclusion

CPI is a useful non-invasive prognostic tool to predict outcomes in patients undergoing TAVR. Future prospective studies using CPI in TAVR are indicated to test its utility for the risk stratification of patients and to guide management strategy.

^*Joint first authors.

From the ¹Department of Cardiovascular Diseases, Mayo Clinic Arizona, Phoenix, Arizona; ²Department of Health Sciences Research, Mayo Clinic Arizona, Scottsdale, Arizona; ³Department of Cardiovascular Diseases, Mayo Clinic Rochester, Rochester, Minnesota; and ⁴Department of Cardiovascular Diseases, Mayo Clinic Florida, Jacksonville, Florida. 

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

Manuscript submitted September 6, 2019, provisional acceptance given September 16, 2019, final version accepted September 20, 2019.

Address for correspondence: Pradyumna Agasthi, MD, Mayo Clinic Arizona, 5777 E. Mayo Blvd, Phoenix, AZ 85054. Email: pradyumna_agasthi@hotmail.com

1. Cotter G, Moshkovitz Y, Kaluski E, et al. The role of cardiac power and systemic vascular resistance in the pathophysiology and diagnosis of patients with acute congestive heart failure. Eur J Heart Failure. 2003;5:443-451.
2. Fincke R, Hochman JS, Lowe AM, et al. Cardiac power is the strongest hemodynamic correlate of mortality in cardiogenic shock: a report from the SHOCK trial registry. J Am Coll Cardiol. 2004;44:340-348.
3. Lang CC, Karlin P, Haythe J, Lim TK, Mancini DM. Peak cardiac power output, measured noninvasively, is a powerful predictor of outcome in chronic heart failure. Circ Heart Fail. 2009;2:33-38.
4. Rosenblum H, Helmke S, Williams P, et al. Peak cardiac power measured noninvasively with a bioreactance technique is a predictor of adverse outcomes in patients with advanced heart failure. Congest Heart Fail. 2010;16:254-258.
5. Williams SG, Cooke GA, Wright DJ, et al. Peak exercise cardiac power output; a direct indicator of cardiac function strongly predictive of prognosis in chronic heart failure. Eur Heart J. 2001;22:1496-1503.
6. Williams SG, Jackson M, Cooke GA, et al. How do different indicators of cardiac pump function impact upon the long-term prognosis of patients with chronic heart failure? Am Heart J. 2005;150:983.
7. Nishimura RA, Otto CM, Bonow RO, et al. 2017 AHA/ACC focused update of the 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2017;70:252-289.
8. Holmes DR Jr, Brennan JM, Rumsfeld JS, et al. Clinical outcomes at 1 year following transcatheter aortic valve replacement. JAMA. 2015;313:1019-1028.
9. Carroll JD, Edwards FH, Marinac-Dabic D, et al. The STS-ACC transcatheter valve therapy national registry: a new partnership and infrastructure for the introduction and surveillance of medical devices and therapies. J Am Coll Cardiol. 2013;62:1026-1034.
10. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2018;71:2199-2269.
11. Zoghbi WA, Adams D, Bonow RO, et al. Recommendations for noninvasive evaluation of native valvular regurgitation: a report from the American Society of Echocardiography developed in collaboration with the Society for Cardiovascular Magnetic Resonance. J Am Soc Echo. 2017;30:303-371.
12. Baumgartner H, Hung J, Bermejo J, et al. Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice. J Am Soc Echo. 2009;22:1-23.
13. Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echo. 2015;28:1-39.e14.
14. Rudski LG, Lai WW, Afilalo J, et al. Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echo. 2010;23:685-713.
15. McLean AS, Needham A, Stewart D, Parkin R. Estimation of cardiac output by noninvasive echocardiographic techniques in the critically ill subject. Anaesth Intensive Care. 1997;25:250-254.
16. Contal C, O’Quigley J. An application of changepoint methods in studying the effect of age on survival in breast cancer. Comp Stat Data Analysis. 1999;30:253-270.
17. Spearman C. The proof and measurement of association between two things. Int J Epidemiol. 2010;39:1137-1150.
18. Tan L-B, Littler WA. Measurement of cardiac reserve in cardiogenic shock: implications for prognosis and management. Heart. 1990;64:121-128.
19. Tan L. Cardiac pumping capability and prognosis in heart failure. Lancet. 1986;328:1360-1363.
20. Mendoza DD, Cooper HA, Panza JA. Cardiac power output predicts mortality across a broad spectrum of patients with acute cardiac disease. Am Heart J. 2007;153:366-370.
21. Holmes DR, Brennan JM, Rumsfeld JS, et al. Clinical outcomes at 1 year following transcatheter aortic valve replacement. JAMA. 2015;313:1019-1028.
22. Cotter G, Williams SG, Vered Z, Tan LB. Role of cardiac power in heart failure. Curr Opin Cardiol. 2003;18:215-222.