Exercise intolerance is the dominant symptom of most chronic cardiac disease and one of its strongest prognostic markers. Cardiopulmonary exercise testing (CPET) quantifies it objectively by measuring gas exchange during incremental exercise, integrating the cardiovascular, pulmonary, muscular, and metabolic responses to a physiological stress and identifying which system limits performance.^1,2 Peak oxygen uptake (VO₂ peak) predicts survival in heart failure and informs transplant candidacy, while the ventilatory efficiency slope (VE/VCO₂) adds independent prognostic information reflecting chemoreflex drive, pulmonary vascular involvement, and filling pressures.^3,4,5,6,7

Yet in routine practice, exercise limitation is still most often attributed to, and monitored by, left ventricular ejection fraction. The correlation between resting haemodynamics and functional capacity is well known to be weak,^8,9 and the reasons are physiologically clear: ejection fraction is a load-dependent, geometrically derived ratio measured at rest, whereas exercise capacity

depends on the ability to augment stroke volume, on chronotropic competence, on ventilatory control, and on peripheral oxygen extraction.^10,11

Two measures might capture these determinants better. Global longitudinal strain (GLS), derived from speckle tracking echocardiography, quantifies the shortening of subendocardial longitudinal fibres and detects contractile impairment while ejection fraction remains normal.^12,13,14 Because longitudinal shortening contributes substantially to stroke volume generation, impaired GLS would be expected to limit stroke volume augmentation during exercise and therefore oxygen delivery at peak. Separately, heart rate variability (HRV) and baroreflex sensitivity (BRS) index the autonomic regulation that governs the chronotropic and vasomotor response to exercise, and autonomic derangement predicts mortality in cardiac populations independently of ejection fraction.^15,16,17

Whether these two measures explain exercise capacity better than ejection fraction — and whether they do so independently of each other — has not been examined comprehensively in a single cohort, particularly in low- and middle-income settings where CPET availability is limited and identifying accessible surrogates has practical value.¹⁸ We therefore assessed autonomic function, myocardial deformation, and cardiopulmonary exercise performance concurrently in patients with established heart disease, with the aim of identifying the independent determinants of VO₂ peak.

2.1 Study design and setting

This was a prospective, observational, cross-sectional study conducted over 18 months in the Department of Cardiology of a tertiary care teaching hospital in India, in collaboration with the Departments of Physiology and Radiology. The protocol was approved by the Institutional Ethics Committee and conducted in accordance with the Declaration of Helsinki (2013 revision) and Indian Council of Medical Research guidelines. All participants gave written informed consent.

 2.2 Participants

Patients aged 18–75 years with established ischaemic heart disease (stable angina or prior myocardial infarction with documented coronary disease), heart failure (reduced, mildly reduced, or preserved ejection fraction, the last with echocardiographic evidence of diastolic dysfunction), or at least moderate valvular heart disease were eligible. Patients had to be medically stable, with no acute decompensation, cardiac hospitalisation, or medication change in the preceding two weeks, and willing and able to complete all three assessments.

 Exclusion criteria were acute coronary syndrome or decompensated heart failure within two weeks; arrhythmias precluding valid HRV analysis (persistent or permanent atrial fibrillation, atrial flutter, ventricular ectopy > 10% of beats, complete heart block); any implanted pacemaker, defibrillator, or resynchronisation device; physical inability to perform cycle ergometry, including limiting orthopaedic, neuromuscular, or peripheral vascular disease; significant non-cardiac comorbidity independently impairing exercise performance (obstructive airway disease with FEV₁/FVC < 70%, haemoglobin < 10 g/dL, uncontrolled thyroid disease, eGFR < 15 mL/min/1.73 m²); known primary autonomic neuropathy with documented orthostatic hypotension; inadequate echocardiographic image quality; and pregnancy or lactation.

 2.3 Sample size

Sample size was estimated for detection of a Pearson correlation of r = 0.30 between GLS and VO₂ peak with two-sided α = 0.05 and 80% power using Fisher’s Z transformation, yielding n = 85. Allowing for 15–20% attrition and to preserve power for subgroup analysis and multivariable modelling with up to four predictors, the target was set at 130 patients.

 2.4 Assessment sequence

All three assessments were completed within a 48-hour window. Patients abstained from caffeine, nicotine, and strenuous activity for at least 12 hours before testing, and all testing was performed between 08:00 and 12:00 to limit circadian variation in autonomic tone.

 2.5 Cardiopulmonary exercise testing

CPET was performed by trained personnel under cardiologist supervision with resuscitation equipment immediately available. Patients fasted for at least two hours beforehand and avoided vigorous exercise for 24 hours.

Testing used an electronically braked cycle ergometer with a metabolic cart capable of breath-by-breath gas analysis. Continuous 12-lead electrocardiography, non-invasive blood pressure every two minutes, and pulse oximetry were monitored throughout. An individualised ramp protocol was used, with work rate increasing continuously at 10–20 W/min, selected to produce an exercise duration of 8–12 minutes, preceded by three minutes of unloaded pedalling and followed by three minutes of active recovery. Testing was terminated for volitional fatigue, significant arrhythmia, ST depression > 2 mm, a fall in systolic pressure > 20 mmHg from peak, oxygen saturation < 85%, or patient request.

Measured variables were VO₂ peak (highest 30-second averaged oxygen uptake, the primary functional outcome); anaerobic threshold, determined by the V-slope method and confirmed by ventilatory equivalents; the VE/VCO₂ slope below the respiratory compensation point; oxygen pulse (VO₂/heart rate) as a non-invasive surrogate of stroke volume; peak work rate and exercise duration; and chronotropic response including heart rate recovery at one minute. Effort adequacy was confirmed by a peak respiratory exchange ratio ≥ 1.10. Data were analysed offline using standardised algorithms, with interpretation following joint ATS/ACCP recommendations.^2,19

 2.6 Autonomic function testing

After at least 10 minutes of supine rest in a quiet room at 22–24°C, a digital 12-lead electrocardiogram was recorded at ≥ 1000 Hz sampling. A 5-minute artefact-free segment was analysed after visual inspection and manual editing of ectopic beats. Time-domain (SDNN, RMSSD, pNN50) and frequency-domain (LF, HF, LF/HF ratio, total power) indices were derived by Fast Fourier Transform following Task Force standards.¹⁵ LF and HF are reported in normalised units.

BRS was assessed by the spontaneous sequence method using beat-to-beat finger plethysmography (sequences of ≥ 3 beats with concordant change in systolic pressure ≥ 1 mmHg/beat and R-R interval ≥ 6 ms/beat, expressed as the regression slope of ΔR-R on ΔSBP) and by the Valsalva manoeuvre (40 mmHg for 15 seconds, with BRS derived from the phase IV overshoot); patients with severe aortic stenosis or active ischaemia did not undergo Valsalva. The mean of available methods was used.

2.7 Echocardiography and deformation imaging

Transthoracic echocardiography was performed by a single trained operator using a commercially available system with a 3.5–5 MHz transducer. Apical four-chamber, two-chamber, and long-axis views were acquired at 50–90 frames per second with three consecutive cycles stored during breath-hold. Studies with fewer than 14 of 16 trackable segments were excluded.

Offline speckle tracking analysis was performed with vendor-independent software by an independent analyst blinded to clinical data and CPET results, on de-identified loops. GLS was the average peak systolic longitudinal strain across 16 segments; values less negative than −16% were regarded as impaired.¹² Strain rate, circumferential and radial strain, left atrial strain, biplane Simpson’s LVEF, and conventional diastolic indices were also recorded.²⁰ Reproducibility was assessed in 20 randomly selected studies by intra- and inter-observer re-analysis, expressed as intraclass correlation coefficients, coefficients of variation, and Bland–Altman limits of agreement.

 2.8 Statistical analysis

Analyses used IBM SPSS Statistics v26.0 and R v4.2.0, with two-tailed p < 0.05 considered significant. Normality was assessed by Shapiro–Wilk and Kolmogorov–Smirnov tests with graphical inspection. Continuous data are presented as mean ± SD or median (IQR); categorical data as counts and percentages.

Bivariate associations used Pearson’s r or Spearman’s ρ as appropriate, interpreted as weak (< 0.3), moderate (0.3–0.5), strong (0.5–0.7), and very strong (> 0.7). Because longitudinal strain is conventionally negative, correlation and regression analyses used the absolute magnitude of GLS; positive coefficients therefore indicate that greater (better) longitudinal deformation is associated with higher VO₂ peak. Group comparisons used the independent-samples t-test or Mann–Whitney U test, and one-way ANOVA with Tukey’s HSD or Kruskal–Wallis with Dunn’s correction across three groups; categorical comparisons used chi-square or Fisher’s exact test.

 Independent predictors of VO₂ peak were identified by multivariable linear regression with hierarchical entry, including variables significant at p < 0.10 on univariate analysis together with the clinically mandated covariates age, sex, LVEF, and NYHA class. Standardised β coefficients are reported. Model assumptions were verified through residual plots, variance inflation factors (VIF < 5, excluding significant multicollinearity), and Cook’s distance for influential observations.

One hundred and thirty patients completed all assessments (Table 1). Mean age was 56.4 ± 11.2 years and 82 (63.1%) were male, with a mean BMI of 27.8 ± 3.9 kg/m². Hypertension was present in 78 (60.0%), diabetes mellitus in 52 (40.0%), and dyslipidaemia in 61 (46.9%). Sixty-eight patients (52.3%) were in NYHA class II and 42 (32.3%) in class III. Ischaemic heart disease accounted for 58 patients (44.6%), heart failure for 47 (36.2%), and valvular disease for 25 (19.2%).

 Table 1. Baseline demographic and clinical characteristics (n = 130)

  3.2 Cardiopulmonary exercise performance

CPET results are presented in Table 2. VO₂ peak was 16.8 ± 4.2 mL/kg/min, substantially below age- and sex-predicted norms and within the range associated with impaired prognosis in chronic heart failure.^3,5 The anaerobic threshold occurred at 11.2 ± 3.1 mL/kg/min. Mean VE/VCO₂ slope was 34.6 ± 6.8, at the threshold conventionally regarded as marking ventilatory inefficiency. Oxygen pulse was 9.1 ± 2.3 mL/beat. A mean peak respiratory exchange ratio of 1.14 ± 0.06 confirmed that maximal or near-maximal effort was achieved, supporting the validity of the peak measurements.

 Table 2. Cardiopulmonary exercise testing parameters (n = 130)

3.3 Autonomic and echocardiographic findings

Autonomic assessment showed reduced overall variability (SDNN 82.6 ± 24.3 ms) and vagal modulation (RMSSD 21.4 ± 10.8 ms), with sympathetic predominance (LF 61.2 ± 12.5 nu, HF 29.6 ± 9.4 nu, LF/HF 2.3 ± 0.8) and depressed baroreflex sensitivity (5.8 ± 2.1 ms/mmHg). On echocardiography, mean LVEF was preserved at 52.4 ± 8.1%, while GLS was impaired at −15.8 ± 3.2% and global longitudinal strain rate was −0.89 ± 0.21 s⁻¹ (Table 3). The cohort therefore exhibited markedly reduced exercise capacity in the presence of an ejection fraction that would conventionally be reported as normal.

 Table 3. Autonomic and echocardiographic parameters (n = 130)

3.4 Correlates of exercise capacity

Bivariate correlations are shown in Table 4. GLS demonstrated the strongest association with VO₂ peak (r = 0.51, p < 0.001), indicating that patients with better preserved longitudinal deformation achieved higher peak oxygen uptake. Autonomic indices were also significantly associated with functional capacity: RMSSD correlated with VO₂ peak (r = 0.38, p < 0.001), as did baroreflex sensitivity (r = 0.36, p = 0.002). The LF/HF ratio correlated positively with the VE/VCO₂ slope (r = 0.44, p < 0.001), linking sympathetic predominance to ventilatory inefficiency. SDNN correlated with GLS magnitude (r = 0.42, p < 0.001).

 Table 4. Correlations between autonomic, mechanical, and cardiopulmonary parameters

Correlations involving GLS use the absolute magnitude of strain; a positive coefficient denotes association between better longitudinal function and the comparator.

 3.5 Independent predictors of peak oxygen uptake

On multivariable linear regression (Table 5), GLS was the strongest independent predictor of VO₂ peak (β = 0.39, p < 0.001), followed by SDNN (β = 0.27, p = 0.003) and age (β = −0.22, p = 0.01). LVEF did not retain independent predictive value (β = 0.09, p = 0.18). Variance inflation factors were below 5 for all retained variables, excluding significant multicollinearity between GLS and LVEF.

Table 5. Multivariable linear regression: independent predictors of VO₂ peak

Model additionally adjusted for sex and NYHA class. NS = not significant.

 3.6 Comparison across disease categories

Functional, mechanical, and autonomic impairment differed significantly by diagnosis (Table 6). Patients with heart failure had the lowest VO₂ peak (14.8 mL/kg/min), most impaired GLS (−14.2%), and lowest SDNN (70.5 ms), compared with ischaemic (17.9 mL/kg/min; −16.4%; 86.2 ms) and valvular (18.2 mL/kg/min; −17.1%; 88.1 ms) subgroups, with p = 0.002, 0.001, and 0.004 respectively. The rank ordering of subgroups was identical across all three domains.

 Table 6. Comparison of functional, mechanical, and autonomic parameters across disease categories

The central finding of this study is that in patients with established heart disease, exercise capacity was determined by myocardial deformation and autonomic tone, and not by ejection fraction. In a multivariable model that deliberately retained LVEF as a clinically mandated covariate, GLS and SDNN emerged as independent predictors of VO₂ peak while LVEF did not.

 4.1 The magnitude of functional impairment

Mean VO₂ peak of 16.8 mL/kg/min represents substantially reduced cardiorespiratory fitness, falling well below age-predicted values and within the band associated with adverse prognosis in heart failure populations.^3,5,21 The mean VE/VCO₂ slope of 34.6 sits at the threshold above which ventilatory inefficiency is conventionally recognised and prognostic risk begins to rise.^6,7,22 That this degree of impairment was present in a cohort whose mean LVEF was 52.4% underlines the limited information ejection fraction conveys about how a patient actually functions. The mean peak respiratory exchange ratio of 1.14 is important methodologically: submaximal effort is the commonest reason for uninterpretable CPET data, and its exclusion here strengthens confidence in the peak values.^2,19

 4.2 Deformation as a determinant of functional reserve

GLS showed the strongest bivariate correlation with VO₂ peak (r = 0.51) and the largest standardised coefficient in the multivariable model. The physiological interpretation is straightforward. Longitudinal shortening of subendocardial fibres, together with the associated atrioventricular plane descent, contributes materially to stroke volume generation.^12,23 When longitudinal function is impaired, the capacity to augment stroke volume during incremental exercise is constrained, and since oxygen delivery at peak is the product of cardiac output and arteriovenous oxygen difference, peak oxygen uptake falls accordingly. GLS thus indexes mechanical reserve in a way that a resting volumetric ratio cannot: two patients with identical ejection fractions may differ substantially in longitudinal function and therefore in their ability to increase output under load. This interpretation is consistent with the observation that oxygen pulse — a non-invasive surrogate of stroke volume — was also depressed in our cohort at 9.1 mL/beat.

 4.3 Autonomic determinants and ventilatory inefficiency

SDNN retained independent predictive value for VO₂ peak after adjustment, and RMSSD and BRS both correlated significantly with it. Autonomic regulation governs the chronotropic response, the redistribution of blood flow to exercising muscle, and venous return, so its impairment limits exercise capacity through mechanisms distinct from those indexed by GLS. This distinctness is what makes the two variables independently informative in the same model.

The correlation between LF/HF ratio and VE/VCO₂ slope (r = 0.44) is of separate interest. Ventilatory inefficiency in cardiac disease is attributed to increased dead space ventilation, heightened chemoreflex and ergoreflex sensitivity, and pulmonary vascular abnormality.^11,24,25 Sympathetic predominance plausibly links these: enhanced peripheral chemosensitivity and muscle ergoreceptor overactivity both drive ventilation and both accompany sympathetic activation in heart failure.^24,25 Our data are consistent with, though cannot establish, this pathway. The LF/HF ratio should nonetheless be interpreted cautiously as a descriptive index rather than a direct measure of sympathovagal balance.

 4.4 The limits of ejection fraction

That LVEF did not retain significance (β = 0.09, p = 0.18) is the study’s most clinically pointed result. It is not that ejection fraction is unrelated to functional capacity, but rather that its contribution is subsumed by measures that capture the same underlying mechanics with greater sensitivity, together with a neural dimension that ejection fraction does not address at all. This parallels the established observation that CPET-derived variables outperform resting LVEF in prognostication.^4,21,26,27 Practically, it argues against using an ejection fraction within the normal range to reassure a symptomatic patient, and in favour of measuring what actually limits them.

 4.5 Subgroup differences

The heart failure subgroup was most impaired in all three domains, and the ordering of subgroups was identical for functional, mechanical, and autonomic indices. This coherence supports the view that these are correlated manifestations of a common disease process rather than independent phenomena. The relative preservation of function in the valvular subgroup should be interpreted with caution given its small size (n = 25) and the heterogeneity of lesions it contains.

 4.6 Clinical implications

CPET remains the reference standard for functional assessment, but its cost, space, and expertise requirements place it beyond the reach of most secondary-level centres in India and comparable settings.¹⁸ Our findings suggest a pragmatic alternative: GLS and SDNN together account for a meaningful proportion of the variance in VO₂ peak and are obtainable from equipment already present in most echocardiography laboratories and from a 5-minute electrocardiographic recording. These measures should not be presented as replacements for CPET, which provides information about ventilatory efficiency, chronotropic response, and effort adequacy that neither can supply. They may, however, serve to identify which patients most need referral for full cardiopulmonary testing, and to provide interim functional information where such referral is not feasible.

 4.7 Limitations

This single-centre cross-sectional study cannot establish causality or prognostic value; longitudinal follow-up would be required to determine whether GLS and HRV predict events, not merely current capacity. The sample, while adequate for the primary analysis, limits subgroup power, particularly for the valvular group. Beta-blocker therapy, near-universal in this population, affects both HRV and the chronotropic response to exercise and was not stratified in the present analysis; this is likely to have attenuated rather than inflated the autonomic associations. Cycle ergometry typically yields VO₂ peak values 5–10% lower than treadmill testing, so absolute values should be compared with modality-matched norms. Inter-vendor variability in strain software limits direct transfer of absolute GLS thresholds between platforms, although a single vendor-independent package and a single blinded analyst were used throughout. Finally, exclusion of patients with atrial fibrillation and implanted devices — necessary for valid HRV analysis — selects against a higher-risk segment of the heart failure population and limits generalisability to that group.

In patients with established heart disease, peak oxygen uptake was independently determined by global longitudinal strain and heart rate variability, while left ventricular ejection fraction retained no independent predictive value. Exercise capacity was substantially reduced despite a preserved mean ejection fraction, and impairment across functional, mechanical, and autonomic domains was greatest in patients with heart failure. These findings support integrating deformation imaging and autonomic assessment into the evaluation of exercise intolerance, and suggest that where cardiopulmonary exercise testing is unavailable, these accessible measures may provide useful information about functional reserve. Prospective outcome studies are required to establish whether a composite of mechanical and autonomic indices improves risk stratification over existing approaches.

1. Wasserman K, Hansen JE, Sue DY, Stringer WW, Whipp BJ. Principles of Exercise Testing and Interpretation. 4th ed. Philadelphia: Lippincott Williams and Wilkins; 2005.
2. Balady GJ, Arena R, Sietsema K, Myers J, Coke L, Fletcher GF, et al. Clinician’s guide to cardiopulmonary exercise testing in adults: a scientific statement from the American Heart Association. Circulation. 2010;122(2):191–225.
3. Mancini DM, Eisen H, Kussmaul W, Mull R, Edmunds LH Jr, Wilson JR. Value of peak exercise oxygen consumption for optimal timing of cardiac transplantation in ambulatory patients with heart failure. Circulation. 1991;83(3):778–86.
4. Arena R, Myers J, Guazzi M. The clinical and research applications of aerobic capacity and ventilatory efficiency in heart failure: an evidence-based review. Heart Fail Rev. 2008;13(2):245–69.
5. Myers J, Prakash M, Froelicher V, Do D, Partington S, Atwood JE. Exercise capacity and mortality among men referred for exercise testing. N Engl J Med. 2002;346(11):793–801.
6. Kleber FX, Vietzke G, Wernecke KD, Bauer U, Opitz C, Wensel R, et al. Impairment of ventilatory efficiency in heart failure: prognostic impact. Circulation. 2000;101(24):2803–9.
7. Chua TP, Ponikowski P, Harrington D, Anker SD, Webb-Peploe K, Clark AL, et al. Clinical correlates and prognostic significance of the ventilatory response to exercise in chronic heart failure. J Am Coll Cardiol. 1997;29(7):1585–90.
8. Francis GS, Goldsmith SR, Cohn JN. Relationship of exercise capacity to resting hemodynamic variables in patients with congestive heart failure. Am Heart J. 1982;104(4 Pt 1):725–31.
9. Clark AL, Poole-Wilson PA, Coats AJ. Exercise limitation in chronic heart failure: central role of the periphery. J Am Coll Cardiol. 1996;28(5):1092–102.
10. Malhotra R, Bakken K, D’Elia E, Lewis GD. Cardiopulmonary exercise testing in heart failure. JACC Heart Fail. 2016;4(8):607–16.
11. Sullivan MJ, Higginbotham MB, Cobb FR. Increased exercise ventilation in patients with chronic heart failure: intact ventilatory control despite hemodynamic and pulmonary abnormalities. Circulation. 1988;77(3):552–9.
12. Marwick TH, Leano RL, Brown J, Sun JP, Hoffmann R, Lysyansky P, et al. Myocardial strain measurement with 2-dimensional speckle-tracking echocardiography: definition of normal range. JACC Cardiovasc Imaging. 2009;2(1):80–4.
13. Mor-Avi V, Lang RM, Badano LP, Belohlavek M, Cardim NM, Derumeaux G, et al. Current and evolving echocardiographic techniques for the quantitative evaluation of cardiac mechanics: ASE/EAE consensus statement. J Am Soc Echocardiogr. 2011;24(3):277–313.
14. Amundsen BH, Helle-Valle T, Edvardsen T, Torp H, Crosby J, Lyseggen E, et al. Noninvasive myocardial strain measurement by speckle tracking echocardiography: validation against sonomicrometry and tagged magnetic resonance imaging. J Am Coll Cardiol. 2006;47(4):789–93.
15. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation. 1996;93(5):1043–65.
16. Kleiger RE, Miller JP, Bigger JT Jr, Moss AJ. Decreased heart rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol. 1987;59(4):256–62.
17. La Rovere MT, Bigger JT Jr, Marcus FI, Mortara A, Schwartz PJ. Baroreflex sensitivity and heart-rate variability in prediction of total cardiac mortality after myocardial infarction. ATRAMI Investigators. Lancet. 1998;351(9101):478–84.
18. Guazzi M, Arena R, Halle M, Piepoli MF, Myers J, Lavie CJ. 2016 focused update: clinical recommendations for cardiopulmonary exercise testing data assessment in specific patient populations. Circulation. 2016;133(24):e694–711.
19. Piepoli MF, Corrà U, Agostoni PG, Belardinelli R, Cohen-Solal A, Hambrecht R, et al. Statement on cardiopulmonary exercise testing in chronic heart failure due to left ventricular dysfunction: recommendations for performance and interpretation. Part II. Eur J Cardiovasc Prev Rehabil. 2006;13(3):300–11.
20. Nagueh SF, Appleton CP, Gillebert TC, Marino PN, Oh JK, Smiseth OA, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography. J Am Soc Echocardiogr. 2009;22(2):107–33.
21. Keteyian SJ, Patel M, Kraus WE, Brawner CA, McConnell TR, Piña IL, et al. Variables measured during cardiopulmonary exercise testing as predictors of mortality in chronic systolic heart failure. J Am Coll Cardiol. 2016;67(7):780–9.
22. Gitt AK, Wasserman K, Kilkowski C, Kleemann T, Kilkowski A, Bangert M, et al. Exercise anaerobic threshold and ventilatory efficiency identify heart failure patients for high risk of early death. Circulation. 2002;106(24):3079–84.
23. Sengupta PP, Krishnamoorthy VK, Korinek J, Narula J, Vannan MA, Lester SJ, et al. Left ventricular form and function revisited: applied translational science to cardiovascular ultrasound imaging. J Am Soc Echocardiogr. 2007;20(5):539–51.
24. Ponikowski PP, Chua TP, Francis DP, Capucci A, Coats AJ, Piepoli MF. Muscle ergoreceptor overactivity reflects deterioration in clinical status and cardiorespiratory reflex control in chronic heart failure. Circulation. 2001;104(19):2324–30.
25. Guazzi M, Reina G, Tumminello G, Guazzi MD. Exercise ventilation inefficiency and cardiovascular mortality in heart failure: the critical independent prognostic value of the arterial CO2 partial pressure. Eur Heart J. 2005;26(5):472–80.
26. Corrà U, Agostoni PG, Anker SD, Coats AJ, Crespo Leiro MG, de Boer RA, et al. Role of cardiopulmonary exercise testing in clinical stratification in heart failure. A position paper from the Heart Failure Association of the ESC. Eur J Heart Fail. 2018;20(1):3–15.
27. Brunner-La Rocca HP, Weilenmann D, Follath F, Schlumpf M, Rickenbacher P, Pfisterer ME. Is cardiopulmonary exercise testing useful to select patients with chronic heart failure for heart transplantation? Analysis of peak VO2 and VE/VCO2 slope. J Heart Lung Transplant. 2004;23(2):184–9.
28. Aaronson KD, Schwartz JS, Chen TM, Wong KL, Goin JE, Mancini DM. Development and prospective validation of a clinical index to predict survival in ambulatory patients referred for cardiac transplant evaluation. Circulation. 1997;95(12):2660–7.
29. Agostoni P, Corrà U, Cattadori G, Veglia F, La Gioia R, Scardovi AB, et al. Metabolic exercise test data combined with cardiac and kidney indexes, the MECKI score: a multiparametric approach to heart failure prognosis. Int J Cardiol. 2013;167(6):2710–8.
30. Arena R, Myers J, Williams MA, Gulati M, Kligfield P, Balady GJ, et al. Assessment of functional capacity in clinical and research settings: a scientific statement from the American Heart Association. Circulation. 2007;116(3):329–43.
31. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG, Coats AJ, et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2016;37(27):2129–200.
32. Thayer JF, Yamamoto SS, Brosschot JF. The relationship of autonomic imbalance, heart rate variability and cardiovascular disease risk factors. Int J Cardiol. 2010;141(2):122–31.