Relationship of resting haemoglobin concentration to peak oxygen uptake in heart failure patients

Transcription

1 Relationship of resting haemoglobin concentration to peak oxygen uptake in heart failure patients Piergiuseppe Agostoni, Elisabetta Salvioni, Chiara Debenedetti, Carlo Vignati, Gaia Cattadori, Mauro Contini, Damiano Magrì, Pietro Palermo, Erica Gondoni, Denise Brusoni, et al. To cite this version: Piergiuseppe Agostoni, Elisabetta Salvioni, Chiara Debenedetti, Carlo Vignati, Gaia Cattadori, et al.. Relationship of resting haemoglobin concentration to peak oxygen uptake in heart failure patients. American Journal of Hematology, Wiley, 0, <.0/ajh.>. <hal-00> HAL Id: hal-00 Submitted on Jan 0 HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

2 American Journal of Hematology Relationship of resting haemoglobin concentration to peak oxygen uptake in heart failure patients Journal: American Journal of Hematology Manuscript ID: AJH--00.R Wiley - Manuscript type: Research Article Date Submitted by the Author: -Feb-0 Complete List of Authors: Agostoni, Piergiuseppe; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Salvioni, Elisabetta; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Debenedetti, Chiara; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Vignati, Carlo; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Cattadori, Gaia; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Contini, Mauro; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Magrì, Damiano; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Palermo, Pietro; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Gondoni, Erica; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Brusoni, Denise; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Fiorentini, Cesare; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Apostolo, Anna; Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy Keywords: Anemia, Exercise capacity, Cardiopulmonary exercise test

3 Page of 0 American Journal of Hematology

4 American Journal of Hematology Page of Relationship of resting haemoglobin concentration to peak oxygen uptake in heart failure patients Piergiuseppe Agostoni*#, MD, PhD, Elisabetta Salvioni*, PhD, Chiara Debenedetti*, MD, Carlo Vignati*, MD, Gaia Cattadori*, MD, Mauro Contini*, MD, Damiano Magrì*, MD, PhD, Pietro Palermo*, MD, Erica Gondoni*, MD, Denise Brusoni*, MD, Cesare Fiorentini*, MD, and Anna Apostolo*, MD. Short title: Anaemia and peak O in heart failure. * Centro Cardiologico Monzino, IRCCS, Dipartimento di Scienze Cardiovascolari, Università di Milano, Milan, Italy, # Division of Critical Care and Respiratory Medicine, Department of Medicine, University of Washington, Seattle, WA, USA. None of the authors have any relationship with industry or any conflict of interest. Key words. Anaemia, exercise capacity, cardiopulmonary exercise test Corresponding Author: Piergiuseppe Agostoni, MD, PhD Centro Cardiologico, IRCCS, Istituto di Scienze Cardiovascolari, Università di Milano Via Parea, 0 Milan, Italy. Phone +000 Fax piergiuseppe.agostoni@unimi.it Abstract word count: Text word count: Tables: Figures:

5 Page of 0 American Journal of Hematology Abstract Anaemia is frequent in chronic heart failure (HF). In order to calculate what change in peak oxygen uptake ( O ) should be expected in the event of changes in haemoglobin concentration, we studied the correlation between peak O and haemoglobin concentration in a large HF population. We carried out retrospective analysis of all cardiopulmonary exercise tests (CPET) performed in our HF Clinic between June 00 and March 00 in HF patients who had a resting haemoglobin concentration measurement taken within days of the CPET. We collected CPETs, 0 tests were considered maximal and analysed. We identified patients (%) as anaemic. Peak O was lower (p<0.00) in anaemic patients (± ml/min) compared to non-anaemic (± ml/min). The slope of the O vs. haemoglobin ratio was ml/min/g/dl at peak exercise. This correlation remained significant also when several confounding variables were analysed by multivariate analysis. As an average, each g of haemoglobin accounts, at peak exercise, for ml/min change in O which is equivalent to 0. ml/min/kg. Therefore, in HF patients anaemia treatment should increase O by ml/min for each g/dl of haemoglobin increase.

6 American Journal of Hematology Page of Introduction Anaemia and reduction of exercise capacity are both frequently reported in chronic heart failure (HF) and are both associated to worsening of HF prognosis (-). The two are linked to each other indirectly having heart failure as a common underlying origin and directly. Indeed, the gold standard variable of exercise capacity evaluation is peak exercise oxygen uptake ( O ), which is a cardiac output x arterio-venous O difference. The latter depends on haemoglobin, which reduction negatively affects the arterial oxygen concentration. Furthermore several reports showed that, in chronic HF patients, anaemia is associated with low exercise capacity (, ). In principle it is possible to calculate the amount of peak O loss due to anaemia if haemoglobin arterial oxygen saturation is normal and if peak exercise cardiac output and peak exercise oxygen extraction are known. In previous studies of our group in HF patients we measured at peak exercise a mean extraction rate ~0% (, ). Therefore because g of haemoglobin binds. ml of oxygen and, at peak exercise extraction rate is ~0%, each g of haemoglobin provides to the muscles ~ ml of oxygen. The latter multiplied by the cardiac output is the amount of O reduction resulting from a g drop in haemoglobin (). For example, if peak exercise cardiac output is L/min the loss of g of haemoglobin is equivalent to a reduction of ~0 ml/min O. Unfortunately, neither cardiac output nor oxygen extraction at peak exercise are usually measured, so that we cannot calculate the reduction of O directly due to anaemia. It would be desirable, however, to understand, at least in a semi-quantitative fashion, the role of anaemia on peak O in HF patients.

7 Page of 0 American Journal of Hematology The present study was undertaken to study the correlation between peak exercise O and haemoglobin concentration in a large HF population. Methods We carried out retrospective analysis of all cardiopulmonary exercise tests (CPET) performed in our HF Clinic between June 00 and March 00 in HF patients in stable clinical conditions, NYHA Class I to III, with left ventricle ejection fraction 0% who had performed a resting haemoglobin concentration measurement in our general laboratory within days of the exercise test. We excluded subjects with history and/or clinical documentation of pulmonary embolism or primary valvular heart disease, pericardial disease, severe obstructive lung disease, primitive pulmonary hypertension or occupational lung disease, asthma, severe renal failure (serum creatinine > mg/dl), significant peripheral vascular disease, advanced atrio-ventricular block, exercise induced angina and/or relevant ST changes. We also excluded patients whose exercise was interrupted as the result of a medical decision before maximal effort was reached, due to severe hypertension or severe arrhythmia. CPETs were performed on a cycle-ergometer using a progressively increasing work load generated by a ramp protocol personalized for each patient, with the aim of reaching maximum exercise in about min. Subjects were asked to pedal at 0 revolutions per minute. The used load for each individual of the personalized ramp protocol was decided by the physician in charge according to previous test results, if available, or to exercise performance as reported by patients. Tests too short (< min) or too long (> min) were repeated with a more appropriate ramp protocol (). Ventilation ( E), O, and carbon dioxide production ( CO ) were measured

8 American Journal of Hematology Page of breath by breath (V-max 00 metabolic cart SensorMedics, Yorba Linda CA for tests performed between June 00 and August 00 and V-max metabolic cart SensorMedics, Yorba Linda CA for tests performed between September 00 and March 00). Twelve-leads ECG and heart rate were monitored continuously. Anaerobic threshold was measured by the V-slope method and confirmed by O and CO ventilatory equivalents and end-tidal O and CO pressure specific changes (). E / CO slope was measured from the beginning of loaded pedalling to the end of the isocapnic buffering period. The latter was identified by an increase of the ventilatory equivalent for CO ( E / CO ) and confirmed by reduction of end-tidal CO pressure. The O /work ratio was measured throughout the entire exercise. Echocardiography was performed in our laboratory on each patient within months of the CPET test. Left ventricular diastolic volume and ejection fraction were measured in the left lateral decubitus position. Left ventricular volume was derived from the conventional apical - and -chamber images and the left ventricle ejection fraction was calculated using biplane Simpson s technique (). Anaemia was defined by WHO criteria as haemoglobin concentration < g/dl in women and < g/dl in men (). The presence of coronary artery disease associated to HF was assessed by standard coronary angiography according to Felker et al. (). We considered only coronary angiography performed within one year of the CPET. In case of multiple exercise tests we selected the CPET used in the present analysis in a random manner. The study and the access of private health information were approved by the local intern review board.

9 Page of 0 American Journal of Hematology Statistical analysis. Data are reported as mean ± SE. Differences between groups were analysed by unpaired t-tests with a p < 0.0 considered as statistically significant. Categorical variables such as NYHA classification, gender differences or treatment were analysed by chisquare test. Correlation between variables was assessed by linear regression analysis. We evaluated vs. peak O several variables including: age, gender, NYHA class, left ventricle ejection fraction, E / CO slope, peak heart rate and work load. Multivariate analysis was performed adding in the model all variables which were correlated to peak O at monovariate analysis. A similar analysis was also performed using O at the anaerobic threshold as the dependent variable. All data were analysed by SPSS.0 software collected in an Excel data base. Results We collected a total of CPETs of HF patients who matched the clinical study criteria, who had both a blood test for haemoglobin concentration determination at rest taken at our central laboratory within one week and an echocardiographic evaluation taken within months. CPETs were discarded because effort was judged submaximal according to peak exercise gas exchange ratio (RQ <.0). Consequently 0 patients were considered in the present report. Population characteristics, treatment and heart failure severity are reported in table I. Anaemic patients were older, had clinically more severe HF and were more frequently treated with diuretics. Haemoglobin was. ± 0. g/dl ranging between. and. g/dl. patients were anaemic (%), representing (%) of women and (%) of males. Non-anaemic

10 American Journal of Hematology Page of patients were (%), representing 0 (%) of women and (%) of males. Anaerobic threshold was not identified in % (n=) of the 0 patients. CPET results are reported in table II for the entire study population and for the anaemic and non-anaemic patients, respectively. Anaemic patients had a lower exercise capacity, with the work load achieved, the O both at peak exercise and at anaerobic threshold being lower. Furthermore, anaemic HF patients had a greater E / CO slope. It should be noted that cardiac function and left ventricular volume at echocardiography were similar for anaemic and non-anaemic patients (table I). Peak O was lower in anaemic male patients compared to non-anaemic but not in women who showed only a trend toward the same difference (p=0.00) (figure ). Coronary angiography was available for of the 0 patients. (%) patients (0 female and 0 male) were classified as having HF associated to coronary artery disease and 0 (%) ( female and male) as not associated to coronary artery disease i.e. with angiografically normal epicardial coronary arteries. The presence of coronary artery disease did not significantly influenced the correlation between O and haemoglobin both at peak exercise and at anaerobic threshold (table III). Table IV shows the correlations between several variables and peak O. At multivariate analysis only gender (p<0.000), haemoglobin (p<0.0), E / CO slope (p<0.000), the work load achieved (p<0.000) and NYHA class (p<0.000) remained significantly correlated to peak O. The correlation between peak exercise O and haemoglobin at rest is reported in figure. Assessing O as ml/min/kg the slope of the O vs. haemoglobin relationship was 0. (ml/kg/min/g/dl). O at anaerobic threshold was significantly correlated to the following

11 Page of 0 American Journal of Hematology variables: age (p<0.000), gender (p<0.000), NYHA class (p<0.000), LVEF (p<0.000), haemoglobin (p<0.000), E / CO slope (p<0.000), peak heart rate (p<0.000) and work load (p<0.000). At multivariate analysis only age (p<0.0), E / CO slope (p<0.000) and work load (p<0.000) remained independently correlated to O. The O vs. Hb relationship at the anaerobic threshold is reported in figure. Discussion The present study shows, in accordance with several previous reports, that anaemia is frequent in chronic HF patients and associated with a lower exercise capacity (-). Indeed, the prevalence of anaemia we observed (%) is in agreement with what has been reported previously (between % and %) (). In HF the presence of anaemia is associated with poor prognosis, with higher mortality risk both in hospitalised and ambulatory patients, higher hospitalisation rate and, overall, with a more severe clinical condition for patients (, -). In our study, on average, peak O and E/ CO slope, both of which are independent strong prognostic indicators of HF (), were % lower and % higher, respectively, in HF patients with anaemia compared to those without anaemia. It is commonly thought that anaemia should be corrected to improve HF patients quality of life and possibly prognosis, however there is still a lack of a scientific evidence; consequently anaemia correction is not considered as mandatory in HF treatment guidelines (, 0). Several studies have focused on the effect of anaemia correction, usually by iron and erythropoietin or darbopoietin, on exercise performance, measured either by peak O, exercise tolerance or by distance walked at the minutes walking test, in both HF and non-hf patients (-). However, only (-) reports all based on small studies, showed the effect of anaemia correction on peak

12 American Journal of Hematology Page of exercise O in HF patients, with discordant findings. Indeed, reports showed that anaemia treatment increased exercise performance (,, ) while one did not (). Several factors influence exercise performance in HF, including, following the Fick principle, cardiac output and the arterio-venous oxygen difference. We limited our analysis only to haemoglobin. To extrapolate the role of haemoglobin we studied the correlation between resting haemoglobin and O at peak and at anaerobic threshold. We observed a correlation between O (ml/min) at peak and haemoglobin with a R = 0., which means that haemoglobin accounts for % of the variance in peak O. This also means that % of peak O variance is not related to haemoglobin. The slope of this ratio (figure ) tells us that, on average, at peak exercise each g of haemoglobin accounts for a ml/min change in O which is equivalent to 0. ml/min/kg. We analysed the correlation between O at peak exercise and resting haemoglobin. Unfortunately peak exercise haemoglobin data is not available. However, exercise-induced hemoconcentration, which is well known and is due to both spleen contraction and fluid shift away from the intravascular space due to an increase in intracellular lactic acid (0, ), probably did not significantly influence our findings. Indeed, exercise-induced hemoconcentration is responsible for approximately 0% of arterio-venous oxygen difference increase in HF subjects, being on average around.0 g/dl in normal subjects and 0. g/dl in HF patients (0). Had the exercise-induced haemoglobin increase been even in patients with different levels of HF severity, this would have only shifted the O /Hb ratio upwards, with no effect on the slope of the ratio and, consequently, with no effect on the predicted O change as a result of haemoglobin

13 Page of 0 American Journal of Hematology changes. Unfortunately this is not the case, because exercise-induced haemoglobin increase is greater in less compromised patients, so the lack of haemoglobin measurements at peak exercise weakens our results. However, when comparing severe and non-severe HF patients the difference in exercised induced haemoglobin increase is trivial (0) and certainly not able to have a significant effect on our results. Data at anaerobic threshold were analysed because exercise-induced hemoconcentration takes place above the anaerobic threshold. The relationship between O at anaerobic threshold and haemoglobin at rest is more shallow than the one observed at peak exercise and ml/min/g/dl, respectively. Furthermore at multivariate analysis the relationship between resting haemoglobin and O AT loses the statistical significance. Indeed, at the anaerobic threshold, haemoglobin has a lower role in determining O compared to O at peak exercise. This finding is not surprising considering that, at the anaerobic threshold, a reduction in oxygen delivery due to anaemia can be still partially counteracted by an increase in cardiac output. The present study has some structural limitations which should be acknowledged. Indeed, the cross sectional nature of the present study does not permit to define, but only to suggest, a causal relationship between haemoglobin levels and exercise capacity. Finally we had described a relationship for a population but this may not be an appropriate prediction in a single individual. In conclusion our data show that, analysing a relevant number of HF patients, a correlation between peak O and haemoglobin exists with a slope equal to ml/min/g/dl. Thus, albeit totally unproven, it is possible to suggest that O should increase by approximately

14 American Journal of Hematology Page of ml/min for each g/dl of haemoglobin increase. Studies analyzing the acute effects of haemoglobin changes on exercise performance in anaemic HF patients are needed.

15 Page of 0 American Journal of Hematology References. Anand IS. Anemia and chronic heart failure implications and treatment options. J Am Coll Cardiol 00;:0-.. Felker GM, Adams KF, Jr., Gattis WA, et al. Anemia as a risk factor and therapeutic target in heart failure. J Am Coll Cardiol 00;:-.. Mitchell JE. Emerging role of anemia in heart failure. Am J Cardiol 00;:D-0D.. Tang YD, Katz SD. The prevalence of anemia in chronic heart failure and its impact on the clinical outcomes. Heart Fail Rev 00;:-.. Kalra PR, Bolger AP, Francis DP, et al. Effect of anemia on exercise tolerance in chronic heart failure in men. Am J Cardiol 00;:-.. Witte KK, Desilva R, Chattopadhyay S, et al. Are hematinic deficiencies the cause of anemia in chronic heart failure? Am Heart J 00;:-0.. Perego GB, Marenzi GC, Guazzi M, et al. Contribution of PO, P0, and Hb to changes in arteriovenous O content during exercise in heart failure. J Appl Physiol ;0:-.. Agostoni PG, Wasserman K, Perego GB, et al. Non-invasive measurement of stroke volume during exercise in heart failure patients. Clin Sci (Lond) 000;:-.. Agostoni P, Cattadori G. Non invasive cardiac output measurement: a new tool in heart failure. Cardiology (in press) 00.. Agostoni P, Bianchi M, Moraschi A, et al. Work-rate affects cardiopulmonary exercise test results in heart failure. Eur J Heart Fail 00;:-0.. Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol ;0: Feigenbaum H AW, Ryan T. Hemodynamics. Feigenbaum's Echocardiography: Ed. Lippincott Williams and Wilkins; 00. pp -.. WHO/UNICEF/UNU. Iron deficency anaemia: assessment, prevention, and control. Geneva, Switzerland: World Health Organization; 00.. Felker GM, Shaw LK, O'Connor CM. A standardized definition of ischemic cardiomyopathy for use in clinical research. J Am Coll Cardiol 00;:-.. McMurray JJ. What are the clinical consequences of anemia in patients with chronic heart failure? J Card Fail 00;:S-.. Ezekowitz JA, McAlister FA, Armstrong PW. Anemia is common in heart failure and is associated with poor outcomes: insights from a cohort of 0 patients with new-onset heart failure. Circulation 00;:-.. Horwich TB, Fonarow GC, Hamilton MA, et al. Anemia is associated with worse symptoms, greater impairment in functional capacity and a significant increase in mortality in patients with advanced heart failure. J Am Coll Cardiol 00;:0-.. Lang CC, Agostoni P, Mancini DM. Prognostic significance and measurement of exercise-derived hemodynamic variables in patients with heart failure. J Card Fail 00;:-.. Dickstein K, Cohen-Solal A, Filippatos G, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 00: the Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 00 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur Heart J 00;:-.

16 American Journal of Hematology Page of Jessup M, Abraham WT, Casey DE, et al. 00 focused update: ACCF/AHA Guidelines for the Diagnosis and Management of Heart Failure in Adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines: developed in collaboration with the International Society for Heart and Lung Transplantation. Circulation 00;:-0.. Mancini DM, Katz SD, Lang CC, et al. Effect of erythropoietin on exercise capacity in patients with moderate to severe chronic heart failure. Circulation 00;:-.. Palazzuoli A, Silverberg D, Iovine F, et al. Erythropoietin improves anemia exercise tolerance and renal function and reduces B-type natriuretic peptide and hospitalization in patients with heart failure and anemia. Am Heart J 00;: e-.. Ponikowski P, Anker SD, Szachniewicz J, et al. Effect of darbepoetin alfa on exercise tolerance in anemic patients with symptomatic chronic heart failure: a randomized, double-blind, placebo-controlled trial. J Am Coll Cardiol 00;:-.. Okonko DO, Grzeslo A, Witkowski T, et al. Effect of intravenous iron sucrose on exercise tolerance in anemic and nonanemic patients with symptomatic chronic heart failure and iron deficiency FERRIC-HF: a randomized, controlled, observer-blinded trial. J Am Coll Cardiol 00;:-.. Toblli JE, Lombrana A, Duarte P, et al. Intravenous iron reduces NT-pro-brain natriuretic peptide in anemic patients with chronic heart failure and renal insufficiency. J Am Coll Cardiol 00;0:-.. van Veldhuisen DJ, Dickstein K, Cohen-Solal A, et al. Randomized, double-blind, placebo-controlled study to evaluate the effect of two dosing regimens of darbepoetin alfa in patients with heart failure and anaemia. Eur Heart J 00;:0-.. Ghali JK, Anand IS, Abraham WT, et al. Randomized double-blind trial of darbepoetin alfa in patients with symptomatic heart failure and anemia. Circulation 00;:-.. Kourea K, Parissis JT, Farmakis D, et al. Effects of darbepoetin-alpha on quality of life and emotional stress in anemic patients with chronic heart failure. Eur J Cardiovasc Prev Rehabil 00;:-.. Van der Meer P, Groenveld H, Januzzi JL, et al. Erythropoietin Treatment in Patients with Chronic Heart Failure: a meta-analysis. Heart Agostoni P, Wasserman K, Guazzi M, et al. Exercise-induced hemoconcentration in heart failure due to dilated cardiomyopathy. Am J Cardiol ;:-0, A.. Agostoni P, Cerino M, Palermo P, et al. Exercise capacity in patients with betathalassaemia intermedia. Br J Haematol 00;:-.

17 Page of 0 American Journal of Hematology Table I HF patients characteristics Total population n=0 Anaemic n= Not-anaemic n= p Age (year). ± 0.. ± 0.. ± 0. <0.00 Gender m (%) (%) (%) N.S. f (%) (%) 0 (%) N.S. NYHA I (%) (%) II (%) (%) <0.00 III 0 (%) (%) LVEF (%). ± 0.. ± 0.. ± 0. N.S. LVDV (ml) ± 0 ± ± N.S. Haemoglobin (g/dl). ±0.0. ± 0.0. ± 0.0 <0.00 Weight (Kg). ± 0.. ±. ± Therapy B-Blockers (%) (%) (%) N.S. Diuretics (%) (%) (%) <0.00 Ace-inhibitor (%) (%) (%) N.S. ARBs (0%) 0 (%) (%) N.S. Spironolactone (0%) (%) (%) N.S. LVEF= Left Ventricular Ejection Fraction; LVDV= Left Ventricular Diastolic Volume

18 American Journal of Hematology Page of Table II Cardiopulmonary exercise test results Total population Anaemic Not-anaemic p Peak O (ml/min) ± ± ± <0.00 Peak O (ml/kg/min). ± 0.. ± 0..0 ± 0. <0.00 Peak O (% of predicted value) 0 ± 0 ± ± <0.00 E/ CO slope ± 0 ± ± 0 <0.00 O /Work (ml/min Watt). ± 0.. ± 0.. ± 0. N.S. O AT (ml/min) 0 ± 0 ± ± <0.00 HR at rest (b/min) ± ± ± N.S. HR max (b/min) 0 ± ± ± <0.00 Delta HR (b) ± ± ± <0.00 Work load (Watt) ± ± ± <0.00 Exercise time (min). ± 0.. ± 0.. ± 0. <0.00 O = oxygen uptake. E / CO = ventilation/carbon dioxide slope. AT = anaerobic threshold. HR = heart rate.

19 Page of 0 American Journal of Hematology Table III Linear regressions between peak O and O at the anaerobic threshold vs. haemoglobin at rest Total population HF+CAD HF no-cad Peak O = Hb - R = 0.0 Peak O = Hb - R = 0. Peak O = Hb - R = 0. O AT = Hb - 00 R = 0. O AT = Hb - 0 R = 0. O AT = Hb - R = 0. O (ml/min) = oxygen uptake. AT = anaerobic threshold. HF = heart failure. CAD = coronary artery disease. Hb (g/dl) = haemoglobin

20 American Journal of Hematology Page of Table IV Monovariate correlations between peak O and several variables Age 0. <0.000 Gender 0.0 <0.000 LVEF(%) 0.0 <0.000 Hb (g/dl) 0. <0.000 R E/ CO slope 0. <0.000 HR peak (b/m) 0. <0.000 Work load (Watts) 0. <0.000 NYHA class 0. <0.000 O = oxygen uptake. E / CO = ventilation/carbon dioxide slope. HR = heart rate. Hb = haemoglobin p

21 Page of 0 American Journal of Hematology Peak oxygen uptake in anaemic (white bar) and not anaemic (black bar) HF patients. Entire HF population (upper panel), HF males (middle panel) and HF females (lower panel). * = p<0.0 xmm ( x DPI)

22 American Journal of Hematology Page 0 of Linear regression between peak VO (oxygen uptake) and haemoglobin concentration in HF patients (n=0). xmm (0 x 0 DPI)

23 Page of 0 American Journal of Hematology Linear regression between VOAT (oxygen uptake at anaerobic threshold) and haemoglobin concentration in HF patients (n=). xmm (0 x 0 DPI)