- Email: [email protected]
Peak oxygen consumption and prognosis in heart failure
International Journal of Cardiology 167 (2013) 157–161
Contents lists available at SciVerse ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Peak oxygen consumption and prognosis in heart failure 14 mL/kg/min is not a “gender-neutral” reference Ugo Corrà a,⁎, Alessandro Mezzani a, Andrea Giordano b, Massimo Pistono a, Marco Gnemmi a, Roberto Caruso a, Pantaleo Giannuzzi a a b
Division of Cardiology, Salvatore Maugeri Foundation, IRCCS, Veruno (NO), Italy Bioengineering Department, Salvatore Maugeri Foundation, IRCCS, Via per Revislate 13, 28010 Veruno (NO), Italy
a r t i c l e
i n f o
Article history: Received 25 July 2011 Received in revised form 18 November 2011 Accepted 17 December 2011 Available online 14 January 2012 Keywords: Women Heart failure Exercise testing Prognosis
a b s t r a c t Background: Peak oxygen consumption (VO2) predictive authority in heart failure (HF) has been established from male cohorts. We evaluated the gender impact on the prognostic meaning of low peak VO2. Methods: We followed 529 HF patients (116 female), with peak VO2 ≤ 14 mL/kg/min, until cardiovascular death (CVD) and urgent heart transplantation. Results: During follow up, 156 (29%) patients had cardiac events. Female gender, age, left ventricular ejection fraction, peak VO2, peak systolic blood pressure, and beta-blocker treatment all contributed to increase the risk ability of the hierarchical multivariate model. Two-year survival was higher in women: 85 vs 66%; χ 2 = 15.7, p b 0.0001. Peculiarly, outcome results were similar when only CVD was considered. Females showed a multivariate adjusted hazard ratio (HR) of 0.46. Since a 1-mL/kg/min increment in peak VO2 was equated with a 12% improvement in prognosis, the same gender adjusted HR was achieved when mean peak VO2 was reduced by 5 units in women: thus, a HF woman with peak VO2 of 9 mL/kg/min has the same 2-year outcome as a HF man with peak VO2 of 14 mL/kg/min. Conclusions: Although HF women have a comparatively lower peak VO2 than men, they live longer. We discovered that the traditional cut point value for peak VO2, i.e. ≤ 14 mL/kg/min is not a “gender-neutral” reference since lumping HF men and women together with the same VO2 value is unlikely to describe the true risk. These preliminary findings do underline the need to assimilate gender-specific issues into clinical practice in HF, when appropriate. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In the past decade, the assumption that women and men react comparably to cardiovascular diseases and drugs has been challenged, and important gender-based differences have been demonstrated [1] indicating that women have been relegated to a Cinderella role. Even though exercise capacity is a strong independent predictor of death among men and women [2,3], in heart failure (HF) the predictive value of peak oxygen consumption (VO2) was established from male-only cohorts [4,5]. Despite the fact that women with HF constitute a significant proportion of heart transplant candidates [6], the peak VO2 reference value for heart transplantation (HT) timing is “male” biased [7]. Hence, the present study was conceived to evaluate the gender impact on the prognostic meaning of low peak VO2 in HF and to
⁎ Corresponding author at: Laboratory for the Analysis of Cardio-respiratory Signals, Divisione Cardioangiologia Riabilitativa, Fondazione Salvatore Maugeri, I.R.C.C.S. Istituto Scientifico di Veruno, Via per Revislate, 13, 28010, Veruno (NO), Italy. Tel.: + 39 0322 884711; fax: + 39 0322 830294. E-mail address: [email protected] (U. Corrà). 0167-5273/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2011.12.055
determine if a different cutoff level may be appropriate for women when considering HT candidacy.
2. Methods 2.1. Study population The study cohort consisted of consecutive patients with chronic HF due to ischemic disease or idiopathic dilated cardiomyopathy, referred to the Salvatore Maugeri Foundation, Veruno Scientific Institute for cardiopulmonary exercise testing (CPET) as part of their functional evaluation. The present report is a retrospective analysis of a prospective study. Data were extracted from the Institute's ergospirometry laboratory database, in which they are prospectively recorded and updated on a regular basis: both clinical characteristics and events at follow up are revised on a yearly basis for each surviving patient, by means of a predetermined institutional protocol designed for internal quality-control activities. The database for this prospective research project was approved by the Institutional Review Board of the Salvatore Maugeri Foundation, and written informed consent was obtained from all patients prior to CPET. Eligibility criteria were: 1) echocardiographic left ventricular ejection fraction (LVEF) ≤40%; 2) ability to perform a symptom-limited CPET, stopped for fatigue and/or dyspnea with a peak respiratory exchange ratio (RER) ≥ 1.05, in order to avoid inappropriate prognostic stratification due to poor patient motivation [8]; and 3) clinical stability defined as no change in NYHA class or absence of hospitalization for HF and stable medical treatment during the month before CPET.
158
U. Corrà et al. / International Journal of Cardiology 167 (2013) 157–161
Medical treatment administered on the day of CPET was recorded. Patients with primary valve disease, myocardial infarction within the previous three months, unstable angina, peripheral vascular disease, chronic lung disease, neuromuscular disease or with orthopedic limitation were proscribed. Echocardiographic evaluations (HewlettPackard, Agilent, Sonos 5000 and GE Vivid 7, Healthcare Technologies Ultrasound, EchoPac) were performed within 5 ± 2 days of CPET in stable clinical and pharmacological conditions. Left ventricular ejection fraction (LVEF) was measured as described elsewhere [9]. CPET was performed on a bicycle ergometer with a ramp protocol of 10 W/min and with breath-by-breath respiratory gas exchange measured by a computerized metabolic cart (Sensormedics, Vmax29, Yorba Linda, CA). Details of the test protocol have been published before [8]. Blood pressure was measured manually at rest and every 3 min during incremental exercise and at peak. The electrocardiogram and heart rate were monitored at rest and throughout exercise at 1-min intervals. Peak VO2 was recorded as the mean value of VO2 during the last 30 s of the test, while percentage of predicted peak VO2 (%ppVO2) was determined using a gender-, age-, height- and weight-adjusted and protocol-specific formula [10]. The ventilatory anaerobic threshold (VAT) was detected using the V-slope method [11]. The slope of minute ventilation versus carbon dioxide production (VE/VCO2 slope) was calculated as a linear regression function from the whole exercise period. Oxygen uptake efficiency slope (OUES), was measured as the relationship between VO2 and log10VE [12]. Peak end-tidal CO2 partial pressure (PET-CO2) was recorded at peak exercise, as a 60-second averaged value [13]. Patients were followed up at the outpatient clinic of our hospital and patients' status was determined from the medical records. The follow-up of those who did not attend their scheduled appointments was obtained by telephone interview of the patient, patient's family or primary care physician. The event censoring was interrupted at 24 months for surviving patients. Urgent status I HT and cardiovascular death (CVD) were considered events. Patients who died of non cardiac-related causes or underwent non-urgent heart transplantation were evaluated as “censored”. 2.2. Statistics Continuous data are expressed as means± SD. Student's t-test for non-paired values was used to compare the means of groups for quantitative variables. For qualitative variables, the chi-squared test with Yates' correction or Fisher's exact test, if necessary, was used. The level of statistical significance was set at a 2-tailed p value of b 0.05. Survival was estimated by the product-limit Kaplan–Meier method; differences between survival curves were evaluated with the log-rank χ2 statistic. The prognostic value of variables was determined using univariate and multivariable hierarchical Cox proportionalhazards regression analysis [14]. Because standard multivariable modeling may not adequately account for differences in baseline characteristics, a gender-specific propensity matching analysis was performed [15]. Using logistic regression, we calculated a propensity score for creating a gender-specific model taking into account the main gender baseline characteristics differences. Adequacy of matching was assessed by comparing baseline characteristics as well as the mean values and variances of the propensity score (AUC of ROC analysis). All calculations were performed using the STATA® 10 system (StataCorp, College Station, Texas USA).
3. Results In the time window July 1995 to January 2009, we screened a total of 618 patients with peak VO2≤14 mL/kg/min: 89 patients were excluded because of: LVEF>40% (n=12), peak RERb 1.05 (n=36), exercise limitation for other reasons than fatigue or dyspnea (n=24), and clinical instability during the 1 month before CPET (n=17). Hence, 529 patients met the inclusion criteria, of which 116 (22%) were female. Most patients were in sinus rhythm (85%), in NYHA classes II and III (92%), and mean LVEF and peak VO2 were 23±7% and 11.2±1.9 mL/kg/min, respectively (Tables 1 and 2): 147 (28%) had a peak VO2≤10 mL/kg/min while 377 (71%) reached a %ppVO2≤50. On average, compared with men, women had lower hemoglobin, higher LVEF and the etiology of HF was predominantly non-ischemic, while HF medications, implanted cardioverter defibrillator (ICD), cardiac rhythm (sinus rhythm, atrial fibrillation, or pacemaker-induced rhythm), and NYHA classification were similar between genders (Table 1). Moreover, mean peak VO2, OUES, and peak PeTCO2 were lower in women, while %ppVO2, resting and peak systolic blood pressure (SBP) were higher in women; finally, VAT identification, VE/VCO2 slope, and peak RER were similar in men and women (Table 2). Collection of follow-up data on patients stopped on March 28st, 2011, and outcome data were obtained for all patients. During follow up, 156 (29%) patients died of cardiovascular reasons or underwent urgent Status I HT: mean follow up was 735 ± 31 for survivors, and
Table 1 Patients' demographic and clinical characteristics according to gender.
Patients Age (yrs) Body mass index (kg/m2) Hemoglobin (gr/dl) Etiology of heart failure: ischemic Implanted cardioverter defibrillator Sinus rhythm Atrial fibrillation Pacemaker-induced rhythm NYHA I class NYHA II class NYHA III class Beta-blockers ACE-inhibitors Angiotensin II receptor blockers Loop diuretics Spironolactone Left ventricular ejection fraction (%)
All
Men
Women
529 60 ± 10 25 ± 4 13.1 ± 1.6 336 (63) 175 (33) 449 (85) 60 (11) 20 (4) 39 (8) 274 (52) 216 (40) 287 (47) 483 (91) 46 (8) 481 (91) 226 (50) 23 ± 7
413 60 ± 9 26 ± 4 13.2 ± 1.6 272 (66) 142 (35) 343 (83) 53 (13) 17 (4) 25 (6) 212 (51) 176 (43) 220 (53) 378 (92) 35 (8) 380 (92) 216 (52) 22 ± 7
116 60 ± 11 25 ± 5 12.5 ± 1.3† 64 (55)⁎ 33 (28) 106 (91) 7 (7) 3 (3) 14 (12) 62 (53) 40 (35) 67 (58) 105 (90) 11(9) 101(87) 50 (43) 27 ± 7†
Data are expressed as mean value ± SD or number (%) of patients. NYHA = New York Heart Association; ACE = Angiotensin Converting Enzyme. ⁎ p b 0.05 between men and women. † p b 0.01 between men and women.
344 ± 226 for non-survivors. Female gender, age, NYHA classification, ICD, LVEF, beta-blockers prescription, peak SBP, peak VO2, %ppVO2, peak RER, non identification of VAT, VE/VCO2 slope, OUES and peak PetCO2 were all significantly (p b 0.01) related to outcome at univariate Cox proportional hazards analysis (Table 3): although %ppVO2 reached the highest χ 2 value, indicating that correction for factors that affect peak VO2, such as age and gender (besides body weight), enhances its predictive accuracy, %ppVO2 was not included in the multivariate model, since the study was designed to address the prognostic correlation between gender and peak VO2. LVEF, peak VO2, peak SBP, female gender, age, and beta-blocker treatment contributed to significantly increase the risk ability of the hierarchical multivariate model (Table 4), while VE/VCO2 slope, OUES, PetCO2 and non detectable VAT did not add predictive information. Since the decision to HT is influenced also by peak VO2, we provide a supplementary analysis with CVD alone as the endpoint. We exclude 17 patients who underwent Status I HT (3 female): 139 patients (14 women and 125 men) died of CVD and predictive results were materially unchanged from those considering the combined end point, either in terms of parameters included in the multivariable model or as hazard of risk.
Table 2 Ergospirometric characteristics according to gender.
Patients Mean peak VO2 (mL/kg/min) Mean % ppVO2 (%) VAT detected (n. patients—%) Peak RER VE/VCO2 slope Peak HR (beats/min) Peak SBP (mm Hg) OUES PetCO2 (mm Hg)
All
Men
Women
529 11.2 ± 1.9 44 ± 12 288 (54) 1.14 ± 0.07 36 ± 9 120 ± 20 143 ± 26 1160 ± 331 31.1 ± 4.9
413 11.4 ± 1.9 42 ± 10 231 (56) 1.14 ± 0.06 37 ± 9 120 ± 21 141 ± 27 1217 ± 335 30.6 ± 4.8
116 10.6 ± 1.9† 53 ± 15† 57 (49) 1.15 ± 0.07 35 ± 8 121 ± 17 148 ± 24⁎ 960 ± 221† 32.5 ± 4.9†
Data are expressed as mean value ± SD or number (%) of patients. VO2 = oxygen consumption, % predicted peak VO2 = %ppVO2, VAT = ventilatory anaerobic threshold, RER = respiratory exchange ratio, VE/VCO2 slope = the slope of minute ventilation versus carbon dioxide production. HR = heart rate; SBP = systolic blood pressure, OUES = oxygen uptake efficiency slope, PET-CO2 = partial pressure of end-tidal CO2. ⁎ p b 0.05 between men and women. † p b 0.01 between men and women.
U. Corrà et al. / International Journal of Cardiology 167 (2013) 157–161
159
Table 3 Clinical and cardiopulmonary parameters related to outcome at univariate Cox regression analysis.
%pp VO2 LVEF Peak SBP NYHA class Peak VO2 Beta-blockers treatment Gender—female Age ICD VE/VCO2 slope VAT—not identified Peak RER OUES Peak PetCO2
χ2
p value
HR
95%CI
67.6 53.9 39.2 25.1 18.2 14.7 14.0 12.4 11.8 14.17 8.60 8.52 18.27 34.13
b 0.0001 b 0.0001 b 0.0001 b 0.0001 b 0.001 0.0001 0.0002 0.0004 0.0006 0.000 0.0034 0.0035 0.0001 0.0001
0.93 0.91 0.97 2.08 0.84 0.53 0.38 0.97 0.50 1.02 1.60 6.9 0.99 0.91
0.92–0.95 0.89–0.93 0.97–0.98 1.56–2.72 0.78–0.91 0.36–0.73 0.23–0.63 0.96–0.99 0.34–0.74 1.01–1.04 1.17–2.19 1.89–25.6 0.99–0.99 0.88–0.94
For abbreviations see Tables 1 and 2. LVEF = left ventricular ejection fraction. ICD = implanted cardioverter defibrillator. HR = hazard ratio. CI: interval of confidence.
With reference to gender, 15% (17/116) of females and 33% (139/ 413) of males had cardiac events, and 2-year survival was significantly higher in women: 85 vs 66%; χ2 = 15.7, p b 0.0001 (Fig. 1). Two-year survival was superior either in females with peak VO2 ≤ 10 mL/kg/min (77% vs 49%; χ2 = 10.03, p = 0.0016), or with peak VO2 > 10 and ≤14 mL/kg/min (92% vs 75%; c2= 10.75, p = 0.001). Of note, 2-year survival was similar in females with peak VO2 ≤ 10 mL/kg/min and in males with peak VO2 > 10 and ≤14 mL/kg/min. In addition, to account for differences in baseline characteristics, we matched 116 men of 413 with 116 women using propensity score matching (1:1 matching without replacement). There was excellent matching of baseline characteristics (other than peak VO2), with an AUC of 0.82 of the model with ROC analysis indicating a strong discrimination. Even after this “correction”, 1- and 2-year outcome was better in female compared to male (p b 0.05). Overall, females showed a multivariate adjusted hazard ratio of 0.46 that means a 54% lower 2-year event risk compared to males. In view of the fact that, when treated as a continuous variable, a 1-mL/kg/min increment in pVO2 was equated with a 12% improvement in prognosis (hazard ratio of 0.88), the same gender adjusted hazard ratio was achieved when mean pVO2 was reduced by 5 units (1 unit = 1 mL/ kg/min) in women. That is to say that a woman with HF and a peak VO2 of 9 mL/kg/min has the same 2-year outcome as a man with HF and peak VO2 of 14 mL/kg/min. 4. Discussion 4.1. Study findings The present prospective study was designed to address the impact of gender differences on prognostic significance of low peak VO2 in HF. Various key points merit comment: first, on average, women had a better outcome compared to men in the present cohort of severe HF with peak VO2 ≤ 14 mL/kg/min; second, female gender Table 4 Multivariate hierarchical Cox proportional-hazards regression analysis for urgent Status I heart transplantation, and cardiac-related death events. Predictive model
Parameter
Likelihood
Harrel C
1 2 3 4 5 6
LVEF Model Model Model Model Model
58.72 70.42 83.21 93.66 99.61 117.30
0.67 0.69 0.71 0.72 0.73 0.74
1 + pVO2 2 + pSBP 3 + gender 4 + age 5 + beta-blockers prescription
For abbreviations see Tables 1–3.
Fig. 1. Two-year Kaplan–Meier survival curves according to gender.
significantly contributed to 2-year prognosis stratification at multivariate hierarchical Cox regression analysis, in addition to traditional risk factors; and third, a reduction of 5 mL/kg/min of peak VO2 in women provides an equivalent adjusted 2-year risk to the peak VO2 value of ≤14 mL/kg/min in men. In HF, women show distinct differences compared to men in regard to epidemiology, risk factors, mechanisms of disease development, response to therapy, and outcome [1,16–18]. Although gender differences in etiology, hemodynamic deterioration and adaptation, renal function, metabolism, thrombotic hemostasis, rennin–angiotensin system activation, inflammatory mechanisms, and co-morbidities may all in a mixture contribute to distinctive disease progression, once patients become symptomatic, women appear to have an overall better outlook [19,20]. In contrast, on average, peak VO2 is lower in women with HF [21,22]. Female exercise impairment, compared to that of men, may be attributed to less physically active life-style, lower fat-free mass, smaller skeletal muscle fiber area, lower oxygen carrying capacity, i.e. lower hemoglobin concentration and blood volume, and smaller degree of stroke volume augmentation [21]. On average, a 4.1 mL/kg/min difference in peak VO2 between males and females after adjustment of age, peak HR, peak RER, LVEF, and etiology of HF has been reported [22]. Hence, women with HF have a significantly different clinical presentation, functional capacity, morbidity and mortality compared to men, and sex differences should be considered in formulating guidelines for risk stratification. In the present study, we focused on the gender impact on shortterm prognosis in patients with severe HF with reduced peak VO2, and we have shown that female gender provides significant predictive information on top of that of traditional risk indexes. In addition, we show that the combination of anthropometric characteristics (age and gender), severity of left ventricular systolic dysfunction (LVEF), and exercise parameters (peak VO2 and peak SBP) offers the best prognostic model. Hence, female gender and peak VO2 give independent outcome advice, and both should be taken into consideration for appropriate risk stratification. Of note, the gender specific predictive difference was confirmed after sample matching by means of propensity score. Gender impact on the prognostic value of peak VO2 in HF has to date been poorly evaluated, insofar as most investigations have been carried out on selected middle-aged male HT candidates [7,8,23,24]. The underrepresentation of women in CPET trials may be related to several facts: namely that, in the past, HF has been viewed as a male disease, that women have a greater incidence of diastolic HF, and that women
160
U. Corrà et al. / International Journal of Cardiology 167 (2013) 157–161
constitute a large proportion of patients in the older age-group outside the candidacy domain for HT. In 1997, Richard et al. [25] demonstrated, in 72 HF patients, that women had a better 12-month survival compared with men, despite mean lower peak VO2. Elmariah et al. [26] confirmed that females had a better 1-year transplant-free survival in 594 HF patients (167 women): 271 patients had a peak VO2≤ 14 mL/kg/ min, and female survival was superior also in this subgroup. Guazzi et al. [27] evaluated 412 HF patients, with 72 women: women had a lower peak VO2, but 1-year survival was comparable. Results were confirmed by Green et al. [28], who studied 552 HT candidates (274 women and 278 men) matched by referral year: again, outcome was similar, although peak VO2 was lower in women. Finally, Hsich et al. [29] evaluated 525 women of a 2105 HF cohort; peak VO2 was a strong predictor of mortality in both women and men, though for any given peak VO2 value women were at lower risk. All together, these studies demonstrate that women with HF have a lower peak VO2 compared to men, and the “relatively” lower peak VO2 was not associated with a worse outcome. We confirm and extend these previous findings, since we focused on the most challenging HF category, where important decisions, including HT candidature, heavily rely on peak VO2 [30]. In this specific HF subset, we confirmed that peak VO2 is a short-term predictor of cardiac events in both women and men, and that, although women had a lower mean peak VO2, they go on better, and less benefit of HT might be seen. Of note, the female outcome superiority was substantiated after excluding HT as an end point to avoid a clinical bias since physicians use peak VO2 to determine eligibility. In practice, the traditional peak VO2 cut-off value for HT candidate selection might be disproportionate in women, and, if adopted, women might be referred to HT too soon. This is a key aspect, since the paucity of donor hearts, mortality rates on the transplant waiting list, high costs, and challenges of post-surgical care render HT a feasible option for only a small fraction of advanced HF patients. Hence, precise and stringent gender-based exercise criteria are absolutely necessary to select those patients who are most likely to derive a survival benefit from HT. Here, we demonstrated that women with HF equate male short-term risk even if their peak VO2 is significantly reduced: a woman with HF and peak VO2 of 9 mL/kg/min has the same 2-year outcome as a man with peak VO2 of 14 mL/kg/min. This finding might indicate the need for a new “exercise” predictive paradigm for HT candidacy, incorporating both peak VO2 and gender: i.e. a man with HF should be considered a HT candidate if peak VO2 is ≤14 mL/kg/min, while a woman should contemplated only when peak is ≤10 mL/kg/min. As ancillary finding, we documented that VE/VCO2 slope, although significantly related to outcome at univariate analysis, was not selected as independent risk predictor using the multivariable screening: the exclusion seems, at first glance, in contra tendency in regards to other experiences [31,32], nonetheless the predictive value of VE/VCO2 slope in the setting of very low peak VO2 has not been investigated, yet. Further studies are needed to address this specific issue. 4.2. Limitations In our relatively modest sample size of females, the number of statistical tests performed could potentially inflate Type I error. In view of this we adopted a restrictive approach, including in the predictive model only pb 0.01 parameters at univariate analysis, and selected only established prognostic indexes in HF to build the multivariate model; indeed, the importance of the parameters chosen strengthens our findings. In addition, recently peak VO2 criteria for HT selection have shifted from 14 to 10 mL/kg/min [30], but, although females had better survival even in this subgroup, the limited number of patients precludes any distinctive conclusion. Moreover, the study population comprised a relatively low percent of patient chronically treated with beta-blockers, due to the extended time span of enrollment. Two-year mortality rates were 22% and
37% in patient with and without beta-blocker treatment, respectively. Although the maintenance of beta-blocker therapy during the follow up was not ascertained, a supplementary analysis according to betablockers prescription would certainly help to define the interaction between gender and therapy, but unfortunately, an accurate predictive analysis was precluded because of the paucity of events in the female subgroup with beta-blockers treatment (8 deaths). Finally, although the study was designed to address the prognostic correlation between gender and peak VO2, we agree that the best way to assess the gender impact on aerobic capacity is described by the use of ppVO2: this relies on the fact that the formula for ppVO2 incorporates both age and gender. As confirmation, the inclusion in of ppVO2 in the predictive multivariable model excludes age, gender and peak VO2 (data not shown). 5. Conclusions Gender consideration is an old issue that has been buried and needs to resurface. Though peak VO2 preserves its prognostic value in both HF men and women with impaired exercise capacity, women – although they have a comparatively lower peak VO2 than men – live longer. We discovered that the traditional cut point value for peak VO2, i.e. ≤14 mL/kg/min is not a “gender-neutral” reference since lumping HF men and women together with the same VO2 value is unlikely to describe the true risk of events. There is a clear need to integrate genderspecific issues into everyday clinical practice, when appropriate. But we have to admit that, while the challenge of modifying prognosticating peak VO2 paradigm for HT candidacy is an intrepid one, changing physicians' management of HF patients in the light of gender may be an even greater challenge. Acknowledgments The authors are grateful to Alfio Agazzone, Elena Bonanomi and Barbara Temporelli for technical support, to Fabio Comazzi for statistical assistance, and to Rosemary Allpress for her careful revision of the English manuscript. The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology. References [1] Jessup M, Pina IL. Is it important to examine gender differences in the epidemiology and outcome of severe heart failure? J Thorac Cardiovasc Surg 2004;127: 1247–52. [2] Ekelund LG, Haskell WL, Johnson JL, Whaley FS, Criqui MH, Sheps DS. Physical fitness as a predictor of cardiovascular mortality in asymptomatic North American men. The Lipid Research Clinics Mortality Follow-up Study. N Engl J Med 1988;319:1379–84. [3] Gulati M, Black HR, Shaw LJ, et al. The prognostic value of a nomogram for exercise capacity in women. N Engl J Med 2005;353:468–75. [4] Piepoli MF, Corrà U, Agostoni PG, et al. Statement on cardiopulmonary exercise testing in chronic heart failure due to left ventricular dysfunction: recommendations for performance and interpretation. Part III: interpretation of cardiopulmonary exercise testing in chronic heart failure and future applications. Eur J Cardiovasc Prev Rehabil 2006 Aug;13(4):485–94. [5] Balady GJ, Arena R, Sietsema K, et al. Clinician's guide to cardiopulmonary exercise testing in adults. A scientific statement from the American Heart Association. Circulation 2010;122:191–225. [6] Lietz KW, Miller LW. Improved survival of patients with end-stage heart failure listed for heart transplantation. Analysis of organ procurement and transplantation network/U.S. United Network of Organ Sharing Data, 1990 to 2005. J Am Coll Cardiol 2007;50:1282–90. [7] Mancini DM, Eisen H, Kussmaul W, Mull R, Edmund LH, Wilson JR. Value of peak oxygen consumption for optimal timing of cardiac transplantation in ambulatory patients with heart failure. Circulation 1991;83:778–86. [8] Mezzani A, Corrà U, Bosimini E, Giordano A, Giannuzzi P. Contribution of peak respiratory exchange ratio to peak VO2 prognostic reliability in patients with chronic heart failure and severely reduced exercise capacity. Am Heart J 2003;145: 1102–7. [9] Giannuzzi P, Imparato A, Temporelli PL, et al. Doppler-derived mitral deceleration time of early filling as a strong predictor of pulmonary capillary wedge pressure in post-infarction patients with left ventricular systolic dysfunction. J Am Coll Cardiol 1994;23:1630–7.
U. Corrà et al. / International Journal of Cardiology 167 (2013) 157–161 [10] Wasserman K, Hansen JE, Sue DY, Casaburi R, Whipp BJ. Measurements during integrative cardiopulmonary exercise testing. In: Wasserman K, Hansen JE, Sue DY, Casaburi R, Whipp BJ, editors. Principles of exercise testing and interpretation. Baltimore, MA: Lippincott Williams & Williams; 1999. p. 62–94. [11] Beaver WL, Wasserman K, Whipps BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol 1986;60:2020–7. [12] Davies LC, Wensel R, Georgiadou P, et al. Enhanced prognostic value from cardiopulmonary exercise testing in chronic heart failure by non-linear analysis: oxygen uptake efficiency slope. Eur Heart J 2006 Mar;27(6):684–90. [13] Arena R, Guazzi M, Myers J. Prognostic value of end-tidal carbon dioxide during exercise testing in heart failure. Int J Cardiol 2007;117:103–8. [14] Cox DR, Oakes D. Analysis of survival data. London, UK: Chapman & Hall; 1984. [15] D'Agostino Jr RB. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med 1998;17:2265–81. [16] Lund LH, Mancini D. Heart failure in women. Med Clin North Am 2004;88: 1321–45. [17] Miller Leslie W. Heart failure: who we treat versus who we study. Cardiol Clin 2008;26:113–25. [18] Petrie MC, Dawson NF, Murdoch DR, Davie AP, McMurray JJV. Failure of women's hearts. Circulation 1999;99:2334–41. [19] Ho KKL, Anderson KM, Kannel WB, et al. Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation 1993;88:107–15. [20] Ghali JK, Pina IL, Gottlieb SS, et al. Metoprolol CR/XL in female patients with heart failure: analysis of the experience in MERIT-HF. Circulation 2002;105:1585–91. [21] Drinkwater BL. Women and exercise: physiological aspects. Exerc Sport Sci Rev 1984;12:24–51. [22] Daida H, Allison TG, Johnson BD, Squires RW, Gau GT. Comparison of peak exercise uptake in men versus women in chronic heart failure secondary to ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 1997;80:85–8.
161
[23] Opasich C, Pinna GD, Bobbio M, et al. Peak oxygen consumption in chronic heart failure: toward efficient use in the individual patient. J Am Coll Cardiol 1998;31: 766–75. [24] Arena R, Myers J, Abella J, et al. Development of ventilatory classification system in patients with heart failure. Circulation 2007;115:2410–7. [25] Richards DR, Mehra MR, Ventura HO, et al. Usefulness of peak oxygen consumption in predicting outcome of heart failure in women versus men. Am J Cardiol 1997;80: 1236–8. [26] Elmariah S, Goldberg LR, Allen MT, Kao A. Effects of gender on peak oxygen consumption and the timing of cardiac transplantation. J Am Coll Cardiol 2006;47:2237–42. [27] Guazzi M, Arena R, Myers J. Comparison of the prognostic value of cardiopulmonary exercise testing between male and female patients with heart failure. Int J Cardiol 2006;113:395–400. [28] Green P, Lund LH, Mancini D. Comparison of peak exercise consumption and the heart failure survival score for predicting prognosis in women and men. Am J Cardiol 2007;99:399–403. [29] Hsich E, Chadalavada S, Krishnaswamy G, et al. Long-term prognostic value of peak oxygen consumption in women versus men with heart failure and severely impaired left ventricular systolic function. Am J Cardiol 2007;100:291–5. [30] Mancini D, Lietz K. Selection of cardiac transplantation candidates in 2010. Circulation 2010;122:173–83. [31] Francis DP, Shamim W, Davies LC, et al. Cardiopulmonary exercise testing for prognosis in chronic heart failure: continuous and independent prognostic value from VE/VCO2 slope and peak VO2. Eur Heart J 2000;21:154–61. [32] Arena R, Myers J, Abella J, et al. Development of a ventilatory classification system in patients with heart failure. Circulation 2007;115:2410–7.
Contents lists available at SciVerse ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Peak oxygen consumption and prognosis in heart failure 14 mL/kg/min is not a “gender-neutral” reference Ugo Corrà a,⁎, Alessandro Mezzani a, Andrea Giordano b, Massimo Pistono a, Marco Gnemmi a, Roberto Caruso a, Pantaleo Giannuzzi a a b
Division of Cardiology, Salvatore Maugeri Foundation, IRCCS, Veruno (NO), Italy Bioengineering Department, Salvatore Maugeri Foundation, IRCCS, Via per Revislate 13, 28010 Veruno (NO), Italy
a r t i c l e
i n f o
Article history: Received 25 July 2011 Received in revised form 18 November 2011 Accepted 17 December 2011 Available online 14 January 2012 Keywords: Women Heart failure Exercise testing Prognosis
a b s t r a c t Background: Peak oxygen consumption (VO2) predictive authority in heart failure (HF) has been established from male cohorts. We evaluated the gender impact on the prognostic meaning of low peak VO2. Methods: We followed 529 HF patients (116 female), with peak VO2 ≤ 14 mL/kg/min, until cardiovascular death (CVD) and urgent heart transplantation. Results: During follow up, 156 (29%) patients had cardiac events. Female gender, age, left ventricular ejection fraction, peak VO2, peak systolic blood pressure, and beta-blocker treatment all contributed to increase the risk ability of the hierarchical multivariate model. Two-year survival was higher in women: 85 vs 66%; χ 2 = 15.7, p b 0.0001. Peculiarly, outcome results were similar when only CVD was considered. Females showed a multivariate adjusted hazard ratio (HR) of 0.46. Since a 1-mL/kg/min increment in peak VO2 was equated with a 12% improvement in prognosis, the same gender adjusted HR was achieved when mean peak VO2 was reduced by 5 units in women: thus, a HF woman with peak VO2 of 9 mL/kg/min has the same 2-year outcome as a HF man with peak VO2 of 14 mL/kg/min. Conclusions: Although HF women have a comparatively lower peak VO2 than men, they live longer. We discovered that the traditional cut point value for peak VO2, i.e. ≤ 14 mL/kg/min is not a “gender-neutral” reference since lumping HF men and women together with the same VO2 value is unlikely to describe the true risk. These preliminary findings do underline the need to assimilate gender-specific issues into clinical practice in HF, when appropriate. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In the past decade, the assumption that women and men react comparably to cardiovascular diseases and drugs has been challenged, and important gender-based differences have been demonstrated [1] indicating that women have been relegated to a Cinderella role. Even though exercise capacity is a strong independent predictor of death among men and women [2,3], in heart failure (HF) the predictive value of peak oxygen consumption (VO2) was established from male-only cohorts [4,5]. Despite the fact that women with HF constitute a significant proportion of heart transplant candidates [6], the peak VO2 reference value for heart transplantation (HT) timing is “male” biased [7]. Hence, the present study was conceived to evaluate the gender impact on the prognostic meaning of low peak VO2 in HF and to
⁎ Corresponding author at: Laboratory for the Analysis of Cardio-respiratory Signals, Divisione Cardioangiologia Riabilitativa, Fondazione Salvatore Maugeri, I.R.C.C.S. Istituto Scientifico di Veruno, Via per Revislate, 13, 28010, Veruno (NO), Italy. Tel.: + 39 0322 884711; fax: + 39 0322 830294. E-mail address: [email protected] (U. Corrà). 0167-5273/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2011.12.055
determine if a different cutoff level may be appropriate for women when considering HT candidacy.
2. Methods 2.1. Study population The study cohort consisted of consecutive patients with chronic HF due to ischemic disease or idiopathic dilated cardiomyopathy, referred to the Salvatore Maugeri Foundation, Veruno Scientific Institute for cardiopulmonary exercise testing (CPET) as part of their functional evaluation. The present report is a retrospective analysis of a prospective study. Data were extracted from the Institute's ergospirometry laboratory database, in which they are prospectively recorded and updated on a regular basis: both clinical characteristics and events at follow up are revised on a yearly basis for each surviving patient, by means of a predetermined institutional protocol designed for internal quality-control activities. The database for this prospective research project was approved by the Institutional Review Board of the Salvatore Maugeri Foundation, and written informed consent was obtained from all patients prior to CPET. Eligibility criteria were: 1) echocardiographic left ventricular ejection fraction (LVEF) ≤40%; 2) ability to perform a symptom-limited CPET, stopped for fatigue and/or dyspnea with a peak respiratory exchange ratio (RER) ≥ 1.05, in order to avoid inappropriate prognostic stratification due to poor patient motivation [8]; and 3) clinical stability defined as no change in NYHA class or absence of hospitalization for HF and stable medical treatment during the month before CPET.
158
U. Corrà et al. / International Journal of Cardiology 167 (2013) 157–161
Medical treatment administered on the day of CPET was recorded. Patients with primary valve disease, myocardial infarction within the previous three months, unstable angina, peripheral vascular disease, chronic lung disease, neuromuscular disease or with orthopedic limitation were proscribed. Echocardiographic evaluations (HewlettPackard, Agilent, Sonos 5000 and GE Vivid 7, Healthcare Technologies Ultrasound, EchoPac) were performed within 5 ± 2 days of CPET in stable clinical and pharmacological conditions. Left ventricular ejection fraction (LVEF) was measured as described elsewhere [9]. CPET was performed on a bicycle ergometer with a ramp protocol of 10 W/min and with breath-by-breath respiratory gas exchange measured by a computerized metabolic cart (Sensormedics, Vmax29, Yorba Linda, CA). Details of the test protocol have been published before [8]. Blood pressure was measured manually at rest and every 3 min during incremental exercise and at peak. The electrocardiogram and heart rate were monitored at rest and throughout exercise at 1-min intervals. Peak VO2 was recorded as the mean value of VO2 during the last 30 s of the test, while percentage of predicted peak VO2 (%ppVO2) was determined using a gender-, age-, height- and weight-adjusted and protocol-specific formula [10]. The ventilatory anaerobic threshold (VAT) was detected using the V-slope method [11]. The slope of minute ventilation versus carbon dioxide production (VE/VCO2 slope) was calculated as a linear regression function from the whole exercise period. Oxygen uptake efficiency slope (OUES), was measured as the relationship between VO2 and log10VE [12]. Peak end-tidal CO2 partial pressure (PET-CO2) was recorded at peak exercise, as a 60-second averaged value [13]. Patients were followed up at the outpatient clinic of our hospital and patients' status was determined from the medical records. The follow-up of those who did not attend their scheduled appointments was obtained by telephone interview of the patient, patient's family or primary care physician. The event censoring was interrupted at 24 months for surviving patients. Urgent status I HT and cardiovascular death (CVD) were considered events. Patients who died of non cardiac-related causes or underwent non-urgent heart transplantation were evaluated as “censored”. 2.2. Statistics Continuous data are expressed as means± SD. Student's t-test for non-paired values was used to compare the means of groups for quantitative variables. For qualitative variables, the chi-squared test with Yates' correction or Fisher's exact test, if necessary, was used. The level of statistical significance was set at a 2-tailed p value of b 0.05. Survival was estimated by the product-limit Kaplan–Meier method; differences between survival curves were evaluated with the log-rank χ2 statistic. The prognostic value of variables was determined using univariate and multivariable hierarchical Cox proportionalhazards regression analysis [14]. Because standard multivariable modeling may not adequately account for differences in baseline characteristics, a gender-specific propensity matching analysis was performed [15]. Using logistic regression, we calculated a propensity score for creating a gender-specific model taking into account the main gender baseline characteristics differences. Adequacy of matching was assessed by comparing baseline characteristics as well as the mean values and variances of the propensity score (AUC of ROC analysis). All calculations were performed using the STATA® 10 system (StataCorp, College Station, Texas USA).
3. Results In the time window July 1995 to January 2009, we screened a total of 618 patients with peak VO2≤14 mL/kg/min: 89 patients were excluded because of: LVEF>40% (n=12), peak RERb 1.05 (n=36), exercise limitation for other reasons than fatigue or dyspnea (n=24), and clinical instability during the 1 month before CPET (n=17). Hence, 529 patients met the inclusion criteria, of which 116 (22%) were female. Most patients were in sinus rhythm (85%), in NYHA classes II and III (92%), and mean LVEF and peak VO2 were 23±7% and 11.2±1.9 mL/kg/min, respectively (Tables 1 and 2): 147 (28%) had a peak VO2≤10 mL/kg/min while 377 (71%) reached a %ppVO2≤50. On average, compared with men, women had lower hemoglobin, higher LVEF and the etiology of HF was predominantly non-ischemic, while HF medications, implanted cardioverter defibrillator (ICD), cardiac rhythm (sinus rhythm, atrial fibrillation, or pacemaker-induced rhythm), and NYHA classification were similar between genders (Table 1). Moreover, mean peak VO2, OUES, and peak PeTCO2 were lower in women, while %ppVO2, resting and peak systolic blood pressure (SBP) were higher in women; finally, VAT identification, VE/VCO2 slope, and peak RER were similar in men and women (Table 2). Collection of follow-up data on patients stopped on March 28st, 2011, and outcome data were obtained for all patients. During follow up, 156 (29%) patients died of cardiovascular reasons or underwent urgent Status I HT: mean follow up was 735 ± 31 for survivors, and
Table 1 Patients' demographic and clinical characteristics according to gender.
Patients Age (yrs) Body mass index (kg/m2) Hemoglobin (gr/dl) Etiology of heart failure: ischemic Implanted cardioverter defibrillator Sinus rhythm Atrial fibrillation Pacemaker-induced rhythm NYHA I class NYHA II class NYHA III class Beta-blockers ACE-inhibitors Angiotensin II receptor blockers Loop diuretics Spironolactone Left ventricular ejection fraction (%)
All
Men
Women
529 60 ± 10 25 ± 4 13.1 ± 1.6 336 (63) 175 (33) 449 (85) 60 (11) 20 (4) 39 (8) 274 (52) 216 (40) 287 (47) 483 (91) 46 (8) 481 (91) 226 (50) 23 ± 7
413 60 ± 9 26 ± 4 13.2 ± 1.6 272 (66) 142 (35) 343 (83) 53 (13) 17 (4) 25 (6) 212 (51) 176 (43) 220 (53) 378 (92) 35 (8) 380 (92) 216 (52) 22 ± 7
116 60 ± 11 25 ± 5 12.5 ± 1.3† 64 (55)⁎ 33 (28) 106 (91) 7 (7) 3 (3) 14 (12) 62 (53) 40 (35) 67 (58) 105 (90) 11(9) 101(87) 50 (43) 27 ± 7†
Data are expressed as mean value ± SD or number (%) of patients. NYHA = New York Heart Association; ACE = Angiotensin Converting Enzyme. ⁎ p b 0.05 between men and women. † p b 0.01 between men and women.
344 ± 226 for non-survivors. Female gender, age, NYHA classification, ICD, LVEF, beta-blockers prescription, peak SBP, peak VO2, %ppVO2, peak RER, non identification of VAT, VE/VCO2 slope, OUES and peak PetCO2 were all significantly (p b 0.01) related to outcome at univariate Cox proportional hazards analysis (Table 3): although %ppVO2 reached the highest χ 2 value, indicating that correction for factors that affect peak VO2, such as age and gender (besides body weight), enhances its predictive accuracy, %ppVO2 was not included in the multivariate model, since the study was designed to address the prognostic correlation between gender and peak VO2. LVEF, peak VO2, peak SBP, female gender, age, and beta-blocker treatment contributed to significantly increase the risk ability of the hierarchical multivariate model (Table 4), while VE/VCO2 slope, OUES, PetCO2 and non detectable VAT did not add predictive information. Since the decision to HT is influenced also by peak VO2, we provide a supplementary analysis with CVD alone as the endpoint. We exclude 17 patients who underwent Status I HT (3 female): 139 patients (14 women and 125 men) died of CVD and predictive results were materially unchanged from those considering the combined end point, either in terms of parameters included in the multivariable model or as hazard of risk.
Table 2 Ergospirometric characteristics according to gender.
Patients Mean peak VO2 (mL/kg/min) Mean % ppVO2 (%) VAT detected (n. patients—%) Peak RER VE/VCO2 slope Peak HR (beats/min) Peak SBP (mm Hg) OUES PetCO2 (mm Hg)
All
Men
Women
529 11.2 ± 1.9 44 ± 12 288 (54) 1.14 ± 0.07 36 ± 9 120 ± 20 143 ± 26 1160 ± 331 31.1 ± 4.9
413 11.4 ± 1.9 42 ± 10 231 (56) 1.14 ± 0.06 37 ± 9 120 ± 21 141 ± 27 1217 ± 335 30.6 ± 4.8
116 10.6 ± 1.9† 53 ± 15† 57 (49) 1.15 ± 0.07 35 ± 8 121 ± 17 148 ± 24⁎ 960 ± 221† 32.5 ± 4.9†
Data are expressed as mean value ± SD or number (%) of patients. VO2 = oxygen consumption, % predicted peak VO2 = %ppVO2, VAT = ventilatory anaerobic threshold, RER = respiratory exchange ratio, VE/VCO2 slope = the slope of minute ventilation versus carbon dioxide production. HR = heart rate; SBP = systolic blood pressure, OUES = oxygen uptake efficiency slope, PET-CO2 = partial pressure of end-tidal CO2. ⁎ p b 0.05 between men and women. † p b 0.01 between men and women.
U. Corrà et al. / International Journal of Cardiology 167 (2013) 157–161
159
Table 3 Clinical and cardiopulmonary parameters related to outcome at univariate Cox regression analysis.
%pp VO2 LVEF Peak SBP NYHA class Peak VO2 Beta-blockers treatment Gender—female Age ICD VE/VCO2 slope VAT—not identified Peak RER OUES Peak PetCO2
χ2
p value
HR
95%CI
67.6 53.9 39.2 25.1 18.2 14.7 14.0 12.4 11.8 14.17 8.60 8.52 18.27 34.13
b 0.0001 b 0.0001 b 0.0001 b 0.0001 b 0.001 0.0001 0.0002 0.0004 0.0006 0.000 0.0034 0.0035 0.0001 0.0001
0.93 0.91 0.97 2.08 0.84 0.53 0.38 0.97 0.50 1.02 1.60 6.9 0.99 0.91
0.92–0.95 0.89–0.93 0.97–0.98 1.56–2.72 0.78–0.91 0.36–0.73 0.23–0.63 0.96–0.99 0.34–0.74 1.01–1.04 1.17–2.19 1.89–25.6 0.99–0.99 0.88–0.94
For abbreviations see Tables 1 and 2. LVEF = left ventricular ejection fraction. ICD = implanted cardioverter defibrillator. HR = hazard ratio. CI: interval of confidence.
With reference to gender, 15% (17/116) of females and 33% (139/ 413) of males had cardiac events, and 2-year survival was significantly higher in women: 85 vs 66%; χ2 = 15.7, p b 0.0001 (Fig. 1). Two-year survival was superior either in females with peak VO2 ≤ 10 mL/kg/min (77% vs 49%; χ2 = 10.03, p = 0.0016), or with peak VO2 > 10 and ≤14 mL/kg/min (92% vs 75%; c2= 10.75, p = 0.001). Of note, 2-year survival was similar in females with peak VO2 ≤ 10 mL/kg/min and in males with peak VO2 > 10 and ≤14 mL/kg/min. In addition, to account for differences in baseline characteristics, we matched 116 men of 413 with 116 women using propensity score matching (1:1 matching without replacement). There was excellent matching of baseline characteristics (other than peak VO2), with an AUC of 0.82 of the model with ROC analysis indicating a strong discrimination. Even after this “correction”, 1- and 2-year outcome was better in female compared to male (p b 0.05). Overall, females showed a multivariate adjusted hazard ratio of 0.46 that means a 54% lower 2-year event risk compared to males. In view of the fact that, when treated as a continuous variable, a 1-mL/kg/min increment in pVO2 was equated with a 12% improvement in prognosis (hazard ratio of 0.88), the same gender adjusted hazard ratio was achieved when mean pVO2 was reduced by 5 units (1 unit = 1 mL/ kg/min) in women. That is to say that a woman with HF and a peak VO2 of 9 mL/kg/min has the same 2-year outcome as a man with HF and peak VO2 of 14 mL/kg/min. 4. Discussion 4.1. Study findings The present prospective study was designed to address the impact of gender differences on prognostic significance of low peak VO2 in HF. Various key points merit comment: first, on average, women had a better outcome compared to men in the present cohort of severe HF with peak VO2 ≤ 14 mL/kg/min; second, female gender Table 4 Multivariate hierarchical Cox proportional-hazards regression analysis for urgent Status I heart transplantation, and cardiac-related death events. Predictive model
Parameter
Likelihood
Harrel C
1 2 3 4 5 6
LVEF Model Model Model Model Model
58.72 70.42 83.21 93.66 99.61 117.30
0.67 0.69 0.71 0.72 0.73 0.74
1 + pVO2 2 + pSBP 3 + gender 4 + age 5 + beta-blockers prescription
For abbreviations see Tables 1–3.
Fig. 1. Two-year Kaplan–Meier survival curves according to gender.
significantly contributed to 2-year prognosis stratification at multivariate hierarchical Cox regression analysis, in addition to traditional risk factors; and third, a reduction of 5 mL/kg/min of peak VO2 in women provides an equivalent adjusted 2-year risk to the peak VO2 value of ≤14 mL/kg/min in men. In HF, women show distinct differences compared to men in regard to epidemiology, risk factors, mechanisms of disease development, response to therapy, and outcome [1,16–18]. Although gender differences in etiology, hemodynamic deterioration and adaptation, renal function, metabolism, thrombotic hemostasis, rennin–angiotensin system activation, inflammatory mechanisms, and co-morbidities may all in a mixture contribute to distinctive disease progression, once patients become symptomatic, women appear to have an overall better outlook [19,20]. In contrast, on average, peak VO2 is lower in women with HF [21,22]. Female exercise impairment, compared to that of men, may be attributed to less physically active life-style, lower fat-free mass, smaller skeletal muscle fiber area, lower oxygen carrying capacity, i.e. lower hemoglobin concentration and blood volume, and smaller degree of stroke volume augmentation [21]. On average, a 4.1 mL/kg/min difference in peak VO2 between males and females after adjustment of age, peak HR, peak RER, LVEF, and etiology of HF has been reported [22]. Hence, women with HF have a significantly different clinical presentation, functional capacity, morbidity and mortality compared to men, and sex differences should be considered in formulating guidelines for risk stratification. In the present study, we focused on the gender impact on shortterm prognosis in patients with severe HF with reduced peak VO2, and we have shown that female gender provides significant predictive information on top of that of traditional risk indexes. In addition, we show that the combination of anthropometric characteristics (age and gender), severity of left ventricular systolic dysfunction (LVEF), and exercise parameters (peak VO2 and peak SBP) offers the best prognostic model. Hence, female gender and peak VO2 give independent outcome advice, and both should be taken into consideration for appropriate risk stratification. Of note, the gender specific predictive difference was confirmed after sample matching by means of propensity score. Gender impact on the prognostic value of peak VO2 in HF has to date been poorly evaluated, insofar as most investigations have been carried out on selected middle-aged male HT candidates [7,8,23,24]. The underrepresentation of women in CPET trials may be related to several facts: namely that, in the past, HF has been viewed as a male disease, that women have a greater incidence of diastolic HF, and that women
160
U. Corrà et al. / International Journal of Cardiology 167 (2013) 157–161
constitute a large proportion of patients in the older age-group outside the candidacy domain for HT. In 1997, Richard et al. [25] demonstrated, in 72 HF patients, that women had a better 12-month survival compared with men, despite mean lower peak VO2. Elmariah et al. [26] confirmed that females had a better 1-year transplant-free survival in 594 HF patients (167 women): 271 patients had a peak VO2≤ 14 mL/kg/ min, and female survival was superior also in this subgroup. Guazzi et al. [27] evaluated 412 HF patients, with 72 women: women had a lower peak VO2, but 1-year survival was comparable. Results were confirmed by Green et al. [28], who studied 552 HT candidates (274 women and 278 men) matched by referral year: again, outcome was similar, although peak VO2 was lower in women. Finally, Hsich et al. [29] evaluated 525 women of a 2105 HF cohort; peak VO2 was a strong predictor of mortality in both women and men, though for any given peak VO2 value women were at lower risk. All together, these studies demonstrate that women with HF have a lower peak VO2 compared to men, and the “relatively” lower peak VO2 was not associated with a worse outcome. We confirm and extend these previous findings, since we focused on the most challenging HF category, where important decisions, including HT candidature, heavily rely on peak VO2 [30]. In this specific HF subset, we confirmed that peak VO2 is a short-term predictor of cardiac events in both women and men, and that, although women had a lower mean peak VO2, they go on better, and less benefit of HT might be seen. Of note, the female outcome superiority was substantiated after excluding HT as an end point to avoid a clinical bias since physicians use peak VO2 to determine eligibility. In practice, the traditional peak VO2 cut-off value for HT candidate selection might be disproportionate in women, and, if adopted, women might be referred to HT too soon. This is a key aspect, since the paucity of donor hearts, mortality rates on the transplant waiting list, high costs, and challenges of post-surgical care render HT a feasible option for only a small fraction of advanced HF patients. Hence, precise and stringent gender-based exercise criteria are absolutely necessary to select those patients who are most likely to derive a survival benefit from HT. Here, we demonstrated that women with HF equate male short-term risk even if their peak VO2 is significantly reduced: a woman with HF and peak VO2 of 9 mL/kg/min has the same 2-year outcome as a man with peak VO2 of 14 mL/kg/min. This finding might indicate the need for a new “exercise” predictive paradigm for HT candidacy, incorporating both peak VO2 and gender: i.e. a man with HF should be considered a HT candidate if peak VO2 is ≤14 mL/kg/min, while a woman should contemplated only when peak is ≤10 mL/kg/min. As ancillary finding, we documented that VE/VCO2 slope, although significantly related to outcome at univariate analysis, was not selected as independent risk predictor using the multivariable screening: the exclusion seems, at first glance, in contra tendency in regards to other experiences [31,32], nonetheless the predictive value of VE/VCO2 slope in the setting of very low peak VO2 has not been investigated, yet. Further studies are needed to address this specific issue. 4.2. Limitations In our relatively modest sample size of females, the number of statistical tests performed could potentially inflate Type I error. In view of this we adopted a restrictive approach, including in the predictive model only pb 0.01 parameters at univariate analysis, and selected only established prognostic indexes in HF to build the multivariate model; indeed, the importance of the parameters chosen strengthens our findings. In addition, recently peak VO2 criteria for HT selection have shifted from 14 to 10 mL/kg/min [30], but, although females had better survival even in this subgroup, the limited number of patients precludes any distinctive conclusion. Moreover, the study population comprised a relatively low percent of patient chronically treated with beta-blockers, due to the extended time span of enrollment. Two-year mortality rates were 22% and
37% in patient with and without beta-blocker treatment, respectively. Although the maintenance of beta-blocker therapy during the follow up was not ascertained, a supplementary analysis according to betablockers prescription would certainly help to define the interaction between gender and therapy, but unfortunately, an accurate predictive analysis was precluded because of the paucity of events in the female subgroup with beta-blockers treatment (8 deaths). Finally, although the study was designed to address the prognostic correlation between gender and peak VO2, we agree that the best way to assess the gender impact on aerobic capacity is described by the use of ppVO2: this relies on the fact that the formula for ppVO2 incorporates both age and gender. As confirmation, the inclusion in of ppVO2 in the predictive multivariable model excludes age, gender and peak VO2 (data not shown). 5. Conclusions Gender consideration is an old issue that has been buried and needs to resurface. Though peak VO2 preserves its prognostic value in both HF men and women with impaired exercise capacity, women – although they have a comparatively lower peak VO2 than men – live longer. We discovered that the traditional cut point value for peak VO2, i.e. ≤14 mL/kg/min is not a “gender-neutral” reference since lumping HF men and women together with the same VO2 value is unlikely to describe the true risk of events. There is a clear need to integrate genderspecific issues into everyday clinical practice, when appropriate. But we have to admit that, while the challenge of modifying prognosticating peak VO2 paradigm for HT candidacy is an intrepid one, changing physicians' management of HF patients in the light of gender may be an even greater challenge. Acknowledgments The authors are grateful to Alfio Agazzone, Elena Bonanomi and Barbara Temporelli for technical support, to Fabio Comazzi for statistical assistance, and to Rosemary Allpress for her careful revision of the English manuscript. The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology. References [1] Jessup M, Pina IL. Is it important to examine gender differences in the epidemiology and outcome of severe heart failure? J Thorac Cardiovasc Surg 2004;127: 1247–52. [2] Ekelund LG, Haskell WL, Johnson JL, Whaley FS, Criqui MH, Sheps DS. Physical fitness as a predictor of cardiovascular mortality in asymptomatic North American men. The Lipid Research Clinics Mortality Follow-up Study. N Engl J Med 1988;319:1379–84. [3] Gulati M, Black HR, Shaw LJ, et al. The prognostic value of a nomogram for exercise capacity in women. N Engl J Med 2005;353:468–75. [4] Piepoli MF, Corrà U, Agostoni PG, et al. Statement on cardiopulmonary exercise testing in chronic heart failure due to left ventricular dysfunction: recommendations for performance and interpretation. Part III: interpretation of cardiopulmonary exercise testing in chronic heart failure and future applications. Eur J Cardiovasc Prev Rehabil 2006 Aug;13(4):485–94. [5] Balady GJ, Arena R, Sietsema K, et al. Clinician's guide to cardiopulmonary exercise testing in adults. A scientific statement from the American Heart Association. Circulation 2010;122:191–225. [6] Lietz KW, Miller LW. Improved survival of patients with end-stage heart failure listed for heart transplantation. Analysis of organ procurement and transplantation network/U.S. United Network of Organ Sharing Data, 1990 to 2005. J Am Coll Cardiol 2007;50:1282–90. [7] Mancini DM, Eisen H, Kussmaul W, Mull R, Edmund LH, Wilson JR. Value of peak oxygen consumption for optimal timing of cardiac transplantation in ambulatory patients with heart failure. Circulation 1991;83:778–86. [8] Mezzani A, Corrà U, Bosimini E, Giordano A, Giannuzzi P. Contribution of peak respiratory exchange ratio to peak VO2 prognostic reliability in patients with chronic heart failure and severely reduced exercise capacity. Am Heart J 2003;145: 1102–7. [9] Giannuzzi P, Imparato A, Temporelli PL, et al. Doppler-derived mitral deceleration time of early filling as a strong predictor of pulmonary capillary wedge pressure in post-infarction patients with left ventricular systolic dysfunction. J Am Coll Cardiol 1994;23:1630–7.
U. Corrà et al. / International Journal of Cardiology 167 (2013) 157–161 [10] Wasserman K, Hansen JE, Sue DY, Casaburi R, Whipp BJ. Measurements during integrative cardiopulmonary exercise testing. In: Wasserman K, Hansen JE, Sue DY, Casaburi R, Whipp BJ, editors. Principles of exercise testing and interpretation. Baltimore, MA: Lippincott Williams & Williams; 1999. p. 62–94. [11] Beaver WL, Wasserman K, Whipps BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol 1986;60:2020–7. [12] Davies LC, Wensel R, Georgiadou P, et al. Enhanced prognostic value from cardiopulmonary exercise testing in chronic heart failure by non-linear analysis: oxygen uptake efficiency slope. Eur Heart J 2006 Mar;27(6):684–90. [13] Arena R, Guazzi M, Myers J. Prognostic value of end-tidal carbon dioxide during exercise testing in heart failure. Int J Cardiol 2007;117:103–8. [14] Cox DR, Oakes D. Analysis of survival data. London, UK: Chapman & Hall; 1984. [15] D'Agostino Jr RB. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med 1998;17:2265–81. [16] Lund LH, Mancini D. Heart failure in women. Med Clin North Am 2004;88: 1321–45. [17] Miller Leslie W. Heart failure: who we treat versus who we study. Cardiol Clin 2008;26:113–25. [18] Petrie MC, Dawson NF, Murdoch DR, Davie AP, McMurray JJV. Failure of women's hearts. Circulation 1999;99:2334–41. [19] Ho KKL, Anderson KM, Kannel WB, et al. Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation 1993;88:107–15. [20] Ghali JK, Pina IL, Gottlieb SS, et al. Metoprolol CR/XL in female patients with heart failure: analysis of the experience in MERIT-HF. Circulation 2002;105:1585–91. [21] Drinkwater BL. Women and exercise: physiological aspects. Exerc Sport Sci Rev 1984;12:24–51. [22] Daida H, Allison TG, Johnson BD, Squires RW, Gau GT. Comparison of peak exercise uptake in men versus women in chronic heart failure secondary to ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 1997;80:85–8.
161
[23] Opasich C, Pinna GD, Bobbio M, et al. Peak oxygen consumption in chronic heart failure: toward efficient use in the individual patient. J Am Coll Cardiol 1998;31: 766–75. [24] Arena R, Myers J, Abella J, et al. Development of ventilatory classification system in patients with heart failure. Circulation 2007;115:2410–7. [25] Richards DR, Mehra MR, Ventura HO, et al. Usefulness of peak oxygen consumption in predicting outcome of heart failure in women versus men. Am J Cardiol 1997;80: 1236–8. [26] Elmariah S, Goldberg LR, Allen MT, Kao A. Effects of gender on peak oxygen consumption and the timing of cardiac transplantation. J Am Coll Cardiol 2006;47:2237–42. [27] Guazzi M, Arena R, Myers J. Comparison of the prognostic value of cardiopulmonary exercise testing between male and female patients with heart failure. Int J Cardiol 2006;113:395–400. [28] Green P, Lund LH, Mancini D. Comparison of peak exercise consumption and the heart failure survival score for predicting prognosis in women and men. Am J Cardiol 2007;99:399–403. [29] Hsich E, Chadalavada S, Krishnaswamy G, et al. Long-term prognostic value of peak oxygen consumption in women versus men with heart failure and severely impaired left ventricular systolic function. Am J Cardiol 2007;100:291–5. [30] Mancini D, Lietz K. Selection of cardiac transplantation candidates in 2010. Circulation 2010;122:173–83. [31] Francis DP, Shamim W, Davies LC, et al. Cardiopulmonary exercise testing for prognosis in chronic heart failure: continuous and independent prognostic value from VE/VCO2 slope and peak VO2. Eur Heart J 2000;21:154–61. [32] Arena R, Myers J, Abella J, et al. Development of a ventilatory classification system in patients with heart failure. Circulation 2007;115:2410–7.