American Heart Journal
Volume 146, Issue 2 , Pages 215-217, August 2003

Hippocrates and prognosis in heart failure: the challenge of static and dynamic measurements

  • Hector O Ventura, MD

      Affiliations

    • Department of Cardiology, Ochsner Clinic Foundation, New Orleans, La, USA
    • Corresponding Author InformationReprint requests: Hector O. Ventura, MD, Ochsner Clinic Foundation, 1514 Jefferson Hwy, New Orleans, LA 70121, USA.
    • ,
  • Mandeep R Mehra, MD

      Affiliations

    • Department of Cardiology, Ochsner Clinic Foundation, New Orleans, La, USA

Article Outline

 

He who would know correctly beforehand those that will recover, and those who will die, and in what cases the disease will be protracted for many days, and in what cases for a shorter time, must be able to form a judgment for having made himself acquainted with all the symptoms, and estimating their powers in comparison with one another.—Hippocrates1

Back to Article Outline

Establishing prognosis in heart failure 

The emphasis on the ability of physicians in establishing the prognosis of a disease state is 1 of the most important teachings of Hippocrates. If one were to critically outline those characteristics that exemplify an astute clinician, the ability to distinguish severity of illness and predict prognosis would rank at the very top. Never has the need for an accurate prediction of outcome been greater in the realm of heart failure. Hand in hand with the increasing prevalence and incidence of heart failure is an ever-growing array of treatment options.2, 3 Most such options require reconciliation of disease severity, becaue their application is intricately dependent on a narrow therapeutic index. Thus, one would not consider destination therapy with mechanical support in a situation in which a bridge to reversal of the heart failure phenotype might represent a viable opportunity. Also, because the cost implications of therapeutic avenues continue to create consternation, the ability to prognosticate and thereby allocate treatment in a “cost-efficient” manner is gaining importance. Therefore, the need to establish both reliable indicators of prognosis and accurate risk stratification parameters is of clinical importance.

Back to Article Outline

Predictors of prognosis in heart failure 

Static predictors 

Although a large body of evidence4 exists that seeks to discriminate prognosis, the typical modality used is a static measurement. Static or “snap-shot” measurements have clearly stood the test of time and represent a powerful differentiator of late outcome. Thus, functional capacity at study entry and severity of systolic dysfunction are touted as excellent arbitrators of prognosis within investigations of therapeutic strategies that seek to enroll patients at “high-risk” of events. Similarly, a snapshot of cardiac reserve as assessed with peak aerobic capacity is also a well-established predictor of prognosis and need for cardiac transplantation.5, 6, 7 However, such static measurements tend to fall short in certain subgroups of individuals and often need refinements that can complicate the value of the modality being used to assess risk in an individual.7 Furthermore, static measurements can be modified by several factors, such as the etiology and stage of heart failure, the biological characteristics of patients, and the therapeutic interventions over time. Thus, the search for an ideal marker for prognosis continues, and in that regard, the assessment of a variety of variables with time in a dynamic measure might represent a viable solution to this challenge.

Dynamic measurements 

Several lines of evidence point to the importance of dynamic shifts in disease stability as an accurate predictor of outcome in heart failure. In the realm of advanced heart failure, hemodynamic alterations with time, especially in response to therapeutic interventions, are becoming important in the prediction of prognosis. In this regard, it has been demonstrated that the dynamic ability of a therapeutic intervention to reduce left ventricular filling pressures is associated with long-term benefits.2 Similarly, the dynamic stability of disease state as evaluated by trends on medical resource use is now being accepted as a prognostic marker. A recent analysis from the Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS) trial8 has shown that as one stacks up the hospitalization profiles of patients with heart failure in the period preceding study entry, one can develop a prognostic profile that differs in outcome from the average study cohort. Thus, in the COPERNICUS trial, subjects who were hospitalized more than twice in the preceding year of study entry were identified as a particularly unstable cohort with a greater mortality rate.8 Data lines such as this suggest that serial dynamic assessments have value and should be considered in the overall algorithm of prognosis assessment.

Few studies are available on how these variables change with time or how changes from a given baseline measurement may be used to assess the subsequent clinical course or predict survival of patients with chronic heart failure.9, 10, 11, 12 In this issue of the Journal, Grigioni et al13 report new information about the use of serial measurement of predictors of survival in heart failure. Before we analyze in detail the study report, we will discuss previous studies on assessing prognosis with serial measurements of parameters. These studies have depicted contradictory results.

It has been reported that patients listed for heart transplantation with peak VO2 values <14 mL/kg/min at baseline evaluation had a 100% survival rate at 2 years when the peak VO2 increased >2 mL/kg/min during the follow-up period.9 In a study by Florea et al10 in 62 patients with chronic heart failure, several parameters of exercise capacity and ventricular structure and function with echocardiography were evaluated at baseline and after a follow-up period of 19 ± 15 months. Patients were observed for a mean of 17 months after the second evaluation. The authors demonstrated that only the changes in peak VO2 per year, independent of changes in left ventricular ejection fraction, were predictive of survival without undergoing heart transplantation. Conversely, another study by Gullestad et al12 of 286 patients referred for heart transplantation demonstrated that serial changes in peak VO2 or changes in ejection fraction and left ventricular dimension did not add any prognostic information in patients with severe heart failure who were stabilized on medical therapy. The follow-up period was 56 ± 0.1 months. The authors conclude that the routine use of these parameters did not seem warranted. The reason for these discrepant results on peak VO2 may be the result of patient selection and specific treatment regimen. Although the study by Florea et al7 included patients with mild to moderate heart failure, the study by Gullestad et al12 enrolled patients who were referred for cardiac transplantation. Both of the studies agreed that echocardiographic parameters, specifically changes in ejection fraction, were not predictive of prognosis.

Grigioni F et al13 report new information about the use of serial measurements of predictors of survival in heart failure. They studied 105 patients with heart failure (New York Heart Association [NYHA] functional class III–IV) who were treated with β-blockers, vasodilators, digoxin, diuretics, and warfarin. Patients were included in the study when several clinical, echocardiographic, exercise, and hemodynamic parameters were assessed serially (mean interval between the 2 assessments, 19 ± 13). Subsequently, patients were observed for 14 ± 11 months, and the impact of this panel of prognostic parameters as predictors of the combined end point of cardiovascular death or cardiac transplantation was evaluated. The authors demonstrated that among the time-related parameters, QRS widening (adjusted RR per 10 msec, 1.21; 95% CI, 1.10–1.48; P = .003) and peak VO2 decrease (adjusted RR per mL/kg/min, 1.11; 95% CI, 1.01–1.22; P = .034) provided incremental information for predicting cardiovascular death/cardiac transplantation for the entire panel of isolated readings. The authors constructed cardiovascular death/cardiac transplant event-free survival curves between the 2 serial points of evaluation and demonstrated that the combination of stable NYHA class, QRS widening <10%, and a pVO2 decreasing <10% was associated with a 92% 1-year event-free survival rate. The prognostic value of the combination of parameters was better that any of the parameters taken alone.

Back to Article Outline

The clinical importance of this study 

First, it demonstrates once more the efficacy of serial measures of peak VO2 in predicting prognosis in patients with heart failure receiving optimal medical therapy, including β-blockers, who are referred for heart transplantation. To the best of our knowledge, compared with previous studies, this is the first study to evaluate parameters of exercise capacity serially in patients with heart failure who are being treated with β-blockers.14 In this study, serial peak VO2 remains a predictor of survival; a recent study has questioned the value not only of exercise parameters but also other neurohormonal and clinical and ventricular function parameters in patients with heart failure treated with β-blockers.

Second, although it has been demonstrated that QRS widening is predictive of adverse events in heart failure, the authors demonstrate that, with time, changes of this simple electrocardiographic measurement are of value in predicting prognosis. This finding is of clinical importance because the rate of QRS widening is easy to obtain serially, and it might be useful to select patients with heart failure who will benefit from the implantation of biventricular pacing.15

Third, changes in NYHA functional class, peak VO2, right atrial pressure, and QRS duration with time remained independently predictive of survival even when they were adjusted for their respective study values, which indicates that the significance of these changes was independent of the initial value. In addition, the authors constructed a time-related prognostic index that took into account clinical parameters (NYHA functional class), electrocardiographic parameters (QRS widening <10%), and exercise parameters (peak VO2 <10% worsening) and were able to illustrate that patients with this combined prognostic index had a better outcome. The future use of this combined prognostic index awaits prospective validation, but it will be relevant in the selection of the most appropriate candidates for heart transplantation.

Finally, as previous studies have shown in the pre β-blocker era, ejection fraction, either single or serially measured, was not predictive of outcome.9, 10, 12 The authors also demonstrate that changes in hemodynamics, in this case right atrial pressure, remain significant when one measures it serially when the initial value is >6 mm Hg.

Back to Article Outline

Shortcomings of the study 

First, as the authors alluded, the data were analyzed retrospectively, and in addition, these findings cannot be extrapolated to a general population of patients with heart failure because most of these patients were referred for heart transplantation.

Second, the authors did not include other new neurohormonal markers such as B-type natriuretic peptide, which seem to be of value in predicting mortality in patients with heart failure.3

Back to Article Outline

Where do we go from here? 

Appropriate risk stratification of patients with heart failure is essential, especially in an era when β-blockers have made an impact in the long-term survival in these patients. As costly therapeutic advances ensue, our ability to prioritize the allocation of such expensive modalities will be contingent on our ability to distinguish prognosis in varying stages of heart failure. The measurement of a parameter at a single point in the progression of heart failure to assess future outcome is flawed because the prognostic value of a single parameter can be modified by several factors. This study is 1 of the first that has attempted to evaluate dynamic changes in parameters of survival in patients with heart failure in the modern era of β-blocker use. A prognostic index, weighing the relative value of several indicators of survival with time, for assessing individualized annual mortality rates, such as the one the authors published in this article, is interesting and useful, but it would require prospective validation. Finally, with a prospectively validated prognostic index, it would be important to use different forms of therapy, such as biventricular pacing, ventricular assist devices, and heart transplantation, to name a few. The latter principle is also included in the teachings of Hippocrates: “The physician who cannot inform his patient what would be the probable issue of his complaint, if allowed to follow its natural course, is not qualified to prescribe any rational treatment for its cure.”16

Back to Article Outline

References 

  1. Hippocrates. The book of prognostics. In: Adams F, editor. The genuine works of Hippocrates, volume I. London: C&J Adlars Printers; 1849. p. 234
  2. Mehra MR, Ventura HO. Heart failure. In:  Civetta JM,  Taylor RW,  Kirby RR editor. Critical care. 3rd ed. Philadelphia: J.B. Lippincott Company; 1997;p. 1887–1903
  3. Hunt S, Baker D, Chin M, et al.  ACC/AHA guidelines for the evaluation and management of chronic heart failure in the adult (a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines). J Am Coll Cardiol. 2001;38:2101–2113
  4. Cowburn P, Cleland J, Coats A, et al.  Risk stratification in chronic heart failure. Eur Heart J. 1998;19:696–710
  5. Mancini D, LeJemtel T, Aaronson K. Peak VO2 (A simple yet enduring standard). Circulation. 2000;101:1080–1082
  6. Milani RV, Mehra MR, Reddy T, et al.  Ventilation/carbon dioxide production ratio in early heart exercise predicts poor functional capacity in congestive heart failure. Heart. 1996;76:393–396
  7. Osman AF, Lavie CJ, Mehra MR, et al.  The incremental prognostic importance of body fat adjusted peak oxygen consumption in chronic heart failure. J Am Coll Cardiol. 2000;36:2126–2131
  8. Packer M, Fowler MB, Roecker EB, et al.  Effect of carvedilol on the morbidity of patients with sever chronic heart failure (results of the carvedilol prospective randomized cumulative survival (COPERNICUS) study). Circulation. 2002;106:2194–2199
  9. Stevenson L, Steimle A, Fonarow G, et al.  Improvement in exercise capacity of candidates awaiting heart transplantation. J Am Coll Cardiol. 1995;25:163–170
  10. Florea, Henein M, Anker S, et al. Prognostic value of changes over time in exercise capacity and echocardiographic measurements in patients with chronic heart failure. Eur Heart J 2000;21:146–53
  11. Levine T, Levine A, Goldberg A, et al.  Reversal of end stage heart failure is predicted by long term therapeutic response rather than initial hemodynamic and neurohormonal profile. J Heart Lung Transplant. 1996;15:297–303
  12. Gullestad L, Myers J, Ross H, et al.  Serial exercise testing and prognosis in slected patients considered for heart transplantation. Am Heart J. 1998;135:221–229
  13. Grigioni F, Barbieri A, Magnani G, et al. Serial versus isolated assessment of clinical and instrumental parameters in heart failure: prognostic and therapeutic implications. Am Heart J 2003; 146: 298–303
  14. Abraham WT, Fisher WG, Smith AL, et al.  Cardiac resynchronization in heart failure. N Engl J Med. 2002;346:1845–1853
  15. Zugck C, Haunstetter A, Kruger C, et al.  Impact of beta-blocker treatment on the prognostic value of currently used risk predictors in congestive heart failure. J Am Coll Cardiol. 2002;15:1615–1622
  16. Hippocrates. The book of prognostics. In: Adams F, editor. The genuine works of Hippocrates, volume I. London: C&J Adlars Printers; 1849. p. 18

PII: S0002-8703(03)00234-5

doi:10.1016/S0002-8703(03)00234-5

American Heart Journal
Volume 146, Issue 2 , Pages 215-217, August 2003

Article Tools