LETTER TO THE EDITOR
Lean tissue adjusted peak oxygen consumption in congestive heart failure
Roland Wensel, MDa and
Stefan D. Anker, MD, PhDa
a Clinical Cardiology, NHLI London, Dovehouse St., London SW3 6LY, UK
s.anker{at}ic.ac.uk
Osman et al. (1) recently reported data on the prognosis of 225 patients with heart failure. In their analysis they sought to test the hypothesis that peak oxygen uptake gains prognostic power when expressed per lean body mass. Taking the present and other available data (2,3), we entirely agree that this may in fact be possible. However, we also see a number of problems in the present article with regard to exercise testing, the patient group itself and its very low event rate, as well as the statistical analysis. Therefore, we cannot consider their article to offer proof of its conclusion.
The mean peak oxygen uptake (VO2) of their population of heart failure patients was 16 ml/kg/min. The mean anaerobic threshold was 12.7 ml/kg/min, and the peak heart rate was only 126 beats/min. Although mentioned in their Methods section, the investigators’ data for respiratory exchange ratio at peak exercise, a marker of metabolic stress, are not given anywhere in their report. The investigators chose to present event rates based on 14 deaths and 15 urgent transplants. In other words, the total mortality rate in this population was about 7% during 19 months of follow-up. This study group appears on average to have been in mild heart failure. This low mortality rate is surprising given the mean peak VO2 of 16 ml/kg/min. Even if urgent transplant is included in the end point, the 12-month event rate in patients with peak VO2  14 ml/kg/min is still only about 15%, which seems very low. Using the same cutoff, Mancini et al. (4) reported a mortality rate of 30% to 53% in 12 months. Taken together, this raises concern about the validity of their exercise tests, suggesting that an adequate VO2 may not have been reached.
Whichever way one examines the data, the total event number of 29 appears too small to perform extensive statistical analyses. The main statistical analysis in the Osman et al. (1) article is concerned with comparing peak VO2/weight versus peak VO2/lean weight as a continuous or dichotomous variable, respectively. Unfortunately, the receiver-operator curves (ROCs) for the two continuous variables are not presented, and the arguments are based on somewhat different chi-square (20.53 vs. 17.17) and p values (0.0001 vs. 0.0007). No statistical comparison for the two continuous variables is provided. Also, the comparison of the two cutoffs appears difficult to interpret.
First, the data in their Table 2 (1) do not specify the follow-up period to which they relate. Second, calculation of the ROC is reserved for variables with many different levels. In their Table 2, because in each case there is only one cutoff, it is not possible to calculate an area under the curve (AUC) for ROC as one has only one data point of corresponding sensitivity and specificity.
Further explanation is necessary. It would be worthwhile to know how the p values for the differences between the AUCs were obtained, as the p values seem very small (Table 2) for the comparison between the two peak VO2 measures.
Additionally, it remains arguable whether the comparison to Mancini’s cutoff is justified at all, as the cutoff was established in a much sicker population, whereas the lean cutoff was specifically designed for this population. Does selection of the optimal peak VO2/weight cutoff change the predictive power of this variable? Finally, we cannot consolidate the fact that the confidence interval for the RR values in their Table 5 (1) encompasses 1.0, yet all p values are highly significant.
The literature reviewed for the Osman et al. article appears somewhat incomplete. Using the DEXA-scan, previous studies have found that absolute peak VO2 (in ml/min) closely relates to lean tissue mass in patients with congestive heart failure (CHF) (5,6). That peak VO2/weight underestimates true exercise capacity in obese patients with CHF (7) and overestimates it in cachectic patients with CHF (8) has been suggested some years ago.
Finally, the investigators (1) conclude that peak VO2/lean weight is the cardiopulmonary exercise parameter that provides "the best risk stratification across the heterogeneous systolic heart failure cohort." The presented data do not justify this conclusion. First, this is due to the above problems, but more importantly this is because the VE/VCO2-slope (which is automatically provided by the equipment the authors used) was ignored for all analyses. For several years now the latter is known to be a strong prognosticator (9,10) independently of peak VO2. In mild CHF, the VE/VCO2-slope is even superior to peak VO2 in predicting prognosis (11). Again, we believe the report by Osman et al. (1) examines an important subject, and we agree that peak VO2/lean weight is better than peak VO2/weight, but we also believe this message has not been proven by these investigators, and that the points we addressed above could have been beneficially optimized during the review process.
 |
References
|
|---|
1. Osman AF, Mehra MR, Lavie CJ, Nunez E, Milani RV. The incremental prognostic importance of body fat adjusted peak oxygen consumption in chronic heart failure. J Am Coll Cardiol. 2000;36:2126–2131[Abstract/Free Full Text]
2. Cicoira M, Zanolla L, Franceschini L, et al. Skeletal muscle mass assessed by DEXA is an independent predictor of exercise capacity in chronic heart failure patients (abstr). Eur Heart J. 2000;21(Suppl):233
3. Davos CH, Doehner W, Rauchhaus M, Anker SD. Evaluation of obese chronic heart failure patients with lean tissue adjusted peak oxygen consumption (abstr). Circulation. 2000;102(Suppl II):II758
4. Mancini DM, Eisen H, Kussmaul W, Mull R, Edmunds LH Jr, Wilson JR. Value of peak oxygen consumption for optimal timing of cardiac transplantation in ambulatory patients with heart failure. Circulation. 1991;83:778–786[Abstract/Free Full Text]
5. Toth MJ, Gottlieb SS, Fisher ML, Poehlman ET. Skeletal muscle atrophy and peak oxygen consumption in heart failure. Am J Cardiol. 1997;79:1267–1269[CrossRef][Medline]
6. Anker SD, Clark AL, Teixeira MM, Hellewell PG, Coats AJ. Loss of bone mineral in patients with cachexia due to chronic heart failure. Am J Cardiol. 1999;83:612–615[CrossRef][Medline]
7. Lang CC, Chomsky DB, Rayos G, Yeoh TK, Wilson JR. Skeletal muscle mass and exercise performance in stable ambulatory patients with heart failure. J Appl Physiol. 1997;82:257–261[Abstract/Free Full Text]
8. Anker SD, Swan JW, Volterrani M, et al. The influence of muscle mass, strength, fatigability and blood flow on exercise capacity in cachectic and non-cachectic patients with chronic heart failure. Eur Heart J. 1997;18:259–269[Abstract/Free Full Text]
9. Chua TP, Ponikowski P, Harrington D, et al. Clinical correlates and prognostic significance of the ventilatory response to exercise in chronic heart failure. J Am Coll Cardiol. 1997;29:1585–1590[Abstract]
10. Kleber FX, Vietzke G, Wernecke KD, et al. Impairment of ventilatory efficiency in heart failure: prognostic impact. Circulation. 2000;10:2803–2809
11. Ponikowski P, Francis DP, Piepoli MF, et al. Enhanced ventilatory response to exercise in patients with chronic heart failure and preserved exercise tolerance—marker of abnormal cardiorespiratory reflex control and predictor of prognosis. Circulation. 2001;103:967–972[Abstract/Free Full Text]
|
Top
References