Prediction of peak oxygen uptake by an endurance test: A wish and a nightmare

Future prognostication is a wish that pertains to several fields of human life. Horoscope forecasts are among the most frequently read parts of newspapers, and the work of wizards and witches has been demanded since the beginning of the human era, and it is still sought after now, regardless of the economic crisis. In medicine, we cannot tell fortunes from cards, coffee grounds, the direction of a black cat, the roll of the dice and the like for prognostic purposes. We need robust data to apply. Indeed, with regard to health, prognosis can be seen either as a population need or as a single individual need. The former is, or should be, used for proper resource allocation, and it is relatively easy work, the latter is almost impossible when the prognosis of a single individual is requested. Indeed, single individuals want to know about themselves, and they usually do not care about other individuals. In other words, when somebody states that the death risk for a specific event is 0.1%, the individual who dies does not care about the 999 who survive. In fact, we usually face one individual, and not a population (Figure 1). Prognosis in cardiovascular medicine is extremely complicated, because there is not usually a single culprit organ or body function. The classic example is chronic heart failure, in which, on top of the heart and vessels, prognosis also depends on the kidney, lung, brain functions, as well as neurohumoral status, and so-called comorbidities such as anaemia, iron deficit, diabetes, hypertension, depression and chronic obstructive pulmonary disease, just to mention the most frequently observed. Moreover, prognostic parameters should ideally be cheap and easy to obtain with no risk for patients and for the community. This is clearly an unmet need, or what the ancient Greeks called a Vimaira. So the first question is, for chronic heart failure do we have a single parameter that comprehends all body functions? In this regard, exercise seems to be the most appropriate tool for a few reasons: (a) heart failure symptoms are present or exacerbated by exercise; (b) patients’ complaints are frequently related to a reduction in physical performance; (c) exercise implies the combined work of the heart, vasculature, muscles, brain, lungs and neurohumoral response. Moreover, in heart failure, a key point during exercise is the transitory redistribution of blood flow, which privileges the working muscles with a reduction of the blood flow to the brain, bowels, skin and kidneys. This blood flow reduction may cause specific symptoms, such as exercise-induced confusion, mental absence, abdominal pain, etc. Therefore, exercise seems a good tool, but which parameters among the plethora of exercisederived parameters should be chosen? Oxygen uptake is likely to be the most appropriate, as it is the product of cardiac output and arteriovenous oxygen content difference. The former is heart rate and stroke volume, thus it reflects sympathetic tone, pre and afterload, and right and left cardiac function. The latter is haemoglobin level, alveolar-capillary gas diffusion, oxygen delivery and uptake by the working muscles, i.e. resistance to oxygen flow across the alveolar-capillary membrane, oxygen transport in the blood, total blood flow and blood flow redistribution, resistance to oxygen flow from the capillary to the mitochondria and muscle performance. Therefore, oxygen uptake is a good indicator of body performance comprehensive of several functions. The second question is how to evaluate oxygen uptake and whether it should be normalised. Indeed, age, gender, height, weight and exercise habits directly influence peak oxygen uptake. Several normalisation formulas for peak oxygen uptake have been proposed and, at present, those of Jones et al. and Hansen et al. are the most applied. The FRIEND-derived peak