Heart rate during haemorrhage: time for reappraisal

Abstract

There is an interesting difference between experimental investigations and clinical practice. While experiments typically report changes in variables following a given intervention, a diagnosis is based on observing manifestation(s) of a disease or on interpretation of one or, more likely, several measured variables. In short, the clinician observes the ‘y-axis’ and has to derive the ‘x-axis’ from it. Yet, it is important that a physiological text accurately describes the typical deviations of variables associated with the intervention. Furthermore, accuracy becomes vital when the text addresses responses to a life-threatening circumstance like hypovolaemic shock caused by not only haemorrhage but also by, e.g., postural stress in the heat and lung recruitment manoeuvres by manual lung inflation, of relevance for intensive care medicine. It remains, therefore, somewhat of a mystery why the heart rate response to haemorrhage, or central hypovolaemia, is not reported accurately in most textbooks of physiology, internal medicine, or even anaesthesia and trauma (Secher & Bie, 1985). The heart rate response to haemorrhage is typically mentioned as being tachycardia, while the bradycardic response to haemorrhage is omitted although it has been recognised since the days of Hunter, and the (patho-) physiology repetitively addressed since the pioneering work of Barcroft et al. (1944). A modest reduction in central blood volume, as established by haemorrhage, but also by just sitting or standing up is associated with a modest ( 120 beats min−1) during haemorrhage and that appears to represent a transition to irreversible shock (Secher & Van Lieshout, 2005). Yet, the premise for bradycardia at a 30% reduction of the central blood volume is that vagal activity is intact and both age and disease, including diabetes, may reduce vagal tone to the heart. Vagal activity may also be overruled as is the case when haemorrhage is associated with pain. Since surgery typically is carried out in older people, the effect of age on the heart rate response to a reduction in central blood volume is important. The paper by Murrell et al. (2009) in a recent issue of The Journal of Physiology focuses on the events leading to fainting following prolonged exercise, but its message has general implications. It is significant not only because of the numerous values provided to describe the events leading to syncope but, in particular, because it illustrates the responses in the healthy elderly (60–80 years of age). The paper demonstrates that in these obviously well-trained athletes, the initial increase in heart rate following a reduction in the central blood volume is so modest that it does not reach an either statistically or clinically significant level. Furthermore, these subjects, in contrast to the young, did not develop bradycardia when their blood pressure decreased during head-up tilt, an established physiological model for investigating the cardiovascular and hormonal responses to haemorrhage. It seems that the current textbook description of the cardiovascular responses to haemorrhage, at least in regard to bradycardia, addresses the symptoms of the elderly patient or the patient in pain. That, however, should not distract the clinician from intervening when heart rate decreases in response to a reduction in central blood volume. As demonstrated in the paper by Murrell et al. (2009), that stage of shock is associated with a reduction in cerebral perfusion and may well pave the way to an irreversible stage of shock. Clearly, we need to tone down the importance of the arterial baroreceptors for control of the circulation when the aim is to maintain a patient in ‘normovolaemia’. Heart rate and blood pressure are imprecise variables for directing fluid therapy and for heart rate this is illustrated by the lack of response in the older people investigated by Murrell et al. (2009). Volume treatment according to blood pressure responses is also imprecise, simply because blood pressure is a regulated variable. Accurate volume treatment can, however, be directed based on Starling's ‘law of the heart’ in so-called individualised goal-directed fluid therapy (Bundgaard-Nielsen et al. 2007). In supine humans, the heart is working on the upper flat part of the Starling curve and normovolaemia may be maintained by evaluating whether cardiac stroke volume, cardiac output, or venous oxygen saturation are responsive to even a small (100–200 ml) volume challenge.