Article Text


The forgotten Barcroft/Edholm reflex: potential role in exercise associated collapse
  1. T D Noakes
  1. UCT/MRC Research Unit for Exercise Science and Sports Medicine, University of Cape Town Sports Science Institute of South Africa, PO Box 115, Newlands 7725, South Africa; tdnoakes{at}

    Statistics from

    Approximately 80% of the athletes who collapse in marathon and other long distance sporting events do so only after they have crossed the finish line.1 Although it is usually taught that “dehydration” explains this phenomenon, logic suggests this to be unlikely because any dehydration induced hypovolaemia should cause cardiovascular failure when cardiac stress is highest—that is, during rather than after exercise when cardiovascular function is returning to the resting state. Instead it is clearly the act of stopping exercise that is the consequent, albeit paradoxical, cause of post-exercise collapse.

    We have previously proposed that this form of exercise associated collapse is caused by the persistence into recovery of a state of low peripheral vascular resistance, compounded by removal of the skeletal muscle pump that maintains the right atrial filling pressure during exercise.1 According to this theory, the combination of a low peripheral vascular resistance and a sudden reduction in venous return would reduce stroke volume and cardiac output acutely, causing hypotension. Indeed this mechanism was used to explain the development of postural hypotension due to “dehydration exhaustion” in military personnel exposed to eight or more hours of exercise in desert heat without fluid replacement, in the classic studies of Adolph.2

    Thus Adolph2 wrote that “the ultimate failure of the circulation (in dehydration exhaustion) is of a peripheral type . . . Dilated blood vessels of the skin require additional blood to fill them; correspondingly, the available compensation for vertical posture are diminished, for the blood vessels of the legs compromise between dilating to carry heated blood and constricting to preserve a greater flow to the head” (p 235). He also reported that: “Lying down promptly relieves the circulation and the symptoms” (p 236). Similarly, Lee3 defined the physiological abnormality causing heat exhaustion as: “A low arterial blood pressure is the critical event, resulting partly from inadequate output by the heart and partly from the widespread vasodilation”. He also noted that: “Treatment is simple – recumbency, fluid and salt administration, gentle cooling, and rectification of any predisposing cause” (p 571).

    However, these explanations fail to explain why a compensatory tachycardia is not a usual feature of this form of hypotension,4 or why the onset of symptoms occurs so rapidly and why they are reversed so rapidly when subjects lie supine in the head down position.1

    An historic study identifying a potentially forgotten reflex and published in The Lancet in 1944 suggests an alternative explanation. Barcroft et al5 studied the effects of rapid venesection of about 1 litre on cardiovascular function measured with right heart cannulation and forearm plethysmography. Studies continued until subjects fainted from the sudden onset of hypotension. Figure 1, redrawn from those historic data, shows the salient findings. Initially, cardiovascular function adapted to progressive blood loss by increasing peripheral vascular resistance and heart rate, each by about 30%. Despite this, average cardiac output, right atrial pressure, and systolic blood pressure all fell by about 40%.

    However, the crucial finding was that fainting occurred only when there was a sudden fall in the peripheral vascular resistance associated with an unexpected increase in forearm blood flow (fig 1).

    A second study evaluated the authors’ hypothesis that it was the fall in right atrial pressure induced by blood loss that activated the reflex reduction in peripheral vascular resistance. A tourniquet was applied to the lower limb and inflated before the beginning of the venesection. Hypotension again developed but after a smaller blood loss of only about 550 ml. Release of the tourniquet immediately corrected the hypotension by reducing forearm blood flow (fig 2). The authors concluded that removal of the tourniquet produced a sudden infusion of blood from the lower limb thereby increasing the right atrial pressure, abolishing the reflex and normalising cardiovascular function.

    Hence they proposed the existence of a potent skeletal muscle vasodilator reflex that is activated when the right atrial pressure either falls below some critical value or begins to fall at a particular rate. This reflex appears atavistic because it compounds rather than corrects the hypotension associated with blood loss. The similarity in the rapidity with which hypotension occurs with either progressive venesection or after the cessation of exercise, and without an associated tachycardia, suggests that this right atrial reflex may also cause exercise associated collapse.1,4

    To my knowledge, some contemporary texts of cardiovascular control during exercise do not specifically mention this reflex. For example, Rowell6 refers to a later 1945 paper in the Journal of Physiology by Barcroft and Edholm,7 in which the physiological basis for this skeletal muscle vasodilation was studied. However, the relevance of this specific reflex to the development of syncope is not directly discussed. Rather, Rowell presents the findings of Barcroft and Edholm as a possible example of neurogenic sympathetic cholinergic vasodilation in skeletal muscle.

    It is of interest that we had empirically discovered some years ago1 that the optimum management of post-exercise collapse is achieved by elevating the feet and pelvis of collapsed athletes above the level of the right atrium. The results of this manoeuvre are usually dramatic, with rapid reversal of hypotension and the symptoms of dizziness. According to the Barcroft/Edholm reflex, the adoption of this head down position would produce this dramatic response by immediately reversing the low atrial pressure that develops on the cessation of exercise in some susceptible athletes. Hence the effect would not result simply by increasing venous return and hence cardiac output, the more usual explanation,6 but also by reversing the skeletal muscle vasodilation induced by the Barcroft/Edholm reflex.

    Finally, I would like to make the obvious point that the Barcroft/Edholm reflex explains why nursing in the head down (Trendellenberg) position would be the more logical treatment for post-exercise collapse than the provision of intravenous fluids.1 The reason is that nursing in the Trendellenberg position rapidly increases the right atrial pressure whereas any effect of intravenous fluids on right atrial pressure is likely to be smaller and much delayed.

    Furthermore the contribution of the Barcroft/Edholm reflex to exercise associated collapse can be very rapidly evaluated by studying the response of the athlete’s blood pressure to changes in posture. Failure of the blood pressure to rise when placed in the head down position must indicate that the Barcroft/Edholm reflex is not active and some other cause for the hypotension, such as myocardial dysfunction or a persistent reduction in peripheral vascular resistance for reasons other than this postulated reflex, must be considered.

    Figure 1

    Changes in blood pressure, total peripheral resistance, forearm blood flow, cardiac output, heart rate, and right atrial pressure in subjects who underwent rapid venesection of about 1 litre of blood in 12 minutes. Note that fainting is caused by a sudden reduction in peripheral vascular resistance resulting from an increase in forearm blood flow as right atrial pressure falls, the Barcroft/Edholm reflex. Redrawn from data in Barcroft et al.5

    Figure 2

    The sudden reduction in systolic blood pressure and the associated increase in forearm blood flow with progressive venesection is immediately reversed by release of a lower limb tourniquet. Redrawn from data in Barcroft et al.5


    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.