«99P J. Physiol. (1972), 222, pp. 99P-118P With 12 text-figure Printed in Grea Britain AN ENQUIRY INTO THE NATURE OF THE MEDIATOR OF THE ...»
J. Physiol. (1972), 222, pp. 99P-118P
With 12 text-figure
Printed in Grea Britain
AN ENQUIRY INTO THE NATURE OF THE MEDIATOR
OF THE VASODILATATION IN SKELETAL MUSCLE IN
EXERCISE AND DURING CIRCULATORY ARREST
Given at the joint meeting of the Physiological and Anatomical
Societies at Nottingham University on 17 and 18 December 1971.
BY H. BARCROFTFrom 44 Wood Lane, Highgate, London, N.6, 5UB I am indeed greatly honoured to have been invited by the Physiological Society to deliver this Review Lecture. I shall never forget that I delivered it at the first meeting of the Physiological Society to be held in the University of Nottingham, where my colleague and friend Professor Greenfield is Professor of Physiology and Dean of the new Medical School.
Many of you were at the International Congress of Physiological Sciences at Munich this summer and will remember the President of the Executive Committee, Professor Kurt Kramer. In the 1930's he and his colleagues did beautiful experiments on the effect of exercise on the blood flow through the isolated perfused dog's gastrocnemius muscle. One of the most important of their findings was that, in the steady state, the blood flow through a skeletal muscle is directly proportional to its rate of oxygen consumption (Kramer, Obal & Quensel, 1939). This is shown in Fig. 1. It has been confirmed by Welch & Stainsby (1967). That is to say we have in our muscles a local mechanism which decreases the resistance of the blood vessels as oxygen demand increases. When the demand for oxygen increases the vessels open up, so that with almost constant arterial blood pressure, that is perfusion pressure, muscle blood flow may increase from 1 to 20-301./ min. How does the oxygen demand open up the vessels ? That is the subject of this lecture.
You will remember that at the beginning of exercise muscles go into debt for oxygen, after a few minutes they reach the steady state, and when exercise is over the debt is paid back in the first few minutes. It will be convenient to note here that Welch & Stainsby (1967) found a direct proportionality between oxygen debt and oxygen consumption. This is seen in Fig. 2, so that blood flow is linearly related to oxygen consumption and to oxygen debt.
Why this oxygen debt? Margaria, Edwards & Dill (1933) suggested that it was probably for oxidative processes needed for the resynthesis of high 12-2
lQQP REVIEW LECTUREenergy phosphates. Fig. 3 shows how phosphoryl creatine in the human quadriceps femoris muscle breaks down at the beginning of moderately severe exercise and reaches the steady state (Hultman, Bergstr6m & Anderson, 1967). It is resynthesized immediately after exercise and oxidative energy is required for this. Margaria's idea has been confirmed by Piiper, di Prampero & Cerretelli (1968). Welch & Stainsby (1967) have calculated that the resynthesis of all the phosphoryl creatine in the dog's,
Fig. 3. Results showing the concentration of phosphoryl creatine (PC) in biopsy samples of the human quadriceps femoris muscle before and during 20 min exercise (Hultman, Bergstrom & Anderson, 1967).
102P REVIEW LECTURE thesis for the regulation of muscle blood vessels in exercise must take account of it.
There are various suggestions for the mediator(s) of hyperaemia in exercising muscle. Anoxia is believed to be the link between blood flow and metabolism by Guyton and his school (Guyton, Ross, Carrier & Walker, 1964); Kjellmar (1965) has suggested increase in potassium; Mellander, Johansson, Sarah Gray, Jonnson, Lundvall & Ljung (1967) have pressed the claim of increase in osmolarity; Khayutin (1968) has suggested that contraction lessens stretching of the arterioles so their smooth muscle relaxes; Forrester & Lind (1969) and Forrester (1972) have drawn our attention to ATP; inorganic phosphate is favoured by Hilton & Vrbova (1970) and Hilton & Hudlicka (1972); and by Skinner & Costin (1970) a combination of anoxia, increase in potassium and increase in osmolarity.
Mellander & Johansson (1968) have reviewed the literature.
In preparing this lecture I found my whole attention absorbed by the claims of anoxia. I believe it deserves rather more consideration. That being so, this lecture will be not so much a review as the submission of evidence that oxygen lack may play a part, and in the steady state, maybe a large part in the regulation of muscle blood flow.
I must begin at the beginning of the story. In point of fact it does not begin with the hyperaemia of exercise at all, but with another condition in which the tissues do not get all the oxygen that they want. In that respect a first cousin of exercise, namely the vasodilatation that takes place in a limb after its blood supply has been cut off. The cousins are active muscle wanting oxygen and resting muscle wanting oxygen. I propose to start with what happens when a resting limb is denied its oxygen.
The classical experiments on this were done by Lewis & Grant in 1925.
'Observations upon reactive hyperaemia in man' was the title of their paper.
They found that the longer the period of arrest of the circulation, (1) the brighter the flushing of the skin after the circulation had been released, and the longer it lasted, (2) the greater the increase in the forearm volume after release, (3) the greater the forearm blood flow after release. For these and other reasons they concluded that the vasodilatation took place during the period of arrest, and that it was due to the action of metabolites, continuously formed and accumulating in the tissues during the period of arrest. The longer the arrest, the greater the accumulation of metabolites, and the larger the vasodilatation during arrest and after release. As the flush was bright arterial red they concluded that reactive hyperaemia was not due to anoxia.
Therefore I was very surprised to see a paper by Fairchild, Ross & Guyton (1966) entitled 'Failure of recovery from reactive hyperemia in the 103P
VASODILATATION IN MUSCLE
-11 --1 10---15
-11 -10 17 18
Fig. 5. Results showing decrease of oxygen tension to 1 mm Hg in the human tibialis anticus muscle during arrest of the circulation. Occlusion for 3 min between the arrows (Kunze, 1968).
circulatory rest with zero flow. On release the record, (b), shows 10 min hind limbs were perfused with blood ventilated with 100 % nitrogen. The rate of flow was about 31 times the resting rate, and remained constant for some 10 min, with no recovery of the blood flow till oxygenated blood perfusion was resumed. It was indeed 'Failure of recovery from reactive hyperemia in the absence of oxygen.' This suggested that vasodilatation in these hind limbs would go along with exhaustion of their oxygen stores. How soon after arresting the circulation would the stores be exhausted? One would have to divide the amount of oxygen in the store by the rate of the oxygen consumption.
Let us make the calculation. Farhi & Rahn (1955) reckon that human muscle myoglobin has a store of about 0-6 ml. oxygen/100 ml. muscle.
Allowing for oxygen in physical solution and in haemoglobin in the blood vessels it might be 1 ml./100 ml. Mottram (1955) estimated oxygen consumption of the forearm muscles to be 0-24 ml./100 ml. min. He told me this should now be increased to 0-28 - make it 0-3 ml. We have then 1 ml./ 100 ml. stored being consumed at 0-3 ml./min. Time for all the oxygen to
VASODILATATION IN MUSCLE 105Pbe consumed is 1/0 3 equals 3 min approximately. This agrees fairly well with the results obtained by Millikan (1937) from his observation of the rate of reduction of the myohaemoglobin (i.e. 1 %/sec) in the cat's soleus muscle following clamping the thoracic aorta. 3 min agrees fairly well for the time taken for oxygen tension in the human tibialis anticus muscle to fall to less than 1 mm Hg after the arrest of its circulation, as may be seen in Fig. 5 (Kunze, 1968). We have no corresponding data for the amount of oxygen stored in the skin or the rate of cutaneous oxygen consumption, but this does not matter, for Evans & Naylor (1967) have shown that following circulatory arrest oxygen tension in the skin declines to zero in less than 2 min, as Fig. 6 shows. In summary the evidence from these diverse sources amounts to the fact that oxygen stores in the forearm would be exhausted in 2-3 min after arresting the circulation.
Time (min) Fig. 6. Results showing that oxygen tension at the surface of the forearm skin decreased to zero in 2 min following occlusion of the circulation in the upper arm (Evans & Naylor, 1955).
Assuming that the time for the oxygen stores to be exhausted in the dog's hind limbs after arresting the circulation would be about the same it seemed likely that a simple switch from oxygenated blood perfusion to nitrogenated blood perfusion should be accompanied by vasodilatation in the first 2-3 min. I wrote to Dr Guyton asking if he had tried this, and if so, with what result. I have his permission to quote his reply which was as follows: 'The two minutes that you calculated is very close to the actual results we obtained in these experiments. That is, upon switching from control arterial blood to anoxic blood, there was a few seconds before any change took place, but as the anoxic blood entered the muscle, blood flow began to increase and reached its maximum level somewhere between 1 and "".O"IEWV LECTURE 106P 2 min. There is reason to believe that some of the oxygen diffuses backwards from the muscle into the anoxic blood, which could make the response somewhat more rapid than one would suspect from calculations of oxygen utilization.' This striking relation between oxygen lack and vasodilatation in the dog's hind limbs prompted me to search the literature to see whether corresponding vasodilatation occurred in the forearm in the first few minutes after circulatory arrest, as oxygen lack made itself felt.
Minutes Fig. 7. Averaged results of six experiments showing reactive hyperaemia in the human forearm after periods of arrest of 3, 5, 10 and 15 min respectively (Patterson & Whelan, 1955).
Fig. 7 has been taken from the paper by Patterson & Whelan (1955).
We see the averaged results of six experiments. Resting blood flow averaged 3 ml./100 ml. min. Peak blood flows averaging 25-5, 28, 28-5 and 30 ml./100 ml. forearm. min were recorded 4 sec after the release of occlusions lasting for 3, 5, 10 and 15 min respectively. That is to say the dilatation following arrest of the circulation was almost complete in 3 min, corresponding to the time at which oxygen stores would have been almost exhausted.
Fig. 8 has been taken from Sir Thomas Lewis's (1927) book, Blood Vessels of the Human Skin and their Responses. Resting blood flow in Sir Thomas Lewis's own forearm was 4 ml./100 ml. min. We see peak blood flows of 15, 34, 49 and 55 ml./100 ml. forearm. min recorded beginning immediately after intervals of occlusion of i, 11, 5 and 15 min respectively. Inspection of the results shows that the vasodilatation must have
VASODILATATION IN MUSCLE 107Pbeen almost complete within 3 min of arresting the circulation. Similar results were obtained five other sets of observations on Lewis's forearm in in two other sets on Grant's forearm (Lewis & Grant, 1925). (As we and have seen already Lewis & Grant used these results as evidence that metabolites accumulated continuously during the period of arrest. In fact this was not very good evidence for continuous accumulation, as the vasodilation was rapid during the first 3 min and very slow from then to the 15th min.)
Fig. 8. Results showing the rates of the inflow of blood into Sir Thomas ate Lewis's forearm immediately after peods of rrest of, 1, 5 and 15 min (Lewis & Grant, 1925; Lewis, 1927).
The data from Fairchild et al. (1966), from Patterson & Whelan (1955) and from Lewis & Grant (1925) have been brought together in Fig. 9. The Figure shows how rapidly cutting off the oxygen supply to the dog's hind legs, and to the human forearm was followed by vasodilatation. The smallest increase was in the dog's hind legs, the reason why it was only threefold is unknown, and is a little disquieting. Of the human results those of Patterson & Whelan, showing a tenfold increase, are probably the nearest to the truth. The reason why Lewis & Grant got a greater increase may have been because the blood flow through the hand was not arrested in their experiments. The thing that all three graphs have in common is that most of the vasodilatation takes place in the first 3 min, as the oxygen
Minutes Results showing: top, forearm blood flows immediately after Fig. 9.
different periods of arrest (Lewis, 1927; Lewis & Grant, 1925); middle, forearm blood flows immediately after different periods of arrest (Patterson & Whelan, 1955); bottom, blood flow through the hind limbs of the dog, perfusion with nitrogenated blood begun at time 0 (A. C. Guyton, personal communication, 1971; Fairchild, Ross & Guyton, 1966).
Minutes Fig. 10. Results showing that stopping the supply of oxygen to the limbs was followed by vasodilatation, reaching 80 % of its final value in 3 min.
The graphs have been obtained from the graphs in Fig. 9. The values on each graph have been expressed as percentages of a final value of 100.
The graphs rise in the following order from left to right: Fairchild et at.
-(1966); Patterson & Whelan (1955); Lewis & Grant (1925).