Experimental Studies in the Myocardial Collateral Circulation

I want to talk about the measurement of collateral flow that Dr. Fulton has so clearly demonstrated. It is not possible to assess this after myocardial infarction in man, and as at present collateral vasculature of the dog resembles that of man we have used this animal to study collateral flow rates immediately following infarction, the changes which accompany recovery, and the effects of drugs and sympathetic blockade. Copyright Royal Medical Society. All rights reserved. The copyright is retained by the author and the Royal Medical Society, except where explicitly otherwise stated. Scans have been produced by the Digital Imaging Unit at Edinburgh University Library. Res Medica is supported by the University of Edinburgh’s Journal Hosting Service: http://journals.ed.ac.uk ISSN: 2051-7580 (Online) ISSN: 0482-3206 (Print) Res Medica is published by the Royal Medical Society, 5/5 Bristo Square, Edinburgh, EH8 9AL Res Medica, April 1967, Special Issue – Lauder Brunton Centenary Symposium on Angina Pectoris: 25-26 doi: 10.2218/resmedica.v5i3-4.480 EXPERIMENTAL STUDIES ON THE MYOCARDIAL COLLATERAL CIRCULATION (Abridged) J. Russell Rees Cardiac Department, Westminster Hospital, London I want to talk about the measurement o f col­ lateral flow that Dr. Fulton has so clearly demon­ strated. It is not possible to assess this after myocardial infarction in man, and as at present collateral vasculature o f the dog resembles that of man we have used this animal to study collateral flow rates immediately following infarction, the changes which accom pany recovery, and the effects o f drugs and sympathetic blockade. T h e infarcts are produced in two ways anterior by ligating the anterior descending coron­ ary artery at operation, and posterior by wedging a catheter in a large branch o f the circumflex artery. T h e latter is inserted via the left carotid artery. In either case moderate sized infarcts result, with the expected electrocardiographic and enzyme changes, and an overall m ortality o f about 20 per cent collateral flow is determined by the clearance o f radioactive xenon 133 injected into the vessels o f the infarct via a fine catheter, and measured with an external scintillation counter. Before producing infarction, we usually meas­ ure the normal myocardial blood flow by the same method measuring the clearance o f xenon 133 injected in solution through a catheter, which does not interfere with normal flow, lying in the mouth o f the left coronary artery. In a normal heart the radioactive xenon diffuses rapidly through the myocardium and is then gradually removed over the next two to three minutes. W ith the aid o f a computer the clearance can be measured in several ways, the clearance rate (K ) calculated, and the m yocardial blood flow derived in ml/mm/gram. By contrast, in an in­ farct 3 days old clearance m ay take as long as 20 minutes, because o f the low rate o f collateral flow. Estimates o f collateral flow immediately after infarction have varied from negligible to ade­ quate, and the current continental and Russian view, recently supported in the Lancet, has been o f interest to us. It is suggested that collateral flow after infarction is at first adequate, but then falls to zero within a few hours as a result o f harmful sympathetic vasoconstrictor reflexes which may be reversible. If this were so, there would be im portant therapeutic implications. In both types o f infarcts in our study we found that immediate collateral flow was established at about a quarter to a third o f normal myocardial blood flow. There is usually a slight rise in col­ lateral flow in the first hour and then a slow fall to the original level, but there is no evidence that it ever falls to zero. Occasionally this fluctuation is very marked. In one instance, the collateral flow was initially quite low, then rose to the highest we recorded, and finally fell until death from ventricular fibrillation after 5 hours. W e have no idea what causes these fluctuations which are unrelated to blood pressure. W e have now recorded collateral flow just before or after resuscitation from ventricular fibrilliation on five occasions, and found that it was always very low, ranging from .05 to . 15 ml/min/gram o f infarct. This is about — 15 per cent o f the normal value for the heart, a rate which is obviously dangerously inadequate. T w o of these animals were successfully resuscitated with sub­ sequent improvement in clearance. W e have examined the effects o f general and local sympathetic blockade as well as surgical sympathectomy without demonstrating any bene­ ficial effect on collateral flow, but it does increase slowly on successive days after recovery from myocardial infarction. There is usually little change over the first 3 days, but then it begins to rise and by about the 8th day reaches approxi­ mately 5 o f the normal value. However this does not always happen. After an initial high level of collateral flow, the rate m ay fall and this may be followed by the death o f the animal. There are o f course marked individual varia­ tions, but the usually inadequate early collateral flow is established through the fine interarteriolar anastomoses which we know do exist in the


EXPERIMENTAL STUDIES ON THE MYOCARDIAL COLLATERAL CIRCULATION (Abridged) J . R ussell Rees
Cardiac Department, Westminster Hospital, London I w ant to talk about the measurement o f col lateral flow that Dr. Fulton has so clearly dem on strated.It is not possible to assess this after m yocardial infarction in m an, and as at present collateral vasculature o f the dog resembles that of man w e have used this anim al to study collateral flow rates im m ediately follow ing infarction, the changes which accom pany recovery, and the effects o f drugs and sym pathetic blockade.
T h e infarcts are produced in two waysanterior by ligating the anterior descending coron ary artery at operation, and posterior by wedging a catheter in a large branch o f the circum flex artery.T h e latter is inserted via the left carotid artery.In either case m oderate sized infarcts result, w ith the expected electrocardiographic and enzym e changes, and an overall m ortality o f about 20 per cent collateral flow is determined by the clearance o f radioactive xenon 133 injected into the vessels o f the infarct via a fine catheter, and measured w ith an external scintillation counter.
Before producing infarction, we usually meas ure the normal m yocardial blood flow by the same method -measuring the clearance o f xenon 133 injected in solution through a catheter, which does not interfere with norm al flow, lying in the mouth o f the left coronary artery.In a normal heart the radioactive xenon diffuses rapidly through the m yocardium and is then gradually removed over the next two to three minutes.
W ith the aid o f a com puter the clearance can be measured in several ways, the clearance rate (K ) calculated, and the m yocardial blood flow derived in ml/mm/gram.By contrast, in an in farct 3 days old clearance m ay take as long as 20 minutes, because o f the low rate o f collateral flow.Estimates o f collateral flow im m ediately after infarction have varied from negligible to ade quate, and the current continental and Russian view , recently supported in the Lancet, has been o f interest to us.It is suggested that collateral flow after infarction is at first adequate, but then falls to zero within a few hours as a result o f harmful sym pathetic vasoconstrictor reflexes which m ay be reversible.If this were so, there would be im portant therapeutic implications.
In both types o f infarcts in our study we found that im m ediate collateral flow was established at about a quarter to a third o f norm al m yocardial blood flow.T here is usually a slight rise in col lateral flow in the first hour and then a slow fall to the original level, but there is no evidence that it ever falls to zero.O ccasionally this fluctuation is very marked.In one instance, the collateral flow was initially quite low, then rose to the highest we recorded, and finally fell until death from ventricular fibrillation after 5 hours.W e have no idea w hat causes these fluctuations which are unrelated to blood pressure.
W e have now recorded collateral flow just before or after resuscitation from ventricular fibrilliation on five occasions, and found that it was always very low, ranging from .05 to .15 ml/min/gram o f infarct.This is about -15 per cent o f the normal value for the heart, a rate which is obviously dangerously inadequate.T w o o f these animals were successfully resuscitated with sub sequent im provem ent in clearance.
W e have exam ined the effects o f general and local sym pathetic blockade as well as surgical sym pathectom y without dem onstrating any bene ficial effect on collateral flow, but it does increase slowly on successive days after recovery from m yocardial infarction.T here is usually little change over the first 3 days, but then it begins to rise and by about the 8th d ay reaches approxi m ately 5 o f the normal value.H ow ever this does not always happen.After an initial high level of collateral flow, the rate m ay fall and this m ay be followed by the death o f the animal.
T here are o f course marked individual varia tions, but the usually inadequate early collateral flow is established through the fine interarteriolar anastomoses which we know do exist in the norm al heart o f both dog and m an, and the subse quent increase by the 4th day is due to progressive enlargem ent o f some o f these channels.A fter 10 days, there are often quite large anastomoses around the obstruction in the artery, around the apex in the septum.H ow responsive is this col lateral vasolature during recovery?W e are per haps most interested in the capacity for vasodila tion, and for this reason w e tested dipyridom e, the most active coronary dilator drug at present available.U nd er its influence, collateral flow in creases, only slightly at first, but substantially more as the infarct ages.T h is response though consider able remains less than in norm al m yocardium .O ur results show that collateral channels are almost m axim ally dilated in the first hours o f infarction, but the greater increments in collateral flow caused on subsequent days indicate that the col lateral vessels are then no longer fully vasodilated.
W e have also found that collateral flow varies with the systemic blood pressure so that reductions in flow accom pany the hypotension o f shock and large doses o f barbiturate.Conversely the use of pressor drugs is accom panied by an increase in collateral flow.T h e action o f nitrites in angina is likely to rem ain controversial until we can decide where in the m yocardium angina rises, and until we can measure blood flow at this focus.M eanw hile it m ay be unwise to dismiss the pos sibility that nitrites act in part by increasing flow.
In fact our studies support the alternative view that an im portant function o f glyceryl trinitrate is to reduce the m yocardial oxygen requirement, even though there is a transient in itial increase in exteonal cardiac work.T h is increased efficiency o f energy utilization is not a specific effect o f the drug but the norm al consequence o f increasing cardiac out put at the same time as a fall in blood pressure takes place.W h at o f the eflfect on flow ?In the normal heart it is easy to dem onstrate a biphasic action on m yocardial blood flow.T h e initial increase is, however, transient and would easily be missed in studies in patients, which are harder to control.
When we measured the effect o f glyceryl trini trate on blood flow to the ischaem ic m yocardium we found an increase o f collateral flow in 4 out o f 6 anim als given the drug 6 hours after infarction.T h e increase was slight, transient and less than in normal anim als, but nevertheless occurred in the most unfavourable setting that could be devised.In those anim als that responded the m ean in crease was 25 per cent.W e were fortunate to come across a natural occurrence in one dog which closely resem bled the situation in a patient with ischaem ic heart disease and previous infarction.T hough spon taneous m yocardial infarction in dogs is rare, this anim al had a large anterior infarct showing fibro sis and new vessel formation.T h e anterior descending artery was sclerotic and much narrowed.In this anim al we were able to measure collateral flow to an ischaemic zone with an old infarct and recent (induced) coronary obstruc tion.N ot surprisingly the flow rates were a good deal higher than in the other dogs.T h ey were in fact sufficient to prevent further infarction, and it is o f particular interest that glyceryl trinitrate and diperidam ole increased the collateral flow.
M ost studies o f the glyceryl trinitrate in patients have been started too late to detect any vasodil ator effect, which anim al studies show to be over within 3 minutes o f giving the drug.B ut with appropriate timing an increase in m yocardial blood flow m ay be found even in patients with angina.In one such, a single tablet o f glyceryl trinitrate taken sublingually increased the m yo cardial flow rate from .6 1 to .70 ml/min/gram.

Conclusion:
In experim ental m yocardial infarction col lateral blood flow is im m ediately established at a rate o f about one quarter o f the norm al resting value.It usually rises to a m axim um after 2 hours, but is then liable to a slight and gradual reduc tion.O ccasionally, for reasons not at present apparent, large fluctuations occur.Neurone blockade has no effect either in preventing in farction or in increasing collateral flow.
A t any time during the early days after in farction, flow m ay fall below the potentially dangerous level o f about .15 m l/m in/gram o f infarct ( 15 per cent o f the norm al resting value) when ventricular fibrillation m ay occur.
O n average, collateral flow begins to increase spontaneously after 4 days, and norm al levels m ay be attained 10 days after infarction.For a few hours the collateral vessels are only slightly responsive to vasodilator drugs, presum ably because they are already m axim ally dilated, but they begin to respond some days before the spon taneous rise in collateral flow.
Finally, though glyceryl trinitrate has other effects, perhaps o f greater im portance, it m ay cause a modest increase in collateral flow to the ischaemic m yocardium .