Postgrad Med J: first published as 10.1136/pgmj.45.526.523 on 1 August 1969. Downloaded from

Postgracl. med. J. (August 1969) 45, 523-527.

Sympatho-adrenal hyperactivity-the key to irreversible ?

M. H. IRVING* Dudley Road Hospital, Birmingham

Summary of agreement has been obtained by workers who There would appear sound evidence on which to have investigated plasma adrenaline and noradrena- incriminate sustained sympatho-adrenal hyper- line levels in states of hypovolaemic and endotoxic activity as a deleterious factor of prime importance shock. Walton et al. (1959) recorded a mean adrena- in hypovolaemic shock. The use of sympatho- line level of 64 ,ug/l in dogs subjected to haemor- mimetics in the treatment of shock of this type is rhagic shock which corresponded with findings irrational and deleterious. Their rational use in subsequently published by Rosenberg et al. (1961). those few cases of clinical shock where they are A similar rise occurs in noradrenaline levels but indicated will have to await knowledge of plasma these invariably constitute a lesser fraction (20-40 Y.) catecholamine levels in such cases. of the total catecholamine content (Lund, 1951). Experimental studies involving a- and P-adrenergic In certain instances the level of catecholamine in blockade, unilaterally and in combination, have shock reaches exceptionally high values. Walton et revealed that, contrary to previous theories, deteriora- al. (1959) observed an adrenaline level of 220 ,ug/l in tion is mediated through both receptor types. a case of experimental haemorrhagic shock whilst Many of the current concepts of the pathophysio- Rosenberg et al. (1959) recorded an adrenaline level logy of shock will need reassessment in the light of as of 457 2 Fg/l in a case of experimental endotoxin by copyright. yet unappreciated facets of catecholamine activity. shock. Sequential studies of catecholamine levels in Introduction experimental haemorrhagic shock have demon- Of all the homeostatic mechanisms initiated by strated that they become maximal soon after haemor- most important rhage commences and remain high throughout the acute hypovolaemic the of & Bragg, 1957), only would appear to be an intense activation of the sym- period hypotension (Watts patho-adrenal system. Outpourings of adrenaline falling on correction of hypovolaemia or at the onset from the adrenal medulla and noradrenaline from of the 'irreversible' stage. con- The question of whether this fall in catecholamine

the post-ganglionic sympathetic nerve-endings http://pmj.bmj.com/ just prior to the irreversible stage, stitute the most immediate and most effective pro- levels, occurring represents an exhaustion of the sympatho-adrenal tective mechanism and are responsible for the majority of the physical signs observed in clinical system and is responsible for death remains contro- shock. versial. In support of such reasoning Hift & Campos This activation of the sympatho-adrenal system (1962) demonstrated the depletion of the myocardial has the of extensive experimental in- noradrenaline stores in dogs dying from irreversible been subject et vestigation, the results of which have led to the in- haemorrhagic shock. On the other hand Walker al. (1959) and Glaviano, Bass & Nykiel (1960) have corporation into current clinical therapy of tech- on September 28, 2021 by guest. Protected niques involving both adrenergic stimulation and shown that plasma catecholamine levels are still blockade. higher in dogs dying from irreversible haemorrhagic adrenergic in animals. It is the purpose of this communication to discuss shock than healthy control stimulate the increased output of some facets of sympatho-adrenal function in shock The factors that to its place in the development of catecholamines and maintain the elevated levels and consider both likewise been the subject of the so called 'irreversible' state and its relationship observed in shock have to clinical therapy. much investigation. The initial stimulus is un- doubtedly neurogenic, hypotension causing reflex The extent of adrenergic activity in experimental autonomic activity which is conveyed to the adrenal shock states medulla via the greater splanchnic nerves (Beck & Dontas, 1955). The adrenomedullary response in Despite the relative unreliability of methods used such circumstances is thus largely abolished by for measuring plasma catecholamines a broad level section of the greater splanchnic nerves (Tachi, *Present address: North Middlesex Hospital, London. 1928). Once hypotension is established other factors Postgrad Med J: first published as 10.1136/pgmj.45.526.523 on 1 August 1969. Downloaded from

524 M. H. Irving develop which stimulate further secretion. Amongst The second observation, stimulated to a large these are said to be metabolic acidosis (Malm et al., extent by the knowledge that infusion of adrenaline 1966), anoxia (Walton et al., 1959) and hypovolaemia caused irreversible hypotension, was that surgically (Fowler, Shabetai & Holmes, 1961). Catecholamine sympathectomized animals, although reduced to a levels remain high throughout shock because of lower level of blood pressure by a measured blood continued release of hormones from the adrenal loss compared with intact animals, did not become medulla and not because of interference with the as acidotic nor did they as readily relapse into irrever- mechanisms that remove them from the plasma. sible shock (Freeman et al., 1938). Other workers Adrenalectomy during hypovolaemia causes the using varying techniques to achieve sympathetic complete disappearance of adrenaline from the denervation subsequently verified these observations plasma (Watts, 1965). (Kleinberg et al., 1942; Eversole et al., 1944). Knowledge of plasma catecholamine levels in The immediate assumption was that the benefit in clinical shock and preterminal states is remarkably these cases was obtained by prevention of compen- scanty. Rhoads & Howard (1963) have, however, satory vasoconstriction with consequent main- documented urinary adrenaline and noradrenaline tenance of tissue perfusion and tissue nutrition. levels in man after trauma and demonstrated that Thus parallel to these somewhat crude surgical the levels are highest 8 hr after injury and may experiments the more refined method of pharma- remain high for several days. cologically induced vasodilatation was investigated. Beginning with ergot, and progressing to dibenamine Adrenergic hyperactivity in shock: is it deleterious? and phenoxybenzamine, considerable investigation It is obvious that the initial adrenergic response of the effects of vasodilator therapy upon experimen- to acute hypovolaemic hypotension is a vital homeo- tally induced haemorrhagic and endotoxin shock static mechanism without which the organism is less was undertaken. able to tolerate the stresses to which it is subjected. Indeed the inhibition of this compensatory response Mechanisms in catecholamine-induced deterioration

in an incompletely resuscitated patient, e.g. by Although Ahlquist (1948) who laid the founda-by copyright. anaesthesia, may precipitate a cardiovascular col- tions of the modern adrenergic receptor concept had lapse. clearly nominated two receptor types (cx and D) as Two principal observations however, aroused the the mediators for adrenergic activity, the question suspicion that sustained sympatho-adrenal activity of adrenergic mechanisms in shock has been domi- in such circumstances could be deleterious. The first nated by the role of the cx receptor. Deterioration has, resulted from investigations ofthe effect ofprolonged apparently quite logically, been attributed to the adrenaline and noradrenaline infusion into healthy vasoconstriction - tissue-underperfusion - tissue- animals. Using this technique several workers, hypoxia cycle (Fig. 1). notably Bainbridge & Trevan (1917), Freeman (1933) Blood loss and Watts & Westfall (1964), produced states of http://pmj.bmj.com/ irreversible hypotension comparable with those seen after protracted haemorrhagic hypotension. The Faol in cardiac output Compensatory sympotho- odrenal similarities between haemorrhagic and adrenaline- induced vasoconstriction infusion shock are noteworthy. In both a profound metabolic acidosis occurs, due principally to a rise in lactic acid, together with anuria and a paralytic ileus. Tissue underperfusion In dogs both stimuli cause portal hypertension, haemorrhagic necrosis of the intestine and a pro- on September 28, 2021 by guest. Protected gressively rising haematocrit in the irreversible stage, Tissue hypoxia said to be the result of plasma loss. The similarity between the two conditions stimu- lated Watts & Westfall (1964) to compare the cate- Metobolic acidosis and cellular damage cholamine levels in each. They found that a con- FIG. 1 tinuous infusion of 3-4 ,ug/kg/min of adrenaline sustained a plasma adrenaline level of 39 ,ug/l and It followed that to prevent or reverse the vaso- produced a state of irreversible hypotension cul- constrictive element would be a logical supportive minating in death. They commented that if this measure in improving the tolerance to shock- plasma level of adrenaline alone was enough to cause inducing stimuli. Support for this concept was forth- a fatal outcome then surely the level of 46 1 pcg/l of coming from Wiggers et al. (1947), who found that endogenous catecholamine observed in haemor- the oc-blocking agent, dibenamine, gave protection rhagic hypotension must in itself be of significance. against the development of irreversible shock if Postgrad Med J: first published as 10.1136/pgmj.45.526.523 on 1 August 1969. Downloaded from

Sympatho-adrenal hyperactivity 525 administered to an animal prior to the onset of blockade was, by and large, ineffectual in preventing haemorrhagic hypotension. the fundamental lesions caused by sympatho- These findings were supported by Remington et adrenal hyperactivity. It thus appeared logical to al. (1948), since when numerous other workers have completely block adrenergic activity by using a- attested the value of this and the newer oc-adrenergic and (-blocking agents in combination. blocking agents in preventing the development of The reasoning appeared to be confirmed when, irreversible shock if given as a premedication prior using both sheep and dogs, we demonstrated that to many forms of shock-inducing stimuli. premedication with ao- and (-adrenergic blocking There is almost universal agreement that adminis- agents in combination so improved the tolerance of tration of ac-blocking agents after the onset of shock the animals that they were able to survive periods of does not give similar protection to that demon- haemorrhagic hypotension and adrenaline infusion strated by premedication, and in many cases causes which would prove fatal to non-medicated animals rapid deterioration. It is probably because of this (Halmagyi et al., 1967a). In addition the experiments that the use of a-blocking agents in clinical shock revealed an adrenergic basis for facets of the shock has, by and large, been so disappointing. syndrome that are not normally connected with Not all workers, however, were able to obtain this catecholamine activity. Thus the rise in plasma benefit by using prophylactic a-adrenergic blockade, lactic acid normally observed in haemorrhagic hypo- some actually recording a deterioration in the tension was markedly curtailed, whilst the anuria resistance of the animals so treated (Ingraham, and paralytic ileus that usually accompany the shock Roemhild & Goldberg, 1948; Brandfonbrenner & state were prevented. It remains to be seen whether, Geller, 1952; Rush, Rosenberg & Spencer, 1965) as we suspect, other aspects of the shock syndrome whilst Baue, Johnson & Parkins (1966) found that such as osmolality and coagulation-mechanism far from reducing the extent of the shock-induced changes are likewise dependent upon catecholamine metabolic acidosis (usually accepted as an index of activity. tissue perfusion) oc-adrenergic blockade made it worse. It was particularly noteworthy that in animals un-

In our own experiments (Halmagyi, Gillett & der combined a- and (-blockade cardiovascular by copyright. Irving, 1967a; Halmagyi et al., 1967b; Irving, 1968; dynamics and responses were more like those of Irving et al., 1968) we have, in common with these non-medicated control animals than in those with latter workers, shown that premedication with unilateral blockade. Thus, the fact that there was phenoxybenzamine does not protect against either near-normal vasoconstrictive activity in the animals the metabolic acidosis or the onset of irreversible under combined blockade suggests that current shock in both severe haemorrhagic and adrenaline- theories implicating sustained vasoconstriction and infusion shock. tissue underperfusion as the cause of metabolic It is only recently, since the advent of the (- acidosis and irreversible shock need reassessment. adrenergic blocking agents, that attention has been In support ofthis our observation that ac-blockade, directed to the role of (-receptor activity in shock. although markedly reducing the vasoconstrictive http://pmj.bmj.com/ Berk et al. (1967) suggest that excessive activity by response to haemorrhage and abolishing it in adrena- creating vasodilatation in the splanchnic and pul- line infusion does not affect the extent of the meta- monary areas is primarily responsible for the bolic acidosis, is undoubtedly of significance. This is deterioration into the irreversible state seen after in striking contrast with the animals under com- sustained haemorrhagic hypotension. They claim bined adrenergic blockade, who in spite of a fall in that (-adrenergic blockade with propranolol is oxygen uptake develop no or minimal acidosis (Halmagyi et al., 1967a). capable of preventing the haemorrhagic pulmonary on September 28, 2021 by guest. Protected and splanchnic vascular congestion observed in dogs If, as we believe, this means that metabolic acidosis subjected to haemorrhagic hypotension, and thus is the result of factors other than tissue underperfu- increases the resistance to irreversible change. sion and tissue hypoxia a review of its allegedly Their results have, however, to be assessed against deleterious effects is indicated. the fact that such profound dilatation in the splanch- Whatever the actual mechanism of adrenergic- nic region is a species-specific character that is not induced deterioration it is clear that it is mediated observed in sheep, primates or man. Additionally it through both receptor types. Future work must look must be remembered that their results were only at the problem of irreversible cellular damage at a achieved after myocardial function had been sup- much more fundamental level than the present vaso- ported with atropine, digitalis and calcium, whilst constriction-tissue underperfusion-theory. profound hypoglycaemia had been corrected with glucose infusion. The clinical application The foregoing results together with our own As long ago as 1895 Oliver & Schafer were quick experiments convinced us that unilateral receptor to advocate the use of their newly discovered extract Postgrad Med J: first published as 10.1136/pgmj.45.526.523 on 1 August 1969. Downloaded from

526 M. H. Irving of adrenal gland in the treatment of circulatory amine probably acts by countering the continuing failure on the grounds that it caused a sharp rise in sympatho-adrenal-induced vasoconstriction and blood pressure. thereby allowing the full extent of the functional This pattern of thinking has in the past dominated hypovolaemia to be revealed and corrected. In this the treatment of shock and sympathomimetics were way 'normovolaemia' and normotension are re- widely used to raise the arterial blood pressure in stored and adrenergic activity curtailed. hypotensive patients. It is only recently that the There would not appear to be any grounds for importance of adequate plasma volume expansion advocating f-adrenergic blockade in the treatment has been stressed and the deleterious effects of pres- of shock. sor therapy fully recognized. Those who still advocate Although combined cx- and r-blockade is already the use of sympathomimetics in shock states claim used in the management of patients with phaeo- they do so more because of their effects upon the chromocytoma it, as yet, has no place in the manage- myocardium than upon the peripheral circulation. ment of the shocked patient. Indeed we have shown There would nowadays be few clinicians who in our experiments that the administration of com- would advocate sustained sympathomimetic in- bined blockade to an animal already in shock is fusion in the treatment of traumatic, endotoxic or associated with cardiovascular collapse and death. hypovolaemic shock. Indeed it may be said that the Its future lies in further experimental work by re- use of such agents in these conditions, where plasma vealing fully the role of catecholamines in the catecholamines have been shown experimentally to physiology and biochemistry of shock. It may find be extremely high, would not only be unphysiologi- clinical application in the premedication of patients cal but frankly deleterious. Their use in cardiogenic who are likely to be exposed for long periods to high shock is more difficult to assess for it cannot be catecholamine levels, such as in cardiac bypass. denied that in some cases their exhibition has been of distinct value. Rational clinical use of sympatho- References mimetics in shock states will have to await know- AHLQUIST, R.R. (1948) A study of the adrenotropic receptors.

ledge of actual plasma catecholamine levels in the Amer. J. Physiol. 153, 586. by copyright. varying types of clinical shock. BAINBRIDGE, F.A. & TREVAN, J.W. (1917) Surgical shock and some allied conditions. Brit. med. J. 1, 381. The unilateral boosting of 3-adrenergic activity BAUE, A., JOHNSON, D.G. & PARKINS. W.M. (1966) Blood holds a special place in this respect. It has been shown flow and oxygen consumption with adrenergic blockade in experimentally by Halmagyi, Horner & Starzecki hemorrhagic shock. Amer. J. Physiol. 211, 354. BECK, L. & DONTAS, A.S. (1955) Vasomotor activity in (1965) and clinically by du Toit et al. (1966), that the hemorrhagic shock. Fed. Proc. 14, 318. short-term selective boosting of 5-activity with iso- BERK, J.L., HAGEN, J.F., BEYER, W.H., DOCHAT, G.R. & LA prenaline is of value in embolic and endotoxin shock. POINTE, R. (1967) The treatment of hemorrhagic shock The beneficial effect appears to stem from the in- by beta adrenergic receptor blockade. Siurg. Gynec. Obstet. crease in cardiac output, combined with pulmonary 125, 311. BRANDFONBRENNER, M. & GELLER, H.M. (1952) Effect of http://pmj.bmj.com/ vascular and bronchial dilatation, in a condition dibenamine on renal blood flow in hemorrhagic shock. where the cardinal pathogenetic factor is a profound Amer. J. Physiol. 171, 482. pulmonary vasoconstriction. This phenomenon Du TOIT, H.J., Du PLESSIS, J.M.E., DoMMISSE, J., RORKE, appears to be in a category of its own and the use of M.J., THERON, M.S. & DE VILLIERS, V.P. (1966) Treatment sustained infusions of isoprenaline in shock states is of endotoxin shock with isoprenaline. Lancet, ii, 143. EVERSOLE, W.J., KLEINBURG, W., OVERMAN, R.R., REMING- probably as illogical as the similar use of other sym- TON, J.W. & SWINGLE, W.W. (1944) The nervous factor in pathomimetics. shock induced by muscle trauma in normal dogs. Amer.

In contrast the use of adrenergic blocking agents J. Physiol. 140, 490. on September 28, 2021 by guest. Protected in the treatment of shock is still a largely untried FOWLER, N.O., SHABETAI, R. & HOLMES, J.C. (1961) Adrenal form medullary secretion during hypoxia, and rapid of therapy. There is little doubt that in some intravenous infusion. Circulat. Res. 9, 427. shocked patients where, after adequate transfusion, FREEMAN, N.E. (1933) Decrease in after pro- a high central venous pressure is associated with longed hyperactivity of the sympathetic nervous system. arterial hypotension and evidence of peripheral Amer. J. Physiol. 103, 185. vasoconstriction, the use of an ax-blocking agent, FREEMAN, N.E., SHAFFER, S.A., SCHECTER, A.E. & HOLLING, such as is H.E. (1938) The effect of total sympathectomy on the phenoxybenzamine, of value. The benefit occurrence of shock from hemorrhage. J. clin. Invest. 17, from phenoxybenzamine in such circumstances is 359. probably obtained in a different manner from that GLAVIANO, V.V., BASS, N. & NYKIEL, F. (1960) Adrenal demonstrated in animal experiments. In the latter medullary secretion of epinephrine and norepinephrine in circumstances blockade is applied to a healthy, dogs subjected to hemorrhagic hypotension. Circtulat. Res. 8, 564. normotensive, normovolaemic animal thereby pre- HALMAGYI, D.F.J., GILLETT, D.J. & IRVING, M.H. (1967a) venting the vasoconstrictive response to haemor- Partial and 'complete' adrenergic blockade in post hemor- rhage. In the clinical setting described phenoxybenz- rhagic shock. J. appl. PhYsiol. 22, 487. Postgrad Med J: first published as 10.1136/pgmj.45.526.523 on 1 August 1969. Downloaded from

Symlpatho-adrenal hyperactivity 527

HALMAGYI, D.F.J., HORNER, G.J. & STARZECKI, B. (1965) ROSENBERG, J.C., LILLEHEI, R.C., LONGERBEAM, J. & ZIMMER- Acute cor pulmonale and shock. Med. J. Aust. 2, 141. MAN, B. (1961) Studies on hemorrhagic and endotoxin HALMAGYI, D.F.J., IRVING, M.H., GILLETT, D.J. & VARGA, shock in relation to vasomotor changes and endogenous D. (1967b) Effect of adrenergic blockade on consequences circulating epinephrine, norepinephrine and serotonin. of sustained epinephrine infusion. J. appl. Physiol. 23, 171. Ann. Surg. 154, 611. HIFT, H. & CAMPOS, H.A. (1962) Changes in subcellular ROSENBERG, J.C., LILLEHEI, R.C., MORAN, W.H. & ZIMMER- distribution of cardiac catecholamines in dogs dying in MAN, B. (1959) Effects of endotoxin on plasma catechol- irreversible haemorrhagic shock. Nature (Lond.), 196, 678. amines and serum serotonin. Proc. Soc. exp. Biol. (N. Y.), INGRAHAM, R.C., ROEMHILD, F. & GOLDBERG, H. (1948) 102, 335. Further observations on the effect of pentobarbital and RUSH, B.F., ROSENBERG, J.C. & SPENCER, F.C. (1965) Effect of an adrenolytic agent upon the survival of animals sub- of dibenzyline treatment on cardiac dynamics and oxida- jected to a procedure resulting in experimental hemor- tive metabolism in hemorrhagic shock. Ann. Surg. 162, rhagic shock. Fed. Proc. 7, 60. 1013. IRVING, M.H. (1968) The sympatho-adrenal factor in TACHI, H. (1928) Influence of hemorrhage upon the epine- haemorrhagic shock. Ann. roy. Coll. Surg. Lond. 13, 300. phrine (Folin) content of the suprarenals of the rabbit. IRVING, M.H., VARGA, D., GILLETT, D.J. & HALMAGYI, Tokohu J. exp. Med. 10, 409. D.F.J. (1968) The effect of adrenergic blockade upon WALKER, W.I., ZILELI, M.S., REUTTER, F.W., SHOEMAKER, shock-induced lacticacidosis in sheep. Brit. J. Surg. 55, 780. W.C., FRIEND, D. & MOORE, F.D. (1959) Adrenal medul- KLEINBERG, W., REMINGTON, J.W., DRILL, V.A. & SWINGLE, lary secretion in hemorrhagic shock. Amer. J. Physiol. W.W. (1942) The nervous factor in the circulatory failure 197, 773. induced in adrenalectomized dogs by intestinal stripping and a single stage bilateral adrenalectomy. Amer. J. Physiol. WALTON, R.P., RICHARDSON, J.A., WALTON, R.P., JR & 137, 362. THOMPSON, W.L. (1959) Sympathetic influences during LUND, A. (1951) Release of adrenaline and noradrenaline hemorrhagic hypotension. Amer. J. Physiol. 197, 223. from the suprarenal gland. Acta. pharmacol. (Kbh), 7, 309. WATTS, D.T. (1965) Adrenergic mechanisms in hypovolemic MALM, J.R., MANGER, W.M., SULLIVAN, S.F., PAPPER, shock. Shock and Hypotension. Grune & Stratton, New E.M. & NAHAS, G.G. (1966) The effect of acidosis on York. sympathoadrenal stimulation. J. Amer. med. Ass. 197, 121. WATTS, D.T. & BRAGG, A.D. (1957) Blood epinephrine levels OLIVER, G. & SCHAFER, E.A. (1895) The physiological effects and automatic reinfusion of blood during hemorrhagic ofextracts from the suprarenal capsules. J. Physiol. (Lond.), shock in dogs. Proc. Soc. exp. Biol. (N. Y.), 96, 609. 18, 230. WATTS, D.T. & WESTFALL, V. (1964) Studies on peripheral

REMINGTON, J.W., WHEELER, N.C., BOYD, G.H. & CADDELL, blood catecholamine levels during hemorrhagic shock in by copyright. H.M. (1948) Protective action of dibenamine after hemor- dogs. Proc. Soc. exp. Biol. (N.Y.), 115, 601. rhage and muscle trauma. Proc. Soc. exp. Biol. (N. Y.), 69, WIGGERS, H.C., ROEMHILD, F., GOLDBERG, H. & INGRAHAM, 150. R.C. (1947) The influence of prolonged vasoconstriction on RHOADS, J.E. & HOWARD, J.M. (1963) The Chemistry of the transition from impending to irreversible hemorrhagic Trauma. Thomas, Springfield, Illinois. shock. Fed. Proc. 6, 226. http://pmj.bmj.com/ on September 28, 2021 by guest. Protected