632

THE PRE-OPERATIVE AND POST-OPERATIVECARE OF PATIENTS RECEIVING CORTISONE

OR OTHER STEROID THERAPYBy ROBERT R. DE MOWBRAY, D.M., M.R.C.P.

Senior Medical Registrar, St. Bartholomew's Hospital, London

Surgical operations may be hazardous onpatients who are being treated with steroid hor-mones or who have recently received such therapy.This is a problem which is likely to occur morefrequently now that cortisone and the A,-steroids,prednisone and prednisolone have been madegenerally available. The situation may arise inone of two ways. First, the patient may be re-ceiving small (physiological) doses of cortisone asreplacement therapy for adrenal insufficiency, dueeither to Addison's disease, to panhypopituitarismor to total adrenalectomy. Secondly, patientsmay be receiving large (pharmacological) doses ofcortisone or A,-steroids for the treatment ofsuch conditions as rheumatoid arthritis, collagendisease or ulcerative colitis. Such patients mayhave to undergo surgical operations on accountof some incidental disease or for specific compli-cations of steroid therapy such as perforation of apeptic ulcer. While they are receiving steroidhormones in large doses, and for some months or ayear or more after steroid therapy has been dis-continued, these patients are suffering frompotential adrenal insufficiency due to suppressionof endogenous adrenal cortical function. Carefulpre-operative assessment and management ofpatients with actual or potential adrenal in-sufficiency is of vital importance. Post-operativecare is but a continuation of the pre-operativeregime; to postpone treatment until after theoperation may be to court disaster, for it is oftendifficult to resuscitate the patient once profoundadrenal failure has supervened.

Physiological BackgroundThe Adrenal Response to Surgical Stress. It has

been known for many years that patients withadrenal insufficiency are incapable of respondingadequately to surgical operations. This applieseven to minor procedures such as dental ex-tractions. Thorn (I951) stated that no majorsurgical operation had been performed success-

fully in a patient with classical signs and symptomsof Addison's disease until deoxycortone acetatewas introduced in 1938. The role of the adrenalcortex in response to stress was elaborated bySelye from I936 onwards and since then evidenceof increased adrenal cortical activity has beendemonstrated during and after surgical operationson otherwise healthy individuals. This is shownby the fall in the eosinophil count in the blood(Laragh and Almy, 1948; Roche et al., I950;Hardy, I955),. by the increase in plasma levels of17-hydroxycorticoids (Tyler et al., I953, 1954;Sandberg et al., 1954; Franksson et al., 1954,I955; Elman et al., 1955; Hammond et al.,I956; Steenburg et al., I956; Helmrich et al.,1957; Virtue et al., I957) and by the increasedoutput of corticoids and 17-ketosteroids in theurine (Venning et al., I944; Browne and Venning,1947; Forbes et al., I947; Thorn et al., I953;Cope and Hurlock, I954; Moncrieff et al., 1954;Tompsett and Smith, I954; Hardy, 1955; Mooreet al., 1955; Hammond et al., 1956; Jepson et al.,I956). The adrenal cortical response to stress isthought to be due to stimulation of thehypothalamo-hypophyseal pathways, resulting inan increased output of corti cotrophin (ACTH) bythe anterior lobe of the pituitary gland (Thorn etal., 1953). Increased corticotrophic activity hasin fact been demonstrated in the blood afteroperations (Taylor et al., I949; Bornstein andTrewhella, I950).

Suppression of Adrenal Cortical Function by theadministration of Steroid Hormones. The ad-ministration of steroid hormones leads to sup-pression of endogenous adrenal cortical activity.This is shown by the occurrence of symptoms ofadrenal insufficiency after sudden withdrawal ofthe hormone, by the fall in output of 17-ketosteroids in the urine during steroid therapy,and by atrophy of the adrenal cortex and depletionof its lipid at autopsy. The capacity of theadrenal cortex to respond to stimulation by

by copyright. on M

ay 11, 2022 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.33.386.632 on 1 D

ecember 1957. D

ownloaded from

DE MOWBRAY: Care of Patients Receiving Cortisone

corticotrophin is also impaired by intensivesteroid therapy. This accounts for the poor re-sponse to surgical stress which patients are liableto show after prolonged and intensive administra-tion of steroid hormones (Sprague et al., I950;Thorn et al., 1953). The risk is probably noless in patients receiving corticotrophin (ACTH)than in those who are being treated with cortisone(Salassa et al., 1953; Hayes, 1956). Prolongedtreatment with corticotrophin leads to degranu-lation and later hyaline changes in the basophilcells of the anterior pituitary lobe, similar to thoseinduced by cortisone therapy, together withbasophil stippling of the chromophobe cells(Golden et al., 1950; Bennett, 1954). Thesechanges suggest that the administration of eithercortisone or corticotrophin interferes with therelease of endogenous corticotrophin by theanterior pituitary lobe. Furthermore, the in-creased secretion of adrenal steroids under theinfluence of exogenous corticotrophin is likely todepress the output of endogenous corticotrophin.However, adrenal insufficiency resulting fromwithdrawal of corticotrophin therapy is likelyto be less prolonged than after withdrawal ofcortisone (Thorn et al., I953). It is possible thatthe operative risk will prove to be greater inpatients receiving treatment with prednisone orprednisolone than in those being treated withcortisone, since the former steroids are meta-bolized more slowly (Slaunwhite and Sandberg,I957) and are more potent in inhibiting the outputof steroids in the urine than are equivalent dosesof cortisone (Nabarro et al., 1955; Spence et al.,I956).

Post-operative Collapse in Patients ReceivingSteroid TherapyThe risk of operating on patients with Addison's

disease has already been emphasized and it stillexists in patients who are receiving replacementtherapy, unless the dose of cortisone is sub-stantially increased over the period of the opera-tion. The same applies to patients with adrenalinsufficiency secondary to hypopituitarism (In-graham et al., I952; Kyle et al., 1953; Troenand Ryneasron, I955). The operative risk is lesseasily appreciated in patients who are receivingpharmacological doses of corticotrophin or steroidhormones, for such patients may show some ofthe features of Cushing's syndrome, as a result ofexcessive endocrine therapy, and yet may fail toproduce an adequate adrenal response to surgeryor other forms of stress (Thorn et al., 1953).Most authors stress the dangers of operating on

patients during or after pharmacological steroidtherapy, yet, as Salassa et al. (1953) point out, manythousands of these patients must have undergone

surgical operations and comparatively few disastershave been reported. The writer has succeeded incollecting from the literature only 33 cases ofpost-operative shock in patients who have receivedcorticotrophin or steroid therapy in pharmaco-logical doses. Fifteen of these cases were reportedby Hayes (1956) and were drawn from a group of28 patients who underwent surgical operationsduring or after endocrine therapy; that is to say,the incidence of post-operative shock in patientswho had received corticotrophin or steroidtherapy in pharmacological doses was just over 50per cent. In contrast, there was no episode ofpost-operative shock among 25 patients who weresuffering from diseases amenable to steroidtherapy but who had not been treated with steroidhormones. Five of the patients developed post-operative shock at intervals of 3 to 24 months(mean i I months) after steroid therapy had beenwithdrawn. One patient collapsed after anoperation while still receiving corticotrophin(ACTH) in unchanged doses and post-operativeshock occurred in another patient from whomcorticotrophin had been withdrawn two monthspreviously. Slaney and Brooke (I957) reportedseven cases, of which six occurred in patientswith ulcerative colitis. These patients had re-ceived total doses of 2 to 9 g. of cortisone overperiods of 2 to 6 weeks. Three of their patientshad no operative cover with steroid hormones,although two of them had received corticotrophinfor 5 to I4 days before operation, and all three diedin post-operative shock. The other four patientsreceived extra doses of cortisone during theoperative period, and although they all collapsedafter operation they were resuscitated by furtherdoses of cortisone or intravenous hydrocortisone;one of them died after a second operation whichwas complicated by a burst wound. Slaney andBrooke subsequently performed ten operations oneight patients with ulcerative colitis who had pre-viously been treated with steroid hormones; inall these patients the operation was systematicallycovered with intensive doses of intramuscularcortisone and intraveneous hydrocortisone and nopost-operative complications ensued. Nicholas etatl. (1955) reported two cases (one of them fatal),Salassa et al. (1953) reported two fatal cases andFraser et al. (1952), Lewis et al. (1953) reportedone fatal case each of post-operative shock, aftertotal doses of 8 to I4 g. of cortisone or more overperiods of four months to five years. Allanby(I957) recorded a similar case after a total doseof 56o mg. of prednisolone over a period of fourweeks. Forbes (1952) reported death soon aftera partial gastrectomy in a patient with a gastriculcer who had recently been given cortisone, ioomg. daily for three weeks, but he did not describe

December .1957 633by copyright.

on May 11, 2022 by guest. P

rotectedhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.33.386.632 on 1 Decem

ber 1957. Dow

nloaded from

POSTGRADUATE MEDICAL JOURNAL

the mode of death. Kern (I957) mentions apatient with rheumatoid arthritis who died ofadrenal failure 24 hours after the incision of aninfected olecranon bursa under local anaesthesia.Only two of these last nine patients receivedadequate doses of steroid over the operative period(Nicholas et al., I955). In three of the casessteroid therapy had been discontinued two days,four days and five months respectively beforeoperation (Fraser et al., 1952; Allanby, I957;Salassa et al., I953). The possibility of adrenalfailure was apparently considered in only four ofthe patients and they were given additionalcortisone intramuscularly and hydrocortisone in-travenously (Salassa et el., 1953; Nicholas etal., 1955). Thus in i i out of I7 cases the outcomewas fatal, only one of these patients having hadadequate steroid therapy over the operative period.Harnagel and Kramer (I955) reportpd acuteadrenal failure in a man aged 21 years withspondylitis, I2 hours after manipulation of thespine under thiopentone anaesthesia. The patienthad received a total of i.9 g. cortisone over theprevious month, the last dose being given i8 hoursbefore the manipulation. The patient wasresuscitated only after four days' intensive therapywith intravenous corticotrophin and corticalextract and intramuscular cortisone. The, I7-ketosteroid output in the urine was still only 2.2mg. per 24 hours at the end of this time. Hayes(1956) does not mention the outcome of his I5cases but Hayes and Kushlan (I956) describefour patients (presumably included in the sameseries) all of whom recovered with the help ofadditional corticotrophin over the period of theoperation.The writer has recently had experience of a

man aged 6i years with rheumatoid arthritis whohad been receiving steroid therapy for 21 years(cortisone 75 mg. daily for the first year and sub-sequently prednisone 25 mg. daily). After i imonths' treatment an arthrodesis of the rightankle and foot was performed uneventfully. Aftera total of 26 months he developed broncho-pneumonia, at which time he ceased to take hisprednisone tablets; he developed anuria andadrenal failure but was revived with ioo mg. in-travenous hydrocortisone and oral cortisone I50mg. daily. Six weeks later amputation of the rightleg was performed on account of a persistent dis-charging sinus in the right ankle. Although hewas given corticotrophin zinc 8o units daily in-tramuscularly for three days before operation, hedeveloped post-operative pneumonia and signs ofadrenal insufficiency. He eventually recoveredwith the aid of antibiotics, 200 mg. intravenoushydrocortisone, prednisone 40 mg. daily and thecontinued administration of corticotrophin zinc

8o units daily. The corticotrophin was thengradually withdrawn and the prednisone graduallyreduced to his usual dose of 25 mg. daily.Hayes (1956) points out that the response to

surgery cannot reliably be predicted from thetotal dosage, from the duration of therapy or fromthe interval between the conclusion of therapy andthe operation. Post-operative collapse occurredat intervals of up to two years after the end oftherapy in his cases and Salassa et al. and Slocumbet al. (I957) also consider that the danger may existfor as long as two years after the withdrawal ofcorticotrophin or steroid hormones. Obviouslythe degree and duration of adrenal suppressionvaries tremendously from patient to patient. Asjudged by the fall in eosinophil count and theincrease in steroid concentrations in the bloodand urine in response to exogenous corticotro-phin, pituitary adrenal inhibition may persist foranything from two to go days after the withdrawalof steroid therapy (Boland and Headley, 1951;McIntosh and Holmes, 1951; Ragan, I953;Thorn et al., 1953; Eik-Nes et al., 1955).

Bennett (I954) found that the weights of humanadrenal glands at autopsy were significantly re-duced if cortisone had been given up till the timeof death and for more than five days altogether butthe adrenal cortex was not atrophic if cortisonehad been discontinued more than 2o days beforedeath. However, the size and histologicalappearances of the glands do not necessarily reflecttheir functional activity. Assessment of theadrenal cortical response to corticotrophin duringlife cannot be entirely relied upon to predict orexclude the likelihood of post-operative collapse.Therefore it is safest to give full supportiveendocrine therapy before, during and after oper-ation in all patients who have been treated withcorticotrophin or steroid hormones, at any ratewithin the previous two years.

Surgical Complications of Peptic Ulcer inPatients Receiving Corticotrophin and SteroidTherapy

This problem deserves special consideration,for the physician and the surgeon are faced withthe dilemma of a patient in whom peptic ulcerationand its complications may be the direct result ofcorticotrophin or steroid therapy, yet who willtemporarily require even higher doses of thesehormones in order to respond to the stress ofperforation or haemorrhage and their respectivesurgical treatment. The writer has collected fromthe literature 2I cases of perforation, and 25 casesof haemorrhage complicating peptic ulcers, andthree cases in which both perforation and hae-morrhage occurred. In eight out of the 14 casesof perforation and in seven out of I2 cases of

634 December 1957by copyright.

on May 11, 2022 by guest. P

rotectedhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.33.386.632 on 1 Decem

ber 1957. Dow

nloaded from

DE MOWBRAY: Care of Patients Receiving Cortisone

haemorrhage in which details are available therewas no history of peptic ulcer before steroidtherapy was started. In I4 cases in which detailsare available perforation occurred after periods of5 to 37 days (mean 2o days) of treatment withcorticotrophin or steroid hormones or both. InI6 cases haemorrhage occurred after periods ofnine days to one year (mean six months) of treat-ment. Thus perforation has occurred consistentlywithin i to 5 weeks of therapy and haemorrhageafter variable and often lengthy periods of treat-ment. This suggests that perforation is a specificcomplication of steroid therapy while in manycases the haemorrhage may have been incidentaland might well have occurred in the absence oftreatment with these hormones. In I4 out of 17cases the perforation was treated surgically, in onecase it was treated conservatively and in two othersit was found unexpectedly at autopsy. Unevent-ful recovery followed in the patient treated con-servatively and in all except one of the patientstreated surgically. One of the patients reportedby Smyth (I95i) died after operation. In fourcases the symptoms and signs of perforation weremasked. In only two of the case reports it statedthat additional steroid therapy was given over theoperative period (Fentress et al., I956; Hess andMacpherson, I957). Of I4 patients with hae-morrhage 13 were treated conservatively and fourof these died; one patient was treated surgicallybut was found to have multiple erosions in thestomach and intestines and died soon afteroperation (Allanby, I957). In at least six of thesepatients steroid therapy was continued despitethe occurrence of bleeding though the doses weregradually reduced.

Judging from this analysis of the literature itwould appear that surgical treatment of a per-forated peptic ulcer is a relatively safe procedurein a patient receiving steroid therapy. This maybe due to the-fact that the perforation has in mostcases occurred within 3 to 5 weAks' treatment,which may not have been long enough to leadto profound suppression of pituitary-adrenalfunction.The writer has had recent experience of per-

forated peptic ulcer in two patients receivingsteroid therapy. One patient was a man aged 49years with rheumatoid arthritis who had receiveda total of 300 mg. of hydrocortisone by intra-articular injection followed by a total of approxi-mately i g. of prednisolone by mouth during thecourse of six weeks. The perforation was suturedwithout additional steroid cover and uneventfulrecovery ensued. The other patient was a womanaged 45 years with exophthalmic ophthalmoplegiawho had received 200 mg. cortisone daily byintramuscular injection for I7 days previously.

In this case cortisone was continued in the samedosage over the operative period and hydro-cortisone was given by intravenous infusion.

In spite of the favourable outcome of surgicaltreatment in most of these cases the writer wouldstill prefer to give additional steroid therapy overthe operative period in every case. The situationis less clear-cut in the case of gastro-intestinalhaemorrhage arising during steroid-therapy. Ifconservative management is decided upon, it isnecessary to weigh up the need for increased dosesof steroid, to meet the stress of haemorrhage(Ward et al., 1953) against the risk of aggravatingthe ulcer still further. If surgical managementis decided upon, increased doses of steroids willagain be required and, in the absence of previousevidence of peptic ulceration, there is the risk ofoperating on a patient with multiple acute pepticerosions or haemorrhagic gastritis, as in the casereported by Allanby (I957).The writer has had recent experience of a male

patient aged 52 years who had been treatedwith a total of approximately 25 g. of cortisoneover a period of eight months, followed by a totalof approximately 5 g. of prednisolone over aperiod of six months and was admitted to hospitalwith massive haematemesis and melaena. Thispatient had had no previous ulcer symptomswhatsoever and the writer hesitated to adv isesurgical treatment in view of the risks of adrenalfailure and the possibility of multiple pepticerosions. The patient improved temporarily withintravenous hydrocortisone but had repeatedmassive haemorrhages and died 24 hours afteradmission. At autopsy a subacute duodenal ulcerwas found which had eroded the superior pan-creatico-duodenal artery. In retrospect it wouldhave been preferable to administer more intensivesteroid therapy and to recommend surgical inter-ference as soon as the patient's general conditionwas adequately restored.

Details of Pre-operative and Post-operativeManagementThe occasional disasters which follow surgical

operations in these patients could be reduceddramatically by fully acquainting the patient andall who are responsible for his care with the factthat he is suffering from adrenal insufficiency orhas been treated with steroids and with thenecessity of taking certain precautions, should theneed for surgery arise. This can be done mostconveniently, as recommended by Forsham andThorn (I955) and by Slocumb et al. (I957), bygiving the patient a card, to be carried at alltimes and bearing his name and address, the nameand address of his private doctor and con-

December 1957 635by copyright.

on May 11, 2022 by guest. P

rotectedhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.33.386.632 on 1 Decem

ber 1957. Dow

nloaded from

POSTGRADUATE MEDICAL JOURNAL

sultant physician, the nature of his disease and thedose of steroid hormone which he is receiving.The patient should always carry a supply of corti-sone tablets with him and instructions should beincluded on the card to give him Ioo mg. cortisoneimmediately in case of emergency and to notify hisprivate doctor or physician at once. The patientor his relatives should be instructed to presenthis card to the surgeon or dental surgeon in caseof accident or impending operation or dentalextraction so that the necessary pre-operativetherapy can be undertaken. Now that steroidhormones are likely to be more generally used aspecific enquiry as to present or past treatmentwith these agents should become an indispensableitem in the taking of all patients' histories.

In view of the difficulty of predicting whichpatients are likely to develop adrenal failure afteroperation, it is wise to give additional doses ofsteroids in all cases over the operation period,regardless of the duration of steroid therapy andof the doses previously administered. The schemeoutlined by Thorn et al. (I953) is to be recom-mended and is applicable to all patients undergoingmajor surgery, whether the patient is sufferingfrom Addison's disease or has undergone totaladrenalectomy or is receiving pharmacologicaldoses of steroid hormones.

SCHEME OF PRE-OPERATIVE AND POST-oPERATIvE THERAPY(Thorn et al., 1953)

Pre-operative: Cortisone, ioo mg. i.m.i. I2 hr. pre-op;100 mg. i.m.i. 2 hr. pre-op.

During operation andfor four to six hours after operation:Hydrocortisone, IO mg. per hour by intravenousinfusion, in dextrose or dextrose saline solution.

Post-operative during first 24 hours: Cortisone, 50 mg.six-hourly i.m.i.

Days 2 and 3: Cortisone, 50 mg. eight-hourly i.m.i.Days 4 and 5: Cortisone, 50 mg. 12-hourly i.m.i.Day 6: Cortisone, 25 mg. six-hourly by mouth.Day 7 onwards: Reduce cortisone gradually to I2.5 to

25 mg. b.d. (plus deoxycortone acetate, i mg. dailyi.m.i., in cases of Addison's disease and post-adrenalectomy state only).

In the case of emergency operations this schememust be instituted immediately and the two pre-operative doses of cortisone should be combinedas a single injection of 200 mg. before operation.In the case of elective surgery the operationshould, if possible, be postponed until a physicianwith special experience of this problem has beenconsulted. In patients who are receiving or havereceived steroid therapy a preliminary course ofcorticotrophin (ACTH) should be given in orderto improve adrenal cortical function and thecapacity to respond to stress. Ideally this shouldbe controlled by estimations of the output ofI7-ketosteroids and/or I7-hydroxysteroids in theurine. To achieve maximal stimulation of the

adrenal cortex, 20 units of corticotrophin shouldbe given by intravenous infusion over a period ofeight hours each day, or Acthar-gel or corti-cotrophin zinc should be given intramuscularly indoses of 40 units 12 hourly. This should be con-tinued for one week, for although some patientson steroid therapy may respond to three days'stimulation with corticotrophin others may requirefour to six days' stimulation (Engleman et al.,1953). (Thorn et al., I953.) The dose of corti-sone should be reduced gradually by 12.5 to 25mg. every second day during the administration ofcorticotrophin (Thorn et al., 1953). The ad-ministration of corticotrophin is only an adjunct tothe pre-operative preparation of the patient andis not a substitute for cortisone and hydro-cortisone, which must still be given as outlinedover the operative period.The writer has had recent experience of an

asthmatic patient who had been receiving predni-sone in doses of 30 mg. daily for one year. Predni-sone was gradually withdrawn during the next twomonths and was followed by corticotrophin zinci85 units over eight days. Three weeks later theoutput of I7-ketostoeroids in the urine was only img. daily but was raised to 14.5 mg. daily by theadministration of corticotrophin-zinc within 48hours, in spite of the continued administration of30 mg. prednisone; when the prednisone wasgradually withdrawn and the corticotrophin wasdiscontinued after a total of io days the 17-ketosteroid output in the urine fell rapidly againto between 2 and 3 mg. daily.Thus treatment with corticotrophin does not

immediately restore the suppressed adrenal cortexto normal nor is the response always permanent,and its effect must be gauged by estimation ofsteroids in the urine, the results of which cannotbe obtained within less than 48 hours. For thisreason corticotrophin must not be relied on as asole means of pre-operative preparation, though itis undoubtedly beneficial.The reason for giving cortisone acetate intra-

muscularly over the operation period is that itsabsorption is slow and prolonged by this route andthe injection thus acts as a depot; it begins to actin 3 to 4 hours, reaches its peak effect in 8 to 12hours and continues for 24 hours or more (Thornet al., 1953). For the same reason this method ofadministration is inadequate in itself to counterthe acute stress of the actual operation; the lattermust be met with an intravenous infusion ofhydrocortisone which acts rapidly but ceases to beeffective after about 4 hours. Cortisone acetateis absorbed more rapidly from the oral than fromthe intramuscular route, and exerts its maximaleffect in 4 to 8 hours (Thorn et al., I953), but itsuse is, ofcourse, limited to the post-operative period

636 December 1957by copyright.

on May 11, 2022 by guest. P

rotectedhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.33.386.632 on 1 Decem

ber 1957. Dow

nloaded from

DE MOWBRAY: Care of Patients Receiving Cortisone

when the patient is able to take food and therapybv mouth.

Intravenous Therapy. Saline solutions shouldbe used with particular caution for the purpose ofintravenous therapy since patients who havealready been treated with cortisone or corti-cotrophin may be suffering from excessive re-tention of sodium and water. They may also bedepleted by potassium if they have been receivinglarge doses of cortisone or corticotrophin over longperiods. Potassium supplements may thereforebe necessary.

Other Measures. The administration of corti-cotrophin or steroid hormones is liable to inducea negative nitrogen balance (Sprague et al., 1950;Thorn et al., 1953). This will exaggerate thenegative nitrogen balance which is liable tofollow any surgical operation. The loss of nitrogenresulting from treatment with corticotrophin orcortisone can be compensated for by doubling theintake of calories and protein (Pearson et al., 1949)by the administration of 4 to 12 g. of potassiumchloride daily (Kinsell et al., 1952; Whitney andBennett, I952) and by androgen therapy. Theadministration of testosterone propionate 25 mg.daily by intramuscular injection has been shownto reverse the negative nitrogen balance inducedby 200 mg. of cortisone daily (Sprague, 1951).

Other Post-operative ComplicationsInfections. Treatment with corticotrophin or

steroid hormones increases the susceptibility ofpatients to infections. The inflammatory responsemay be masked even though the infection con-tinues unabated (Thorn et al., I953). Adrenalfailure is liable to supervene unless the doses ofsteroids are increased (Page, 1954; Jacobs andClifford Rose, 1955; Shaper and Dyson, 1955;Allanby, 1957).

Antibiotic therapy is therefore desirable over theperiod of operation, and provided this is given inadequate doses the risk of the increased doses ofsteroid hormones will be met (Thorn et al., 1953).

Delayed Healing of Wounds. This complicationis not as common or as serious as was originallyfeared. In the early days of treatment with corti-cotrophin and cortisone delayed healing was re-ported in a few instances in biopsy wounds, in-cisions into abscesses and bed sores (Ragan etal., 1949; Plotz et al., 1950 a.c.; Behrman andGoodman, I950; Creditor et al., I950; Videbacket al., I950; Levin et al., 1953). Delayed healingof wounds and fractures has also been reportedin numerous animal experiments but some workershave not obtained this effect. It is clear that thereis considerable variation in the effects on differentspecies and the effect is also dependent upon thedose of hormones used and upon the state of

nutrition (Ragan et al., I953). In many of theanimal experiments the doses were out of all pro-portion to those used clinically. Ragan (1953)found that healing was delayed in open granulat-ing wounds and in wounds with intestinal anasto-moses when cortisone was given in doses of morethan ioo mg. daily; on doses of less than ioo mg.daily healing was only slightly prolonged and thiswas overcome by leaving the sutures in place fora slightly longer period than usual. Thorn et al.(I953) state that serious interference with healingrarely occurs unless the doses of steroid are undulyhigh (more than I50 mg. cortisone daily) and ex-cessively prolonged and unless the intake ofprotein is inadequate.

Thrombo-embolic Complications. Views as to theinfluence of steroid therapy on thrombo-embolismare controversial. Cosgriff et al. (1950, 195I)originally reported that the clotting time ofvenous blood was shortened and recorded 48episodes of thrombosis and embolism in 700patients treated with corticotrophin or cortisone;this was a higher incidence than would have beenexpected without the use of endocrine therapy.Coste et al. (195i) reported five similar cases.Subsequent workers have not confirmed theseresults. Fahey (1951) found no significant effectson the clotting time. Variable effects on theclotting mechanisms of the blood were reportedby Monte et al. (I950), Smith et al. (1950) andMargulis et at. (1950). Russek et al. (I954) foundno instance of thrombo-embolic complicationsamong ioo patients with cardiovascular disease, in-cluding patients with angina pectoris and previouscardiac infarction. Allanby (I957) states that theincidence of thrombo-embolism amongst patientstreated with corticotrophin and steroid hormonesat Guy's Hospital ' has not appeared abnormal.'Hume and Moore (I95I) considered that thrombo-phlebitis was perhaps more liable to follow with-drawal of steroid therapy than it was to occurduring treatment; they suggested that the generalliability of surgical patients to post-operativevenous thrombosis might be related to the fallingoff of adrenal cortical activity after the increasedsecretion during the operation period. Steroidtherapy might therefore discourage rather thanprovoke thrombo-embolic complications. Indeed,striking subjective and objective improvement hasbeen reported in patients with thrombophlebitiswithin 24 hours' treatment with corticotrophin orcortisone (Hume and Moore, 1951; McGraw etal., 1952; Flickinger and Henderson, 1956).

Compltications in Ulcerative Colitis. Instancesof perforation of the colon have been reported inpatients with ulcerative colitis during the ad-ministration of steroid therapy. (Tulin and Almy,I952; Kirsner and Palmer, I951; Texter et al.,

December i9;- 637by copyright.

on May 11, 2022 by guest. P

rotectedhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.33.386.632 on 1 Decem

ber 1957. Dow

nloaded from

638 POSTGRADUATE MEDICAL JOURNAL December 1957

I953; Hayes and Kushlan, I956). Since thiscomplication is liable to occur in the absence ofsteroid therapy, it is not necessarily justifiable toattribute it to the use of hormones. On occasionsthe colon is said to have disintegrated duringmanipulation in patients who have been treatedwith steroids (Larkin and Flannery, I953; Brooke,1956).Some surgeons may therefore be disinclined to

use endocrine therapy when surgery is contem-plated for patients with ulcerative colitis. Un-fortunately, many of their patients are likely tohave been treated with steroids in the past and theymay be incurring a serious risk unless they covertheir operations in such patients with additionaldoses of cortisone and intravenous hydrocortisone.Slaney and Brooke (I957), who were formerlydisinclined to use steroid cover, have been forcedto reverse their policy. They point out that themortality of surgically treated patients who havepreviously received steroid therapy was 32 percent. in Truelove and Witts' series (I955), ascompared with a surgical mortality of io to I2per cent. in the pre-cortisone era (Brooke, 1956;Swinton, 1956), but they suggest that adrenal in-sufficiency due to previous steroid therapy mayhave played a part in the high post-operative deathrate reported by Truelove and Witts.

SummarySurgical operations are hazardous on patients

who are being treated with corticotrophin (ACTH)or steroid hormones, or who have previously re-ceived such treatment. This risk applies both topatients who are taking small replacement dosesof cortisone for the treatment of adrenal insuffi-ciency due to Addison's disease, hypopituitarismor previous adrenalectomy and to patients whohave been given pharmacological doses of corti-cotrophin or steroids for the treatment of suchcondition as rheumatoid arthritis, collagen diseasesand ulcerative colitis. In all such patients en-dogenous cortical adrenal function and the capacityto respond to surgical stress are absent or impaired.Post-operative adrenal failure is liable to occurand is often fatal, unless the doses of hormone areincreased over the period of the operation. Therisk applies to minor as well as major operationsand may still exist as long as two years aftersteroid therapy has been withdrawn.Although comparatively few cases of post-

operative adrenal failure have been reported afterpharmacological steroid therapy, there is noreliable means of predicting which patients arelikely to be susceptible. Therefore, full supportiveendocrine therapy, with additional doses of intra-muscular cortisone acetate and intravenous hydro-cortisone, should be given over the period of the

operation in all cases. In the case of electivesurgery the operation should be deferred until anexpert physician has been consulted and preli-minary stimulation with corticotrophin has beencarried out.

Patients who require operations for perforatedor bleeding peptic ulcers also need additionaldoses of steroids, even though these compli-cations may be the direct result of steroid therapy.An enquiry as to present or past steroid therapy

should become a routine item in the taking of casehistories and every patient receiving such treat-ment should carry written information as todiagnosis and steroid dosage, with instructions topresent this information immediately to thesurgeon attending him.Sodium retention and potassium depletion are

liable to occur and the usual post-operativenegative nitrogen balance is liable to beaccentuated. Avoidance of excessive sodiumchloride, an increased intake of calories, proteinand potassium and, if necessary, the administra-tion of testosterone, will obviate the semetabolicdisturbances.

Susceptibility to infections is increased and anti-biotic cover is always desirable.

Healing of wounds seldom presents any seriousproblem and there is little evidence of an increasedtendency to thrombo-embolic complications.

BIBLIOGRAPHYALLANBY, K. D. (I957), Lancet, i ,II04.BEHRMAN, H. T., and GOODMAN, J. J. (I950), J. Amer. med.

Ass., I44, 2i8.BENNETT, W. A. (I954), J. Bone Jt. Surg., 36a, 867BOLAND, E. W., and HEADLEY, N. E. (i95i), Calif. Med., 74,

4I6.BORNSTEIN, J., and TREWHILLA, P. (i950), Lancet, ii, 678.BROOKE, B. N. (I956), Ibid., ii, 532, 1175.BROWNE, J. S. C., and VENNING, E. H. (947), Trans. Ass.

Amer. Phys., 6o, I6.COPE, C. L., and HURLOCK, B. (I954), Clin. Sci., I3, 69COSGRIFF, S. W. (i9sI), J. Amer. med. Ass., 147, 924.COSGRIFF, S. W., DIEFENBACH, A. F., and VOGT, W.

(I950), Amer. J. Med., 9, 752.COSTE, F., GALURICHE, P., PIGNET, B., DELBARRA, F.,

and LAURENT, F. (195I), Rev. rhum., I8, I89.CREDITOR, M. C., BEVANS, M., MUNDY, W. L., and

RAGAN, C. (1950), Proc. Soc. exp. Biol. (N.Y.), 74, 245.EIK-NES, K., SANDBERG, A. A., MIGCON, C. J., TYLER,

F. M., and SAMUELS, L. T. (I955),Y. clin. Endocr., I5, 13.ELMAN, R., WEICHSELBAUM, T. E., MONCRIEFF, J. C.,

and MARGRAF, H. W. (I9s5), Arch. Surg. (Chicago), 7I, 697.ENGLEMAN, E. P., KRUPP, M. A., JOHNSON, H. P., WELSH,

J. E., WRENN, H. T., and KING, W. R. (I953), Arch. intern.Med., 9I, I.

FAHEY, J. L. (I95I), Proc. Soc. exp. Biol. (N.Y.), 77, 491.FENTRESS, V., FIRNSCHILD, P., and REVENO, W. S. (1956),

New Engl. .7. Med., 254, 657.FLICKINGER, F. M., and HENDERSON, I. W. D. (I956),

Surg. Gynec. Obstet., I02, 66.FORBES, J. (1952), Lancet, ii, 555.FORBES, A. P., DONALDSON, E. C., REIFENSTEIN, E. C.,

and ALBRIGHT, F. (I947),.7. clin. Endocr., 7, 264.FORSHAM, P. H., and THORN, G. W. (is55), 'Textbook of

Endocrinology,' ed. R. H. Williams, Philadelphia, W. B,Saunders Co., ch. 4.

FRANKSSON, C., and GEMZELL, C. A. (955), J. clin. Endocr.,I5, I069.

FRASER, C. G., PREUSS, F. S., and BIGFORD, W. D. (1952),Y. Amer. med. Ass., I49, 1542.

by copyright. on M

ay 11, 2022 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.33.386.632 on 1 D

ecember 1957. D

ownloaded from

December 1957 DE MOWBRAY: Care of Patients Receiving Cortisone 639

GOLDEN, A., BONDY, P. K., and SHELDON, W. H. (I950),Proc. Soc. exp. Biol. (N. Y.), 74, 455.

HAMMOND, W. G., ARONOW, L., and MOORE, F. D. (1956),Ann. Surg., 144,715.

HARDY, J. D. (1955), ' Surgical Physiology of the Adrenal Cortex,'C. C. Thomas, Springfield.

HARNAGEL, E. E., and KRAMER, W. G. (I955), _. Amer. med.Ass., z58, I5r8.

HAYES, M. A. (I956), Surgery, 90, 945.HAYES, M. A., and KUSHLAN, S. D. (1956), Gastroenterology,

30, 75-HELENRICH, M. L., JENKINS, D., and SWAN, M. (I957)

Surgery, 4I, 895.HESS, E. V., and MACPHERSON, M. E. (I957), Brit. med. J7.,

i, 27I.HUME, D. M., and MOORE, F. D. (1951), Proc. 2nd clin. ACTH

Conf., ed. J. R. Mote, London, J. & A. Churchill Ltd., 2, 289.INGRAHAM, F. D., MATSON, D. D., and McLAURIN, R. L.

(1952), NeW Engl. 37. Med., 246, 568.JACOBS, J. H., and CLIFFORD ROSE, F. (1955), Tubercle,

36, II3.JEPSON, R. P., JORDAN, A., and LEVELL, M. J. (1956), Brit.

.7. Surg., 43, 390.KERN, R. A. (1957), Amer. 3'. med. Sci., 233, 430.KINSELL, L. W., PARTRIDGE, J. W., BOLING, L., and

MARGEN, S. (1952), Ann. intern. Med., 37, 921.KIRSNER, J. B., and PALMER, W. L. (I95I), J'. Amer. med. Ass.,

147, 541-KYLEY, L. H., WALSH, W. P., and DOOLAN, P. D. (1953),

Arch. intern. Med., 91, 283.LARAGH, J. H., and ALMY, T. P. (1948), Proc. Soc. exp. Biol.

(N.Y.), 69, 499.LARKIN, M. A., and FLANNERY, M. G. (I953), Arch. Surg.

(Chicago), 66, 386.LEVIN, M. H., RIVO, J. B., SCOTT, W., FIGUEROA, W. G.,

FRED, L., and BARNETT, T. F. (I953), Amer. 3'. Med.,14, 265. 0

LEWIS, L., ROBINSON, R. F., YEO, J., HACKER, L. A.,and EISEN, G. (I953), Ann. intern. Med., 39, ii6.

McGRAW, A. B., MARGULIS, R. R., and BRUSH, B. E. (1952),Arch. Surg. (Chicago), 65, 8I.

McINTOSH, H. W., and HOLMES, C. B. (195I), Lancet, ii, Io6I.MARGULIS, R. R., and SMITH, R. W. (1950), 3. lab. clin. Med.,

36, ioo8.MONCRIEF, J. A., WEICHSELBAUM, T. E., and ELMAN, R.

(I954), Surgical Forum-Proc. 39th Clin. Congr. Amer. Coil.Surg. Chicago, October I953, W. B. Saunders Co., Phila-delphia, London, 4, 469.

MONTO, R. W., BRENNAN, M. J., MARGULIS, R. R., andSMITH, R. W. (I950), 3. lab. clin. Med., 36, ioo8.

MOORE, F. D., STEENBURG, R. W., BALL, M. R., WILSON,G. M., and MYRDEN, J. A. (I955), Ann. Surg., I4I, 145.

NABARRO, J. D. N., STEWART, J. S., and WALKER, G.(I955), Lancet, ii, 993.

NICHOLAS, J. A., BURSTEIN, C. L., UMBERGER, C. J.,and WILSON, P. D. (I955), Arch. Surg. (Chicago), 71, 737.

PAGE, J. A. (1954), Brit. med. 3'., ii, 1334.PEARSON, 0. H., ELIEL, L. P., RAWSON, R. W., DOBRINER,

K., and RHOADS, C. P. (I949), Cancer, 2, 943.PLOTZ, C. M., BLUNT, J. W., and RAGAN, C. (I950), Arch.

Derm. Syph. (Chicago), 6I, 913.RAGAN, C. (I953), Bull. N.Y. Acad. Med., 29, 353.RAGAN, C., GROKOEST, A. W., and BOOTS, R. H. (I949),

Amer. 3'. Med., 7, 741.

RAGAN, C., LATTES, R., BLUNT, J. W., DE GUISE, V.,JESSAR, R. A., and EPSTEIN, W. (i953), 'The Effect ofACTH and Cortisone upon Infection and Resistance,' Sym-posium on Sect. Microbiol., N.Y. Acad. Med. No. 6, March1952, ed. Shwartzman, G., N.Y., Columbia Univ. Press,ch. 5.

ROCHE, M., HILLS, A. G., and THORN, G. W. (I950), Proc.Ist cin. ACTH Conf., ed. J. R. Mote, p. 55.

RUSSEK, A. S. (I954), Amer. Heart J., 47, 653.SALASSA, R. M., BENNETT, W. A., KEATING, R. F., and

SPRAGUE, R. G. (I953), J. Amer. med. Ass., 152, I509.SANDBERG, A. A., EIK-NES, K., SAMUELS, L. T., and

TYLER, F. H. (I954), J. clin. Invest., 33, I509.SELYE, H. (1936), Brit. J. exp. Path., I7, 234.SELYE, H. (I946), J. clin. Endocr., 6, 117.SHAPER, A. G., DYSON, H. (I955), Lancet, i, 887.SLANEY, G., and BROOKE, B. N. (I957), Ibid., i, I I67.SLAUNWHITE, W. R., and SANDBERG, A. A. (I957), J. clin.

Endocr., 17, 395.SLOCUMB, C. H., POLLEY, H. F., WARD, L. E., and HENCH,

P. S. (I957), Ann. intern. Med., 46, 86.SMITH, R. W., MARGULIS, R. R., BRENNAN, M. J., and

MONTO, R. W. (I950), Science, 112, 295.SMYTHI, G. A. (I95I), 7. Amer. med. Ass., 145, 474.SPENCE, A. W., JENKINS, J. S., and ROBINSON, A. M. (I956),

Brit. med. 7., ii, 1348.SPRAGUE, R. G. (I95i), Amer. Y. Med., 10, 567.SPRAGUE, R. G., POWER, M. H., and MASON, H. L. (I950),

.7. Amer. med. Ass., 144, I341.STEENBURG, R. W., LENNIHAN, R., and MOORE, F. D.

(1956), Ann. Surg., i43, i8o.SWINTON, N. W. (1956), Proc. roy. Soc. Med., 49, 945.TAYLOR, A. B., ALBERT, A., and SPRAGUE, R. G. (I949),

Endocrinology, 45, 335.TEXTER, E. C., LEGERTON, C. W., RUFFIN, J. M., BOYD,

J. A., and SMITH, A. G. (I953), Arch. intern. Med., 9I, 744.THORN, G. W. (I95I), 'The Diagnosis and Treatment of Adrenal

Insufficiency,' 2nd edition, Charles C. Thomas, Springfield,Illinois.

THORN, G. W., JENKINS, D., LAIDLAW, J. C., GOETZ,F. C., DINGMAN, J. F., ARONS, W. L., STREETEN,D. M. P., and McCRACKEN, B. (1953), New Engl. 7. Med.,248, 284, 323, 369, 588, 632.

TOMPSETT, S. L., and SMITH, D. C. (X954),.7. clin. Endocr.,14, 922.

TROEN, P., and RYNEARSON, E. H. (i955), Ibid., I5, 857.TRUELOVE, S. C., and WITTS, L. J. (955), Brit. med. .7.,

is, 1041.TULIN, M., and ALMY, T. P. (1952),j. Amer. med. Ass., 150, 559.TYLER, F. H., SANDBERG, A. A., and EIK-NES, K. (1953),

Y. clin. Invest., 32, 6o8.TYLER, F. H., SCHMIDT, C. D., EIK-NES, K., BROWN, H.,

and SAMUELS, L. T. (I945), Ibid., 33, 1517.VENNING, E. H., HOFFMAN, M. M., and BROWNE, J. S. L.

(1944), Endocrinology, 35, 49.VIDEBACK, A. A., ASBOC-MAUSEN, G., ASTRUP, P.,

FABER, V., HAMBURGER, C., SCHMITH, K., SPRECH-LER, M., and BROCHNER-MORTENSEN, K. (1950), Actaendocr. (Kbh.), 4, 245.

VIRTUE, R. W., HELMRICH, M. L., and GAINZA, E. (1957),Surgery, 41, 549.

WARD, L. E., POLLEY, H. F., SLOCUMB, C. H., and HENCH,P. S. (1953),. Amer. med. Ass., 152, Ix9.

WHITNEY, J. E., and BENNETT, L. L. (1952), Endocrinology,50, 567.

NOTICE OF SPECIAL INTEREST TO SUBSCRIBERS: VNOT'WHY NOT HAVE YOUR COPIES OF THIS

JOURNAL BOUND INTO YEARLY VOLUMES?' HA Y U

You can have your twelve monthly issues fully bound in dark green pin headcloth. lettered in gilt on spine with name of Journal, Volume Number and year,complete with index at front, for 21s. Od. post free.. A limited number of out ofprint lournals are available to bind into volumes and tnake your library complete.Price on application giving details of issues required to complete back volumes.

TH4E FELLOWSHIP OF POSTGRADUATE MEDICINE BOU60 PORTLAND PLACE. LONDON. W.l

by copyright. on M

ay 11, 2022 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.33.386.632 on 1 D

ecember 1957. D

ownloaded from