Annual of 1972 - BMJ

Postgraduate Medical Journal (September 1973) 49, 644-655.

Annual review of surgery 1972

HAROLD ELLISD.M., M.Ch., F.R.C.S.

Westminster Hospital, London, S. W. 1

hitroductionIn a short review such as this, only a few of many

fascinating points of progress in surgery can betouched upon, and it is the author's prerogative tochoose those which have particularly interested himduring 1972. This survey has been prepared with amind to those practioners who do not normallyclosely peruse the surgical journals, and for thisreason broad aspects ofadvance have been abstractedrather than minutiae of operative technique. A fewkey references are given to earlier publications, butin the main we quote papers which have appearedduring the past year.

Carcinoma of the BreastBreast cancer remains the commonest killing

tumour of women in the Western World; of 1000females, fifty will develop breast cancer and thirty-five will die of this disease. The mortality has re-mained almost unchanged for the last 30-40 years.The incidence is higher in the upper social classes,it is commoner in single women and its incidence isless as parity increases; however, this is related to theage of the woman at the birth of her first child, sothat if the first baby is born before the age of 18, thewoman has one-third less chance of developing cancerthan if the first baby is born at 35. Lactation doesnot seem to be a factor. Earlymenopause decreases therisk of cancer, and thus hysterectomy and bilateralsalpingo-oophorectomy before the age of 40 reducesthe woman's chance of developing cancer of thebreast to one-quarter of that of the average femalepopulation. The risk is increased in women with apositive family history (Campbell, 1972).

It might be imagined that by now clinicians wouldhave determined the best way to manage a patientwho presents with an early carcinoma of the breastone that is not yet clinically disseminated beyond theconfines of the breast tissue and adjacent lymphnodes. Surprisingly enough there is more controversyabout this particular subject than almost any othertopic in tumour therapy, and this is more so todaythan ever before. All grades of operation are beingperformed in a spectrum which ranges from simpleremoval of the lump to super-radical mastectomy;

these techniques might or might not be combinedwith radiotherapy or with ovarian ablation. Eachtreatment has its enthusiastic group of advocatesand yet, despite hundreds of reports, there is little totell us which is the 'best buy' for the patient (Leadingarticle, 1972).One fallacy is the clinical staging of the disease on

which many protocols of treatment are based and onwhich some trials have depended. Mere palpation ofthe axillary lymph nodes is grossly inaccurate;Wallace & Champion (1972), for example, correlatedhistological findings with clinical staging and foundthat 25°/ of patients with impalpable axillary nodeshad tumour present on microscopic study, whereas55'/O of patients with clinically palpable nodes werefree of histological evidence of node metastases. Thesame authors confirmed that it was the histologicalgrading of the lymph nodes and not their palpabilitythat determined the 5-year survival of the patient.The internal mammary chain of lymph nodes is, ofcourse, completely impalpable and therefore un-available for clinical staging of breast cancer.Handley (1972) has carried out biopsy of this chainin 900 patients undergoing mastectomy. Biopsy waspositive in 95/s of cases in which the axillary nodeswere clear and 35°o of patients in which the axillarynodes were histologically invaded. With increase insize of the primary tumour there was an increasingincidence of positive internal mammary chainbiopsies. This study provided some importantevidence in prognosis; of 400 patients followed upfor 10 years or more there was a 71°/ survival inthose in whom neither-the axillary nor mediastinalnodes were involved. This fell to only 11% whereboth groups of nodes were positive. In patients whohad either axillary or mediastinal nodes aloneinvolved the survival rate was 40%°. Unfortunately,only histological examination can, as yet, give usaccurate information about the lymph node status;the internal mammary chain cannot be demon-strated by lymphography, nor is this techniqueaccurate in the early detection of axillary nodemetastases (Kett, Lukacs & Varga, 1972). Even moredisconcerting than the inaccuracy of lymph nodeassessment is the finding by Galasko (1972) that no

Protected by copyright.

on January 13, 2022 by guest.http://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.49.575.644 on 1 Septem

ber 1973. Dow

nloaded from

http://pmj.bmj.com/


less than twelve of fifty women (24%) who hadapparently early mammary cancer with normal radio-graphic skeletal survey had evidence of bony depositson gamma scanning after the administration ofradio-active "8fluorine. Undoubtedly, many patients whopresent to us with apparently early and 'curable'breast tumours already have wide dissemination ofdisease detectable by these more sophisticatedtechniques.

Controlled trialsThe most important problem to date has been the

rarity of carefully controlled trials of the manydifferent methods of treatment. In the few instanceswhere these have been carried out it has been apparentthat the choice of treatment has little effect onsurvival; thus, expectation of life following simplemastectomy combined with radiotherapy is littledifferent from that resulting from extended radicalmastectomy, and survival rates following radicalmastectomy have not been affected by post-operativeradiotherapy. There is urgent need for further pros-spective controlled trials to determine, first, whetherany form of local therapy gives a survival advantageover others. If all suggested techniques in fact givethe same expectation of life, it is then necessary todetermine which of them gives the best chance oflocal control of the disease, so that even if the patientmust eventually die of disseminated cancer, she mayat least be saved the pain and suffering of localulceration. Finally, it is necessary to determinewhich method of treatment, given comparablesurvival rate and local control, is the kindest to thepatient, with the least morbidity as regards armoedema, shoulder stiffness and minimal disfigure-ment.No-one should under-estimate the difficulty of

such trials. Atkins et al. (1972) reported their trialcarried out over the past 10 years to compare widelocal excision of the tumour combined with radio-therapy and radical mastectomy combined withradiotherapy. The trial was confined to patients overthe age of 50 with clinical stage I and stage IItumours. No significant difference was foundbetween survival in the two groups in clinical stage Icases, but those with stage II tumours showed astatistically better survival in the radical mastectomygroup at 10 years. A significantly greater rate of localrecurrence was observed in the local excision group,but there was a higher incidence of lymphoedema ofthe arm in the radical mastectomy patients. Theproblems of clinical trials in the treatment of breastcancer are well illustrated by the subsequent exten-sive correspondence which followed the publicationof this paper. Thus, initially thiotepa was adminis-tered in addition to radiotherapy, but this adjunctivechemotherapy was not continued after 1968; the

regimen of radiotherapy was different in the groupundergoing the wide local excision and in the groupsubjected to radical mastectomy; moreover, the doseemployed was criticized as being less than optimalfor the destruction of residual tumour after surgery(Lee et al., 1972).Very careful prospective multicentre trials are now

in progress, and undoubtedly the next few years willsee the resolution of many of our present problems.One national trial in the United Kingdom has beenin progress for 2 years; it is designed to studywomenunder the age of 70 with clinical stage I or stage IIcarcinoma of the breast. A simple mastectomy isperformed and in the post-operative period thepatients are allocated to either a control group or agroup receiving a standardized course of radio-therapy. Documentation and pathological reportingis standardized and the information is stored on acentral computer (Baum, Edwards & Magarey,1972). Already 1050 patients are included in thestudy. It is too early for any preliminary results withregard to differences between thetwoforms oftherapyto be evident, but already some interesting observa-tions have been made; of the first 161 patients in the'watch-policy' group, forty were judged to be instage II, and here no less than 75°. of the axillarylymph nodes have regressed. The regression wasapparent at the first follow-up examination 3 monthsafter mastectomy and has persisted. It may be that aproportion of the nodes which have regressed did sobecause of the removal of the tumour and, with it,the stimulus to sinus histiocytosis, but it remains tobe seen whether in other cases lymph nodes contain-ing malignant cells are still capable of exerting someprotective process against further growth (Edwards,Baum & Magarey, 1972).

Locally advanced breast cancerOccasionally patients present with locally ad-

vanced carcinoma of breast but without clinical orradiological evidence of disseminated disease. Thesewomen form a rather special group as regards treat-ment, since it is important, if possible, to obtaincontrol of the local lesion and to prevent the miseryof ulceration, discharge, odour and bleeding fromthe fungation of the primary tumour. The conven-tional treatment for advanced local breast cancer isundoubtedly radiotherapy, but although remarkableshrinkage of the tumour mass and healing of ulcera-tion may occur, local recurrence is far from un-common (Leading article, 1973). Sonneland (1972)treated an elderly lady with a fungating carcinomaof the breast by means of zinc chloride paste followedby skin graft to the raw area, after biopsies had shownthat the tumour had been completely cleared. Theadvantages of chemosurgery, coagulation or cryo-surgery in these advanced lesions still remain to be

645P

rotected by copyright. on January 13, 2022 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.49.575.644 on 1 S

eptember 1973. D

ownloaded from

http://pmj.bmj.com/

Harold Ellis

assessed. We have reported twenty-four patientswith locally advanced breast cancer treated by super-voltage radiotherapy to a maximum tumour dose ofapproximately 6000 rads, followed 4-16 weeks laterby simple, extended or radical mastectomy (Stoker &Ellis, 1972). Primary wound healing was achieved ineach case. In a follow-up of from 2 to 10 years, threepatients have to date developed local recurrence.Seven patients are dead and in six there was no localtumour present at the time of death. Twelve patientshave survived 3 years or more following surgeryand three of these have no signs of disseminateddisease. This study suggests that toilet mastectomymay be a reasonable procedure in selected casesfollowing radiotherapy for advanced local breastcancer, but obviously a multicentre controlled trialis required to establish whether or not mastectomyimproves the quality of survival in these patientscompared with those treated by supervoltage radio-therapy alone.

Changing fashions in peptic ulcer surgeryVagotomy has become increasingly more popular

in the surgical treatment of duodenal ulcer over thepast quarter of a century, but the procedure hasundergone repeated modification. The originaloperation of total or truncal vagotomy entailedcomplete division of the vagi at the oesophagealhiatus. This denervates not only the stomach butalso the gut as far as the mid-colon and the gallbladder, pancreas and liver. Gastric stasis, whichaccompanies this operation, requires the addition ofa gastric drainage procedure-either gastroenteros-tomy, pyloroplasty or antrectomy. Selective vago-tomy was then introduced, which aimed at denerva-tion ofthe stomach but not the other intra-abdominalviscera. Now, highly selective vagotomy (otherwisecalled selective proximal vagotomy or parietal cellvagotomy) has recently been introduced. This aimsat denervation of the acid-secreting cells of thestomach but preservation of the nerve supply of theantrum, so that it is claimed that the addition of adrainage procedure is not necessary. It remains tobe seen, however, whether this procedure will standthe test of time. The results of numerous clinicaltrials have been reported during the year.Humphrey et al. (1972) compared the incidence of

dumping after either truncal or selective vagotomywith pyloroplasty, and highly selective vagotomywithout a drainage procedure. Dumping was foundto be significantly less frequent in patients who hadundergone the highly selective procedure withoutdrainage. Johnston et al. (1972b) studied the inci-dence of diarrhoea after these three operations;diarrhoea was recorded in 24%/ of patients aftertruncal vagotomy and pyloroplasty but in only 2%,

of patients after highly selective vagotomy withoutdrainage procedure. These authors suggest that post-vagotomy diarrhoea can be attributed both to un-regulated gastric emptying due to the drainageoperation and to the extra-gastric denervation pro-duced by truncal vagotomy.

Claims that selective vagotomy can be carried outwithout a drainage procedure have been made in anumber ofpublications,but Clark et al. (1972) reporttwo independent trials carried out in Birminghamand Liverpool; ten of the fifteen patients studiedhad their gastric emptying times measured pre- andpost-operatively, and all showed significant delayafter selective vagotomy without drainage. Threepatients in this study have developed a gastric ulcerand one of these, in addition, had recurrence of theduodenal ulcer. The operation of selective proximalvagotomy (highly selective vagotomy)without gastricdrainage, although associated with early encouragingresults (Hedenstedt et al., 1972; Johnston et al.,1972a), has had even less time for long-term assess-ment. Here a note of caution has been introducedby Wastell et al. (1972); in sixteen patients treatedby selective proximal vagotomy without gastricdrainage two definite recurrent ulcers had occurredwithin a year of surgery and both required re-operation; a further patient had symptoms and abarium meal appearance strongly suggestive ofrecurrent ulceration. These authors advise surgeonsto await the results of further trials before carryingout many of these operations without gastric drain-age. A report of a case of gastric ulceration afterhighly selective vagotomy without drainage (Hall,1972) adds further weight to this advice.The reason for the almost universal current

popularity of vagotomy and pyloroplasty is un-doubtedly its simplicity compared with the moreradical procedure of partial gastrectomy, and alsoits appeal as being a more 'physiological' surgicalapproach to the problem of duodenal ulceration.A most surprising and important report fromGoligher et al. (1972) has compared total vagotomyand pyloroplasty with vagotomy and gastro-enterostomy, vagotomy and antrectomy and sub-total gastrectomy. Recurrent ulceration was com-moner after vagotomy and pyloroplasty than afterall the other operations. Moreover, the overallassessment by the Visick functional grading gavepoorer results after vagotomy and pyloroplasty thanafter any of the other operations. These authorsconclude that truncal vagotomy and pyloroplastyhas not lived up to expectations and its place as thecurrently most popular procedure in the electivesurgical treatment of duodenal ulcer should bereconsidered. Under both headings of recurrence andfunctional grading antrectomy and subtotal

6466P


http://pmj.bm

j.com/

Postgrad M


eptember 1973. D

ownloaded from

http://pmj.bmj.com/


gastrectomy gave the best results in this particularstudy. Undoubtedly, in expert hands the immediateand long term results of partial gastrectomy forduodenal ulcer can be extremely good; McKeown(1972) has reported a personal series of 800 casesfollowed up for a 10-20 year period. The hospitalmortality in 778 elective procedures was 07°/, and intwenty-two emergency gastrectomies forhaemorrhagewas 22 7%4; 79V/ of cases were graded as Visick I and15%. as Visick II. Stomal ulceration occurred insixteen cases (2%); all were re-operated on andthirteen were converted to a satisfactory status. Noestablished case of post-gastrectomy bone diseaseoccurred in the whole series. Such magnificent resultscannot be expected to be achieved in average hands.Postlethwait & Johnson (1972) reviewed 2819patients undergoing elective surgery of all types forduodenal ulcer in American Veterans Administra-tion Hospitals with a 2.8%/ mortality. Interestinglyenough the mortality was two and a half timeshigher in obese subjects than those of normal build.

Although in average hands vagotomy is a simplerand safer procedure than a partial gastrectomy, itnevertheless has some complications peculiar toitself. Price, Powis & Morrissey (1972), for example,report oesophageal rupture as a complication oftruncal vagotomy in 0 8% of a collected series of4026 patients in the Midland region. Of these forty-one cases, there were four deaths; death is especiallylikely to occur, not unnaturally, when the perforationis overlooked at the time of surgery and onlydiagnosed because ofsubsequent pneumoperitoneum,empyema, or pneumothorax. Musgrove (1972) hasreported an example of the unusual complication ofa chylous fistula during an abdominal truncalvagotomy, probably by division of an aberrantlacteal. Early exploration and ligation of the offend-ing lymphatic is recommended in such a case.

Any student of the history of the surgery of duo-denal ulceration is impressed by the initial enthusiasmwith which each operation is introduced, with itsdisadvantages only becoming apparent as patientsare followed-up year after year. Factors such as thesurgeon's skill and experience and the selection ofpatients undoubtedly play an important role,although one that is difficult to measure. Mostsurgeons find that their best results follow surgeryon patients with severe local disease and a lesshappy outcome when there is 'a small ulcer in theduodenum and a large ulcer in the mind'. We havefound, for example, that uniformly excellent resultsfollow vagotomy and drainage for those patientswhose duodenal ulcer has reached the stage of grossorganic pyloric stenosis (Ellis, 1972). McColl et al.(1971) have shown that psychological testing pre-operatively can give a fairly accurate prediction of

those patients who are likely to achieve a poor resultfrom their gastric surgery.

Vagotomy for gastric ulcerA small proportion of gastric ulcers are due to

gastric retention, either secondary to pyloric stenosisor to atony of the stomach following vagotomywithout drainage. The aetiology of gastric ulcerationin the great majority of cases is quite unknown, butthis has not deterred surgeons from carrying out thevarious types of vagotomy, with or without drainage,in the treatment of this condition; after all, did notJohn Hunter advise Edward Jenner, 'Do not think,try the experiment'? Clarke, Lewis & Williams(1972) report a series of sixty-eight patients withgastric ulcer who were treated by vagotomy andpyloroplasty. In a follow-up period averaging 3 years,four patients have developed recurrent pepticulceration requiring further surgery. Of fourteenpatients who presented with acute haemorrhage, onesubsequently required gastrectomy for continuedbleeding, but the others have obtained satisfactoryresults to date. Of twenty-one patients with a highgastric ulcer, one has suffered recurrence. Theauthors conclude that vagotomy and pyloroplasty isa satisfactory form of treatment for a high or bleed-ing gastric ulcer, but that for all other gastric ulcerssome form of gastric resection is preferable. Kennedy,Kelly & George (1972), in a 3 year average follow-upperiod, report four recurrent gastric ulcers occurringamong thirty-three patients with an ulcer situated onthe lesser curve or body of the stomach. However,among forty-two patients with an ulcer situated inthe antrum or with an associated duodenal lesionthere was only one recurrence. These authors alsoconsider that gastrectomy is to be preferred in thestraightforward case of gastric ulceration. Miguel(1972) has had four recurrent ulcers in a series offorty-two patients undergoing vagotomy and drain-age for gastric ulceration. A review of recent pub-lished figures by Bynhum, Hartsuck & Jacobson(1972) finds an average recurrence rate of 6%4 follow-ing this procedure-clearly inferior to the results ofpartial gastrectomy. Johnston et al. (1972c) havetreated fourteen patients with gastric ulcer and fivewith combined gastric and duodenal ulcers by meansof highly selective vagotomy without drainage in thepast 3 years. It is interesting that spontaneous acidoutput and acid response to meat extract was low inall cases and none had evidence of stenosis. Eachpatient had the ulcer excised entirely at operationin order to confirm its benign nature. Post-operativebarium meal examination in eleven patients followedup for 9 months or more demonstrated satisfactoryhealing and effective gastric emptying and, to date,all the patients are symptomless. Although this studyis encouraging it must be remembered that the

647P


http://pmj.bm

j.com/

Postgrad M


eptember 1973. D

ownloaded from

http://pmj.bmj.com/

Harold Ellis

follow-up period is short and all experienced surgeonsare well aware of the dangers of early good resultsin gastric surgery. In particular, one must rememberthat an old operation for gastric ulcer was merely itslocal excision, a procedure which was carried out inall Johnston's cases in addition to the parietal cellvagotomy!

Post-operative gastric suctionBecause of the fear that the denervated stomach

would fail to empty after vagotomy, nearly allsurgeons in the early days of this procedure em-ployed either post-operative nasogastric suction orelse gastrostomy drainage of the stomach for a

period of up to 10 days following surgery. Withincreasing experience, more and more surgeons inthe United Kingdom have given up this practice,adopting what Hendry (1962) called 'tubeless gastricsurgery'. We found, for example, that even in thepresence of gross pyloric stenosis post-operativenasogastric suction was not required followingvagotomy and either pyloroplasty or gastrojejunos-tomy (Ellis, 1972). Miller et al. (1972) have carriedout a randomized trial comparing no tube, naso-gastric aspiration for 48 hr, and gastrostomy drainagefor 8 days in patients undergoing vagotomy andeither pyloroplasty or gastroenterostomy. Of forty-three patients treated without gastric decompression,only one required the passage of a nasogastric tubepost-operatively. In addition to the fact that thepatients in the 'tubeless' group were much morecomfortable, there was a significant decrease in post-operative chest complications, wound infection anddysphagia in this group. This agrees well with a

previous study carried out by Barnes & Williams(1967) on similar groups of patients who were not,however, subjected to a randomized controlled trial.American surgeons have been more conservative

in their approach to this problem, but now more andmore are following the British practice. Kerry,Hoshal & Ruiz (1971) report on 200 patients dividedinto two groups of 100 treated with and withoutnasogastric suction following vagotomy. Of the 100treated without a tube seven were intubated in thepost-operative period, four probably unnecessarily!There was a 16°/ post-operative pulmonary com-plication rate in the tubeless group compared with23%Y in the intubated patients. Herrington (1972)reports his cautious trial of tubeless gastric surgery.

The nasogastric tube was employed, but removed8-12 hr post-operatively; however, no fluids at allwere given by mouth for 72 hr following surgery. In815 elective patients following vagotomy sixty-tworequired post-operative suction (7-6y.), the indica-tions being a difficult technical operation, post-operative distension, nausea or vomiting. Nachlaset al. (1972), as a result of post-operative barium

studies, conclude that nasogastric suction is onlyneeded in patients with obstruction, gastric dilatationor where it is necessary to observe for post-operativehaemorrhage. In other cases its use can be avoided,to the great relief of the patient.There is little doubt that in the early stages of

evolution of major gastric surgery a nasogastric tubemight well have been useful or even lifesaving. Undermodern conditions, where relatively non-toxicanaesthetics are used, where fluid replacement is wellunderstood, where the patient is electrolyticallyrestored pre-operatively, where relatively atraumatictechniques of anastomosis have been developed, andwhere drugs like chlorpromazine are available toprevent nausea in the post-operative period, theincidence of post-operative gastric dilation or emesisis so low that routine gastric suction, with all itsdiscomfort to the patient and embarrassment to hisbreathing, can now be eliminated.

Liver and biliary systemLiver traumaRoad accidents are responsible for most cases of

abdominal injury in this country, and the incidenceof damage to the liver has risen with the increase ofhigh-speed traffic. Better techniques of resuscitationand surgery have undoubtedly improved the oncevery serious prognosis in this situation. Blumgart &Vajrabukka (1972) reviewed forty-four cases ofclosed injury of the abdomen requiring laparotomy.Thirty-five of these were due to motor vehicleaccidents; twenty of the forty-four cases had liverinjuries, and in eighteen of these there were associ-ated injuries, particularly to the head chest andskeletal system. The authors classified their casesinto minor and major. The minor cases were thosein which the liver laceration could be closed bysuture with complete control of haemorrhage. Thisaccounted for fourteen of their patients, only one ofwhom died, and that from severe associated injuries.Major liver injuries were defined as ones with alaceration so extensive that the hepatic wound couldnot be closed by simple suture, and was oftenassociated with internal disruption of liver tissue. Ofthe six cases of major injury in this series, the firstdied of haemorrhage after packing the wound. Thesecond was treated by simple suture, bleeding con-tinued and a right hemi-hepatectomy had to be per-formed, with subsequent death. Following this, theauthors' policy changed to partial hepatectomy, andin the subsequent four cases there were threesurvivals, the death being associated with othermultiple injuries.The post-operative complications in these seriously

injured patients include jaundice, for which no com-plete explanation is available, and a haemorrhagic

648



j.bmj.com

/P


j.49.575.644 on 1 Septem

ber 1973. Dow

nloaded from

http://pmj.bmj.com/


diathesis which results from disseminated intra-vascular clotting which produces an acute defibrina-tion syndrome. Intravascular clotting is presumedto be activated by the release of tissue thrombo-plastin as the result of massive liver injury.

Mercadier, Clot & Cady (1972), from Paris, reporttheir experience of ten patients who underwent righthepatectomy in the treatment of liver trauma. Therewere six survivors. These authors confirm that hemi-hepatectomy for trauma to the right lobe of the liveris justified in severe bursting injuries, uncontrollablehaemorrhage resulting from a tear to the right hepaticvein or an injury to the inferior vena cava, or incases where a hepatic laceration has been suturedwithout realizing the extent of the injury, and wheresubsequent haemorrhage, necrosis, secondary infec-tion or haemobilia make further surgery obligatory.These authors favour the simple technique of clamp-ing the portal pedicle and then fracturing throughthe liver substance with the fingers, tying off bloodvessels and bile ducts as encountered, rather than theso-called anatomical hepatectomy with preliminarydissection of the hepatic vessels and bile ducts in theporta hepatis, since the latter is a time-consumingprocedure.The experience of European surgeons fades into

insignificance compared with that of their Americancolleagues in the field of abdominal trauma. Lim,Knudson & Steele (1972) report 285 cases of liverinjury, of which 182 were open (ninety-nine gunshotwounds and eighty-three stabbings) and 103 were dueto blunt trauma. Major resections were required in14°/, the indications being identical to those alreadydescribed. White & Cleveland (1972) give an accountof 126 patients treated for liver trauma over a 3-yearperiod at their hospital in Los Angeles. Seventy-sevenhad penetrating liver wounds. In spite of the factthat fourteen patients required emergency hepaticresection, the mortality was only eight in the wholeseries (6-3%). The presence of associated injuries wasthe greatest factor adversely affecting survival, sinceseven of the eight deaths were victims of multipletrauma.Not all surgeons consider that partial hepatectomy

is necessarily the treatment for severe liver trauma.May (1972) advocates ligation of the hepatic arterysupplying the traumatized lobe as a means of con-trolling haemorrhage. After 7-10 days the lobe orsegment is revascularized by collaterals from theuninjured side. Morton, Roys & Bricker (1972)review an enormous experience from Houston,Texas, of 1068 liver injuries treated between 1939and 1970. The mortality rate has declined from 64%initially to a present rate of 11 70/. Hepatic lobec-tomy was used rarely. Hepatic venous bleeding isnow controlled by a new technique in which aninternal shunt is introduced into the inferior vena

cava through the right atrium; this was utilizedsuccessfully in five patients.

It is well known that the liverhas remarkable powersof regeneration. Blumgart, Leach & Karran (1971)have made some interesting observations on theirfive patients after right hepatic lobectomy, whicheffects an approximate 50%. resection of liver sub-stance. One patient who came to necropsy 10 daysafter operation was found to have a liver which hadquite remarkably already achieved its estimated pre-operative weight. Serial scans of the liver in threesurviving patients demonstrated an early increase inthe size of the liver during the first 10 post-operativedays, with little or any further increase in size in theensuing 9 months. This work suggests a much morerapid restoration of liver mass occurs in man thanprevious estimates, which computed that completerestoration of the liver might take 4-6 months toachieve.

Malignant obstructive jaundiceMalignant obstruction of the bile ducts by either

an intrinsic cancer or from carcinoma of the head ofthe pancreas is occasionally amenable to radicalcurative surgery. Unfortunately, the most we canachieve in the majority of cases is temporary pallia-tion. Elmslie & Slavotinek (1972), from Adelaide,discuss some of the problems presented by patientswith inoperable cancer of the head of the pancreas.It is first necessary to try and make an accuratedifferential diagnosis from non-malignant conditions;the operative finding of a grossly dilated thin walledcommon bile duct is typical of obstruction by acancer and in marked contrast to the white, opaqueand thick walled duct which is so often seen inobstruction due to chronic pancreatic or biliaryinflammatory disease. Cholangiography is also ofvalue, but the surgeon may have to resort to trans-duodenal biopsy. Biliary decompression by means ofa cholycyst-enterostomy should be combined with agastro-enterostomy if the operative findings suggestthat duodenal obstruction from the invading canceris likely to occur. If the upper level of the growth isencroaching on the site of entry of the cystic ductinto the common duct, then the common duct itselfrather than the gall bladder should be used for de-compression in order to prevent recurrence ofjaundice as tumour extension takes place. Pain reliefmay require coeliac ganglion alcohol block by thedorsal route.

Terblanche, Saunders & Louw (1972), from CapeTown, review their experience with twenty-one casesof carcinoma situated at the junction of the commonhepatic with the main hepatic ducts. Although radicalsurgical therapy is not possible, worthwhile pallia-tion may be achieved by opening the common bileduct below the obstruction, dilating the malignant

649P


http://pmj.bm

j.com/

Postgrad M


eptember 1973. D

ownloaded from

http://pmj.bmj.com/

0Harold El/is

stricture and keeping it open by means of an indwell-ing tube. Unfortunately, these tubes tend to becomeoccluded, but the authors describe an ingenuoustechnique of threading a U-tube from the abdominalwall which passes through the common bile duct,traverses the growth, and emerges through the sub-stance of the liver onto the abdominal wall. Ifoccluded or dislodged it can be replaced by simplerailroading without the need for further operativeprocedure. Seven of their patients survived forlonger than a year, and three are still alive, 2-3 yearsafter the first operative intervention. Niloff (1972)has described a flanged vitallium prosthesis which hehas found useful in the palliative intubation ofobstructing cholangiocarcinomas.

If the tumour of the bile duct is more distallyplaced, resection may be possible. Warren, Moun-tain & Lloyd-Jones (1972), from the Lahey Clinic,have reviewed seventy-seven cases of malignanttumour of the bile ducts and found that radicalsurgery was possible in twenty-four of these, inwhich seventeen survived for more than a year, andfive are still alive over 3 years later. Surgery com-prised either pancreaticoduodenectomy, or resectionwith -end-to-end anastomosis. Since 1965 hepaticartery infusion chemotherapy has been used at theLahey Clinic for non-resectable lesions of the bileducts, and this was employed in fifteen patients.Nine were still alive after 1 year and one patient isalive more than 3 years later. Chemotherapy is com-bined in these cases with palliative dilatation of themalignant stricture and insertion of a rubber stentin order to overcome the obstructive jaundice.Of course, the ultimate treatment of intrahepatic

primary cancer is total hepatectomy and liver trans-plantation, but current results are far from encourag-ing. Williams (1972) has reviewed his experience ofnineteen cases from King's College Hospital, ofwhich five were hepatic duct carcinomas and fiveprimary hepatomas. Eleven died without leavinghospital and two are recent cases. Six patients lefthospital, but four of these died up to 11 monthspost-operatively and only two are alive in theirseventh and seventeenth month.

Gall stones and their dissolutionRecent exciting studies have suggested that

cholesterol stones may be associated with the pro-duction by the liver of bile which is supersaturatedwith cholesterol with respect to phospholipid andbile salts, and with a diminished bile salt pool(Admirand & Small, 1968). Mackay et al. (1972)report a study of the bile of seventeen patients withgall stones compared with twenty-one patientssuffering from duodenal ulcer, the bile in the latterbeing obtained by duodenal aspiration. The gallbladder bile from the gall stone patients contained

significantly more cholesterol than the bile obtainedfrom the ulcer patients, and the hepatic bile in thegall stone group contained significantly more choles-terol than the gall bladder bile. Not all the patientsin this study, however, had supersaturated cholesterolin their bile; of the seventeen stone patients, five hadhepatic bile whose relative composition lay outsidethe limits of cholesterol solubility, three were at thelimits, and the remaining nine patients had bile whichwas undersaturated with cholesterol. It may be, ofcourse, that the hepatic bile is not always super-saturated with cholesterol (as has been demon-strated by Smallwood, Jublonski & Watts, 1972),and may vary, for example, with diet, or that super-saturation had been present some time in the pastand had initiated gall stone formation. It may alsobe that supersaturation of bile with cholesterol isonly one factor in stone formation and that othercircumstances, such as the presence of mucus, infec-tion, protein or stasis in bile, may render mixedmicelles unstable, thus resulting in precipitation ofcholesterol and the initiation of stone production.

This chemical basis of the production of at leastsome gall stones leads to the possibility of theirchemical dissolution. Bell et al. (1972) have shown byin vitro experiments that rapid dissolution of humangall stones will take place in monkey's bile if associ-ated with undersaturation of the bile, a high choles-terol content in the stone, and a high surface area ofstone compared with its weight. The presence of aradio-opaque outer rim of calcium on the stonedelays its solution. Danziger et al. (1972), from theMayo Clinic, have shown that chenodeoxycholicacid induces the formation of unsaturated bile inpatients with cholesterol stone and increases the bileacid pool in these subjects. Six women with choles-terol stones were treated with oral chenodeoxycholicacid with disappearance of the stones in one patientand progressive diminution in size in three othersafter 6-18 months. Bell, Whitney & Dowling (1972),at the Postgraduate Medical School, London, nowreport the treatment of seventeen patients with gallstones contained within a functioning gall bladder.After 6 months treatment with chenodeoxycholicacid, repeat radiography showed disappearance ofthe stones in four cases, reduction in size in a furtherfour and no change in the remaining nine patients.Three further patients with non-functioning gallbladders on cholecystography showed no return infunction.These preliminary reports are naturally received

with considerable interest because of the definitepossibility of removal of cholesterol stones inpatients whose symptoms are mild enough to enablethe clinician the long period of time required forstone dissolution to take place. Where symptoms arepressing, of course, surgery still remains obligatory.

650



j.bmj.com

/P


j.49.575.644 on 1 Septem

ber 1973. Dow

nloaded from

http://pmj.bmj.com/

Annual review ofsurgery 1972

Congenital anomalies of the testisMany of the misconceptions which surrounded the

subject of maldescent of the testis have yielded inthe face of careful scientific studies (Scorer &Farrington, 1971). It is now agreed that a testiswhich has failed to enter the scrotum within a fewmonths of birth will not do so spontaneously; thosecases in which this was said to occur we nowrecognize as examples ofretractile testis misdiagnosedas cases of maldescent. It is easy enough by patientexamination to be able to coax such a testis into thescrotum and to reassure the family that the child isperfectly normal. It is now also agreed that theoperation in cases of maldescent should be carriedout at about the age of 5 years, since histologicaldevelopment of the testis is only slightly delayed atthis age, whereas the tissues are by now sufficientlydeveloped to render surgical handling relatively easy,and there is, moreover, minimal interference with thechild's schooling. There is certainly no point indelaying the operation in the misguided hope thatdescent might take place at puberty. In cases withan associated troublesome inguinal hernia, surgerymay indeed be indicated at an earlier stage than5 years of age. All sorts of complicated techniqueshave been described for maintaining the testis in thescrotum at surgery, but many now favour the verysimple method in which the testis is held in thescrotum simply by placing it in a superficial pouchfashioned between the skin of the scrotum and thedartos muscle. This has been well described byPryn (1972), who reports satisfactory results intwenty-five patients.

It is well recognized that the undescended testishas a much greater chance of undergoing malignantchange than a normal organ (and this applies alsoto the undescended testis which has been broughtdown into the scrotum). Riegler (1972) reports anexample of torsion of an intra-abdominal testis inwhich a seminoma had developed. He was able tocollect thirty-three other examples of this situation,in twenty-two of which malignant change hadoccurred. He noted that diagnosis of intra-abdominaltorsion was rarely made pre-operatively, in spite ofthe fact that one or both testis were missing fromthe scrotum. It is a wise surgeon who makes a testis-count on all his male patients!The subject of torsion of the testis has been of

considerable recent interest, not that there is any-thing fresh to say on the treatment of the subject,but there is at least an increasing awareness of thelarge number of young people in whom the diag-nosis is made too late, with resultant hopeless in-farction of the organ. As we have mentioned, torsionmay occur with an intra-abdominal testis, but thevast majority take place when there is a congenitalabnormality known as the 'bell-clapper' deformity,

in which the testis and epididymis hang freely insidethe tunica vaginalis on the spermatic cord. This intra-vaginal torsion of the testis may occur at any age,although it is commonest in adolescence. Extra-vaginal torsion of the whole spermatic cord is con-fined to new-born and this is rare. Chapman &Walton (1972) have reviewed 100 cases of torsion ofthe testis or its appendages at the London Hospital.In eighty-five instances there was torsion of thetestis, in fourteen torsion of the hydatid of Morganiand in one there was torsion of a lipoma of thescrotum. One case of testicular torsion was bilateral.The important finding was that in thirty-eight of thepatients with testicular torsion there had been oneor more preceding attacks of pain and swelling dueto a torsion which had spontaneously untwisted. Innineteen cases operative fixation was performedbecause of this preceding story and all these testeswere saved. In the other nineteen cases, the testeswere not explored until the patient presented with asevere episode, and seven of these testes were lostfrom either gangrene or delayed atrophy. Twentypatients were misdiagnosed and initially treated asepididymo-orchitis, although there was no evidenceof urinary tract infection. These authors point outfirst, the importance of 'warning' attacks of pain,and second, the advisability of exploring any acutelyswollen and painful testicle which is not accompaniedby definite evidence of urinary or urethral infection.The other testicle is explored and fixed at the sametime, and in most of the cases there will be found tobe a similar anatomical abnormality to that of theside in which the torsion has occurred. Krarup (1972)underlined the importance of early accurate diag-nosis; in eleven cases where this was achieved, thetestis was saved, but in eleven further patients wherediagnosis was wrong or delayed, often with anti-biotic therapy for misdiagnosed 'epididymitis',orchidectomy had to be performed when the patientcame to eventual surgery.A useful diagnostic point is stressed by Moore

(1972), who points out that the original pain feltwith torsion of the testis is in the lower abdomen justabove the internal inguinal ring; only when itsoverlying coverings are affected is the testis itselfpainful. This referral of the pain is responsible forsome early cases being misdiagnosed as appendicitis,or some other abdominal problem. Although the'bell-clapper' deformity cannot be detected clinicallyin the painful swollen torted testis, the abnormalitymay be seen with considerable frequency in theopposite, unaffected testis, which is liable to the samecongenital anomaly. This was pointed out by Angell(1963). Corriere (1972) now reports that in eighteenconsecutive cases of torsion of the testis he foundthat the opposite testis lay with its long axis hori-zontal; this can only be detected accurately when the

651P


http://pmj.bm

j.com/

Postgrad M


eptember 1973. D

ownloaded from

http://pmj.bmj.com/

Harold Ellis

patient is examined in the standing position. Thisclinical test is an important pointer to the diagnosis.A rare, but interesting, entity is noted by Phillips

Holmes (1972), who report nineteen patients withtorsion of an ectopic testis in the superficial inguinalpouch, immediately outside the external ring of theinguinal canal. No less than eighteen of these patientshad spastic paralysis. Since many of these werementally retarded, diagnosis was difficult and oftendelayed.The doctor who first sees a new patient with tor-

sion of the testis should immediately attempt mani-pulation which may be aided, if necessary, by apreliminary injection of morphia. Burton (1972), asMedical Officer of the Student Health Service inSheffield, reports six successful manipulations onyoung men between the ages of 15 and 22. Sub-sequently, both testes should undergo operativefixation as a prophylactic against subsequentepisodes.

Post-operative wound infectionPost-operative wound infection is still a common

and important complication of general surgicalprocedures. Under the conditions which exist inmost modern operating theatres, the majority ofwound infections follow either potentially con-taminated operations, which include the surgery ofthe hollow muscular organs (gut, biliary system,urinary system, etc.), or where there is frank infec-tion, such as purulent peritonitis, present at the timeof operation. The so-called 'clean' operations, suchas hernia repair, have a low infection rate, in theregion of 2-4o/, but this rises to the region of 20°/ inoperations on the gastro-intestinal and biliary tracts;this is particularly the case when the large bowel isopened or when biliary surgery is carried out in thepresence of infected bile (Engstrom et al., 1972).The use of plastic drapes appears to have no effectat all on lessening the rate of this endogenous infec-tion, as was shown by the careful controlled trialcarried out by William et al. (1972).

Since wound sepsis is now so often associated withintestinal organisms (including Escherichia coli,Streptococcus faecalis and Bacillus proteus) spilt intothe wound at the time of surgery, there is strongtheoretical support to the use of local antibiotics inthese potentially contaminated wounds in order todeal with the contaminating bacteria before infectioncan become established. Particularly in operationswhere the gastrointestinal or biliary tracts have beenopened, topical ampicillin in a dosage of 1 g inpowder form appears to be the antibiotic of choice.Andersen, Korner & Ostergaard (1972) report arandom trial in 240 patients undergoing colonicresection or abdomino-perineal excision of therectum. The infection rate in those receiving topical

ampicillin was 2 5°/, compared with 18.3%/ in thecontrol cases. There was no difference in the rate ofwound breakdown in the two groups. In our ownstudy (Stoker & Ellis, 1972) ampicillin was comparedwith a penicillin and sulphadiazine combination as atopical wound application in patients undergoingbiliary or gastro-intestinal surgery; eleven of thefifty-three patients having topical penicillin andsulphadiazine developed wound sepsis (20 8%y) com-pared with four of the fifty-nine patients receivingampicillin (6 8%). This difference is statisticallysignificant. In another study we have shown thatampicillin is superior in this respect to noxythiolinpowder (Stoker & Ellis, 1971). Indeed, noxythiolinin powder form appears to act as a chemical irritant,since there were ten examples of infection in onlytwenty-five patients in which this powder was used,eight of the infections comprising sterile pus orgrowing Staphylococcus albus only. The advantageof topical antibiotic powder is that an extremely highconcentration of this substance is achieved at thevery time when potential bacterial contaminationtakes place, the concentration of antibiotic being farhigher than can be obtained by systemic administra-tion, and yet being without the disadvantages andpotential dangers of a full systemic antibiotic course,which include Candida infections (Gaines &Remington, 1972). It is interesting in this contextthat Pollock & Tindal (1972) found no statisticaldifference in the infection rate of patients receivinga single intravenous injection of 500 mg of ampicillinimmndiately pre-operatively and untreated controls,whether these were clean, potentially contaminated,or frankly contaminated cases. However, Fullen,Hunt & Altermeier (1972), in a study of prophylacticantibiotics in penetrating abdominal wounds, founddramatic differences in their cases of colon perfora-tion. Patients where antibiotics were commencedpre-operatively developed wound infection in 6°/,while those whose therapy was started either pre- orpost-operatively had a wound infection rate respec-tively of 35 and 60%°.

It is interesting that many aspects of surgical ritualare coming under close scientific, and particularlybacteriological, scrutiny. Doig (1972) has recentlycompared the aerial contamination which followsthe use of conventional theatre clothes, specialtheatre garments, which include elastic arm bandsand the addition of cotton bloomers for females, andthe use of ordinary outdoor clothes. Surprisinglyenough, contamination was least when ordinary out-door clothes were worn. All three types of dress hadbacterial dissemination reduced by about half whengowns, masKs, caps and boots were worn. Aside fromcomfort, this study suggests that outdoor clothes neednot be removed in theatre or any clean area providedthat gown, mask, etc. are worn. A recent study from

652P


http://pmj.bm

j.com/

Postgrad M


eptember 1973. D

ownloaded from

http://pmj.bmj.com/


Westminster Hospital (Selwyn & Ellis, 1972)investigated the bacteriology of biopsy specimens ofskin. These enabled a very complete harvest ofsurface organisms to be obtained. Information aboutdeeply placed skin organisms was collected by ex-amining sections of skin biopsies after incubating for6 hr or more on agar. The predominant flora con-sisted of non-pathogenic Staphylococcus/Micro-coccus strains and diphtheroid bacilli. The bacteriawere found to be situated in the hair follicles, andafter incubation they could be seen exuding up themouths of these follicles. These residents are rarelypathogenic, and indeed in over 20% of subjectsproduced antibiotics against a wide range ofpathogens. These biopsy studies confirmed that thepatient's skin can be readily cleared of virtually allnon-sporing contaminants by swabbing with 0-5y.chlorhexidine or 1-50/ iodine in alcohol allowed toact for 30 sec. Only in the special case of tissueischaemia, for example, amputation of a gangrenouslimb in advanced arteriosclerosis, does it appear thatmore vigorous skin preparation is necessary in orderto remove spore-bearing organisms, and here alsoprophylactic systemic penicillin is required to preventclostridial infection.

Starch granuloma within the peritoneumA large number of substances which find their way

into the peritoneal cavity may stimulate granulomaformation, a sterile peritonitis, extensive intra-abdominal adhesions, or any combination of thethree. The first material to be implicated was talc(magnesium silicate), which was introduced as adusting powder for surgical gloves. This was there-fore abandoned and replaced by starch powder in1947, in the hope that this would be completelyabsorbed within the peritoneum. Unfortunately, insubsequent years isolated examples were reported ofgranuloma formation following the use of this sub-stance. The diagnosis is readily confirmed by micro-scopic examination of the tissue under polarizedlight, which demonstrates the characteristic Maltesecross appearance of the birefringent starch granules.Holmes & Eggleston (1972) have reported a seriesof thirteen patients with starch granulomatousperitonitis. The syndrome is typically seen between10 and 30 days after laparotomy. The symptoms andphysical findings give a picture suggesting eitherintestinal obstruction, intra-abdominal infection, orboth. Typically there is a pyrexia between 1000 and102°F with an elevation of the white blood count.Operative findings show a thickened nodular omen-tum, ascitic fluid, small miliary nodules scatteredover the serosal surfaces and dense adhesions. Unlessthe condition is borne in mind, the surgeon maydiagnose miliary tuberculosis, or, more dangerousstill, carcinomatosis peritonii. The disease appears

to be self-limiting, and there are no documentedcases of a second episode. Coder & Olander (1972)report three cases, two of whom were given steroidsfollowing operative confirmation of the diagnosis;both showed rapid clinical response and were wellwithin 48 hr. Ignatius & Hartmann (1972) reportseven confirmed cases of the starch peritonitis syn-drome. One of these was symptom-free, but multiplenodules were found at re-operation for an aorticaneurysm 7 months following cholecystectomy.These authors encountered six more clinically sus-pected cases, five of which settled within 2-6 weeksof conservative treatment, although a sixth patientcontinued to have a pain for 5 months. Three ofthese patients were given steroids and there was aclinical impression of response to this therapy.However, the same authors (Hartmann & Ignatius,1972), found that a single injection of hydrocortisonefailed to produce any observable effect on thegranulomatous reaction of intra-peritoneal starch inmice. Obviously, further clinical and laboratoryinvestigations are required to study the role ofsteroids in this context.An important help in diagnosis of this condition

may be abdominal paracentesis. Warshaw (1972)carried out this procedure in nine suspected cases ofstarch peritonitis. Ascitic fluid was obtained whichwas clear, contained mononuclear white cells, and,in seven examples, contained starch shown by iodinestaining or polarized light. All aspirates were sterileon culture. Ten control post-operative patientsunderwent paracentesis; the fluid contained smallernumbers of white cells and no starch. Re-explorationwas avoided in all nine cases, with recovery in eachinstance.The detailed pathology of this condition has been

well described by Davies & Neely (1972). Theystudied nine cases, eight following surgery, the ninthresulting from three aspirations for ascites. Thematerial was obtained from 12 days to 11 monthsafter surgery. The pathologist may encounter starchunder three circumstances; extraneous particlesfound indiscriminately in the mounting medium,freshly implanted starch in blood clots and on thesurface of the specimen but with no phagocyticresponse, and intrinsic, where the granules are almostall found within histiocytes, giant cells, or areas ofnecrosis. Only the last, of course, is of clinicalsignificance.

Naturally, most surgeons presenting series of casesof starch granuloma advise special precautions ofglove washing and drying in order to remove anytrace of glove powder. However, in a series ofexperiments which we reported at the Associationof Surgeons in 1972 (Jagelman & Ellis, 1973), wefound that careful washing of the gloves did notremove the starch powder completely but merely

653P


http://pmj.bm

j.com/

Postgrad M


eptember 1973. D

ownloaded from

http://pmj.bmj.com/

654 Harold El/is

clumped the starch granules together. In animalexperiments we noted that small amounts of starchpowder could be completely absorbed from thenormal peritoneal cavity of the rat, although largeamounts (between 05 and 1P5 g) would result inextensive granuloma formation. However, the samesmall amount of starch powder (0 01 g), which wasinnocuous in the intact peritoneal cavity, wouldproduce granulomas and adhesion formation in thepresence of peritoneal injury, presumably in associa-tion with the fibrinous exudation resulting from thistrauma. Possibly only small amounts of starchpowder are necessary in the traumatized peritonealcavity in order to precipitate adhesion formation.It is obvious that the ideal lubricant for the surgicalglove remains to be discovered.

ReferencesCarcinoma of the breastATKINS, H., HAYWARD, J.L., KLUGMAN, D.J. & WAYTE, A.B.

(1972) Treatment of early breast cancer: a report after 10years of a clinical trial. British Medical Journal, 2, 423.

BAUM, M., EDWARDS, M.H. & MAGAREY, C.J. (1972)Organisation of clinical trial on national scale: manage-ment of early cancer of the breast. British Medical Journal,4, 476.

CAMPBELL, H. (1972) A review of studies of breast cancer inhuman populations excluding endocrine studies. Proceed-ings of the Royal Society of Medicine, 65, 641.

EDWARDS, M.H., BAUM, M. & MAGAREY, C.J. (1972)Regression of axillary lymph-nodes in cancer of thebreast. British Journal of Surgery, 59, 776.

GALASKO, C.S.B. (1972) Skeletal metastases and mammarycancer. Annals of the Royal College of Surgeons, 50, 3.

HANDLEY, R.S. (1972) Observations and thoughts on cancerof the breast. Proceedings of the Royal Society of Medicine,65, 437.

KETT, K., LUKACS, L. & VARGA, G. (1972) Diagnostic valueof lymphography of the arm in the pre-operative diagnosisof early metastases in breast cancer. American Journal ofSurgery, 123, 712.

LEADING ARTICLE (1972) Treatment of early carcinoma ofbreast. British Medicat Journal, 2, 417.

LEADING ARTICLE (1973) Locally advanced breast cancer.British Medical Journal, 1, 372.

LEE, E.S., NEWTON, K.A., WESTBURY, G. & SPITTLE, M.F.(1972) Treatment of early breast cancer. British MedicalJournal, 2, 71 1.

SONNELAND, J. (1972) The 'inoperable' breast carcinoma. Asuccessful result using zinc chloride fixative. AmericanJournal of Surgery, 124, 391.

STOKER, T.A.M. & ELLIS, H. (1972) Post-irradiation toiletmastectomy in the management of locally advancedcarcinoma of the breast. British Journal of Radiology, 45,851.

WALLACE, I.W.J. & CHAMPION, H.R. (1972) Axillary nodesin breast cancer. Lancet, i, 217.

Peptic ulcer surgeryBARNES, A.D. & WILLIAMS, J.A. (1967) Stomach drainage

after vagotomy and pyloroplasty. American Journal ofSurgery, 113, 494.

BYNUM, T.E., HARTSUCK, J. & JACOBSON, E.D. (1972) Gastriculcer. Gastroenterology, 62, 1052.

CLARKE, R.J., LEWIS, D.L. & WILLIAMS, J.A. (1972) Vagotomyand pyloroplasty for gastric ulcer. British Medical Journal,2, 369.

CLARKE, R.J., MCFARLAND, J.B. & WILLIAMS, J.A. (1972)Gastric stasis and gastric ulcer after selective vagotomywithout a drainage procedure. British Medical Journal, I,538.

ELLIS, H. (1972) Pyloric stenosis due to duodenal ulceration.In: Chronic Duodenal Ulcer (Ed. by C. Wastell), chap. 15.Butterworths, London.

GOLIGHER, J.C., PULVERTAFT, C.N., IRVIN, T.T., JOHNSTON, D.,WALKER, B., HALL, R.A., WILLSON-PEPPER, J. & MATHE-SON, T.S. (1972) Five to eight year results of truncalvagotomy and pyloroplasty for duodenal ulcer. BritishMedical Journal, 1, 7.

HALL, R. (1972) Gastric ulcer after highly selective vagotomy.British Medical Journal, 4, 789.

HEDENSTEDT, S., LUNDQUIST, G. & MOBERG, S. (1972)Selective proximal vagotomy (SPV) in the treatment ofduodenal ulcer. A preliminary report. Acta chirurgicascandinavica, 138, 591.

HENDRY, W.G. (1962) Tubeless gastric surgery. BritishMedical Journal, 1, 1736.

HERRINGTON, J.L. (1972) Additional experience with elimina-tion of routine nasogastric suction following gastricoperations. Surgery, 71, 132.

HUMPHREY, C.S., JOHNSTON, D., WALKER, B.E., PULVERTAFT,C.N. & GOLIGHER, J.C. (1972) Incidence of dumping aftertruncal and selective vagotomy with pyloroplasty andhighly selective vagotomy without drainage procedure.British Medical Journal, 3, 785.

JOHNSTON, D., GOLIGHER, J.C., PULVERTAFT, C.N., WALKER,B.E., AMDRUP, E. & JENSEN, H.E. (1972a) The two to fouryear clinical results of highly selective vagotomy (parietal-cell vagotomy) without drainage procedure for duodenalulcer. Gut, 13, 842.

JOHNSTON, D., HUMPHREY, C.S., WALKER, B.E., PULVERTAFT,C.N. & GOLIGHER, J.C. (1972b) Vagotomy withoutdiarrhoea. British Medical Journal, 3, 788.

JOHNSTON, D., HUMPHREY, C.S., SMITH, R.B. & WILKINSON,A.R. (1972c) Treatment of gastric ulcer by highly selectivevagotomy without a drainage procedure: an interimreport. British Journal of Surgery, 59, 787.

KENNEDY, T., KELLY, J.M. & GEORGE, J.D. (1972) Vagotomyfor gastric ulcer. British Medical Journal, 2, 371.

KERRY, R.L., HOSHAL, V.L. & RuIz, J.A. (1971) Gastricdecompression following vagotomy and drainage pro-cedures. Archives of Surgery, 102, 248.

MCCOLL, I., DRINKWATER, J.E., HULME-MUIR, I. & DONNAN,S.P.B. (1971) Preoperative prediction of success or failureof gastric surgery. Gut, 12, 856.

MCKEOWN, K.C. (1972) A prospective study of the im-mediate and long-term results of polya gastrectomy forduodenal ulcer. British Journal of Surgery, 59, 849.

MIGUEL, J. DE (1972) Conservative surgical approach to thetreatment of gastric ulcer. British Journal of ClinicalPractice, 26, 205.

MILLER, D.F., MASON, J.R., MCARTHUR, J. & GORDON, I.(1972) A randomised prospective trial comparing threeestablished methods of gastric decompression aftervagotomy. British Journal of Surgery, 59, 605.

MUSGROVE, J.E. (1972) Post vagotomy abdominal chylousfistula. Annals of Surgery, 175, 67.

NACHLAS, M.V., YOUNIS, M.T., RODA, C.P. & WITYK, J.J.(1972) Gastrointestinal motility studies as a guide to post-operative management. Annals of Surgery, 175, 510.

POSTLETHWAIT, R.W. & JOHNSON, W.D. (1972) Complica-tions following surgery for duodenal ulcer in obesepatients. Archives of Surgery, 105, 438.

PRICE, J.J., POWIS, S.J.A. & MORRISSEY, D.M. (1972)Oesophagal perforation during abdominal truncal vago-tomy for duodenal ulcer. British Journal of Surgery, 59,936.



j.bmj.com

/P


j.49.575.644 on 1 Septem

ber 1973. Dow

nloaded from

http://pmj.bmj.com/

Annual review of surgery 1972 655

WASTELL, C., COLIN, J.F., MACNAUGHTON, J.I. & GLEESON,J. (1972) Selective proximal vagotomy. British MedicalJournal, 1, 28.

Liver and biliary systemADMIRAND, W.H. & SMALL, D.M. (1968) The physicochemical

basis of cholesterol gallstone formation in man. Journal ofClinical Investigation, 47, 1043.

BELL, C.D., WHITNEY, B. & DOWLING, R.H. (1972) Gallstonedissolution in man using chenodeoxycholic acid. Lancet,ii, 1213.

BELL, G.D., SUTOR, D.J., WHITNEY, B. & DOWLING, R.(1972) Factors influencing human gallstone dissolution inmonkey, dog and human bile. Gut, 13, 836.

BLUMGART, L.H. & VAJRABUKKA, T. (1972) Injuries to theliver: analysis of 20 cases. British Medical Journal, 1, 158.

BLUMGART, L.H., LEACH, K.G. & KARRAN, S.J. (1971)Observations on liver regeneration after right hepaticlobectomy. Gut, 12, 922.

DANZIGER, R.G., HOFMANN, A.F., SCHOENFIELD, L.J. &THISTLE, J.L. (1972) Expansion of decreased bile acid pooland dissolution of gallstones by chenodeoxycholic acid.Gut, 13, 321.

ELMSLIE, R.G. & SLAVOTINEK, A.H. (1972) Surgical objectivesin unresected cancer of the head of the pancreas. BritishJournal of Surgery, 59, 508.

LIM, R.C., KNUDSON, J. & STEELE, M. (1972) Liver trauma;current method of management. Archives of Surgery, 104,544.

MCKAY, C., CROOK, J.N., SMITH, D.C. & MCALLISTER, R.A.(1972) The composition of hepatic and gallbladder bile inpatients with gallstones. Gut, 12, 759.

MAY, E.T. (1972) Lobar dearterialisation for exsanguinatingwounds of the liver. Journal of trauma, 12, 397.

MERCADIER, M.P., CLOT, J.P. & CADY, J.-P. (1972) Righthepatectomy in the treatment of liver trauma. AmericanJournal of Surgery, 124, 353.

MORTON, J.R., Roys, G.D. & BRICKER, D.L. (1972) Thetreatment of liver injuries. Surgery, Gynaecology andObstetrics, 134, 298.

NILOFF, P.H. (1972) A prosthesis for palliative treatment ofobstructive jaundice due to cholangiocarcinoma. Surgery,Gynaecology and Obstetrics, 135, 611.

SMALLWOOD, R.A., JABLONSKI, P. & WATTS, J. MCK. (1972)Intermittent secretion of abnormal bile in patients withcholesterol stones. British Medical Journal, 4, 263.

TERBLANCHE, J., SAUNDERS, S.J. & Louw, J.H. (1972)Prolonged palliation in carcinoma of the main hepaticduct. Surgery, 71, 720.

WARREN, K.W., MOUNTAIN, J.C. & LLOYD-JONES, W. (1972)Malignant tumours of the bile ducts. British Journal ofSurgery, 59, 501.

WHITE, P. & CLEVELAND, R.J. (1972) The surgical manage-ment of liver trauma. Archives of Surgery, 104, 785.

WILLIAMS, R. (1972) Aspects of liver transplantation.Journal of the Royal College of Surgeons of Edinburgh, 17,148.

Congenital anomalies of the testisANGELL, J.C. (1963) Torsion of the testis. A plea for diag-

nosis. Lancet, 1, 19.BURTON, J.A. (1972) Atrophy following testicular torsion.

British Journal of Surgery, 59, 422.CHAPMAN, R.H. & WALTON, A.J. (1972) Torsion of the testisand its appendages. British Medical Journal, 1, 164.

CORRIERE, J.N. (1972) Horizontal lie of the testicle: a diag-nostic sign in torsion of the testis. Journal of Urology, 107,616.

KRARUP, T. (1972) Torsion of the testis. ScandinavianJournal of Urology and Nephrology, Supplement 15, 6, 165.

MooRE, T. (1972) Torsion of testis. British Medical Journal,1, 374.

PHILLIPs, N.B. & HOLMES, T.W. (1972) Torsion infarction onectopic cryptorchidism: a rare entity occurring mostcommonly with spastic neuromuscular disease. Surgery,71, 335.

PRYN, W.J. (1972) The maintenance of the maldescendedtesticles within the scrotum using a dartos pouch. BritishJournal of Surgery, 59, 175.

RIEGLER, H.C. (1972) Torsion of an intra-abdominal testis.An unusual problem in diagnosis of the acute surgicalabdomen. Surgical Clinics of North America, 52, 371.

SCORER, C.G. & FARRINGTON, G.H. (1971) CongenitalDeformities of the Testis and Epididymis. Butterworths,London.

Post-operative wound infectionsANDERSEN, B., KORNER, B. & OSTERGAARD, A.H. (1972)

Topical ampicillin against wound infection after colorectalsurgery. Annals of Surgery, 176, 129.

DOIG, C.M. (1972) The effects of clothing on the dissemina-tion of bacteria in operating theatres. British Journal ofSurgery, 59, 878.

ENGSTROM, J., GROTH, C.G., LUNDH, G. & LONNQVIST, B.(1972) Infectious complications after surgery for biliarycalculus. Acta chirurgica scandinavica, 138, 357.

FULLEN, W.D., HUNT, J. & ALTERMEIER, W.A. (1972)Prophylactic antibiotics in penetrating wounds of theabdomen. Journal of Trauma, 12, 282.

GAINES, J.D. & REMINGTON, J.S. (1972) Disseminated can-didiasis in the surgical patient. Surgery, 72, 730.

POLLOCK, A.V. & TINDAL, D.S. (1972) The effect of a single-dose parenteral antibiotic in the prevention of woundinfection. A controlled trial. British Journal ofSurgery, 5998.

SELWYN, S. & ELLIS, H. (1972) Skin bacteria and skin dis-infection reconsidered. British Medical Journal, 1, 136.

STOKER, T.A.M. & ELLIS, H. (1971) Prophylaxis of woundinfection. British Medical Journal, 3, 769.

STOKER, T.A.M. & ELLIS, H. (1972) Wound antibiotics ingastro-intestinal surgery. Comparison of ampicillin withpenicillin and sulphadiazine. British Journal of Surgery,59, 187.

WILLIAMS, J.A., OATES, G.D., BROWN, P.P., BURDEN, D.W.,MCCALL, J., HUTCHISON, A.G. & LEES, L.J. (1972)Abdominal wound infections and plastic wound guards.British Journal of Surgery, 59, 142.

Starch granulomaCODER, D.M. & OLANDER, G.A. (1972) Granulomatous

peritonitis caused by starch glove powder. Archives ofSurgery, 105, 83.

DAVIES, J.D. & NEELY, J. (1972) The histopathology ofperitoneal starch granulomas. Journal of Pathology, 107,265.

HARTMANN, W.H. & IGNATIUS, J.A. (1972) The starch-peritoneal reaction: an experimental study. Annals ofAllergy, 175, 398.

HOLMEs, E.C. & EGGLESTON, J.C. (1972) Starch granuloma-tous peritonitis. Surgery, 71, 85.

IGNATIUS, J.A. & HARTMANN, W.H. (1972) The glove starchperitonitis syndrome. Annals of Surgery, 175, 388.

JAGELMAN, D. & ELLIS, H. (1973) The role of starch in intra-peritoneal adhesion formation. British Journal of Surgery,60, 111.

WARSHAW, A.L. (1972) Diagnosis of starch peritonitis byparacentesis. Lancet, ii, 1054.



j.bmj.com

/P


j.49.575.644 on 1 Septem

ber 1973. Dow

nloaded from

http://pmj.bmj.com/

Documents

Annual of 1972 - BMJ