228 POST-GRADUATE MEDICAL JOURNAL July, I945 THE COMPLICATIONS OF GASTRIC AND DUIODENAI, ULCER By A. K. MONRO, M.A., M.D., F.R.C.S.Eng. (Surgical Registrar, Southend General Hospital, Hunterian Projessor, Royal College of Surgeons.) i. Haemorrhage Haemorrhage is the commonest complication of gastric and duodenal ulceration. It has been estimated to occur, in gross form, in about 2o per cent of all cases. It may be of any degree, from the slight haemorrhage which only discolours the stools or the persistent small haemorrhages which cause a severe anaemia, to the more common, profuse and dramatic haemetemesis which at times is so severe as in a few minutes to endanger the patient's life. Pathology In acute gastric erosions haemorrhage is a common complication but, being caused by a superficial lesion, it does not endanger life. In more penetrating acute gastric or duodenal ulcers haemorrhage may be severe and even, on occasion, fatal; this is, however, a rare event. In all active chronic peptic ulcers minute haemorrhages are almost constant as shown by the presence of occult blood in the stools. Such bleeding comes from small vessels in the walls of the ulcer. Larger vessels, however, may be involved, particularly the pancreaticoduodenal, left gastric and more rarely the splenic arteries and their branches. As the arterial wall becomes eroded, a time comes when the wall is unable to contain the pressure of the arterial pulse and haemorrhage occurs. In the case of a large vessel brisk haemorrhage continues probably for IO-I5 minutes in which time 2-3 pints of blood may be lost. From rapid distension of the stomach the patient is likely to vomit, whilst from loss of blood he feels weak or faints. Various factors then come into play which tend to stop the hae- morrhage, namely: (a) Fall of blood pressure. (b) Retraction of the inner coat of the vessel. (c) The coagulability of the blood. In favourable circumstances a clot forms in the open vessel. If this clot is allowed to become firm, danger is, at least temporarily, past. Any variation of the above three factors may affect unfavourably the development and retraction of the clot. Rest- lessness or movement of the patient, or-injudiciously *rapid intravenous infusion may raise the blood pressure. Retraction of the inner coat may be prevented by arteriosclerosis of the vessel's wall or by its fixation in thick scar tissue, whilst the coagulability of the blood may be diminished by vitamin deficiency, by starvation or as the result of repeated haemorrhages. These three last are especially noteworthy, because they may all three be present under the starvation regime. Under normal conditions the coagulability of the blood is increased after a single haemorrhage (Moon 194I), whilst blood transfusion has a siimiilar effect. Methods of Treatment and Results Until recent years the recognised method of treatment of peptic ulcer haemorrhage was medi- cal, comprising a starvation, morphia, gradually increasing diet regime. Later blood transfusion was added, at first given rapidly, then by the slow continuous drip method of Marriott and Kekwick (I935). Surgery was undertaken only for those cases which were despaired of under medical treatment, and, not unexpectedly, was attended by a huge mortality. The results of this treatment have been widely discussed and figures for mortality reported from all parts of the world. They showed a wide varia- tion, ranging from 4 * 2 per cent (Crohn and Lerner, New York, I939) to 58 per cent (Ross, Melbourne, I930), but the average figure lay between 9 and 12 per cent. The following are representative: Mortality Statistics. Starvation Regime London, Aitken I934 ii per cent Birmingham, Bulmer I932 I0.7 Denmark, Christiansen I934 799 Norway, Frostad I934 9.4 Sweden, Mossberg I933 9.0 Germany, Umber I935 9-5 Boston, Jankelson I938 90O San Francisco, Goldman I937 II-5 This method of treatment therefore has a very considerable mortality. It is furthermore open to the grave objection that a number of deaths and considerable morbidity result from the treat- ment rather than from the disease. Chest compli- cations are largelv responsible for this whilst Meulengrach (I937) vividly describes the miseries of a patient who died from ascending parotitis after 8-Io days under strict medical regime. The reversal of the starvation policy in the treatment of ulcer haemorrhage was first suggested by Andresen in I927. His routine included the 2-hourly administration of a gelatin-water copyright. on March 26, 2021 by guest. Protected by http://pmj.bmj.com/ Postgrad Med J: first published as 10.1136/pgmj.21.237.228 on 1 July 1945. Downloaded from
Transcript
228 POST-GRADUATE MEDICAL JOURNAL July, I945
THE COMPLICATIONS OF GASTRICAND DUIODENAI, ULCER
By A. K. MONRO, M.A., M.D., F.R.C.S.Eng.
(Surgical Registrar, Southend General Hospital,Hunterian Projessor, Royal College of Surgeons.)
i. HaemorrhageHaemorrhage is the commonest complication of
gastric and duodenal ulceration. It has beenestimated to occur, in gross form, in about 2o percent of all cases. It may be of any degree, fromthe slight haemorrhage which only discolours thestools or the persistent small haemorrhages whichcause a severe anaemia, to the more common,profuse and dramatic haemetemesis which at timesis so severe as in a few minutes to endanger thepatient's life.
PathologyIn acute gastric erosions haemorrhage is a
common complication but, being caused by asuperficial lesion, it does not endanger life. Inmore penetrating acute gastric or duodenal ulcershaemorrhage may be severe and even, on occasion,fatal; this is, however, a rare event.
In all active chronic peptic ulcers minutehaemorrhages are almost constant as shown bythe presence of occult blood in the stools. Suchbleeding comes from small vessels in the walls ofthe ulcer. Larger vessels, however, may beinvolved, particularly the pancreaticoduodenal,left gastric and more rarely the splenic arteriesand their branches. As the arterial wall becomeseroded, a time comes when the wall is unable tocontain the pressure of the arterial pulse andhaemorrhage occurs. In the case of a largevessel brisk haemorrhage continues probably forIO-I5 minutes in which time 2-3 pints of bloodmay be lost. From rapid distension of the stomachthe patient is likely to vomit, whilst from loss ofblood he feels weak or faints. Various factorsthen come into play which tend to stop the hae-morrhage, namely:
(a) Fall of blood pressure.(b) Retraction of the inner coat of the vessel.(c) The coagulability of the blood.
In favourable circumstances a clot forms in theopen vessel. If this clot is allowed to become firm,danger is, at least temporarily, past. Any variationof the above three factors may affect unfavourablythe development and retraction of the clot. Rest-
lessness or movement of the patient, or-injudiciously*rapid intravenous infusion may raise the bloodpressure. Retraction of the inner coat may beprevented by arteriosclerosis of the vessel's wallor by its fixation in thick scar tissue, whilst thecoagulability of the blood may be diminished byvitamin deficiency, by starvation or as the resultof repeated haemorrhages. These three last areespecially noteworthy, because they may all threebe present under the starvation regime. Undernormal conditions the coagulability of the bloodis increased after a single haemorrhage (Moon194I), whilst blood transfusion has a siimiilar effect.
Methods of Treatment and ResultsUntil recent years the recognised method of
treatment of peptic ulcer haemorrhage was medi-cal, comprising a starvation, morphia, graduallyincreasing diet regime. Later blood transfusionwas added, at first given rapidly, then by the slowcontinuous drip method of Marriott and Kekwick(I935). Surgery was undertaken only for thosecases which were despaired of under medicaltreatment, and, not unexpectedly, was attendedby a huge mortality.The results of this treatment have been widely
discussed and figures for mortality reported fromall parts of the world. They showed a wide varia-tion, ranging from 4 * 2 per cent (Crohn and Lerner,New York, I939) to 58 per cent (Ross, Melbourne,I930), but the average figure lay between 9 and12 per cent. The following are representative:
This method of treatment therefore has a veryconsiderable mortality. It is furthermore opento the grave objection that a number of deathsand considerable morbidity result from the treat-ment rather than from the disease. Chest compli-cations are largelv responsible for this whilstMeulengrach (I937) vividly describes the miseriesof a patient who died from ascending parotitisafter 8-Io days under strict medical regime.The reversal of the starvation policy in the
treatment of ulcer haemorrhage was first suggestedby Andresen in I927. His routine included the2-hourly administration of a gelatin-water
copyright. on M
arch 26, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.21.237.228 on 1 July 1945. D
mixture but he did not publish the results of hismethod until I939 when he reported a mortalityrate of 2.5 per cent in I20 patients. It remainedto Meulengracht (934) to produce a profoundand widespread impression by his advocacy of animmediately liberal diet in the treatment of pepticulcer haemorrhage. He advocated 3-hourly feedsof a light, soft diet including tea, cocoa, bread andbutter, a variety of soups, minced meat and fish,representing about 2,300 calories, together withalkalies, extract of hyoscyamus and iron. Theworse his condition, the greater the importancelaid on the patient taking his feeds; from thefirst day he was allowed to move about in bed.In I937 Meulengracht reported 368 consecutivecases of haematemesis and melaena from pepticulcers, acute and chronic, treated on these lines,with a mortality of I-3 per cent.These claims have been fully substantiated by
subsequent writers. Witts (I937) in a mostexcellently reasoned article states that patientsso treated look, feel and do much better than anyprevious series he has seen. The following figures,from various sources, are representative of theresults obtained:
Mortality Statistics. Prompt FeedingRegime
Jones, St. Bartholomew'sHospital I939 20 per cent
Lineberry, Birmingham 1937 3 6Scott, Glasgow 1940 3.3Woldman, St. Luke's Hosp. I94I 2 ,-Thotsted, Detroit I942 2-8
A comparison of the results shown in these twotables leaves little doubt as to the efficacy of theprompt feeding regime. The following suggestionshave been put forward as reasons for its success ascompared with the starvation regime:
(i) Peristaltic action is diminished when thestomach contains food.
(2) Food neutralises the gastric juices whichtherefore do not come into contact, undiluited,with the ulcer.
(3) The deleterious effects of deprivation offluid, food and vitamins on shock, clottingof blood, blood regeneration andconvalescenceare avoided.
(4) The patient's morale is much improved.
Thus the most effective medical treatment atthe present time is a combination of prompt feedingwith a moderate calorie, high vitamin value diettogether with iron, alkalies, antispasmodics,moderate doses of morphia and blood transfusion
by the drip method. In the last io years with thedevelopment of this treatment, the mortality rateof peptic ulcer haemorrhage has been very muchreduced, in fact to about 3 per cent. It is rarenow, therefore for a patient to die of this con-dition and there is an inclination to accept theremaining mortality as inevitable. Meulengracht(I937) voices this by stating that with so low amortality the question of operation does not arise.When figures for surgical treatment of this
condition in past years are quoted with mortalityrates up to I00 per cent (Hurst and Ryle. I937)there is a tendency to take the view that thesecases represent the complete failure of surgery inhaemorrhage from gastric and duodenal ulcers.It can be argued with greater truth that, as surgerywas invoked only as a last desperate venture,the failure lay primarilv in not recognising theseverity of the haemorrhage earlier in its course,before the patient had entered the stage of irre-versible shock.
Is it possible to predict which cases are likely'todie from peptic ulcer haemorrhage?
Statistics from many sources have shown thatulcer haemorrhage is rarely fatal (a) in women,(b) under the age of 45, (c) from acute ulcers, andthat in fatal cases arteriosclerosis is a commonpost-mortem finding (Blackford and Allen. I942,Thorsted. I942). Cullinan and Price (I932) showedthat a recurrent haemorrhage is much more likelyto be fatal than a first haemorrhage, althoughBlackford (I942) shows that first haemorrhagesare responsible for 75 per cent of the fatalities.Frpm these facts it appears that there is a groupin which the mortality is considerably higher thanthe average figure, namely, in men (i), over 45years of age (ii), known to have chronic ulcers(iii), which have previously caused gross haemor-rhage (iv), the outlook being further worsened bythe presence of arteriosclerosis. The writer is ofthe opinion that the mortality figures in this groupwould be of very considerable interest and maywell prove to be in the region of 30 per cent.Can anything be learned by division of the cases
into two groups, mild and severe? Variouscriteria have been put forward to differentiate thetwo. The earlier suggestions such as a red cellcount below 2,000,000 per cmm. or a haemoglobinbelow 40 per cent have been shown to be unreliableowing to the variability of haemodilution afterhaemorrhage. Such cases are now, in any case,routinely transfused, which further vitiates theresult. Witts (I937) regards as serious any casein which the blood urea rises above 75 mgms. percent, but this has not been widely followed up.Recent methods of examination of the bloodpromise more reliable information as to the severityand progress of the case. They include deter-
copyright. on M
arch 26, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.21.237.228 on 1 July 1945. D
mination of the blood specific gravity by thehanging-drop method of Barbour and Hamilton(I926), blood haematocrit readings and the esti-mation of plasma protein values. Scudder (1939)showed that in conditions of massive haemorrhagethe blood specific gravity falls from the normalI056.6 to I040 or less. From these or similarestimations new criteria may emerge by which itwill be possible early in the event to differentiatebetween mild and severe haemorrhage.
Hinton (I939) proposed a therapeutic test.He suggested that, if the haemoglobin level andthe blood pressure remain low after the admini-stration of two blood transfusions of 500 c.cs.each, a large vessel has been involved and surgeryis indicated.Has surgery then a place in the treatment of
ulcer haemorrhage? Firstly, let us agree that,as a last resort where prolonged haemorrhage hasfailed to respond to the best medical treatment,surgical intervention holds out very slender hope.If, however, this same case be brought to operationearlier in the course of the haemorrhage the outlookis very different. Finsterer (I939), who firstadvocated immediate operation for massive ulcerhaemorrhage reported a mortality of 5 'I per cent in78 cases operated upon within 48 hours of the onsetof haemorrhage. After 48 hours his mortality raterose to 29.7 per cent. Gordon-Taylor (I935) insimilar circumstances records a mortality ofI9 per cent in 32 cases. Rankin (I939) is of theopinion that surgery is advisable in massivehaemorrhage from a known chronic ulcer whichhas failed to respond to medical treatment, aswell as in recurrent haemorrhage, but most writersavoid absolute indications stating merely thatoperation should be considered.On the surgical side advances have been made
in the fields of rapid and controlled shock therapy,in anaesthesia and in post-operative care. Theoperation of choice is partial gastrectomy withremoval of the ulcer; if the patient's conditiondoes not permit this, the ulcer is exposed byopening the stomach or duodenum and haemorrhageis controlled either by sutures passed under anobvious bleeding vessel, or by mattress suturesunderrunning the whole ulcer. The chief dangersare shock and haemorrhage, ileus and post-operativechest complications. Rapid and continuous bloodtransfusion, the indwelling gastric or jejunal suctiontube and cyclopropane anaesthesia with its smoothinduction, rapid elimination and absence of post-operation vomiting, together with chemotherapyand suction-bronchoscopy if necessary, have donemuch to diminish these risks. Provided, therefore,firstly that the patient is in other respects a goodanaesthetic risk; secondly that he has not enteredan irreversible state of shock, and thirdly that a
competent surgical team, anaesthetist and bloodtransfusion facilities are immediately available,surgery holds out good hope. The problem revolvesround the possibility of bringing the serious caseto operation during the early stage of the haemor-rhage, that is, within 24 or perhaps 48 hours ofits onset.
ConclusionThe treatment of haemorrhage from chronic
peptic ulcers of the stomach and duodenum iswithout doubt medical. Since the introductionof the prompt feeding regime by Meulengrachtresults have been encouraging and the mor-tality has been greatly reduced. Deaths aretherefore uncommon but they are none the lessdistressing.
If the fatal tendencies of the serious cases can beforetold during the early stages of the haemorrhage,surgery, in otherwise healthy individuals, holdsout good hope. In a certain group of cases themortality under medical regime remains high,namely in recurrent haemorrhages in men over45 years of age known to have chronic ulcers,especially in the presence of arteriosclerosis. Inthis group, provided that the patient is otherwisehealthy and that competent surgical, anaestheticand transfusion teams are at hand, operation isindicated within 24-48 hours of the onset ofhaemorrhage.
In first haemorrhages, which are responsiblefor a high proportion of the fatalities in this con-dition, surgery has, been advocated, not un-reasonably, in cases of known chronic ulcer whichin spite of full medical treatment proceed tobleed (Rankin). In general, however, we are asyet unable to form an opinion early in the diseaseof the potentialities of such cases. With furtherresearch, perhaps by a combination of bloodexamination and of a therapeutic test, we may beable to pick out the seriouis cases within 24 hoursor less of their onset. Surgery may then be ableto play a further part in the reduction of themortality from ulcer haemorrhage.
2. PerforationGeneral ConsiderationSince Mikulicz first closed a perforated ulcer
in i88o, the treatment of perforation of pepticulcers has been the subject of perennial discussionand enquiry. More especially has this been so inrecent years because the incidence of the conditionhas been shown to be increasing, whilst in spiteof all efforts the mortality in the last ten yearshas shown little change, remaining for general
GASTRIC AND DUODENAL ULCER2hospital series in the region of 20-25 per cent.For the patient, perforation is one of the mostagonising catastrophies; for the surgeon, it isespecially a challenge because the majority ofhis cases are useful members of the community inyoung middle life.
Historically the death at the Court of Louis XIV,of the Duchess of Orleans, daughter of KingCharles I of England, from a perforated gastriculcer was most vividly described by Madame dela Fayette, and was ^epitomised by Bossuet inthese famous words: "O nuit effroyable, oi retentittout a coup, comme un eclat de tonnerre, cetteetonnante nouvelle: Madame se meurt! Madameest morte"' (Bossuet, I670).The first successful operation for closure was
performed by Heussner in I892 (Kriege, i892).Braun (I897) introduced the addition of gastro-enterostomy, whilst von Haberer (I9I9) advisedpartial gastrectomy in suitable cases, a procedurenow popular in certain clinics on the continent.It has long been known, however, that all casesdo not necessarily die without operation. Hall(I892) reported 6 cases of spontaneous recovery,whilst Blackford (I942) mentions 28 cases treatedby gastric suction through a Levine tube, withoutoperation, with three deaths. Occasional caseshave been similarly treated in this country, withcomparable results (Turner 1945).
It appears to be impossible to estimate satis-factorily what percentage of ulcers perforate, butthere can be no doubt that the incidence of per-foration has greatly increased in the last 20 years(de Bakey I940). Furthermore, from being mainlyan affection of women (Brinton I856: 68*4 percent females) it is now largely restricted to men(92 .2 per cent, collected figures, de Bakey. go percent, Southend General Hospital). The diminishedincidence of acute ulceration in young womenaccounts in part for this change, but for the higherincidence in men some factor in our everydaylife would appear to be responsible. Much evi-dence points to increased tobacco smoking as theimportant factor. Bager (I929) draws interestingparallels between the annual tobacco consumptionand the perforated ulcer incidence in Sweden.Records of the percentages of smokers and non-smokers developing perforations would be ofconsiderable interest.
All age groups are affected, from early infancy(2 days, Stern, I929) to old age (84 years, KellyI939), but the majority of cases occur between theages of 20 and 50 years. Judine (1939) showsthat the average age of duodenal perforationsin his series was 32 years, whilst that of gastricperforations was 46 years. A slight seasonalvariation was noticed by him possibly due to anincrease after influenza epidemics.
There has been much discussion as to whetherperforation of gastric or duodenal ulcers is com-moner. Widely differing figures have been pub-lished. It would appear that this divergence ofopinion has arisen in part at least, from the diffi-culty at operation of defining the exact positionof the pylorus. It is a common experience to findthat a perforation, first thought to be pyloric(and therefore gastric), on closer inspection provesto be duodenal. In 54 cases at the SouthendGeneral Hospital 6 were gastric giving a ratio of8 : i in favour of duodenal perforations.
Judine (1939) from careful macro- and micro-scopical examination of stomachs resected forperforation states that 87-5 per cent of 928 caseswere duodenal. Other recent figures bear out thispreponderance (de Bakey 1940).
Controversy has raged also over the question,is it the acute or the chronic ulcer which perforates?At operation, with the whole area grossly oedema-tous, it may be difficult to establish the exactnature of the underlying ulcer. The patient'sprevious history is therefore a more accurate guideand should be enquired into a second time duringconvalescence. In the agony of perforation a manmay well forget earlier comparatively trivialattacks of indigestion. In 54 cases at the SouthendGeneral Hospital 38 gave a definite history ofattacks of dyspepsia (70 per cent). Higherfigures are common (go per cent, Graves I933.95 per cent, Judine I939). Walton (1930) statesthat of 42 cases which came to autopsy, 4I showedchronic ulcers. Chronic ulcers are therefore re-sponsible for the majority of perforations, butwithout doubt acute ulcers can perforate and, inso doing, produce their first symptom.The only point of agreement amidst these con-
troversial matters is that it -is the anterior ulcerwhich perforates. Posterior ulcers penetrate butrarely perforate. They perforate only (a) intothe lesser sac, which is rare as the sac is usuallyobliterated near the ulcer by adhesions, or (b)by involving the superior surface of the stomachor duodenum, and so the general peritonealcavity.The diagnosis is not as a rule difficult. Di-
minished liver dullness, elicited with the chestraised, is an important sign. It can frequentlybe confirmed by an X-ray taken with the patientsitting, when a crescent of gas is seen under thediaphragm. A duodenal perforation not in-frequently simulates acute appendicitis owing toescape of duodenal content down the right para-colic gutter; this provides one of the commonestsources of error in abdominal diagnosis. It is ofinterest that although the inflammation producedby perforation of peptic ulcers is in close proximityto the diaphragm and is a fruitful source of sub-
phrenic abscess, pain referred to the shoulders isnot common (de Bakey, I940).Moynihan (I928) pointed out that patients
suffering from perforation are ra-rely in a state oftrue surgical shock as shown by rise of pulse rateand fall of blood pressure. He refers to theirstate as one `of prostration. Soutter (194I) statesthat only 20 of his series of 335 cases showed signsof true surgical shock. If true shock is presentthe prognosis is grave.The bacteriology of the peritoneal exudate has
been extensively studied and has been shown tobe valuable in prognosis. It is now also of con-siderable value in influencing the choice of chemo-therapy. In the first 6 hours the fluid is oftensterile, in the second 6 hours it is usually, andthereafter always, infected. The sterility in theearly stages is probably due to the gastric acidity.As the acid is neutralised after perforation, sothe bacteria are able to multiply. The prognosisis more serious in infected cases. Trout (I935)advocated a study of the bacterial content of themouth, teeth and throat, pointing out that thebacteria in the stomach will correspond to thosein the-mouth. This bears out the clinical observa-tion that the prognosis of a perforation in a patientwith dental sepsis is grave.
Treatment.The surgical treatment of a perforation of a
peptic ulcer is only an incident, though a majorincident, in a full course of medical treatment.During the last six years the following routine
has been developed in the treatment of perforationsat the Southend General Hospital:
(i) Pre-operative.As soon as the diagnosis has been made morphia
gr. i/4th, or in robust individuals gr. 1/3rd isgiven. A Ryle's tube is then passed through thenose to lie exactly in the lowest point of thestomach. The patient is turned and made to lieon his left side, so that any further escape throughthe perforation is only gaseous. The stomach isthen emptied by suction through the Ryle's tube.If solid food is present which cannot be aspiratedthrough a Ryle's tube a small stomach tube,size I2 or I4 E, replaces it. An intravenous dripinfusion is then started, most conveniently intoa vein in the middle of the left forearm, the patientlying on his left side. The fluid administered atfirst is plasma, as being near in composition to thefluid lost into the peritoneal cavitv. It is continuedlater with normal saline and 5 per cent glucose, orby I /5 N. Saline with 4 - 3 per cent glucose accordingto the estinated needs of the patient. The stomach
is kept empty by frequent aspirations, the mouthis cleaned, and, with the drip running, the patientis given 45 minutes' rest. If further morphia isthought necessary, small doses are injected withoutdisturbance into the drip near the vein.Although the average case is not suffering from
true surgical shock, after such treatment hepresents a very different picture. In place ofagonised anxiety he shows a drowsy, reassuredindifference. A considerable rise of blood pressurehas on occasion been observed, but, in all cases,the mouth is clean, the stomach empty, theblood chemistry partially righted and the mentalcondition changed beyond recognition.
(2) OperativeThe patient is brought to the theatre with his
eyes covered, the Ryle's tube in his stomach andthe intravenous drip running. Throughout theoperation the stomach is aspirated at short inter-vals by the anaesthetist.
Local anaesthesia, consisting of an upperabdominal field block with i per cent novocain(without adrenalin), has been the anaestheticof choice in this series. In exceptionally muscularor nervous cases this has been supplemented writhcyclopropane.The incision is made through the inner margin
of the right rectus, centred one half inch to theright of the mid point between the umbilicus andthe sterno-xiphoid junction, that is, over thesurface marking of the pvlorus, and is of theshortest convenient length. An average is 3Ainches. Escape of gas is looked for. Fluid isremoved from the peritoneal cavity by suctionand the liver margin is raised with a Deaver'sretractor in the right upper part of the incision,held by the assistant's right hand. The pylorusnow lies exposed in the centre of the wound, thestomach being empty and flaccid. A gauze stripis placed on the pylorus and, with the indexand middle fingers of the left hand on this gauze,the pylorus is retracted to the patient's left.The perforation is thereby brought into view.Without moving the left hand three sutures are
passed through the whole thickness of the duodenalwall, one above, one through and one below theperforation. A suitable piece of omentum isbrought up and laid within the sutures which arethen tied, thus firmly closing the perforation. Bothcatgut and fine silk have been used for these sutures.
Fluid is further removed by suction from theaccessible parts of the abdomen which is closedwithout further manipulation of any kind. Inthe earlier part of this series closure was effectedby buried interrupted sutures of No. 8 plaited silkincluding all layers except the skin. The results
left room for improvement; wound infection andseparation of the skin edges were not uncommon.A number of stitch sinuses resulted which weretroublesome but were easily dealt with by theimmediate removal, under local anaesthesia, ofthe offending sutures. Other methods have beenused including strong silk, linen thread and wirethrough all layers including the skin, as in therepair of a burst abdomen. These wounds arestrong, heal well, but produce unsightly scars.
Drainage of the abdominal cavity has not beenemployed in any case of less than I2 hours dura-tion. Drainage of the wound has been carriedout in an increasing number of cases in view of thefrequent occurrence of fluid collections. It seemsa rational procedure. After closure the wound iscovered with an Elastoplast dressing which sealsit completely. No bandage or binder is used inorder to be certain that chest expansion is notimpeded.
(3) Post-operative.The patient returns to bed with the Ryle's tube
still in his stomach and the intravenous driprunning. Being conscious, after only local anaes-thesia, he is placed immediately in Fowler'sposition and at hourly intervals he is made tomove his arms and legs, to breathe deeply and,with his hand firmly supporting his wound, tocough. Within 48 hours he is expected to sitforward without assistance and to move down thebed and back for his pillows to be arranged.Morphia is given in small but repeated dosesinto the intravenous drip (gr. i/8th.-i/6th). Inthe early stages in order to lessen the pain caused,the patient does his exercises 30 minutes after aninjection of morphia. His mouth is cleaned every2 hours, but for 48 hours he is allowed to swallowonlv sips of clear fluid. These are immediatelyaspirated through the Ryle's tube by which meansthe stomach is kept empty and at rest.
Aspirations are continued until the fluid obtainedfrom the stomach is clear and auscultation of theabdomen shows that bowel sounds have returned.This occurs usually in 36-48 hours, when the Ryle'stube is removed. During this time the patientreceives 5 pints of fluid intravenously each 24hours. For the most part this consists of I/5 N.saline with 4`3 per cent glucose: to this are addedvitamins B and C in full doses, chemotherapy,and at least i pint of blood or plasma. Morerecently a sterile solution of amino acids in theform of a casein hydrolysate solution has beenadded to supply, at least in part,.the patient'sprotein requirements.
Physical signs of poor aeration of the bases ofthe lungs has been a constant finding. If signs
of collapse of an area of lung develop the patientis put on the "stir-up regime" (A.M.A. Subcom.on Anaesth. I942). He is made to move hisarms and legs, to breathe deeply and to coughevery I5 minutes. Firm percussion over theaffected area has been tried. If the signs ofcollapse are not considerablv improved in 6 hours,suction-bronchoscopy is performed.
Chemotherapy. In the early part of this seriesno chemotherapy was used. Later, intramuscularsulphapyridine was given to those cases whichshowed evidence of respiratorv infection. Recentlysoluble sulphathiazole has been given routinelyintravenously (2 grammes, followed by i grammefour-hourly) the course being completed by mouthafter the infusion has been discontinued. Peni-cillin has not until now been available.
Diet. After 48 hours, bowel sounds being presentand stomach aspirations clear, two-hourly feedsare started together with alkalies, antispasmodicsand phenobarbitone in sufficient dose to keep thepatient restful. Alternatively, if the patient hastolerated his nasal tube well, it is reinserted intothe stomach and a milk drip is statted. This formsthe start of a full course of medical treatment.It is needless to say that smoking is strictly pro-hibited.Owing to bed shortage patients in this series
have been allowed home in 2-3 weeks to completetheir treatment at home.
DiscussionAlthough patients after perforation are, as
Moynihan said, in a state of prostration ratherthan of true surgical shock, there can be no ques-tion that I-2 hours spent in pre-operative prepara-tion effects a great improvement in their condition.Provided that no further leakage is permitted intothe peritoneal cavity, the writer is of the opinionthat this time is time well spent. Leakage isprevented by aspiration of the stomach contentand by turning the patient on to his left side. Bythis means any remaining stomach content liesto the left, in contact with the greater curvature,and cannot escape through a perforation on thelesser curvature, at the pylorus or in the duodenum.Furthermore, at operation, the stomach is emptyand flaccid.
Local anaesthesia is preferred because by itsmeans the cough reflex is never lost and the patient,being conscious throughout, is able to be proppedup at once on return to bed, to move, to breathedeeply and to cough. If, on account of restless-ness, failure of relaxation or the presence of agastric perforation away from the pylorus, localanaesthesia is unsatisfactory it is supplementedby general anaesthesia. Cyclopropane is the
anaesthetic of choice owing to smoothness ofinduction, rapid return of consciousness and absenceof postoperative vomiting. It has the furthergreat advantage that within a few hours thepatient has lost all feeling of having been anaes-thetised.
In this series the chances have been 8 to X thatthe perforation would be in the first part of the duo-denum. The incision has accordingly been madethrough the inner margin of the right rectus muscle,centred over the pylorus. In most cases, a lengthof 31 inches has proved ample. The incision isplaced to the right of the midline for these reasons,(a) it is placed directly over the pylorus, (b) a fewfibres of the rectus muscle left on the inner side ofthe incision add strength to the closure, and (c)the resulting scar does not impede subsequentsurgery should this be necessary. If difficulty isencountered the incision is at once prolonged.The method of closure of the perforation de-
scribed is the simplest and causes the least possibletrauma. No exteriorisation of the stomach orpylorus, and no further exploration of the abdomenhas been permitted. The object of the operationhas been solely to close the perforation. For thisreason* neither gastro-enterostomy nor partialgastrectomy has been performed. It is impossibleto cause stenosis by the application of an omentalplug.
Practical, rather than theoretical evidence hasyet to be produced that closure by an omentalplug is more likely to be followed by persistenceof the ulcer than other methods. In either casethe ulcer heals by granulation, and so by scartissue. Admittedly this may be slightly less inamount. if the size of the ulcer is diminished bysuture. At the same time, sutures introducetension and so more fibrosis.
Strong silk and through and through sutureswere used in closure of the wounds in order to givethe strength particularly needed for early move-ment, breathing exercises and coughing. Themethod used prevents efficient wound drainage.The penalty has been some increase of woundinfection.
In the post-operative treatment morphia hasbeen used freely, but in small doses. Used judi-ciously, the patient being stimulated to move,breathe and cough every hour, the writer holdsthe view that morphia can help to prevent ratherthan to promote chest complications. The im-mediate institution of movements and exercisesis of the first importance in the prevention ofchest complications.
Rest. The cause of the condition having been,as far as possible, removed at operation by theclosure of the perforation, the acutely inflamedstomach is kept at complete rest until it shows a
return of function. This is achieved by continuoussuction or by repeated aspirations through theRyle's tube which is maintained until (a) thefluid aspirated from the stomach is clear, and (b)auscultation shows that intestinal peristalsis hasrecommenced. Forty-eight hours is an average time.Even without fluids by mouth considerable quan-tities of fluid mav be aspirated from the stomach,having reached it by regurgitation from theduodenum or by being poured out from theinflamed stomach wall. Fluids given by mouthduring this time only tend to increase this accu-mulation, to stimulate an inflamed organ andpossibly to produce vomiting which, at all costs,must be avoided.
Fluids. For-48 hours or more under this regimethe patient's gastrointestinal tract is kept entirelyat rest. During this time therefore he must relyentirely for his intake on his intravenous infusion.As far as is possible this should supply all hisrequiremrients. Fluids are given according to thepatient's needs, 5 pints being an average dailyallowance, but this may be increased if the patientis dehydrated. Plasma is given initially, butthereafter the standard fluid used in this serieshas been I/5 N. Saline with 4 3 per cent glucosewhich supplies the patient's water, salt andglucose requirements (Naunton Morgan and AveryJones, I938). To this are added full doses of theshort term vitamins B and C and, as alreadymentioned, chemotherapy. The protein require-ments have been in part met by the inclusionof i pint of plasma or better, of whole bloodeach 24 hours. Recently the protein requirementshave been more fully supplied by the additionof amino acids in the form of a sterile solution ofcasein hydrolysate (Gaunt, I943). Much researchis being carried out into the fascinating field ofprotein requirements in health and disease.For instance, Croft and Peters (I945) have shownthat methionine is particularly in demand fortissue repair. Such work may bear fruit in thenear future in the production of a fluid synthesizedto contain all the requirements for metabolism,which may be of the greatest value in helping thepatient in his return to health.
Chemotherapy. The two chief causes of deathin this condition are (i) peritonitis and (2) lungcomplications (57-2 per cent and 20 8 per centrespectively of 952 deaths. de Bakey, I940).Both are bacterial in origin. There is good reason,therefore, for the routine use of chemotherapy.In this series the drug of choice has been sulpha-thiazole, the soluble form being used in the intra-venous drip, the course being completed by mouthafter the drip has been discontinued. Penicillinmay be valuable. From present knowledge, acombination of penicillin and sulphathiazole may
well prove to be a very potent weapon againstinfection.
Medical Treatment. It cannot be over-empha-sized that the surgical treatment of a perforatedpeptic ulcer is only a prelude to a full course ofmedical treatment. Early in the convalescence,therefore, a physician is consulted with regard tomedical treatment which consists essentially of agradually increasing diet of frequent milk feeds,together with alkaline powders, antispasmodicsand sedatives in sufficient dose to make the patientrestful. Phenobarbitone is extremely useful inthis respect. WVhen the'patient is convalescenthe is transferred to the care of the physician whoundertakes the further treatment. After 3 monthsthe patient returns for a routine X-ray exami-nalion, subsequent treatment depending on itsfindings. In I,525 cases treated by closure withan omental plug collected from the literatureby de Bakey (1940) more than half remainedsymptom-free; the remainder developed subsequentulcer symptoms, and I7 per cent of the totalrequired further surgical treatment. From thesefigures there is therefore a more than 50 per centchance of cure following simple closure of aperforation. Strict medical treatment will improvethis chance.
SummaryThe key to success in the treatment of perforation
of peptic ulcers still lies in the hands of the generalpractitioner in the shape of early diagnosis.
Surgical treatment is the prelude to a full courseof medical treatment.One to two hours are well spent in pre-operative
preparation.Local anaesthesia supplemented if necessary by
cyclopropane, is the anaesthetic of choice.The operative procedure should be the minimum
necessary to close the perforation securely.Drainage of the peritoneal cavity is rarely neces-
sary in cases of less than I2 hours duration.Chemotherapy should be used routinely.Early movements and exercise are of the greatest
importance in the prevention of chest complications.Morphia, judiciously used, can prevent rather
than promote chest complications.Over 50 per cent of cases of perforation 'subse-
quently remain symptom-free. For those caseswhich subsequently develop symptoms, furthertreatment should be decided upon after dueconsideration, not at the time of perforation.
3. CarcinomaSince early last century it has been recognised
that carcinoma of the stomach may be associated
with a chronic gastric ulcer. At necropsy ofpatients dving from carcinoma of the stomachunmistakable signs have from time to time beenfound of the presence of a chronic gastric ulcerwhich have be-en confirmed by a long history ofulcer symptoms. The growth has usually been tooadvanced to state that it arose from the ulcerbut appearances have been suggestive. With thedevelopment of gastric surgery specimens wereobtained showing carcinomata arising obviouslyin the edges of chronic ulcers, a picture morerecently seen in vivo by means of the gastroscope.Histologists have gone further by showing thepresence of carcinoma in the edge of ulcers whichto all macroscopic appearance were benign.
In I909, MacCarty aroused widespread discussionby his statement that 68 per cent of the gastriculcers resected at the Mayo Clinic were associatedwith carcinoma, and that 7I per cent of the resectedcarcinomata of the stomach :were associated withchronic gastric ulcers. Moynihan (I926) lent someweight to these views by his statement that 2/3rdsof his cases of carcinoma of the stomach gave along history of dyspepsia. In the following yearsother views were expressed. Morley (I923), forinstance, was frankly critical of the Mayo Clinicfigures. In his series of 50 gastric ulcers thoughtmacroscopically to be benign, histological exami-nation showed 5 to be definitely malignant anda further 6 to be "possibly undergoing malignanttransformation," making a total of 22 per centcertainly or possibly malignant. Cabot and Adie'(I925) in a severely critical article showedthe enormous variation in the published estimatesof ulcer-cancer incidence and stated that manyfigures were entirely valueless owing to laxity inpathological diagnosis. In their own series of56 cases diagnosed as simple ulcers, 5 proved to becancerous (9 per cent). Stewart (I925) reportingwith meticulous histological criteria on 2I6 opera-tion specimens, found 134 simple ulcers, 68 carcino-mata and I4 ulcer-cancers. He concluded that9 5 per cent of the resected ulcers had thereforebecome malignant. Klein (1938) was completelysceptical and on most critical pathological groundsstated that ulcer cancer is rare, the main difficultvbeing the establishment of the existence of a pre-existing ulcer. More recently Kirklin and Mac-Carty Jnr. (1942) state that between io and I2 percent of all gastric ulcers radiologically and macro-scopically benign are in fact malignant. Finsterer(I939) states that I5 2 per cent of 487 apparentlybenign gastric ulcers resected by him proved to bemalignant. Allen and Welch (94I) state that of69 ulcers resected by them as benign, 30 (43 percent) proved histologically to be malignant.Maingot (I942) found that in an operative seriesof 50 apparently benign ulcers in women, ii
236 POST-GRADUATE MEDICAL JOURNAL July, 1945proved to be undergoing malignant change (22per cent).These figures concern only cases treated surgi-
cally. They are likely to comprise therefore onlvthe more severe, chronic ulcers which have resistedmedical treatment. Such operative series shouldby no means be confused with total series whichcontain all ulcers referred for treatment. In thislatter group, Allen and Welch (I94I) report thatof 277 cases originally diagnosed as gastric ulcer,39 (14 per cent) finally proved to have cancer,whilst Judd and Priestley (I943) report on thesubsequent history of I46 cases of gastric ulcertreated, apparently successfully, medically andfollowed up for at least 5 years. Of these I46cases:
46 5 per cent were cured15 *8 per cent well on medical regime.4 8 per cent ulcer shown to persist.
Ii O per cent underwent operation forgastric ulcer.
9-6 per cent developed carcinoma ofstomach.
o 7 per cent died from haemorrhage.ii 6 per cent died from unrelated causes.
These figures are of considerable value. Theyshow the subsequent course of a series of cases ofgastric ulcer treated medically at the Mayo Clinicin which a follow-up study was possible, namelythat 46 per cent were cured, that a further I5per cent remained well on diet whilst 9-6 per centdeveloped carcinoma of the stomach. It can beobjected that a number of these carcinomatamay have been unconnected with an ulcer. Thismay have been go, but the figures still show theincidence of carcinoma of the stomach in a seriesof cases of gastric ulcer. To sum up, therefore,(i) There is clear evidence that an ulcer whichclinically, radiologically and even macroscopicallyappears benign, may in fact be malignant. Theaverage percentage error in diagnosis is at leastIO per cent.
(2) To prove the actual development of carci-noma in a previously benign ulcer of the stomachis impossible. It would involve microscopy ofthe complete original ulcer in order to exclude thepossibility of malignancy from the start. Circum-stantial evidence is, however, very strong. Thedevelopment of carcinoma in 9*6 per cent ofJudd and Priestley's series of 146 cases treatedoriginally medically with apparent healing isnoteworthy.
Therapeutic Test: In view of these facts atherapeutic test has been employed in the treat-ment of gastric ulcers. The diagnosis beingestablished by clinical, radiological and gastro-
scopical evidence the patient is given a 6 weeks'course of strict medical treatment. The investi-gations are then repeated. If the ulcer is healedit is considered benign. If improved but not yethealed, a further 6 weeks' treatment is prescribedand the investigations are repeated a third time.If still unhealed, the ulcer is probably malignantand surgery is indicated. If the ulcer is healed,the patient is allowed up but must continue hismedical regime and must understand that recur-rence of the ulcer is possible and may be serious.He should report at once any return of symptomsand should return for further examination at3 and later 6-monthlv intervals. These examina-tions are physical, radiological and gastroscopical.Herein lies one of the most useful spheres ofgasfroscopy. By its means a recurrence of theulcer or an irregular or nodular appearance at thesite of the scar, indicating an early carcinoTna,may be seen before either symptoms develop orradiological signs become positive. If surgicalintervention is delayed until anorexia, wasting anda palpable mass are present, the chances of cureare tragically small.
In all such cases therefore it is of the utmostimportance that routine examination be carriedout at regular intervals. Any recurrence of theulcer indicates (i) that it is malignant or (2) thateven if the ulcer is benign, further medical treat-ment is unlikely to be successful. Recurrence istherefore an indication for surgery.
This therapeutic test combined with medicaltreatment is, however, by no means infallible.From the writer's experience it is often difficultto persuade a patient of middle or advancing yearsto attend for routine gastroscopy. He may feelwell and be disinclined to see the importance of therepetition of an, at least, unpleasant experience.The writer has seen four such cases, apparentlycured by medical treatment, return later withinoperable carcinomata of the stomach. Again,improvement under medical regime by no meansrules out the possibility of malignancy, as Walters(I942) confirms. Furthermore, a carcinoma maydevelop rapidly between the routine examinations,whilst by no means all patients are' capable ofadhering to a regime which includes dieting andrestraint from tobacco and alcohol. Lastly Juddand Priestley (I943) in a careful follow-up surveyshow that the chances of cure by medical.treat-ment are less than 50 per cent.
Consideration.-Under a medical regime there is,in the first place, the danger that the lesion, thoughapparently benign,- may in fact be malignant.Errors of diagnosis even in the best hands are notuncommon and amount to some IO per cent of all
copyright. on M
arch 26, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.21.237.228 on 1 July 1945. D
cases. Secondly carcinoma may develop subse-quently. In addition there are risks from hae-morrhage, perforation, and stenosis, together withonly a 50 per cent chance of cure.
In surgical clinics in which the mortality forpartial gastrectomy for gastric ulcer is between3 and 5 per cent other considerations obtain.The risk of ulcer cancer is here greater thanthat of operation. The combined risks ofhaemorrhage, perforation, and cancer are muchgreater than those of operation in otherwisehealthy cases. The treatment of chronic gastriculcer in such cases should therefore, it is held, besurgical. Exceptions are made in the case of(i) patients under 45 years of age, in whom therisk of cancer and of other serious complications issn,ll, and (2) patients of advancing years or ofinfirmity from other causes in whom the risks ofoperation are great. In these groups medicaltreatment is advised but careful follow-up exa-minations must be carried out.Summary.-Evidence from a number of sources
shows that, in the diagnosis of chronic gastriculcer, with our present methods of examinationeven in the best hands there is an error of IO percent. At least I in IO of such ulcers is in factmalignant. Strong circumstantial evidence existsthat others will become malignant, especially inpatients over the age of 45. The total risk fromthese two sources is without doubt more than IOper cent and may well prove to be double thatfigure.Where the mortality of gastrectomy is 20 per
cent, the risk of ulcer cancer is of little significancein the choice between medical or surgical treat-ment. Where the mortality of gastrectomy isbetween 3 and 5 per cent the risk of ulcer cancerbecomes of very considerable significance. Underthese circumstances the combined risks of hae-morrhage, perforation, and cancer in patients ofmiddle life materially exceed, in otherwise healthyindividuals, the risks of gastrectomy. For suchpatients the treatment of choice is surgical, andonly by means of surgery can the death rate fromulcer cancer be reduced.
In patients under the age of 45, the risks ofcancer as well as those of other serious complica-tions are small. Medical treatment may thereforebe advised but careful follow-up examinations areessential. Recurrence is an indication foroperation.
In patients of advancing years and in those ofpoor health from other causes the risks of surgeryare greater than those of complications andtreatment should again be medical.
4. The Complications of Operation forGastric and Duodenal Ulcer
For convenience of discussion the complicationsof operation for gastric and duodenal ulcer areconsidered in three groups, (i) immediate, oc-curring at or within 24 hours of operation, (2) inter-mediate, occurring during the remainder of thepatient's hospital stay, and (3) remote, occurringsubsequently.
Immediate Complicationsi. Collapse under Anaesthesia.-During the
course of anaesthesia collapse may occur fromshock, as- in gross primary haemorrhage, fromdeficiency of oxygen which may be due either toan obstructed airway or to an actual failure of theoxygen supply, or from anaesthetic abnormalities.Shock is not a prominent feature in gastric surgeryand with good technique haemorrhage should beminimal. If either is anticipated, an intravenousinfusion should be started before operation, andthrough it sufficient suitable fluid should be givento anticipate the onset of shock. Any blood lossis immediately replaced. At the conclusion ofanaesthesia danger arises from the sudden drop inoxygen tension of the gases respired when the maskis removed. Sudden collapse from anoxia mayfollow. It is wise to bring the patient's B.L.B.mask and oxygen cylinder to the theatre and tocontinue the administration of oxygen forthwith.During the period of recovery from anaesthesia
there is a risk of aspiration of vomitus if anyfluid is allowed to collect in the stomach. Theuse of cyclopropane for anaesthesia greatly shortensthe period of recovery and almost rules out post-anaesthetic vomiting. The removal of all fluidfrom the stomach by means of frequent aspirationthrough an indwelling gastric suction tube removesthe risk.
2. Haemorrhage.-Severe primary haemorrhageshould not occur in non-urgent gastric surgery.It is prevented by careful ligation of vessels. Anyblood lost during operation lies either in Ruther-ford-Morison's pouch or under the left side of thediaphragm. An ooze, for instance, from the leftpart of the divided gastrocolic omentum collectsnear the spleen. If such collections are not care-fully sought and removed at the close of operationthey are likely to cause at least post-operativepyrexia and malaise, if not frank subphrenicinfection. If infections and adhesions are to beavoided, haemostasis must be complete.
Reactionary haemorrhage is possible. If clampsare used, their release before the insertion of theanterior sutures of the anastomosis diminishes this
risk. Anastomosis yithout clamps, however, com-bined with under-running the vessels in the gastricsubmucosa after dividing the seromuscular coatsdoes everything that is possible to eliminate it.By this method any haemorrhage is immediatelyseen and is immediately controlled. That it issuccessful is shown by the fact that stomachaspirations after I2 hours are almost invariablyfree from blood.
3. Peritonitis.-This may result either fromsoiling at the time of operation or from subsequentleakage from the anastomosis. For the first, theperfection of the aseptic method of anastomosisas advocated by O'Hara (I900), Wangensteen(I942), and by Pannett (I945) may entirely ruleout operative soiling. If this is not attainable,proper preparation of the stomach, repeated aspira-tion during operation by means of the indwellinggastric suction tube, and good technique serve tominimise the chances of infection. Leakage fromthe anastomosis results either from faulty suturingor from stretching of the anastomosis by sub-sequent ileus. The use of three layers of suturesfor the anastomosis, the outer layer consisting ofinterrupted sutures, is a very considerable safe-guard at the time of operation.Acute dilatation of the stomach and upper
intestine which was a fruitful cause of tension onthe anastomosis and so of leakage, is now almosteliminated by means of the indwelling gastricsuction tube. This, a Ryle's tube, is passedthrough the nose two hours before operation andis then used to give the stomach a final wash out.It remains in situ during operation when it enablesthe anaesthetist to keep the stomach empty, andafter operation when it serves to keep the stomachat rest, to prevent distension and to provideevidence of the stomach's return to function. Itis removed only when aspirations are clear, whenfluids taken by mouth are evidently passing onfrom the stomach and when auscultation showsthe return of bowel sounds, in fact usually after36-48 hours.
After a gastro-enterostomy in a case of pyloricobstruction from a duodenal ulcer in a man of 68,seventeen pints of fluid were -aspirated from thestomach by means of an indwelling suction tubein 3 days. The question of the nature of thefluids to be given intravenously to replace thisloss was an interesting one. The patient re-covered.
4. Vomiting and Acute Intestinal Obstruction.-With the use of continuous gastric aspiration,vomiting after gastric operations has become rare.It may be troublesome after gastro-enterostomy.It is largely prevented by a well-planned stoma.If it occurs, it is best treated by keeping the stomachempty and clean and by supplying the patient's
fluid, glucose and vitamin requirements intra-venously. Acute upper intestinal obstruction mayoccur at any time after operation. Its onset isdramatic. The patient collapses, his pulse rateis rapid, his eyes become sunken. The differentialdiagnosis from haemorrhage or general peritonitismay be impossible. Likely causes are a retrogradeintussusception of the jejunum through the stoma,obstruction of a loop of jejunum passing throughthe aperture in the transverse mesocolon or rota-tion of the ascending or descending loops in ananterior anastomosis.
In a recent case after a difficult., partial gas-trectomy with an anterior Polya anastomosis fora large chronic gastric ulcer, the patient's progressfor. 2o hours was good. He then suddenly col-lapsed. His face was blanched, his eyes sunken,his pulse rate i6o; he complained of pain in theepigastrium. The abdomen revealed only someepigastric tenderness. He died four hours later.Autopsy showed that the proximal loop, from theduodeno-jejunal flexure to the anastomosis, hadfallen back and to the left behind the distal loop,and was acutely obstructed behind the mesenteryof the jejunum forming the anastomosis. It thusformed a closed loop obstruction. This could havebeen prevented by the use of the Y-shaped suctiontube suggested by Wangensteen (1942), or bypassing the tube at operation beyond the anasto-mosis into the proximal loop and subsequentlyusing a second tube to aspirate the stomach;
5. Chest Complications.-These are includedadvisedly in the immediate group. It is whilstthe cough reflex and the respiratory movementsare impaired that the foundations of almost allchest complications are laid.
Before operation it is important to rule out pre-existing chest disease and to eliminate septic focifrom the teeth. Throughout anaesthesia the air-way must be unimpeded; this is best achieved bymeans of an intratracheal tube. The effects ofanaesthesia should preferably be quickly dispelledso that the patient may sit up soon after his returnfrom the theatre and the aeration of the bases ofhis lungs may be thereby improved. There shouldbe no post-operative vomiting. in these tworespects cyclopropane is invaluable.The operation itself (i) must be sound and should
not give rise to a severe peritoneal inflammationwhich will impair the movement of the diaphragm,and (2) the wound must be well and stronglysutured to allow early movement and exercises.Mimpriss (I944) emphasises these points.
After operation as soon as the patient becomesrestless morphia is given freely but in moderatedoses, usually gr. ith at once followed by gr. *th,repeated when necessary. Thirty minutes afterhis injection the patient is made to move his knees
GASTRIC AND DUODENAL ULCERup and down as in cycling, to do deep breathingexercises'and, with his hands firmly supportinghis wound, to cough. At 2-hourly intervals he isstimulated to move, to breathe and to cough upany sputum which may be present. Within24 hours' of operation he is expected to be able tosit forward by himself. This immediate movementregime is carried out largely by the nursing staffand involves a certain amount of bullying. If theexercises are carried out at first half an hour afterinjections of morphia, however, the pain is re-lieved. After doing his exercises and clearinghis bronchi the patient sinks back into a soundand refreshing sleep.
In the opinion of the writer morphia used in thisway prevents, rather than promotes chest com-plications.
In the immediate post-operative period signs ofpoor air entry to the bases of the lungs are tobe expected. Signs of an area of collapse mustbe carefully sought. If they appear, the full"stir-up" regime (A.M.A. Subcom. on Anaesthesia,I942) is employed, the patient being made tomove, breathe, and cough every I5 minutes. Ifthe signs of collapse do not rapidly improve,suction-bronchoscopy is employed.
Provided that the lungs were previously healthythe response to this routine is satisfactory. Inheavy smokers and in cases of chronic bronchitisthe patient has difficulty and pain in coughing upthick, purulent sputum and sinks back exhaustedafter the effort. In such cases it is better to spend3 weeks in preoperative treatment for the chestcondition, than to spend 3 months coughing up alung abscess and developing an incisional hernia.
6. Venous Thrombosis.-Here, again, the foun-dations of trouble are laid during the period ofimmobilisation. To prevent thrombosis, there-fore, as soon as the patient is conscious he is madeto move his arms and legs and, as soon as possible,the is expected to help himself up and down thebed whilst his pillows are rearranged. No knee-pillow is permitted because
(i) it impedes the venous return in the poplitealveins during sleep, and
(ii) it prevents full movement of the legs duringexercise.
Admittedly the patient tends to slip down fromthe full Fowler position whilst asleep, but this iscounteracted by nursing care at least 2-hourly,and the exercise involved in returning to thecorrect position is by no means harmful.
Intermediate Complicationsi. Chest Complications.-In this group, chest
complications are of first importance. Mimpriss
(1944) gives an excellent account of the compli-cations he met in IOO consecutive cases of gas-trectomy. He divides them into (i) bronchitis,(2) lobar atelectasis, and' (3) lobular atelectasis.Bronchitis is the commonest post-operative chest.complication and is usually an exacerbation of apre-existing condition. It may cause the patientmuch pain in coughing but, by all possible means,he should be encouraged to cough up any sputumwhich may be present. Lobar atelectasis is rare,but its presence should be carefully sought at leasttwice daily. If it develops it is treated at onceby the "stir-up " regime previously described and,if necessary, by suction-bronchoscopy. Lobularatelectasis, as Brock (I936) pointed out, is acommon post-operative development, the fre-quency of its diagnosis depending on whether thesurgeon is looking for it. Mimpriss found it in29 of his IOO cases, but shows that in 25 it provedto be mild and transitory. In the remaining 4cases it progressed to more serious conditions,bronchopneumonia in three and empyema in one,but three of these cases were associated withserious infection below the diaphragm. In onlyone case uncomplicated by abdominal infectiondid the lobular atelectasis proceed to a fatalbronchopneumonia.
If signs of infection or atelectas'is develop,chemotherapy is instituted immediately. Sul-phathiazole, in the writer's experience, has beenthe drug of choice and is given in full dosesintravenously or, later, by mouth. It has entirelyaltered the prognosis of infective chest com-plications.
2. The Wound.-Wound infections result fiomlack of care at operation in allowing spillage of gas-tric content, in imperfect protection of the woundedges with impermeable packs or in incompletehaemostasis. Serous collections in the wound aremore common with catgut but are less seriousthan with silk, which may cause a sinus until theoffending stitch is removed. Provided that thewound is well and truly sutured a burst abdomenis a rare complication of operations for pepticulcer. If it occurs it is the result of coughingcombined with wound infection. Where the woundis covered with an- occlusive dressing the first signof separation of the wound edges is an out-pouringof fluid which soaks through the dressing. Itssignificance may not be immediately appreciated.Treatment 'consists of immediate resuture ofthe wound with strong silk or wire through alllayers including the skin. Such wounds heal wellbut leave unsightly scars with "cross-hatching."Strong arguments can be advanced for the primaryclosure of the wound by this method in all casesof carcinema, of chronic bronchitis and in elderly,or for other reasons frail patients.
POST-GRADUATE MEDICAL JOURNAL3. Subphrenic Abscess.-This is a rare compli-
cation. It may arise from (i) subsequent infectionof blood collections not removed at operation,(2) spillage of content from a badly preparedstomach, (3) the base of an ulcer left in thepancreas or liver, (4) injury to the common bileduct, (5) leakage from an imperfect anastomosisor (6) bursting of the duodenal stump. A duodenalfistula may follow. Such fistulae may heal spon-taneously; continuous suction is of the greatestvalue in treatment.
4. Diarrhoea.-This is a common sequel ofgastric operations and may cause the patient muchinconvenience. It commonly starts 5-7 daysafter operation and lasts 3 days or more. If itwere due to absence of acid from the stomach,the administration of hydrochloric acid by mouthshould be effective. If purely bacterial in origin,it might be susceptible to sulphaguanidine orsulphasuxidine. All three are useless. It appearsmore likely that the diarrhoea is due to an un-accustomed irritation of the small bowel, eitherphysical or chemical, by food not previouslypartially digested in the stomach. This wouldexplain the fact that the best treatment at presentappears to be a temporary limitation of dietcombined with kaolin and a soothing mixture.Predigested-foods may prove to be of value.
5. Thrombophlebitis.-With movement and sti-mulation of the patient from the earliest possiblemovement after operation, this complication israre. When it has occurred it has resulted fromintravenous therapy and has been caused byimperfect asepsis either at the time of insertionof the needle or cannula, or in subsequent care.If it occurs Heparin, or possibly Dicoumarin,should be used to prevent further thrombosis,the blood coagulation time being carefully checked.
6.. The Anastomosis.-In any non-urgent opera-tion for a gastric or duodenal ulcer the stomachshould be well prepared and its walls should behealthy. There is therefore no excuse for leakage.The greatest safeguard in this respect is a three-layer anastomosis, the outer layer consisting ofinterrupted, preferably mattress, non-absorbable,seromuscular sutures which take the weight of thejejunum, thus preventing any tension on theinner catgut layers. Great care is taken at theupper and lower angles of the stoma to [see thatthe inner sutures are complete and reach wellabove and below the incisions in the jejunum andstomach.A well-planned anastomosis should function
satisfactorily as soon as the stomach and jejunumreturn to activity. Occasionally after gastrectomythe patient complains of discomfort after food inthe left upper quadrant of the abdomen. Thisusually subsides within IO days and has been
ascribed to jejunitis. X-ray confirmation of thisdiagnosis may be obtained in the shape of "feather-ing" of the upper jejunum.
Remote ComplicationsI. Anastomotic Ulcer.-This complication may
follow any operation in which gastric and jejunalmucosa become continuous. Its incidence ishighest in operations in which the bile is divertedfrom the anastomosis, as in the Roux-in-Y opera-tion, now abandoned. This has been confirmedin animal experiments in which the bile has beendiverted to the ileum, in which case ulceration islikely to occur throughout the jejunum.
After gastrojejunostomy the incidence of anasto-motic ulceration has been estimated varyingly upto 32 per cent (Marshall, I942), occurring partic-ularly in young men having a high gastric acidity.After partial gastrectomy, it was found to be notuncommon after operations in which the mucousmembrane of the pylorus and of the pyloric antrumwere not removed (Ogilvie, 1938; Kiefer, I942).It may occur after more complete gastrectomiesbut its incidence is rare. If it does occur it isprobably an indication that an insufficient amountof stomach has been removed, leaving sufficientacid-producing area to produce further ulceration.An anastomotic ulcer is situated on the line of
the anastomosis or immediately beyond it in thedistal loop of the jejunum. It has all the charac-teristics of a gastric or duodenal ulcer, is likely tobe associated with inflammation and spasm of theanastomosis and is liable to haemorrhage, to per-foration and, by adhesion to the transverse colon,to form a gastro-jejuno-colic fistula.
Clinically, if anastomatic ulceration is to occur,it does so usually within a year of the primaryanastomosis. It gives rise to typical symptoms,namely, pain to the left of the umbilicus, severeand gnawing in character, not necessarily relatedto food, but making the patient afraid to eat andcausing considerable loss of weight. The pain maymake life a burden. Occult blood is constantlypresent in the stools. Frank haemorrhage mayoccur at any time. If perforation occurs, theanastomosis being posterior, the escaping fluid liesbeneath the shelf of the transverse mesocolon andpasses down to the left iliac fossa, where tender-ness and rigidity are most marked. The upperabdomen is protected by the mesocolon and maybe relatively little involved.
In the presence of a chronic stomal ulcer, theoverlying transverse colon is liable to be involvedat first in adhesions, later in the actual ulcerativeprocess. The results of the formation of a gastro-jejuno-colic fistula are serious. The patient deve-lops a persistent diarrhoea with the passage of
GASTRIC AND DUODENAL ULCERundigested food in the stools, he has an unbearablehalitosis, loses weight rapidly, and may developnutritional oedema. 'There are few states morepitiable. Barium taken by mouth may be shownto pass immediately into the colon, whilst aBarium enema may at once outline the stomach.The latter is often better seen.
Treatment.-The medical treatment of anasto-motic ulceration has proved disappointing. Im-provement and relief of symptoms may be obtainedbut even with complete- adherence to a medicalregime, symptoms are likely to recur and thepatient to become a semi-invalid.- He is unableto work and his life revolves round his diet, hissleep, and his bowels.
Surgical treatment, on the other hand, offers agood chance of cure. The operation of choice issub-total gastrectomy, removing a large proportionof the acid-bearing area of the stomach, andclosing the jejunal stoma with or without resection.If the gastro-jejunostomy has been done withouta loop, the separation of the proximal portion ofjejunum may be the most difficult part of theoperation. If a loop is present, separation of theanastomosis may be much simpler. Great caremust be taken not to damage the middle colicvessels. The jejunal opening may be closed mostsimply by repairing it transversely, in ratherlonger time by the method of Pauchet (I934) or,if necessary, by resection and end-to-end anasto-mosis. The jejunum distal to this closure is usedfor the subsequent anastomosis which in view ofthe necessarily wide resection of stomach is bestof the anterior Polya type. With good anaesthesiaand transfusion facilities the mortality of thisoperation in skilled hands is low. Marshall (I944)quotes 62 cases with 3 deaths, representing amortality rate of 4.8 per cent.Haemorrhage from an anastomotic ulcer is
treated as from a gastric or duodenal ulcer. Theoutlook of recurrent haemorrhages is poor; resectionis advisable.The serious results of a gastro-jejuno-colic fistula
are not due to the loss of gastric content into thecolon, but are due to persistent gastro-enteritiscaused by faecal contamination of the stomach.The whole bowel is in a constant state of acuteinflammation. This is proved by the strikingimprovement effected by a defunctioning colostomy.At a preliminary operation the hepatic flexure ismobilised and brought out as a Paul-Mikuliczcolostomy through a right-sided oblique or trans-verse incision. If the area of the fistula can beinspected with retraction, the whole area is seento be grossly inflamed and oedematous. For aweek or two after opening this colostomy the stillinflamed intestine pours forth fluid faeces, and theskin around the colostomy may become red unless
this possibility is carefully foreseen. The patientand his attendants may lose heart and feel that thecolostomy was useless. With daily washouts,however, both from the colostomv and from therectum, the distal loop of colon gradually becomesclean and contains only a small amount of fluidwhich has passed through the fistula. No furtherfaecal contamination of the stomach occurs andgradually the inflammation of the whole intestinesubsides, colostomy actions become fewer, andthe patient's condition improves. He loses hishallitosis and oedema, his appetite becomesexcellent, he puts on weight and feels vastlyimproved. It should be noted that the improve-ment after colostomy is not immediate.
Six to eight weeks after the preliminary colo-stomy, the second operation is performed. Anintravenous infusion is maintained throughout theoperation. The inflammation noted at the timeof colostomy has subsided. The first problem isthe separation of the colon. Only a small areaof it is usually involved and can be dealt withsimply by inversion. The bowel being defunc-tioned, little risk arises from this source. Theremainder of the operation is as for an uncom-plicated gastro-jejunal ulcer.Ten days after the gastrectomy the spur of the
colostomy is crushed, restoring the continuity ofthe colon, and three weeks later the colostomy isclosed extraperitoneally, with careful inversion ofits margins. Delayed primary suture has beenfound of much value in closing these wounds.
Objections to this routine of treatment are,firstly that it involves at least 3 months in hospital,secondly that it involves the unpleasantness of acolostomy, and thirdly that it requires three stageoperations. These are, however, the price thepatient has to pay for safety in the shape of agreatly reduced mortality, as compared to thatof the one-stage operation. There is no moregrateful patient than a man who has been cured ofa gastro-jejuno-colic fistula.
2. Post-gastrectomy Dyspepsias.-The conditionof patients after gastrectomy for gastric andduodenal ulcer has been the subject of much dis-cussion. It is generally agreed that, for gastriculcer, the results are almost uniformly satisfactory.For duodenal ulcer, after the Polya type of anasto-mosis occasional cases occur in which the patientcomplains of a feeling of distension soon after foodand even retching and vomiting may occur. ABarium meal in these cases usually shows that theanastomosis is lying transversely and that a con-siderable part of the meal has passed into the proxi-mal blind loop of the duodenum. Such patients oftenfind that after a meal they have to lie down for atime, preferably on the left side, after which the dis-tension is relieved and they feel well again. The
242 POST-GRADUATE MEDICAL JOURNAI, July, r945discomfort may make the patient afraid to eat.This symptom tends'to diminish with time as thepatient learns to adjust his diet. It may beavoided by placing the stoma obliquely, proximaljejunum to lesser curvature, with the proximalloop brought up to the upper part. of the lessercurvature well above the anastomosis. Food isthereby diverted into the distal loop of the jejunumand so, without obstruction, into 2I feet of smallintestine. The effect may be increased by closingthe upper half of the cut surface of the stomach,which acts as a valve directing the food distally(Hofmeister, I905). The results of this operationhave proved satisfactory, the patients being sub-sequently well and able to eat, drink, and smokeas they please, provided that the resection hasbeen sufficiently radical in removing the acid-bearing portion of the stomach.
3. Post-Gastrectomy Anaemia.-Since the recog-nition of the fact that, provided the resection hasbeen sufficiently radical, subsequent dieting isunnecessary, anaemias, both micro- and macro-scopic, have been almost unknown. Occasionallyadministration of iron for a few weeks after opera-tion may help the patient to restore his haemo-globin, but its continued administration is seldomnecessary. The writer has yet to see a case ofmacrocytic an-aemia following gastrectomy.
4. Incisional Hernia.-The great cause of in-cisional hernia is wound infection. With goodtechnique this should be rare. With the evolutionof aseptic anastomosis it may be eliminated. Inorder to be able to move the patient immediatelyafter operation and to make him do breathing andcoughing exercises, a strong closure' is essential.Buried interrupted silk sutures through all layersexcept the skin give the strongest possible closure.With this method of closure one can with con-fidence, assure the patient that he will'not damagehis wound by coughing. Silk is, however, liableto cause sinuses which persist until the silk isremoved. Catgut is not liable to sinus formation,but does not give such a strong closure.
SummaryIn undergoing any operation for gastric or
duodenal ulcer, the patient runs the risk of anumber of complications. These may prove fatalor they may cause merely a hitch in the smoothprogress of his convalescence. In either case withproper knowledge they should-be largely avoidable.
In the prevention of chest complications which
are the patient's greatest enemy, four factors areof prime importance:
(i) Careful preoperative preparation with specialcare to rule out pre-existing chest disease.
(2) Good anaesthesia, allowing rapid return ofconsciousness and not being followed byvomiting.
(4) The "stir-up" regime, which aims at rapidand complete re-expansion of the lungs. Itfurther has the advantage of restoring thevenous circulation and so of reducing theincidence of venous thrombosis. Lastly theindwelling gastric suction tube has eli-minated the danger of tension -on the sutureline caused by upper intestinal ileus.
In this field, above all, prevention is betterthan cure.
REFERENCkSI.HAEMORRHAGE
AITKEN, R. S. (I934), Lancet, 1, 839.ANDRESEN, A. F. R. (I939), Am. J. Digest. Dis. 6,641.BARBOUR, H. G., and HAMILTON, W. F. (1926), Jour. Biol. Chem., 69,
625.BLACKFORD, J. M., and ALLEN, A. (1942), JA.M.A., 120, 8ii.BULMER, E. (1932), Lancet, 2, 720.CHRISTIANSEN, T. (1935), Acta Med. Scand., 84, 374.CROHN, B. B., and LERNER, H. H. (I939), Am. J. Digest Dis., 6, I5.CULLINAN, E. R., and PRICE, R. K. (I932), St. Barts. Hosp. Rep., 65,
I85.FINSTERER, H. (I939), Surg., Gyn. Obst., 69, 291.FROSTAD, S. (1934), Norsk. Mag. Laegevidensk, 95, 578.GOLDMAN, L. (I938), Am. J. Surg., 40, 545.GORDON-TAYLOR, G. (1935), Lgncet, 2, 8 ii.HINTON, J. W. (I939), Ann. Surg., 110, 376.HURST, A. F., and RYLE, J. A. (I937), Lancet, 1, I.JANKELSON, I. R., and SIEGEL, M. A. (1938), New Eng. J. Med.
219, 3.JONES, F. A. (1939), B.M.J., 1, 915.LINEBERRY, E. D., and ISSOS, D. N. (I937), South Med. Jour., 30,
I228.MARRIOTT, H. L., and KEKWICK, A. (I935), Lancet, 1, 977.MEULENGRACHT, E. (I934), Act. Med. Scand., 59, 375.MEULENGRACHT, E. (I937), Munch. Med. Wchschrft., 40, 1565.MOON, H. V., et al. (I941), J.A.M.A., 117, 2024.MOSSBERG, 0. (1933), Hygeia., 95, 898.RANKIN, F. W., et al. (I939), South. Surgeon., S, 298.ROSS, K. (1930), M. J. Australia, 1, i68.SCOTT, L. D. W. (I940), Edinb. Med. Jour., 47, 49.SCUDDER, J. (J. B. Lippincott, 1940), Shock: Blood Studies as a Guide
to Therapy.THORSTAD, M. J. (1942), Surgery, 12, 964.UMBER, F. (I935), Deutsch. Med. Wchnschrft., 61, 1265.WITTS, L. J. (I937), B.M.J., 1, 847.WOLDMAN, E. E. (I941), Am. J. Digest. Dis., 8, 39.
2. PERFORATION
A.M.A. SUB-COMMITTEE ON ANAESTHESIA (Am. Med. Ass. Press,Chicago. 1942), Fundamentals of Anaesthesia, 149.
BAGER, B. (1929), Acta Chir. Scand., Suppl. II, 64, 5.BLACKFORD,. J. M. (I942), J.A.M.A., 120, 825.BOSSUET. Oraisons Funebris (Hachette. Boulevard, St. Germain, Paris).BRAUN, H. (I897), Zentralblf. Chir. 24, 739.
copyright. on M
arch 26, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.21.237.228 on 1 July 1945. D
July, I945 GASTRIC AND DUODENAL ULCER 243BRINTON, W. (I856), Brit. and For. Med.-Chir. Rev., 17, 159.CROFT, P. B., and PETERS, Prof. R. A. (March 3rd, I945), Lancet, 266.DE BAKEY, M. (1940), Surgery, 8, 852 and 1028.GAUNT (I943-44), Nutrition Abstracts and Reviews, 13, 50I.GRAVES, A. (I933), Ann. Surg., 98, 197.HALL, W. W. 4I892), Brit. Med. Jour., 1, 64.JUDINE, S. (I939), J. internat. de Chir., 4, 219.KELLY, M. W. (I939), Surgery, 6, 524.KRIEGE, H. (I892), Berlin Klin. Wchnschr;, 29, 1244.MIKULICZ, J. (I897), Zentralbl. f. Chir., 24, 69.MORGAN, C. N., and AVERY JONES, F. (I939), Lancet, 2, 6ii.MOYNIHAN, SIR BERKELEY (W. B. Saunders Co. I926), Abdominal
Operations.SOUTTER, L. (I94I), Surgery, 10, 233.STERN, M. A., et al. (1929), Lancet, 49, 492.TROUT, H. H. (1935), J.A.M.A., 104, 6.TURNER, E. W. BEDFORD (March 31st, 1945), Brit. Med. Jour., 457.VON HABERER, H. (I9I9), Wien, Klin. Wchnschr., 32, 413.WALTON, SIR JAMES (Arnold & Co., London. 1930), Textbook of the
Surgical Dyspapsias, 64.
3. CARCINOMA
ALLEN, A. W., and WELCH, C. E. (1941), Ann. Surg., 114, 498.CABOT, HUGH and ADIE, G. C. (1925), Ann. Surg., 82, 86.FINSTERER, Prof. H. (I939), Proc. Roy. Soc. Med., 32, I83.JUDD, E. S., and PRIESTLEY, J. T. (1943), Surg. Gyn. Obst., 77, 21.KIRKLIN and MAcCARTY (I942), Jour. A.M.A., 210, 733.KLEIN, S. H. (I938), Arch. Surg., 37, I55.MACCARTY, W. C., and WILSON, L. B. (I909), Am. Jour. Med. Sci.,
138, 846.MACCARTY, W. C. (I9I0), Surg. Gyn. Obst., 10, 449.MAINGOT, RODNEY (1942), Postgrad. Med. Jour., 18, 93.MORLEY, JOHN (I923), Lancet, 2, 823.MOYNIHAN, SIR BERKELEY (W. B. Saunrders Co. 1926), Abdominal
Operaticns.STEWART, Prof. M. J. (I925), Brit. Med. Jour., 2, 882.WALTERS, W. (I942), Arch. Surg., 44, 520.
4. THE COMPLICATIONS OF OPERATION FOR GASTRIC ANDDUODENAL ULCER
A.M.A. SUBCOM. ON ANAES. (A.M.A. Press, Chicago. 1944) ,Funda-mentals of Anaesthesia, I65.BROCK, R. C. (I936), Guy's Hosp. Rep., 86, I9I.HOFMEISTER. Quoted by Spivak (1936) in The Surgical Technic of
A bdominal Operations, 431-436.KIEFER, E. D. (1942), Jour. A.M.A., 120, 8I9.MARSHALL, S. F. (i942), Surg. Clin. North Am., 24,624.MIMPRISS, T. W., and ETHERIDGE, F. G. (Oct., I944), Brit. Med.
Jour., 466.OGILVIE, W. H. (1938), Lancet, 2, 295.O'HARA, J. W. (I900), Am. Jour. Obst. (St. Louis), 42, 8i.PANNETT, C. A. (I945), Brit. Jour. Surg., 32, 4I8.PAUCHET. (I934), Pratique chir., 19, 132.WANGENSTEEN, Prof. 0. H. (Chas. Thomas, Springfield, Illinois
I942), Intestinal Obstructions, 232.
NEW PUBLICATIONSJ. B. LIPPINCOTT COMPANY will shortly be publishing
the following:LIPPINCOTT'S QUICK REFERENCE BOOK FOR MEDICINE ANDSURGERY: A Clinical, Diagnostic, and TherapeuticDigest of General Medicine, Surgery, and the Specialitis,Compiled Systematically from Modern Literature, bvGeorge E. Rehberger, A.B., M.D. 12th edition revised.Approximately I,400 pages, 40o illustrations, 50 colouredplates, price 4 IOS. net.The new edition of this all-inclusive encyclopedia is
completely up-to-date. The coloured illustrations havebeen almost entirely changed, and all phases of treatment,
particularly chemotherapy, have been brought right upto the minute and the most recent diagnostic aids areincluded. The book covers general medicine and surgery;gynaecology, obstetrics, diseases of the male genito-urinary system; the skin, eye, ear, nose and throat; andorthopaedics.
THE ART OF ANAESTHESIA, by Paluel J. Flagg, M.D.,519 pages. I66 illustrations. 7th edition revised.Price 36s. net.
The changes in this new edition of this widespread workhave been widespread. A few of them concern the latestviews relative to the barbiturates, avertine, paraldehyde.Table positions have been revised and the dangers of theprone position noted. Special reference is made to anaes-thesia for brain and thoracic surgery. The usefulnessof helium is stressed. New methods of cardiac stimula-tion are reported and recent advances in oral surgery arediscussed.
SURGICAL ERRORS AND SAFEGUARDS, by Dr. Max Thorek.io85 pages. 4th edition. 794 illustrations, manycoloured. Price £4 ios. net.
The many pitfalls to which the surgeon is heir are ablydescribed and discussed in this unique work. In additionto the many changes and additions in the part of thisbook devoted solely to surgery an entirely new chapteron " Legal Responsibility for Surgical Practice." has beenadded, written by Dr. Hubert W. Smith.
J. & A. CHURCHILL LTD.
Books published May 1945
A SHORT PRACTICE OF MIDWIFERY FOR NURSES. 13thedition. By Henry Jellett, M.D., F.R.C.P.I. SizeCrown 8vo. 468 pages. Price about Ios.
Short description: This new edition of a favouritestandard handbook for nurses has been brought thor-oughly up-to-date.
REPRINTS
THE RADIOLOGY OF BONES AND JOINTS. By J. F. Brails-ford, M.D., F.R.C.P. 3rd Edition. 45s.
TRAINING FOR CHILDBIRTH; From the Mother's Point ofView. By Minnie Randell, S.R.N., T.M.M.G. 3rdEdition. ios. 6d.
FUNDAMENTALS OF PSYCHIATRY
By EDWARD A. STRECKER, M.D., Sc.D., F.A.C.P., MedicalPublications Ltd: London. I944. Price I2S. 6d.
This compact vade mecum will be of great interest andvalue to all psychotherapists, students and practitionersalike. In rather under two hundred pages Dr. Streckerhas covered practically the whole field of modernpsychiatric medicine, tabulating his subject under appro-priate headings and sub-headings. Excellent use hasbeen made of different kinds of type which has contributeda great deal to the clarity and readability of the book,and an excellent glossary of psychiatric terms is appended.The author states in his preface that his aim has been
to supply "essential data organised in systematic andusable fashion " In this he has succeeded and in so doinghas done a service which will be appreciated in these days
copyright. on M
arch 26, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.21.237.228 on 1 July 1945. D
