TRAUMATIC DIAPHRAGMATIC HERNIA - Thoraxthorax.bmj.com/content/thoraxjnl/16/2/99.full.pdf · Thorax(1961), 16, 99. TRAUMATIC DIAPHRAGMATIC HERNIA BY W. R. PROBERT AND C. HAVARD From

Thorax (1961), 16, 99.

TRAUMATIC DIAPHRAGMATIC HERNIABY

W. R. PROBERT AND C. HAVARDFrom Sully Hospital, Glamorgan

(RECEIVED FOR PUBLICATION FEBRUARY, 1961)

Rupture of the diaphragm and herniation ofthe abdominal viscera into the chest are well-recognized results of crushing or penetratinginjuries.Gordon Bryan's study (1921) of both types of

injury is a surgical classic. The second worldwar rekindled interest in traumatic diaphragmatichernia, and several important papers dealt withthose due to penetrating injury (Edwards, 1943;Mackey and Bingham, 1945; Morgan, 1945).In peacetime reports from the United States(Hughes, Kay, Meade, Hudson, and Johnson,1948; Bugden, Chu, and Delmonico, 1955;Desforges, Strieder, Lynch, and Madoff, 1957),from Australia (Sutherland, 1958), and fromGreat Britain (Evans and Simpson, 1950) havedrawn attention to the frequency of traumaticdiaphragmatic herniae caused by motor accidentswhere crushing against the steering-wheel was thecommon mechanism.

In hospitals serving mining and industrial areastraumatic diaphragmatic herniae are more likelyto result from falls of coal and rock, or fromcrushing by trucks and pit cages. They areespecially serious because the respiratory functionof the injured men is often already impaired bypneumoconiosis and emphysema.We report 15 cases of traumatic diaphragmatic

hernia from a thoracic surgical centre in SouthWales. Fourteen were in men whose ages rangedfrom 16 to 67 years (average 48 years): in fourthe herniae resulted from penetrating wounds, in10 from crush injuries. These 10 were coal-miners, all but one of whom had been injured at

TABLE ITRAUMATIC DIAPHRAGMATIC HERNIA

Total number.15Right sided .. .. .. .. 3Left 12Due to crushing injury.

, ,, penetrating. 4Operations performed.10Mortality 1(10%)

work. The only case seen in a woman was dueto a crushing injury in an air-raid. Nine of the11 herniae due to crushing, and three of the fourherniae due to penetrating wounds, were left-sided.

CASE REPORTSCASE 1.-B. M., a coal-miner aged 52 years, had

been buried under a fall of rock in 1932 while atwork in a kneeling position. He sustained fracturesof the third, fourth, and fifth left ribs and a compoundfracture of the right leg. He coughed a little bloodsoon after his injury. Sixteen years later he com-plained of " wind and rumbling " in the abdomen buthad no pain. Chest radiographs (Fig. 1) and bariumenema showed the colon to be in the left side ofthe chest, and a diagnostic pneumoperitoneum pro-duced a pneumothorax.At left thoracotomy on November 12, 1948, there

was a 9 in. tear in the diaphragm extending from thecostal margin towards the oesophageal hiatus. Aloop of colon had herniated but there was no sac.The gut was reduced and the tear repaired withinterrupted thread sutures. Post-operatively the rightlower lobe collapsed, but recovery was satisfactory.Four years later he was free from symptoms andradiography showed a sound diaphragm (Fig. 2). InFebruary, 1961, he reported that he was well andworking.CASE 2.-W. R., aged 40 years, was wounded in the

left side of the chest in 1943 and nine accessiblemetal fragments were removed soon afterwards. Sixweeks later he developed intestinal obstructionrequiring laparotomy and a colostomy which wasmaintained for three months. In 1950 a left diaphrag-matic hernia was recognized and repair was attemptedby the abdominal route. His attacks of abdominalpain and vomiting were not relieved thereby, andcontinued till his admission to Sully Hospital in 1956.Peristalsis could be heard at the base of the left chest,and radiography showed the hernia clearly (Fig. 3).At left thoracotomy on February 17, 1956, thestomach, transverse colon, and omentum were exten-sively adherent to the lung and the margin of thediaphragmatic tear. The abdominal viscera were

reduced and the defect repaired in two layers. Hislater progress was satisfactory. In February, 1961,he reported that he was well and working as a builder.

on 20 April 2018 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.16.2.99 on 1 June 1961. D

ownloaded from

http://thorax.bmj.com/

W. R. PROBERT and C. HAVARD

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.-. ........:

FIG. 2.-Case 1: Postero-anterior radiograph of chest showing stateof affairs after repair of the hernia.

CASE 3.-W. H., a coal-miner aged 44 years, wasgradually crushed beneath a slowly descending pitcage and the bottom of the shaft on December 9,1956. He was released a few minutes later andimmediately complained of pain in the chest, difficultyin breathing, and dimness of vision. He was admittedto hospital soon afterwards. On examination he wasgrossly dyspnoeic; bowel sounds were heard overthe left side of the chest. He showed subconjunctivalhaemorrhages and petechiae over the upper half ofthe trunk. Lateral chest radiography, which wasdifficult owing to his restlessness, showed evidence ofbowel herniation anteriorly (Fig. 4). It was not tillthree days later that he was judged to be fit for leftthoracotomy. Lengths of small bowel and part of thetransverse colon and omentum entered the chesttrugh a large transverse tear in the front of the

diaphragm. On its costal aspect only a narrow rimof peripheral muscle remained. There were severalother tears, but they did not involve the full thicknessof the diaphragm, the peritoneum remaining intact ineach. The lateral portion of the main tear could onlybe closed by passing sutures round the eighth rib.The medial portion, where the defect extended a littleway under the intact pericardium, could only berepaired in one layer. Three weeks later he was ingood general condition, the petechiae had disappeared,and his vision had returned to normal. In February,1961, he was complaining of effort dyspnoea, and hadnot worked since the accident. Radiology showedthe diaphragmatic repair to be sound.CASE 4.-D. A. S., aged 16 years, sustained an

accidental wounid of the left chest with a .410 shot-FiG. la andb.-F-Case1: PosteroGanteriorand left lateralraraographsgun while out shooting in the country on Decemberof chest showing herniation ofcolon through left hemidiaphragm. 27, 1949. He could hear a gurgling sound from his

100



j.com/


ownloaded from


TRAUMATIC DIAPHRAGMA TIC HERNIA

FIG. 4.-Case 3: Left lateral radiograph showing poorly definedpattern of small bowel in the chest.

Fic.. 3h

FIG. 3a and b.-Case 2: Postero-anterior and left lateral chestradiographs after barium swallow showing part of stomach inthe left diaphragmatic hernia.

chest, but managed to run the length of two fields forhelp, after which he had a brisk haemoptysis. Hewas taken to Swansea General Hospital, where leftanterior thoracotomy was performed. The left lungwas found to be lacerated and there was a 4-in. tearin the cupola of the diaphragm. Much blood was

aspirated, the diaphragm was sutured, and the chestclosed. Laparotomy was then performed, a rupturedspleen removed, a gastric perforation sutured, and a

tear of the left lobe of the liver controlled. Post-operative chest aspirations were needed, followed bya period of postural treatment in a right-sided plastercast before re-expansion of the left lung was achieved.Within eight months he was playing professional foot-ball.

In November, 1958, he was working as a painterand decorator, he had no dyspnoea even aftervigorous exercise, and had no intestinal symptoms.His chest movements were excellent, all wounds were

soundly healed, and radiography showed the leftdiaphragm to be intact but a little high.CASE 5.-A. W. M., a coal-miner aged 67 years, was

admitted as an emergency with haematemesis and sub-sternal pain. Ever since being crushed by a coal trucksix months before, when fractures of the left ninth,tenth, and eleventh ribs had been seen, he had com-

plained of "indigestion pain" behind the sternum.He continued to vomit altered blood for some timeafter admission. There was dullness to percussion atthe base of the left chest. Chest radiography and a

barium meal examination showed the greater part of

FIG. 3a

101



j.com/


ownloaded from



the stomach in the left side of the chest. At leftthoracotomy on April 14, 1955, un-united rib fractureswere seen. The part of the stomach which hadherniated was congested but its vessels pulsated. Thiswas returned to the abdomen, and the defect, whichlay anteriorly, was sutured. The patient made a goodrecovery. In February, 1961, he reported that he waswell and free from dyspepsia.

CASE 6.-C. W., aged 47 years, had been woundedat Dunkirk in 1940, and a metal fragment had beenremoved from the left chest wall at an early operation.For 15 years he had suffered from fleeting pains inthe abdomen and chest unrelated to food. Oninvestigation in June, 1955, there were no abnormalphysical signs. Chest radiography showed an opacityabove the left diaphragm; barium meal examinationwas not helpful, but the presence of a diaphragmatichernia was confirmed following induction of apneumoperitoneum (Fig. 5). Operation was refused.In 1956 he needed admission to hospital forpneumonia, and in September, 1960, reported that hehad effort dyspnoea, that he had lost weight, andthat he had frequent bouts of indigestion.

CASE 7. J. C. M., a coal-miner aged 33 years, hadsuffered for four years from pain in the right side ofthe chest, relieved by lying down. Ten years beforehe had been crushed by a fall of coal at work, andremained in hospital for 12 months with a fractureof the skull and a fracture of the lumbar spine withparaplegia. He was admitted to Sully Hospital onDecember 13, 1950, following an attack of dyspnoeaand right-sided chest pain. Chest radiography andbarium studies showed a loop of colon reaching up tothe apex of the right pleural cavity (Fig. 6). Hisacute symptoms subsided and he refused furtherinvestigation and treatment. In November, 1958, hereported that he was working as a productioninspector in an engineering factory, but that hesuffered from effort dyspnoea and from frequentbouts of dyspepsia.

CASE 8.-D. T., a coal-miner aged 55 years, wascrushed beneath a slowly descending pit cage. Histrunk was acutely flexed, and later he was found tohave a fracture of the base of the skull, fracture dis-location of the third and fourth lumbar vertebraewith paraplegia, and diastasis of the pubic symphysis.His cerebral condition improved by the next day, buthe became cyanosed and dyspnoeic. The media-stinum was displaced to the right, the left side of thechest was dull to percussion, and breath sounds wereabsent. There were no bowel sounds in the chest.There was a marked difference between the chestradiograph taken on the day of injury (Fig. 7) andthat of the following day (Fig. 8). Because of hiscondition films could not be taken in the erect position,and it was impossible to distinguish between visceralherniation through the diaphragm and haemothorax.At left thoracotomy on November 18, 1955, a large

anterior diaphragmatic tear was found. Most of thestomach had herniated together with some of the

small bowel, transverse colon, and omentum. Thestomach was tightly constricted at the defect but wasviable; the herniated portion was tensely distendedwith much fluid and a little gas. The viscera werereturned to the abdomen and the defect sutured. Hemade a good recovery from the thoracotomy and was

FIG. 5a and b.-Case 6: Postero-anterior and left lateral radiographsof chest after induction of artificial pneumoperitoneum. Someair has entered the hernia, but there is no pneumothorax. Inthe lateral film gastric rugae can be seen passing up into thehernia. The spleen is well shown.

102

......



j.com/


ownloaded from



FIG. 6a

FIG. 7.-Case 8: Postero-anterior radiograph of chest on day ofinjury showing abnormal left diaphragm and a little mediastinaldisplacement to the right.

FIG. 6.-Case 7: a, Postero-anterior radiograph of chest showinglarge size of traumatic hernia of right hemidiaphragm. b, Rightlateral radiograph of chest during " follow-through" of bariummeal examination. Contrast agent has only partially filled theherniated portion of colon which extends to the apex of the chest.

FIG. 8.-Case 8: Antero-posterior radiograph of chest taken on dayfollowing injury with the patient recumbent. There is a littleclarity in the middle of the left-sided opacity. This was seen atoperation to be due to the stomach being distended with fluidand gas. No fluid level was demonstrated because the patient'scondition precluded taking an erect film.

103



j.com/


ownloaded from


FIG. 9a and b.-Case 9: Postero-anterior and left lateral radiographsof chest showing traumatic hernia of left hemidiaphragm.Barium meal showed that the herniated viscus was part of thestomach. The right-sided rib fractures and cross-union are wellshown.

soon able to co-operate in the treatment of his otherserious injuries. In January, 1961, he was walkingwith the aid of a calliper. He reported that he hadno dyspepsia.CASE 9.-W. H., a coal-miner aged 58 years, had

been crushed in 1944 between the pit roof and a truckon which he was travelling. He became very ill,needed transfusion, and recovery was slow. Fouryears later he began to become increasingly short ofbreath. Ten years after the accident he began tosuffer from epigastric pain after food, relieved bylying down, and 16 months later consulted his doctor.His chest movements were poor and he became shortof breath after walking 50 yards on the flat. Radio-graphy showed part of the stomach in a left-sidedhernia (Fig. 9). At left thoracotomy on March 9,1957, a large central defect was found, without a sac.It extended back from the phrenic nerve, with itsmedial margin near the attachment of the peri-cardium, nearly to the oesophageal hiatus; laterallyit extended into the muscular part of the diaphragm.The herniated parts of the stomach, transverse colon,omentum, and left lobe of the liver were returned tothe abdomen. The defect could not be completelyclosed by suture and was reinforced with a free graftof parietal tissue as described below. He appeared tomake a good immediate recovery, but died suddenlyon the nineteenth post-operative day. At necropsythe diaphragmatic repair was shown to be sound.The lungs were emphysematous and there was hypo-static pneumonia of the lower lobe of each lung. Thesuddenness of his death was unexplained.CASE 10.-T. B., a coal-miner aged 63 years, had a

haemoptysis five months before admission. He hadbeen wounded in the right side of the chest in 1917and had had an operation on the chest wall. Radio-graphy showed an anterior opacity above the rightdiaphragm; fluoroscopy showed poor movements.Bronchoscopy was normal; bronchography showedmild right lower lobe bronchiectasis.At right thoracotomy on July 28, 1954, the lung

was universally adherent. There was an anteriordefect in the diaphragm, three inches in its longestdiameter, forming the margins of a firm fibrous sac.Part of the right lobe of the liver projected into thesac but was not adherent to it. The sac was excisedand the defect sutured. He made a good recovery.In April, 1957, he was well; radiologically thediaphragmatic repair was sound and the lung fieldsclear. He died at home in December, 1957, after anillness in which he lost weight and had a copioushaematemesis.CASE 11.-D. D., a coal-miner aged 53 years, was

injured in a motor accident in 1951, when he wastreated for fractured right ribs and a fractured pelvis.From that time he complained of dyspnoea, and wassent for investigation following mass miniature radio-graphy at work. There was a rounded opacity abovethe right diaphragm (Fig. 10). Barium studies werenormal. Induction of a pneumoperitoneum resultedin a small pneumothorax. At right thoracotomy on



j.com/


ownloaded from


FIG. Ia

F .IcOa

tf

FIG. llb

FIG. 10b

FIG. 10.-Case 12: a, Right lateral radiograph of chest showingherniation of part of the liver through the right hemidiaphragm.b, Postero-anterior radiograph of chest after induction ofpneumoperitoneum. A shallow pneumothorax could be seenin the original film. There is a healed fracture of the rightfifth rib.

FIG. 1 la and b.-Case 13: Postero-anterior and left lateral radio-graphs showing distended stomach, small intestine, and colonin the left side of the chest. The left lung is compressed and apneumothorax is present.



j.com/


ownloaded from


W. R. PROBERI' and C. HAVARD

October 13, 1954, a "mushroom" of liver projectedthrough a defect in the anterior part of thediaphragm; there was no sac. The defect wasenlarged a little to allow the liver to be reduced, andwas sutured in two layers. Recovery was uneventful.In February, 1961, he reported that he was well andworking as a boilerman on a housing estate.

CASE 12.-O. D., a coal-miner aged 46 years, hadslipped from his support while operating a drill atshoulder height and had fallen heavily backwardsagainst a pile of coal while at work in 1938. Twohours later he felt faint for a few minutes, but wasable to return to work the next day. A few dayslater he became ill with bronchitis and remained athome for nine weeks. Thereafter he was in asanatorium for 14 weeks complaining of cough, painin the left chest, and night sweats. Sputum examina-tion was negative for tubercle bacilli. In the succeed-ing years he suffered from time to time from pain inthe left side of the chest and shortness of breath.Sixteen years after injury radiography showed a highleft diaphragm which moved paradoxically. Inductionof a pneumoperitoneum was not helpful, butdiaphragmatic herniation was diagnosed withoutcontrast radiography.At left thoracotomy on January 26, 1955, the lung

was adherent to the diaphragm, and there was ahernia with a sac. Its neck was wide and its anteriormargin was formed by a prominent ridge of musclenear the ribs. It contained the spleen and most ofthe stomach. Most of the sac was excised and thedefect sutured with overlapping. Histological exami-nation showed the sac to be entirely fibrous. Thoughbronchospirometry before and after operation showedlittle difference, the patient noticed great improvementin his breathing. In February, 1961, he was well andworking, and the diaphragmatic repair appearedsound radiologically.CASE 13.-C. L., a coal-miner aged 51 years, had

been injured at work 17 years previously when a fallof rock fractured the left tibia and fibula and tworibs on the left side. He was treated in hospital fortwo weeks, and was off work for six months.

For five years after the accident he suffered fromattacks of colicky substernal and epigastric pain last-ing about 10 minutes and occurring about once amonth. He remained well for the next 11 years; butsix months before admission to hospital he developedright lower abdominal pain, lassitude, and loss ofappetite. Two weeks before admission he experiencedpain in the left upper chest with an unproductivecough; one week before admission he suddenlybecame short of breath and collapsed while out walk-ing, and remained in bed at home with fever. Onexamination the chest moved poorly, the trachea wasdisplaced to the right, the left chest was hyper-resonant and breath sounds were absent on this side.Bowel sounds could be heard all over the left chest.There was a palpable lump in the right iliac fossa.There was intermittent fever up to 102° F. and thehaemoglobin was 64%. Chest radiography (Fig. 11)

showed the stomach, together with colon and smallbowel, occupying the left side of the chest. Therewas a small left-sided pneumothorax, and the lateralfilm showed a little gas under the left diaphragm.

Diagnostic aspiration at the base of the left chestwas negative. He was given two tranfusions of packedred cells. The fever was not influenced by antibiotics.Air was aspirated from the left pleura and later anapical drain inserted. Three weeks after admissionhis condition deteriorated rapidly: he complained ofabdominal pain, and was tender and rigid over thewhole abdomen. He developed peripheral circulatoryfailure and died.At necropsy there was a defect 6 in. in diameter

in the dome of the left hemidiaphragm. There wasno hernial sac; the oesophago-gastric junction wasnormally placed, but most of the stomach lay in theleft pleural cavity together with much of the trans-verse colon and small bowel. There were noadhesions, but the viscera were covered with a smallamount of fibrinous exudate. There was a carcinomaof the caecum with gross invasion of mesenteric andpara-aortic lymph nodes. There was an acute ulcerof the anterior wall of the first part of the duodenumwhich had perforated causing generalized peritonitis.

It was concluded that the abdominal symptoms hehad experienced in the last six months of his life weredue to the caecal carcinoma. The acute episode threeweeks before admission was attributed to leakagefrom the duodenal ulcer; the pneumothorax wasexplained by the passage of gas into the chest throughthe diaphragmatic defect, and the fever by the peri-toneal and pleural exudates.

CASE 14. S. J., a housewife aged 56 years, hadbeen treated in hospital after an air-raid 17 yearspreviously when the house in which she was livingcollapsed and she had been buried up to her shouldersin debris. Her vision was impaired for three weeks,but she was able to go home free from symptomsat the end of that time. Though four monthspregnant at the time she later had a normal delivery,though her child only survived three weeks.

After this she continued to live a normal life, givingbirth to healthy children in 1944 and 1947. Shehad noticed increasing shortness of breath in the sixyears before her admission to hospital. She wasadmitted to hospital following an acute episode inwhich she experienced numbness of the back of thehead and arms and weakness of both arms; the attackrapidly subsided, leaving no residual signs. She saidthat she became short of breath when climbing stairsor walking fast, and that she could not lie on herright side because of difficulty in breathing. Takinglarge meals or copious drinks made her short ofbreath, but she had no dyspepsia. She was an obesewoman; the blood pressure was 180/80 mm. Hg;at the base of the left chest there was dullness topercussion and poor air entry; bowel sounds couldbe heard all over the left side of the chest. Chestradiography (Fig. 12) showed a hernia through theleft hemidiaphragm with much of the stomach, small

106



j.com/


ownloaded from



F.G. 12c

FIG. 12.-Case 14: a and b, Postero-anterior and left lateral radio-graphs of chest showing herniation ofstomach and colon throughthe left hemidiaphragm. There is spinal osteoarthritis, but norib fractures were seen. c, The herniated colon shown bycontrast radiography.

bowel, and colon in the chest. These findings wereconfirmed by contrast radiography.

This patient is now being treated for obesity, and isawaiting operation for the hernia.

CASE 15.-H. J., a coal-miner aged 57 years, hadbeen crushed between two coal trucks 16 years pre-viously and was admitted to hospital where his chestwas strapped. After this accident he suffered fromsubsternal .pain after meals, and 12 years beforeadmission, while being investigated elsewhere bybarium meal, a left-sided diaphragmatic hernia wasdemonstrated. Ten years before admission hedeveloped a persistent and productive cough; twoyears before admission he had a disabling attack ofleft-sided chest pain, and from that time onwardscomplained of occasional dysphagia, regurgitation,and substernal pain related to effort.On admission he had a moderate kyphoscoliosis

and his chest moved poorly. His maximum voluntaryventilation was 32 litres/minute. Chest radiography(Fig. 13) confirmed the presence of a hernia throughthe left hemidiaphragm; a barium meal examina-tion did not reveal any gastric herniation ordemonstrate oesophageal reflux. Oesophagoscopyshowed no abnormality. Electrocardiographyindicated ischaemic heart disease.The herniation was thought to involve the liver or

spleen only, and as his symptoms were mainly thoseof cardiac ischaemia and oesophageal reflux, andbecause of his poor cardiac and respiratory function,operation was not advised.

Fic;. I2 i

FIG. 121'

107



j.com/


ownloaded from



He was treated with antacids with much improve-ment, but reported in February, 1961, that he wasstill troubled by dyspnoea and pain in the left side ofthe chest.

DIAGNOSISThe diagnosis of traumatic diaphragmatic

hernia is rarely obvious, and in practice is madeeither soon after injury or following a longinterval during which symptoms are slight orabsent.EARLY DIAGNOSIS.-In penetrating injuries

diaphragmatic injury frequently follows a smallexternal wound. Diaphragmatic rupture is always

TABLE IIHERNIAE DUE TO PENETRATING INJURY

Total number.4 (all men)Right sided 1Left 3Hernia-diagnosedearly. 1, , late 3

Average delay period in herniae diagnosed late-19 yearsShortest- 6Longest-37

Average age at time of treatment -41

TABLE IIIHERNIAE DUE TO CRUSHING INJURY

Total number.11 (10 men, I woman)Right sided 2Left 9Hernia diagnosed early 2

,,l ,, late 9

Average delay period in herniae diagnosed late-10+ yearsShortest- 6 monthsLongest-17- years

Average age at time of treatment -52 years

I

FIG. 13a and b.-Case 15: Postero-anterior and left lateral radio-graphs of chest showing herniation of a solid viscus (probablyspleen) through left hemidiaphragm. There are healed fracturesof the seventh and eighth right ribs. (No operation was per-formed and proof of herniation is lacking in this case. It isadmitted that a similar appearance may be seen in cases whereno hernia exists.)

a possibility where there has been crushing injuryto the chest or abdomen, and its presence is notexcluded by the absence of rib fractures (Evansand Simpson, 1950; present series, Cases 3, 7, 8,11, 12, and 14). Radiography may be difficult dueto the patient's poor condition and associatedinjuries, as it was in our Case 8. Sutherland (1958)stressed the danger of missing a visceral fluidlevel by radiography in the supine position, andthe value of antero-posterior radiography in thelateral position in patients in poor generalcondition.Evans and Simpson recorded distressing

dyspnoea and cyanosis in two of their sevenpatients with herniae due to crushing. Dyspnoeawas marked in Cases 3 and 8 of our series, butas it is a fact that miners suffering from pneumo-coniosis and emphysema may become desperatelyill from this cause following fracture of a few ribsonly, the symptom is not of great diagnostic help.

108



j.com/


ownloaded from



It has now become our practice to make a criticalexamination of the diaphragm in all cases of crushinjury of the chest or of the abdomen.Though not observed in our series, intestinal

obstruction may give the first indication ofdiaphragmatic injury (Dugan and Samson, 1948;Meyer, 1950).

LATE DIAGNOSIS. - Diagnosis of traumaticdiaphragmatic hernia may be made a long timeafter the injury that caused the tear in thediaphragm. It may be an " accidental " findingduring mass surveys (Cases 1, 6, 11, and 12), orduring the investigation of coexisting respiratorydisease (Cases 10 and 14). Often the diagnosis ismade when investigating ill-defined abdominalsymptoms (Cases 2, 7, 9, and 15). In Case 5 thepatient had suffered from substernal pain afterfood from the time of his injury to his emergencytreatment for haematemesis; while in Case 13 thecomplications of serious intestinal disease broughtthe hernia to light. In a patient treated by R. H.Gardiner there was a long interval withoutsymptoms after wounding in the first world war.Acute obstruction of the colon developed 33-years later, preceded only by a few months ofmild premonitory symptoms (Probert, 1959).

In cases following penetrating wounds it is notuncommon for the symptoms to be attributed tomalingering or hysteria for many years Thereason is that the initial hernia is small and escapesnotice; the condition is wrongly diagnosed andthe diagnosis persists. Diaphragmatic herniaetend to be ingravescent; they increase in sizecontinuously and their later discovery may explainsymptoms that have long been attributed topsychological causes. Tudor Edwards diagnosedone such case after 25 years (Barrett, 1959).DIAGNOSTIC DIFFICULTY DUE TO PHRENIC

PARALYSIs.-There is a danger of confusing acute,or transient, phrenic paralysis (such as commonlyoccurs in all sorts of war injuries of the chest)with diaphragmatic hernia. Although rare, thismistake has been reported by Barrett (1959). Itmay seem even stranger that phrenic paralysisshould cause difficulty in cases treated after along interval. In the following case, thoughphrenic paralysis was thought to be a very likelyeffect of the original injury, the diaphragmaticabnormality mimicked a hernia closely.A man aged 44 years had received multiple injuries

in a motor-cycle accident nine years previously. Thecervical spine, right scapula, and right humerus hadbeen fractured and the right brachial plexus severelydamaged. His right arm was later amputated, butsevere shoulder pain persisted. Five weeks before

K

Fir. I4.l

FIG. 14b

FIG. 14.-a, Postero-anterior radiograph showing malunited fractureof clavicle. The right hemidiaphragm is high. There is atranslucency corresponding to the hepatic flexure of the colon.b, Corresponding right lateral film. The opacity in the rightlung was interpreted as an area of atelectasis. The appearancewas thought to indicate a hernia; operation revealed a paralysedbut intact diaphragm.

109



j.com/


ownloaded from



admission he developed epigastric pain. Radiographysuggested herniation through the right hemidiaphragm(Fig. 14), but at thoracotomy the diaphragm wasfound to be high and atrophic due to phrenic nerveinjury, and there was no indication of localized injuryto the diaphragm. The abnormality was treated byplication.A similar case was even more confusing.

FIG. 15.-a, Postero-lateral radiograph of chest showing high lefthemidiaphragm. b, Left lateral film. The stomach, colon, andspleen are shown; the opacities in the left lung were interpretedas areas of atelectasis. The irregularity anteriorly was thoughtto indicate herniation; operation revealed a paralysed but intactdiaphragm.

A coal-miner aged 46 years had been crushedbetween two trucks and had been off work for 16weeks in 1956. Two years later he noticed increasingshortness of breath, and two years later still he com-plained of pain in the left side of the chest. Radio-graphy showed the left hemidiaphragm to be highwith paradoxical movement (Fig. 15), but hemiationwas considered likely. At left thoracotomy there wasno defect in the diaphragm, which was thin andatrophic. A localized area of irregularity in thephrenic nerve suggested a probable site of formerinjury.

THE HERNIAL CONTENTSThe stomach, small intestine, colon, and

spleen call for no further comment. Rarelythe kidney may enter a hernia due topenetrating injury (Barrett, 1945; Paul, Uragoda,and Jayewardene, 1960). The liver may passas a whole into the right chest (Case 7), butit is not generally appreciated that part of theliver can protrude like a mushroom through a

defect (Case 10). This is because the organ is not,as many imagine, a solid, wooden sort of structurewhose shape is fixed, but is capable of beingmoulded by stresses around it. Part of the livermay occupy a left-sided hernia as it probably didin Case 15. We have also seen at operation a" mushroom " of the left lobe of the liver occupy-ing a congenital hernia through the anterior partof the left diaphragm. (Sully Hospital Case No.02670, to be reported.)

ACCOMPANYING INJURIESOther injuries resulting from the original

trauma are often serious and are shown in TableIV. Barrett has reported the case of a miner inwhom rupture of the diaphragm without fractureof ribs was associated with a hyperextensionfracture of the thoracic spine and a chylothorax.In the experience of many surgeons the patientsmost likely to suffer from rupture due to increased

TABLE IVACCOMPANYING INJURIES

Crush injuries(3 patients had more than one accompanying injury)

Fracture of lumbar spine .. .dislocation of lumbar spine with paraplegia 2ofpelvis. 2

skull 1ribs .5tibia and fibula .. 2

Petechial haemorrhages of head and trunk with tem-porary loss of vision. I

No known accompanying injury I

Penetrating injuriesCompound fracture of radius and ulna ..Sciatic nerve lesionWounds of spleen, lung, and stomach ..No known accompanying injury 1

110



j.com/


ownloaded from



intra-abdominal pressure are infants and youngchildren whose abdomens have been run over.Rather surprisingly there have been no herniaedue to this cause in our series. In Evans andSimpson's series five out of seven herniae due tonon-penetrating injury followed accidents of thiskind; four of these patients had fractures of thepelvis, none had fractured ribs, and all escapedvisceral injury. Abdominal visceral injuries werefew among our patients, but their menace wasimpressed on us by the following case (not one ofthis series):A coal-miner aged 44 years was crushed at work.

Three days later his condition rapidly deterioratedand left thoracotomy was needed for haemothoraxdue to two ruptured intercostal arteries. Explorationshowed a tear 1 in. long in the muscular part of thediaphragm in the line of its fibres but not involvingits whole thickness. Extending this incision into theperitoneal cavity revealed a thin film of blood due toan unsuspected rupture of the spleen. Splenectomywas performed.

In two herniae due to crushing (Cases 3 and 14)there were widespread petechial haemorrhages atthe time of injury. Though not rare in our experi-ence of crushing injuries of the chest, thisphenomenon is transient and seems unimportant.

Dr. Roger Seal, pathologist to Sully Hospital,has recently drawn our attention to an associationof injuries in two further cases of diaphragmaticrupture (not included in the above analysis).A car driver and his passenger were killed instantly

in an accident. Necropsy on each of them showedsimilar findings: a tear through the left hemi-diaphragm parallel with the pericardial attachment,accompanied by complete transection of the aorta atthe beginning of its descending portion.

THE MECHANISM OF DIAPHRAGMATIC TEARING INCRUSHING INJURIES

Only one of our cases followed a car accident,one was the result of an air raid, and the remain-ing nine followed accidents at work. In many ofthese the crushing mechanism was slow, and itis possible that the diaphragm became more liableto rupture as a result of violent respiratory effortsduring the period of compression.

In depicting the mechanism of ruptureDesforges et al. (1957) show a tear in the cupoladue to the abdominal viscera being compressedagainst the diaphragm. If such compression werethe most important feature in these accidents itwould be reasonable to expect visceral injury tooccur more often than it does. It is noteworthythat Carlson, Diveley, Gobbel, and Daniel (1958)in a review of the English language literature on

traumatic diaphragmatic hernia from 1946 to1957 found visceral injury in 22 of 99 patients,but mining accidents accounted for only five ofthe entire series. Furthermore, in our Cases 3, 8,and 9 each defect lay close to the chest wall, andthe narrow rim of tissue at the periphery in eachcase suggested avulsion of an active muscle ratherthan bursting of a passive sheet of tissue. Somelight may be thrown on this matter by thepossibility of rupture of the diaphragm duringlabour. Though rare, such a case has beendescribed by Barrett (1959). After a prolongedlabour resulting in the birth of twins, severevomiting, at first misdiagnosed as " toxaemia," ledto the diagnosis of strangulation of bowelherniated through a recent diaphragmatic tear.Recovery following operation was complete.Another possible mechanism considered by

Desforges and his collaborators was that forcescompressing the chest from opposite directionscould rupture the diaphragm like the membraneof a drum. Under these conditions the tear wasexpected to occur in the line of compression andagain to be centrally placed. It is difficult toreconcile this concept with the arched shape ofthe diaphragm and its laxity. Furthermore, inCases 3, 8, and 9, where we have evidence fromthe clinical history that the chest was compressedthis way, the tears were peripheral.We are unable to explain the mechanisms

involved, but we think that forced respiratorymovements of the diaphragm during crushing areimportant.

PARTIAL TEARSIn Case 3, in addition to a large defect

anteriorly in the diaphragm, two separate peri-pheral tears of muscle and pleura were foundunder which the peritoneum was intact.

It has long been realized that the differencebetween thoracic and abdominal pressures cancause delayed herniation by acting on a thindiaphragmatic scar or an omentum-plugged gap.If this mechanism acted upon an incomplete teara hernial sac might be expected. Hollander andDugan (1955) found a sac proved by microscopyin one of their cases of traumatic diaphragmatichernia, and in another demonstrated the herniatedliver covered by stretched diaphragmatic pleura.

THE PRESENCE OF A HERNIAL SACHernial sacs were found in two of our patients.

In one (Case 10), where the hernia resulted frompenetrating injury, there was a thin fibrous sac,and microscopy showed collagenous tissue withno muscle fibres. In the other patient (Case 12),

III



j.com/


ownloaded from



where the hernia resulted from a crushing injury,the central part of the diaphragm was seen atoperation to be replaced by a thin layer of tissue;microscopically this was shown to be collagenousand muscle fibres were absent.Hughes and others (1948) found sacs in only

two of their 28 cases, but stated that " there wasoften a membrane covering a part of the herni-ated structure, but it was not complete." Thisobservation suggests that the membrane is a newlyformed structure. The organization of blood andpleural exudate on the surface of herniated visceracould result in the formation of a fibrous orcollagenous sac, and the finding of sacs in herniaedue to penetrating injury suggests that they can beformed in this way even when the initial injuryinvolves all layers of the diaphragm. An explana-tion of sac formation in penetrating injuries hasbeen advanced by Barrett (1945). Within a fewhours of injury a tiny knuckle of omentuminsinuates itself through a hole, particularly ifthe hole is small. The resulting mushroom ofomentum gradually increases in size and becomesinvaginated by gut. The limiting layer ofomentum forms a " false sac " as opposed to a"true sac" formed by peritoneum.

RESPIRATORY HAZARDSOf the cases diagnosed early respiratory

embarrassment was present in Case 8 and wasa prominent feature in Case 3. Of the casesdiagnosed late dyspnoea was complained of byfive patients (Cases 9, 11, 12, 14, and 15), in oneof whom (Case 9) it was severe. In this patientthe maximum ventilatory volume was only 40litres/minute, but bronchospirometry showed nodifferential loss of lung function. The overall lossof function was attributed to pneumoconiosis andemphysema, and this contributed to the fatal resultin his case.Three patients (Cases 1, 12, and 15) had been

under review from time to time by the Pneumo-coniosis Panel of the Ministry of Labour andNational Insurance. It is significant that thoughthe diagnosis of diaphragmatic injury was madelate in Case 12 the patient had had a severe chestinfection soon after his original injury. Webelieve that this was the result of retention ofbronchial secretions in a patient whose injuriesand impaired lung function made coughing diffi-cult. We have already commented on the dangerof chest injury in patients with impaired lungfunction. In them excessive analgesia can behazardous, and we are convinced of the need tolimit analgesics to a minimum. Doses tolerated

by patients with normal lung function may bedisastrous in those with poor lung function.These patients must not be allowed to lie drowsyand immobile, but with the chest firmly supportedmust be vigorously encouraged to cough up allretained bronchial secretions.

SELECTION OF TIME FOR OPERATIONIn those cases diagnosed early Edwards (1943)

and Tubbs (1955) advise delaying operation tillthe patient has recovered sufficiently from hisother injuries. This principle was followed inCase 3, when it was three days before the patientwas fit for operation; fortunately the herniatedbowel was amply viable. On the other hand,Dugan and Samson (1948) and Bugden et al.(1955) recommend early operation to preventintestinal obstruction, while Sutherland (1958) inhis series of 10 cases of indirect traumatic ruptureof the diaphragm found two examples of strangu-lation of the stomach. In both of them grossdistension of the body of the stomach had drawnthe left gastric vessels tightly against the intactmuscular margin of the oesophageal hiatus whichwas not involved in a neighbouring tear. In ourCase 8 at an operation not lightly undertakenbecause of the serious spinal injury, the stomachwas found tightly constricted by the defect in thediaphragm on the day after the accident.

OPERATIONThe approach was by thoracotomy in all our

patients, but laparotomy was added in Case 4.Separation of adhesions in late cases and repairof the diaphragmatic defect is easier by thethoracic route. Apart from the main defect ourexperience suggests that an attempt should bemade to repair any partial tears, treating them asweak spots liable to develop into herniae. Anattempt was made in all our patients to approxi-mate the edges of the main defect itself with adouble layer of interrupted linen thread sutures.Difficulties were encountered when the defect layclose to the costal margin or extended beneaththe attachment of the pericardium, and in thesepart of the repair was in one layer only. In onepatient (Case 9) there was difficulty in repairingthe diaphragm. The edges were brought togetheras far as possible, but a central gap remained.This was covered by a free graft of tissue fromthe chest wall as described below.

A METHOD OF TISSUE GRAFTINGAt Sully Hospital grafts of muscle, fascia, and

periosteum from the chest wall were formerly

112



j.com/


ownloaded from



FIG. 16.-Photograph showing thickness of flap of tissue taken fromchest wall. In this illustration the flap was used for reinforcingthe roof of an osteoplastic thoracoplasty, but a similar flap canbe used to repair other types of defect. The method of raisingthis flap is described in the text.

used to strengthen the roof of the refashionedthoracic cage in osteoplastic thoracoplasty, andare now used from time to time to repair costaldefects after excision of tumours. The graft istaken from the outer aspect of the ribs mostconveniently related to the defect. The diathermyneedle is used to outline the extent of the graft,the base of which will remain attached to thelower border of a rib. The periosteum is raisedfrom the whole outer surface of this rib with a

curved rugine. With the freed periosteum andoverlaying fascia held under moderate tension,another diathermy cut is made from within theperiosteum through its attachment to the upper

border of the rib into the muscle of the inter-costal space. A layer of muscle, probablyrepresenting the greater part of the thickness ofthe external intercostal muscle, is dissected upuntil the lower margin of the rib above is reached.The periosteum of this rib is next incised alongthe extent of the graft and separated as in thecase of the previous rib. The process is repeateduntil the graft is big enough for its purpose (Fig.16), when it is turned upwards as a flap andsutured into position. In Case 9 this type of graft

was used for diaphragmatic repair for the firsttime, but differed in that it was completely free.

SUMMARYFifteen cases of traumatic diaphragmatic

hernia are reported, and associated injuries arementioned.The mechanism of diaphragmatic rupture in

crush injuries and the significance of partial tearsof the diaphragm are discussed.The importance of respiratory disease in

patients with crush injuries and the dangers ofexcessive analgesia are stressed.A method of tissue grafting for diaphragmatic

repair is described.

Our thanks are due to Dr. E. A. Danino, Dr. A. J.Thomas, Messrs. J. Elgood, J. F. E. Gillam, J. RussellHughes, R. H. B. Mills, Melbourne Thomas, and E.Meurig Williams, who referred patients tc thisCentre; also to Dr. H. M. Foreman, physiciansuperintendent, Dr. L. R. West, who performedrespiratory function tests in some cases, and Dr.R. M. E. Seal, pathologist.Our interest in this subject was kindled by Mr.

Dillwyn Thomas, who was responsible for the treat-ment of all these patients; we thank him for hisencouragement and advice.

Finally, we thank Miss Patricia Morse forsecretarial help.

REFERENCESBarrett, N. R. (1945). Brit. J. Surg., 32, 421.- (1959). Personal communication.Bryan, C. W. G. (1921). Brit. J. Surg., 9, 117.Bugden, W. F., Chu, P. T., and Delmonico, J. E. (1955). Ann.

Surg., 142, 851.Carlson, R. I., Diveley, W. L., Gobbel, W. G., and Daniel, R. A.

(1958). J. thorac. Surg., 36, 254.Desforges, C., Strieder, J. W., Lynch, J. P., and Madoff, 1. M. (1957)

Ibid., 34, 779.Dugan, D. J., and Samson, P. C. (1948). Ibid., 17, 771.Edwards, A. T. (1943). Brit. J. Surg., 31, 74.Evans, C. J., and Simpson, J. A. (1950). Thorax, 5, 343.Hollander, A. G., and Dugan, D. J. (1955). J. thorac. Surg., 29, 357.Hughes, F., Kay, E. B., Meade, R. H., Jr., Hudson, T. R., and

Johnson, J. (1948). Ibid., 1';, 99.Mackey, W. A., and Bingham, D. L. C. (1945). Brit. J. Surg., 33,

135.Meyer, H. W. (1950). J. thorac. Surg., 20, 235.Morgan, C. N. (1945). Brit. J. Surg., 32, 337.Paul, A. T. S., Uragoda, C. G., and Jayewardene, F. L. W. (1960)-

Ibid., 47, 395.Probert, W. R. (1959). Postgrad. med. J., 35, 153.Sutherland, H. D'Arcy (1958). Ibid., 34, 210.Tubbs, 0. S. (1955). In Modern Operative Surgery, 4th ed, vol. 1,

p. 431, ed. G. G. Turner and L. C. Rogers. Cassell, London.

113



j.com/


ownloaded from


Documents

TRAUMATIC DIAPHRAGMATIC HERNIA - Thoraxthorax.bmj.com/content/thoraxjnl/16/2/99.full.pdf · Thorax(1961), 16, 99. TRAUMATIC DIAPHRAGMATIC HERNIA BY W. R. PROBERT AND C. HAVARD From