J. clin. Path. (1960), 13, 432.

THREE CASES OF ACUTE MESENTERIC LYMPHADENITISDUE TO PASTEURELLA PSEUDOTUBERCULOSIS

BY

N. S. MAIR, HELENE J. MAIR, E. M. STIRK, AND J. G. CORSON

From the Public Health Laboratory Service, Isolation Hospital, Leicester, and theRoyal Infirmary, Leicester

(RECEIVED FOR PUBLICATION JUNE 16, 1960)

Seventeen cases of mesenteric adenitis were investigated between May and October, 1959,for the presence of virus as well as for evidence of Pasteurella pseudotuberculosis infection.Specimens examined included mesenteric glands, appendix, throat swab, faeces, and serum,

although glands were received from only 12 patients. Virus was not isolated from any of thespecimens, but evidence of infection with Pasteurella pseudotubercuilosis was obtained in threepatients.The three cases of acute mesenteric lymphadenitis due to Pasteurella pseudotuberculosis are

reported.Pasteurella pseudotuberculosis Type IA was isolated from the mesenteric glands of two of the

cases, and all three cases showed serological evidence of infection with the organism. Themesenteric glands of two of the cases showed histological changes characteristic of pseudotuber-culosis.

Acute mesenteric lymphadenitis, mimickingappendicitis, is not uncommon in this country.Aird (1945) reports that in the Royal Hospital forSick Children, Edinburgh, during the year 1944,there were admitted in his charge 37 patientssuffering from non-specific mesenteric adenitiscompared with 83 suffering from acute appen-dicitis and four suffering from abdominal tuber-culosig. During the period May to October, 1959,of 93 children admitted for suspected appendicitisto the Leicester Royal Infirmary, 20 were found tobe suffering from acute mesenteric adenitis.Many causes have been suggested, including

infection with organisms of the dysentery group(Felsen, 1935), infection with a hypothetical virusderived from the upper respiratory tract (Aird,1945), and hyperinfestation of the intestine withroundworms (Kirthi Singha, 1959).

In 1953, Masshoff and D6lle observed certainpathological changes in the mesenteric glands ofyoung children operated on for suspected appen-dicitis. At operation the appendix was found tobe normal or only slightly inflamed while the ileo-caecal glands were enlarged. The changes in theglands were characterized by the formation of oneor more follicles consisting mainly of reticulumcells. Because of the tendency of the follicles toundergo necrosis the authors described thepathological process that they had observed as an

" abscess-forming reticulocytic lymphadenitis,"and suggested that it might be of viral origin, sincethe histological changes in the affected glandsresembled those found in cat-scratch fever andlymphogranuloma inguinale. In the followingyear, however, Knapp (1954) and Knapp andMasshoff (1954) reported the isolation of Past.pseudotuberculosis from the enlarged mesentericglands of two children operated on for appendicitis.In both cases the appendix was normal. The histo-logical changes in the glands were identical withthose described by Masshoff and Dolle. From thenuntil 1957, 117 cases of Past. pseudotuberculosiswere diagnosed by Knapp and his co-workers atthe Hygiene-Institut in Tubingen (Knapp, 1958).

Early in 1959 Dr. G. S. Wilson drew our atten-tion to the work of Knapp and his colleagues, andas we were already making an investigation intothe possible viral origin of mesenteric adenitis, itwas decided to include Past. pseudotuberculosisin the scope of our inquiry.Arrangements were made to collect glands,

appendix, faeces, throat swab, and serum fromeach patient. In the six months from May, 1959,we received the full complement of specimensfrom 12 patients, and from -five others all thematerial with the exception of the glands.Bacteriological and histological examinationswere made of the glands and appendices, and the

copyright. on 10 M

ay 2018 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.13.5.432 on 1 S

eptember 1960. D

ownloaded from

ACUTE MESENTERIC LYMPHADENITIS

sera were examined for antibodies to Past. pseudo-tuberculosis. Frozen glands, faeces, and throatswabs were cultured for viruses on amnion andMK2 cells-a continuous monkey kidney cellline (Westwood, Macpherson, and Titmuss, 1957).We were unable to isolate viruses from any of

the specimens examined. On the other hand, wewere able to find evidence of infection with Past.pseudotuberculosis in three of the 12 childrenfrom whom both gland and serum were received.There was no serological evidence of infectionwith the bacterium in the other five patients.We report here the three cases of pseudo-

tuberculous mesenteric lymphadenitis. As far aswe are aware they are the first to be recorded inthis country.

Case ReportsCase l.-R. Sturgeon, a boy aged 10, was admitted

to the Leicester Royal Infirmary on May 11, 1959,with a five-day history of central abdominal pain,described as a dull ache. He had had similar attackspreviously, but never one so long or so severe as thepresent one. He felt sick during the illness but didnot vomit. He also had mild diarrhoea of two days'duration.On examination the temperature was 99.8' and

the pulse rate 88. There was tenderness and guardingin the right iliac fossa. The rebound sign was equi-vocal. No masses were felt in the abdomen norwas anything abnormal detected per rectum. Theabsence of vomiting made the diagnosis somewhatuncertain, but since there appeared to be more tender-ness in the right iliac fossa than in the left anappendicectomy was performed.

Operation (May I 1).-A normal retrocaecalappendix was present. There was no Meckel'sdiverticulum. The mesenteric glands were enlarged.The appendix and a mesenteric gland were removedfor examination.Progress.-The temperature became normal on the

first post-operative morning. On the second day thepatient was restless and irritable, but eventuallysettled down. No diarrhoea occurred after theoperation.Laboratory Findings.-The mesenteric gland was

about I in. in diameter. No organisms were seen inthe direct smears, and culture was sterile. Histo-logical examination showed early features ofPasteurella pseudotuberculosis infection. Serumtaken on May 12, the day after operation, agglutinatedPasteuirella pseudotuberculosis Types IA and IB to a

titre of 1: 300. Serum taken five months later showedno agglutinins for Pasteurella pseudotuberculosis.

Case 2.-K. Schofield, a girl aged 15, was admittedto the Infirmary on May 13, 1959, with abdominalpain of three days' duration. The pain, whichremained localized in the umbilical region, was

dragging in nature. There was no nausea or vomiting.

Appetite had been poor for three days. The patientcomplained of slight frequency with some suprapubicpain on micturition. She had been constipated fora week before admission. A similar attack threemonths previously had passed off in one week.On examination the temperature was 102.4°, the

pulse rate 112, and respirations 20. She did not lookill. The tongue was clean. The abdomen was softwith no guarding. There was tenderness on the leftside and right hypochondrium. The descending colonwas loaded. Bowel sounds were normal. Urine,S.G. 1012, was acid, straw-coloured, without albumin,sugar, or acetone. A provisional diagnosis ofmesenteric adenitis or subacute appendicitis was made.

Operation (May 13).-The appendix was mildlyinflamed and not adherent. The pelvis was normal.There was no Meckel's diverticulum. The terminalileal glands were enlarged. Appendicectomy andgland biopsy were performed.Progress.-On the day following operation the

temperature remained at 100°. It fell to normal onthe morning of the 15th, but rose to 100' that evening.The following evening it shot up to 102'. On theevening of the 18th it was 101° and then it fell tosubnormal, and remained thus until discharge onMay 20. On the day of the high fever the scar wastender, but this settled in 24 hours.

Laboratory Findings.-A direct Gram-stained smearof the gland revealed numerous Gram-negative cocco-bacilli. Culture yielded a profuse growth ofPasteurella pseudotuberculosis (Schofield strain).Serum taken on May 13 agglutinated the homologousorganism and Pasteurella pseudotuberculosis TypesIA and IB to a titre of 1:500. A second specimenof serum taken on October 17 showed no trace ofagglutinins. Histological examination of the glandwas not made in this case.

Case 3.-G. Burke, a boy aged 12, was sent tohospital by his practitioner on October 19, 1959, withthe provisional diagnosis of mesenteric adenitis. Hehad a two-day history of central abdominal pain,which moved over to the right side on the day beforeadmission. He vomited once on the night ofOctober 18. Bowel movements and micturition werenormal. He had slight nausea and anorexia. Hehad had no previous attacks.On examination temperature was 101.6' and pulse

rate 120. The tongue was slightly coated and moist,and foetor was present. The abdomen was not dis-tended, but tender with muscle guarding in the rightiliac fossa. There was no rebound tenderness. Nomasses were felt. Rectal examination revealedtenderness in the right side over the peritoneal surface.A diagnosis of acute appendicitis was made.

Operation (October 19).-The appendix lookednormal. There was no Meckel's diverticulum.Appendicectomy was performed and a very largegland in the.ileocaecal angle was removed.Progress.-Apart from a slight pyrexia of 99' on

October 22 the boy made an uneventful recovery.

433

copyright. on 10 M

ay 2018 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.13.5.432 on 1 S

eptember 1960. D

ownloaded from

N. S. MAIR AND OTHERS

Laboratory Findings.-No organisms were seen inthe direct smears of the gland. Culture yieldedPasteurella pseudotuberculosis, and the organism wasalso isolated from a guinea-pig inoculated with asuspension of the gland (Burke strain). Histologicalexamination showed the characteristic features ofPasteurella pseudotuberculosis infection. Serumtaken on October 19 agglutinated the homologousorganism to a titre of 1: 5,000 and Pasteurellapseudotuberculosis Types IA and IB to a titre of1:3,000. Opportunity was taken to study the declinein agglutinins in this patient and the results aredescribed later in the section on serology.

BacteriologyExamination of Mesenteric Glands.-Each gland

was examined as follows: Smears were stained byGram's method and by a modified acid-fast techniquerecommended by Cook (1952) for the recognition ofPasteurella pseudotuberculosis in tissues. The glandfrom Case 2 was the only one to show Gram-negativeand acid-fast bacilli in the smears.The gland was then ground up in 5 ml. digest broth,

and loopfuls of the suspension were inoculated on5% horse-blood digest agar and into digest broth.Duplicate cultures were incubated at 220 C. and 370 C.for 48 hours. The Schofield strain was obtained ondirect culture at 37° C. On the other hand, Burke'sstrain was isolated only after preliminary incubationin broth at 220 C. for 24 hours.

Guinea-pig inoculation was carried out only withBurke's gland suspension. The animal became ill onthe seventeenth day after inoculation and was killedon the twentieth day. At necropsy there was a largecaseating abscess at the site of injection, caseousenlargement of the regional gland, and generalizedsepticaemic spread with multiple nodules in the liver,spleen, and lungs. Pasteurella pseudotuberculosiswas isolated in large numbers from the local lesion,organs, and heart blood.Examination of the Appendix.-The interior of

each appendix was exposed and inoculated on bloodagar plates, which were incubated at 220 C. and 37°C. and examined after 24 and 48 hours with negativeresults for pasteurella.

Identification of Schofield and Burke Strains.Both strains showed similar morphological, cul-tural, and biochemical characteristics. Bothconsisted of Gram-negative cocco-bacilli, motilein broth culture at 220 C. and non-motile at 370C. They grew readily on the usual media, pro-ducing on blood agar, after 24 hours at 370 C.,flat, dry, non-haemolytic colonies, with crenatededges and a dull granular surface. After 24hours' incubation at 22° C., small moist colonieswith entire edge and smooth shining surface wereproduced. Colonies grown at 370 tended toagglutinate spontaneously in normal saline, incontrast with those grown at 220 C. which showed

no tendency to auto-agglutination. Minutecolonies appeared on MacConkey's medium after24 hours at 370 C. In broth growth was diffuse at220 C., and tended to be more viscous at 370 C.Acid without gas was produced after 18 hours

at 370 C. in dextrin, glucose, laevulose, maltose,mannitol, rhamnose, and trehalose, after 48 hoursin galactose and glycerol, and after four days insalicin. Trace reactions were obtained insorbitol and xylose. Arabinose, dulcitol, inulin,lactose, raffinose, and sucrose were not fermentedafter 14 days.The methyl-red reaction was positive and the

Voges-Proskauer reaction negative. Indole andH2S were not produced and gelatin was not lique-fied. Urea was decomposed, litmus milk wasrendered alkaline, and nitrates were rapidlyreduced. Catalase was present.Guinea-pigs injected intramuscularly with 0.5

ml. of 24-hour broth cultures of the Schofieldand Burke strains died in nine and 18 daysrespectively. Post-mortem findings were similarto those obtained with Burke's infected glandsuspension.

SerologyStrains of seven serotypes and subtypes of

Pasteurella pseudotuberculosis were obtained fromthe National Collection of Type Cultures. Theantigenic constitution of the different types isshown in Table I.

TABLE IANTIGENIC CONSTffUTION OF PAST. PSEUDOTUBER-

CULOSIS ACCORDING TO KNAPP (1956)

Type Subtype Somatic Antigen Flagellar AntigenIA 1 2 3 aIB 1 2 4 a

II IJA 1 5 6 aIIB 1 5 7 a

III 1 8 aIV 1 9 bV 1 10 a

Living suspensions were used throughout foragglutination tests as recommended by Knapp (1956),who reported that heat-killed (1000 C.) antigen wasnot agglutinated by human antisera. Agglutinatingantigen was prepared by growing the organism ontryptose agar at 220 C. After 48 hours the growthwas harvested in normal saline, centrifuged,resuspended in saline, agitated with glass beads ina shaker for 20 minutes, and finally diluted to adensity equivalent to No. 10 Brown opacity tube.Agglutination tests were carried out in '-in. wideround-bottom tubes. One drop of the concentratedsuspension was added to 0.5 ml. of each serum dilu-tion and the tubes were incubated for two hours at370 C., left in the refrigerator overnight, and read

434

copyright. on 10 M

ay 2018 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.13.5.432 on 1 S

eptember 1960. D

ownloaded from

ACUTE MESENTERIC LYMPHADENITIS

two hours after removal next morning. Parallel tests,using Dreyer's technique and incubating the suspen-sion for 24 hours at 56° C., gave similar results. Forabsorption, serum diluted 1:10 was poured over thecentrifuged deposit of a dense suspension of theabsorbing organism and thoroughly mixed. Themixture of serum and bacteria was incubated for twohours at 37° C. followed by 20 hours at refrigeratortemperature (+5° C.).

In preparing 0 antisera it was necessary to takespecial precautions in order to obtain suitable heat-killed antigens. Many authors have noted the ten-dency of Pasteurella pseudotuberculasis to agglutinatespontaneously in normal saline, especially when thesuspension is heated. Recommendations to overcomethis difficulty include reduction of salt content to 0.3°'(Topping, Watts, and Lillie, 1938), strong shaking ofthe suspension before boiling (Preston and Maitland,1952), and the use of cultures grown at 22° C. (Knapp,1956). By using a combination of these methodswe were able to obtain stable heat-killed suspensions.Smooth colonies of a 22° C. culture were sown ontryptose agar and incubated at 22° C. for 24 hours.The growth was harvested in 0.5% saline, agitatedwith glass beads in a shaker for 20 minutes, washedwell in 0.5% saline, shaken again, and then exposedto flowing steam in a Koch sterilizer for two and ahalf hours or autoclaved at 1200 C. for two hours;0.25%,' phenol was added as a preservative.

Agglutination Tests with Patients' Sera.-Pre-liminary slide agglutination tests indicated thatagglutinins to Types IA and IB were present inthe sera of all three patients. No agglutinationwas observed with the other serotypes. Becauseonly limited amounts of sera were available tubeagglutination tests had to be restricted to reactionswith the Burke and Schofield strains and theNational Collection of Type Cultures serotypesIA and IB. The results of the agglutination testswith unabsorbed and absorbed sera shown inTable II suggest that the Schofield and Burkestrains are antigenically similar and probablybelong to Type IA. The results with Sturgeon'sserum also indicate infection with Pasteurellapseudotuberculosis Type I, but because of therelatively low titres obtained with the unabsorbedserum, we are unable to demonstrate residualtitres after absorption with the four antigens.Typing of Schofield and Burke Strain&-Typing

of both strains was confined to the identificationof the somatic factors present. Recognition ofthe flagellar antigen is of little value in the deter-mination of the different types because, with theexception of Typs IV, the antigen is common toall members of the group. 0 agglutinatingantisera were prepared in rabbits for Types IAand IB as well as for Schofield and Burke strains.Autoclaved antigen was inoculated intravenously

in six doses (0.25 ml., 0.5 ml., 1 ml., 1 ml., 1 ml.,1 ml.) at four-day intervals. The animals werebled on the twenty-eighth day. The results of

TABLE IIRESULTS OF AGGLUTINATION TESTS WITH ABSORBED

AND UNABSORBED PATIENTS' SERA

Agglutination Titres of Sera againstFour Strains of

Past. pseudotuberculosisSera

Agglutinating Antigens

Schofield| Burke Type IA Type IB

Schofield serumUnabsorbed .. .. 500 500 500 500Absorbed with 1 1 1

Schofield strain.. .. 2 L<5 2Burke strain .. L2 < 5~ <25Type IA f.. S f5 2Type IB 25.. 25

Burke serumUnabsorbed .. .. 5,000 5,000 3,000 3,000Absorbed with 1 1 1 1

Schofield strain. .. <25 L <25 <25Burke strain .. . .<25Type IA . .J J J J_Type IB .. 200 200 200

Sturgeon serumUnabsorbed .. 300 300 300 300Absorbed with 1 1 1

Schofield strain.. .. <25

Burke strain .<25} <25 <25 <25

TABLE IIIRESULTS OF AGGLUTINATION TESTS WITH ABSORBED

AND UNABSORBED RABBIT ANTISERA

Agglutination Titres of Sera againstFour Strains of

Past. pseudotuberculosisSera

Agglutinating Antigens

Schofield Burke Type IA Type IB

Schofield antiserumUnabsorbed .. .. 2,003 2,000 2,000 2,000

traceAbsorbed with 1 1

Schofield strain.. . L <25L<2 <5Burke strain .. .. f <25Type IA f. fType IB .. .. 200 200 100 J

Burke antiserumUnabsorbed .. .. 5,000 5,000 5,000 5,000Absorbed with ) 1 1 1

Schofield strain. .. L <25 <25 <25Burke strain .. . .<25Type IA .. .J J J J_ _Type IB .. 500 500 200

Type IA antiserumUnabsorbed .. 10,000 10,000 10,000 10,000Absorbed with 1 1 1

Schofield strain. . 25 L <25 L 25Burke strain . < r <5 <25Type IA f.. JJType IB .. .. 500 500 500 J

Type IB antiserumUnabsorbed .. .. 10,000 10,000 5,000 20,000

trace

Schofield strain. .. 1 2,000Burke strain . .. _ <25 <25 <25 2,000Type IA . ..i F r 1,000TypeIB . .. J J<25

435

copyright. on 10 M

ay 2018 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.13.5.432 on 1 S

eptember 1960. D

ownloaded from

N. S. MAIR AND OTHERS

the cross-absorption tests show that the Schofieldand Burke strains are antigenically similar toType IA, thus confirming the findings made withthe patients' sera (Table III). Both strains werelater sent to Professor Knapp, of Tubingen, whokindly informed us that they belonged to Type IA.

Post-operative Decline in Antibody Titres.-Serum taken from Schofield and Sturgeon fivemonths after operation showed no evidence ofagglutinins to Pasteurella pseudotuberculosis.

Burke's serum showed a rise in antibody titretwo weeks after operation, but thereafter therewas a steady decline until agglutinins had all butdisappeared five months after onset of the illness(Table IV).

TABLE IVRESULTS OF AGGLUTINATION TESTS WITH BURKE'S

SERA SHOWING DECLINE IN ANTIBODIES

Day after Onset of Illness Serum Titre

4th 5,00017th 10,00033rd 2,00077th 500160th 25 trace

Antibiotic Sensitivity Tests.-These were carried outon the Schofield and Burke strains by the flood platetechnique using "multodisk" (Oxoid) paper discs.Both cultures were fully sensitive to chloramphenicol(10 pig.), streptomycin (10 Mg.), tetracycline (10 pg.),polymyxin B (10 pLg.), neomycin (10 Mg.), and nitro-furantoin (200 pg.). Resistance was shown to penicillin(1.5 jug.), erythromycin (10 Mg.), oleandomycin (5 Mg.),novobiocin (5 ,Ag.), bacitracin (5 units), and sulpha-furazole (100 Mg.).

HistologyBiopsy specimens of appendices and mesenteric

glands were fixed in formol saline and embedded inparaffin wax. Sections were stained withhaematoxylin and eosin.Case 1 (Sturgeon).-In the appendix there was

no evidence of inflammation. The lymphoidfollicles were large with prominent Flemmingcentres.

In the lymph node there was considerable sinushyperplasia and the follicles showed well-definedcentres. Towards the periphery of the node therewas a single small follicle consisting of a groupof polymorphonuclear pus cells and swollen cellsresembling reticulum cells. The latter were quitelarge and an occasional cell appeared to havetwo nuclei. It was a very unusual type of follicleto be seen within a lymph node (Figs. 1 and 2).No organisms could be seen in Gram-stained,Cook-stained, or Ziehl-Neelsen-stained sections.

Case 2 (Schofield).-In this case the lymphoidfollicles formed an almost continuous bandaround the lumen of the appendix and there wasa plasma cell infiltration of the mucosa itself.Within the lumen of the appendix there wasgranular debris, containing fairly large numbersof lymphocytes, and a smear taken from thisdebris showed a large number of Gram-negativediplobacilli. They were also to be seen withinthe lymphoid follicles. No lymph node wasexamined in this case.

Case 3 (Burke).-In the appendix in this case,lymphoid follicles were very prominent and theytended to be two or three layers in depth.

Sinus hyperplasia in the lymph node was dis-tinct. In this instance there were two follicleswhich appeared to be slightly later in stage thanthe one observed in Case 1 (Sturgeon). Thefollicles showed central necrosis with the presenceof eosinophilic-staining debris. This necrotic areawas infiltrated with plasma cells and polymorphs.It was surrounded by a layer of reticulum-cellsshowing somewhat haphazard palisading aboutfour cells in depth. Beyond it the reticulum cellsshowed no orderly arrangement. In this outerzone there was an occasional foreign-body typeof giant cell, polymorph and lymphocyte (Figs. 3and 4). No organisms could be seen in Gram-stained, Cook-stained, or Ziehl-Neelsen-stainedsections.The lesions noted in the lymph glands were

not found in the lymphoid follicles of any of theappendices.

DiscussionUntil a few years ago human infection

with Pasteurella pseudotuberculosis was generallyregarded as rare. The first authentic case wasdescribed by Albrecht (1910). During the periodbetween 1910 and 1952 some 15 cases werereported in the medical literature. Of these, 13were characterized by a severe typhoid-like illnesswhich terminated fatally in 11 cases.

In contrast to this rare septic type of infectionthe more frequent benign appendiceal form isnow recognized. In addition to numerous casesmentioned in the German literature, other caseshave been reported in Hungary (Podhragyai andFodor, 1956) and in France (Ingelrans, Soots,Poupard, and Vitse, 1957; Girard, Leger, Paulhet,and Duffau, 1959). No cases appear to have beenrecorded in the English language.

This new disease complex shows a predilectionfor children and young people (2 to 23 years)and occurs more often in males than in females.The clinical picture is that of acute or subacute

436

copyright. on 10 M

ay 2018 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.13.5.432 on 1 S

eptember 1960. D

ownloaded from

FIG. 1.-Case 1: Follicle consisting of swollen reticulum cells. The FIG. 2.-Case 1: Centre of follicle. (Haematoxylin and eosin x 425.)centre of the follicle is infiltrated with polymorphs. (Haema-toxylin and eosin x 150.)

FIG. 3.-Case 3: Follicle showing haphazard palisading of reticulumcells and area of central necrosis infiltrated with polymorphs.(Haematoxylin and eosin x 350.)

FIG. 4.-Case 3: Foreign-body type of giant cells in marginal zoneof reticulum cells. (Haematoxylin and eosin x 350.)

copyright. on 10 M

ay 2018 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.13.5.432 on 1 S

eptember 1960. D

ownloaded from

N. S. MAIR AND OTHERS

appendicitis with pain in the middle or right lowerquadrant of the abdomen. The temperature iselevated to 1000 or even higher, and there isinvariably a leucocytosis. Laparotomy shows, inmost cases, a normal-looking appendix. Thelymphatic glands in the ileocaecal angle and inthe mesentery are enlarged either singly or ingroups. The peritoneum over the lymph nodesis congested and a clear or slightly purulentexudate is often present. Inflammatory infiltra-tion of the terminal ileum has also been described.

In our first two cases there was some pre-operative doubt as to the correct diagnosis, butboth patients were too unwell to justify with-holding an exploratory laparotomy. The thirdcase (Burke) was considered by his own doctorto have mesenteric adenitis, but on admission hissymptoms and signs were such that there seemedto be no doubt about the diagnosis of acuteappendicitis.The findings at operation on all three patients

were essentially similar. After the first two hadbeen seen-and the diagnosis of pasteurella infec-tion subsequently established-the third case wasstrongly suspected at operation and later con-firmed pathologically. In all three it was obviousthat the appendix was not the cause of thetrouble; such exploration as could be carried outthrough a muscle-splitting incision failed to revealany abnormality other than in the affectedileocaecal glands. Owing to the limited exposureobtained, the extent of the glandular enlargementcould not be determined. Suppuration was notseen in any of the three cases and there was nogross oedema of the mesentery around any ofthe affected nodes. Peritoneal exudate was notobserved.The naked-eye appearance of the glands

suggests that it may be possible to distinguishbetween those cases due to infection withPasteurella pseudotuberculosis and those due toother causes. The glands seen in the non-pasteurella group appear to be pale, flattenedrather than round, and uniformly enlarged. In ourthree cases associated with pasteurella infectionthe glands were dark in colour with one or twowhite streaks across the surface, almost round, andvariable in size, the largest being about the sizeof a cherry and the smallest no bigger than adried pea. All were firm, not hard, fixed withinthe mesentery but not adherent to it. Removalwas easy and there was no undue tendency tobleeding.The definitive diagnosis, however, is established

by the isolation of the organism from thelymphatic glands, by the demonstration of specific

antibodies in the serum, and through histologicalexamination of the lymph nodes.

Pasteurella pseudotuberculosis is not readilyisolated from infected glands. Only 12 of the117 cases mentioned by Knapp, and one of ninecases reported by Schmidt (1959), were confirmedby isolation of the organism from the lymphnodes. That incubation at 370 C. may not besufficient is shown by the behaviour of Burke'sstrain, which was isolated only after preliminaryincubation in broth at 220 C. Identification ofthe organism itself is not difficult. Its motilityin 18-hour broth cultures at 220 C., its ability togrow on MacConkey medium and to decomposeurea, and the caseous nodules produced in guinea-pigs after experimental inoculation, help to dis-tinguish it from other members of the pasteurellagroup.

Antibodies to Pasteurella pseudotuberculosiscan be demonstrated in the blood during theacute phase of the illness. By means of agglutina-tion tests with live, smooth variants of Pasteurellapseudotuberculosis Types I-V, titres of 1 :80 to1:12,800 have been recorded. In our own casestitres varied from 1: 300 on the sixth day ofSturgeon's illness to 1: 5,000 on the fourth day ofBurke's attack. The tests are specific only forTypes I, III, and V. The specificity of theagglutination test for Types II and IV is affectedby the antigenic relationships which exist betweenPasteurella pseudotuberculosis Type II and factorIV of the salmonella B group and Pasteurellapseudotuberculosis Type IV and factor IX of thesalmonella D group (Kauffmann, 1933; Knapp,1955). These give rise to fairly strong cross-reactions. In order that a serological diagnosiscan be made it is necessary to absorb the patient'sserum with the corresponding strain of the B orD group. However, such absorptions are notoften necessary, since more than 90% of pseudo-tuberculous infections in man are caused byPasteurella pseudotuberculosis Type I. The speci-ficity of the agglutination test was confirmed byKnapp (1956), who was unable to find antibodiesagainst Pasteurella pseudotuberculosis in the bloodof 76 children with typical appendicitis. In 1,601sera obtained for Wassermann and Widal testshe found antibodies to Types I, III, and V inonly nine patients, all of whom gave a historysuggestive of mesenteric adenitis. If complicationsdo not occur, agglutinins disappear within a fewmonths. All our three cases, who remained wellafter operation, showed little or no antibody inthe blood five months later. In Burke's case therewas a rise in titre a fortnight after operation.Two and a half months later the titre had fallen

438

copyright. on 10 M

ay 2018 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.13.5.432 on 1 S

eptember 1960. D

ownloaded from

ACUTE MESENTERIC LYMPHADENITIS

considerably, and at the end of five months onlya trace reaction was obtained with a serumdilution of 1:25.The histological findings confirm the observa-

tion made by Knapp and Masshoff (1954). In ourthird case, the appearance of the two abnormalfollicles consisting of reticulum cells with centralnecrosis was most striking. The arrangement ofthe reticulum cells around the periphery of thenecrotic area was much more haphazard than onesees in sarcoidosis, and the central eosinophilictissue did not resemble caseation of tubercle.The same phenomenon was observed in Case 1,but to a lesser degree. These changes are notspecific for pseudotuberculosis since they havealso been observed in tularaemia, lymphogranu-loma inguinale, and cat-scratch fever. Neverthe-less, they appear to be a characteristic feature ofinfection with Pasteurella pseudotuberculosis andwere noted by Knapp (1959) in 86% of his cases.The epidemiology of human pseudotuberculosis

remains obscure. Pasteurella pseudotuberculosisis known to cause epizootics in guinea-pigs andturkeys, and sporadic infections have beenreported in rabbits and hares, cattle, chickens,goats and pigs, cats and dogs, pigeons, canaries,sparrows, and blackbirds. It has not been possibleto relate human cases to contact with domesticanimals or to the use of their products. In thepresent cases the only history of contact withanimals was that a budgerigar was kept by theBurke family and that 12 days before the boywas taken ill the bird sickened and remainedoff-colour for seven days, when it recovered,apparently without any ill effects. It was notpossible to obtain the bird for examination,but droppings from the cage were culturedfor Pasteurella pseudotuberculosis with negativeresults.

Little has been reported on the treatment ofhuman pseudotuberculosis. Four cases of thesevere septic type of infection were treated withsulphonamides and antibiotics and two of themsurvived (Knapp, 1959). Although antibioticshave been used for the treatment of the benignappendiceal form and as a prophylactic measureto prevent possible post-operative complications,it is difficult to assess their value, as most of thesecases run a benign course, and those treated

surgically usually make an uneventful recovery.Should the administration of antibiotics be con-sidered necessary, the results of tests in vitrosuggest that the broad-spectrum antibiotics wouldbe the drugs of choice.

It is evident from the results of this investiga-tion that pseudotuberculous mesenteric adenitisoccurs in this country as well as on the Continent.While the definitive diagnosis rests on the isolationof the organism from the lymphatic glands thepossibility of adhesions arising from routine glandbiopsy must be taken into consideration (Aird,1958). In those cases where biopsy is deemedinadvisable, or where provisional diagnosis ofacute mesenteric adenitis is made, it may bepossible to arrive at a diagnosis in manycases of pasteurella infection by carrying outagglutination tests with the patient's serumagainst the five serotypes of Pasteurella pseudo-tuberculosis.

The investigation would not have been possible butfor the interest of the consultant surgeons at theLeicester Royal Infirmary and the Leicester GeneralHospital; all kindly agreed to their patients withmesenteric adenitis being included in the survey. Foraccess to the records of Cases 1 and 2 we are indebtedto Mr. D. McGavin, and for Case 3 to Mr. E. R.Frizelle.

Dr. E. Milford Ward's observations on the histo-logical findings were of considerable value in con-firming the diagnoses. We are especially grateful toMr. B. J. Fowler for his interest and advice.

REFERENCESAird, I. (1945). Brit. med. J., 2, 680.

(1958). A Companion in Surgical Studies. Livingstone,Edinburgh.

Albrecht, H. (1910). Wien. klin. Wschr., 1910, 991.Cook, R. (1952). J. Path. Bact., 64, 228.Felsen, J. (1935). Amer. J. Dis. Child., 50, 661.Girard, G., Leger, H., Paulhet, J., and Duffau, A. (1959). Presse

med., 61, 249.Ingelrans, Soots, Poupard, and Vitse (1957). Lille chir., 12, 201.Kauffmann, F. (1933). Z. Hyg. Infekt.-Kr., 114, 97.Kirthi Singha, H. S. (1959). Brit. med. J., 2, 220.Knapp, W. (1954). Zbl. Bakt., 1. Abt. Orig., 161, 422.

(1955). Ibid., 164, 57.- (1956). Z. Hyg. InfektKr., 143, 261.

(1958). New Engl. J. Med., 259, 776.(1959). Ergebn. Mikrobiol., 32, 196.

- and Masshoff, W. (1954). Dtsch. med. Wschr., 79, 1266.Masshoff, W., and Ddlle, W. (1953). Virchows Arch. path. Anat.,

323, 664.Podhragyai, L., and Fodor, I. (1956). Orv. Hetil., 97, 277.Preston,N. W.,and Maitland, H. B. (1952). J. gen. Microbiol.,7, 117.Schmidt, J. (1959). Arch. Hyg. (Berl.), 143, 262.Topping, N. H., Watts, C. E., and Lillie, R. D. (1938). Publ. Hlth

Rep. (Wash.), 53, 1340.Westwood, J. C. N., Macpherson, I. A., and Titmuss, D. M. J. (1957).

Brit. J. exp. Path., 38, 138.

439

copyright. on 10 M

ay 2018 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.13.5.432 on 1 S

eptember 1960. D

ownloaded from