385

TYPHUS FEVER:

OBSERVATIONS ON A SEROLOGICAL TEST WEIL-FELIXREACTION.)

BY C. M. CRAIG, M.D. MANCH., D.P.H.,CAPTAIN, R. A.M.C.: O.C. — MILITARY LABORATORY;

AND

N. HAMILTON FAIRLEY, M.D.,MAJOR, A.A.M.C.; PATHOLOGIST, — AUSTRALIAN GENERAL HOSPITAL.

THE following is a brief account of our experience of the Weil-Felix reaction during the early months of 1918 in aEgypt and Palestine. The reaction is an amplification of ythe work done by Dr. Wilson,2 of Belfast, who first showed dthat typhus serum possessed the power of agglutinatingcertain intestinal organisms. During the past year it has been favourably reported on by Austrian and Germanworkers.3

Characters of organism uxed in this investigation.-This Iorganism of the strain known as X 19 is a short, B

Gram-negative, slightly motile bacillus, which growsunder aerobic conditions only. On ordinary media itthrives, producing on agar a.nd MacConkey’s media asomewhat delicate growth, rather resembling that of the a

typhoid group; it liquefies gelatin. This organism pro- duces acid and gas in glucose, saccharose, maltose, and cmannite, but does not ferment dulcite or lactose. It pro- duces indol freely, and acid without clot formation in milk.In our experience it has been non-lethal to guinea-pigs sub- a

cutaneously in doses of 2 c.cm. of 24 hours’ broth culture. 2Results of culture in typhus cases.-Cultures of blood in 12 t

cases yielded negative results. The urine, fsecen, and sputum c

have been examined in 21 cases. An organism resembling the one described above has only once been isolated-i.e., in the urine of a patient on about tenth day of illness. This differed from the stock culture (X 19) in possessing much <

feebler agglutinating properties. c

Agglutinating Phenomena in Definite Typltics Fever. :In all 90 cases were examined, 67 of which were British

and 23 Egyptians. cCertain cases yielded rather anomalous results: 1. In one <

case agglutinins were not noted until two days after feverhad subsided. 2. In the second case positive agglutination was not obtained until ninth day of disease. 3. In the thirdcase two separate examinations of serum during second week of disease yielded negative results. In all other casesagglutination reactions were noted in the first examination.

(a) Forty-five British cases (in whom time incidence of illness was definitely known) were examined during thepyrexial period. The titres of the agglutinations obtainedwere as follows :-

Day. Cases. Titre. 4th ... 6... 1/10, 1/10, 1/40, 1/50,

1/100. 1/1000.5th... 10 ... 1/10. 1/10, 1/10. 1/40,

1/50, 1/50, 1/50, 1/50,1/80, 1/160.6th... 6 ... 1/50, 1/50. 1/50, 1/80. 1/100, 1/320.7th ... 3...1/10, 1/20, 1/320.

Day. Cases. Titre. 0

I 8th ... 6 ... 1/40. 1/80, 1/100, 1/640, 0

1/1000, 1/1280.’ 9th...2 2 ... 1/20,1/40. o

10th ... 6 ... 1/160, 1/200. 1/320, a

1/320, 1/640, 1/1280.11th ... 1 ... 1/40.12th ... 2 ... 1/L0, 1/640. 15th...3 3 ... 1/80, 1/640, 1;280.

(b) The sera of 21 native cases (exact day of disease not 1definitely ascertained, probably all in second week of illness)

5

gave positive reactions as follows :-Titre ...... 1/10 120 l,’50 1/100 1’200 1/ICOO 1/2000 Cases ...... 5* 3t 2 6 2 2 1

One of these cases four days later, showed agglutination with 1/50.t All showed an agglutination of 1/50 two days later.

(c) During convalescence 24 cases were examined andagglutination reactions in the following titres were

obtained :-

Day. Cases. Titre.2nd ... 1...1/205th ... 1... 1/406th ... 2... 1/1280, 1/25607th ... 1 ... 1/3208th ... I... 1/6409th ... 2... 1/160.1/164012th ... I... 1/320

Day. Cases. Titre. Il3th ... 1 ... 1/50l4th ... 1 ... 1/20015th ... 1 ... 1,64021st ... 1... 1/10022nd ... 2 ... 1/50.1/32026th ... 1 ... 1(10027th ... 2... 1/100, 1/100

’ Day. Cases Titre. 29th 1. 1/1031st 1 . 1/100 c33rd ... 1 ... 1/50 ]34th 1. 1/100 37th ... 1 ... 1/100 t

60t .,. 1 ... Nil. (

Control reactions.&mdash;The sera of 70 known negative caseswere examined for agglutination against this proteus-likeorganism. They included cases of undulant fever, typhoid,

syphilis, relapsing fever (30), pneumonia, and malaria..

In three cases a slight reaction was obtained in a dilutionof 1/10; these occurred (1) in a healthy, much inoculatedorderly; (2) in a case of acute pneumonia; (3) in an

indefinite pyrexia, probably enterica (the serum gave acomplete reaction in dilution of 1/10 against B. para-typhosus B, and in 1/100 against B. typhosus).

Thp. sera of 10 convalescent typhus patients were testedagainst stock cultures of B. Shiga, B. Flexner-Y, B. typhosus,B. paratyphosus A and B, Cholera vibrio, M. melitensis, andtwo strains of B. coli, one true to type, the other atypical.The serum of one case agglutinated the typhoid bacillus in adilution of 1/100 (inoculated against T.A.B.). One serumagglutinated 31. melitensis in a dilution of 1/10. These werethe only exceptions; the remainder failed to give anyreaction even in the lowest dilutions.In a group of 7 additional cases of typhus which were

tested against stock cultures of B. typhosus and B. para-typlaosus A and B during the pyrexial period 6 yielded nega-tive results. In the other case, who had been inoculatedwith T.A.B. nine months previously, B.paratyphosus A and Byielded a positive agglutination with patient’s serum indilution of 1/40. B. typhosus yielded agglutination in 1/1280.An autopsy infection with B. typhosus was definitely excludedin this case.Note.-Tested in a dilution of 1/10 against standard

immune serum for B. Shiga, B. Flexner, B. typhosus, B. para-typhosus A and B, and the cholera vibrio (prepared by theLister Institute), emulsions of this bacillus gave completelynegative results.

Cumlvlement-fixatioza Reaction.As the sera of typhus patients agglutinated in high titre

a saline suspension of this proteus-like organism, an investi-gation by means of the complement-fixation method,originally devised by Bordet and Gengou,4 was made.The antigen used was a fresh saline (0’85 per cent. saline

and 0’5 per cent. phenol) suspension prepared from a

24 hours’ growth of this organism on agar slopes. Thetechnique employed was similar to that used in the ordinaryquantitative Wassermann reaction. Six and 12 minimumhsemolytic doses of complement were used in this test. Aserum was regarded as yielding a positive result if in thepresence of this bacterial antigen it fixed at least six M.H.D.of complement. In the first stage of the reaction quantitiesof antigen, (?) immune serum, and complement were mixedtogether for one hour at 37&deg; C. Subsequently, sensitisedsheep’s corpuscles were added, and final readings were madeafter another hour’s incubation at 37&deg; C. The amount ofcomplement fixed by pooled negative sera plus the amountof bacterial antigen used in the test did not exceed two M.H.D.of complement.The results were recorded as follows:&mdash;1. P ++++ fixation

of 12 M.H.D. complement. 2. P +++ fixation of 9 M.H.D.l 3. P ++ fixation of 6 M.H.D. 4. P + fixation of 3 M.H.D.l Only the first three readings are regarded as positive.;

In all, 29 sera collected from typhus cases in different! stages of the disease were examined.

Their agglutination reactions were as follows: 1 case,1/2560; 5 cases, 1/1280 ; 7 cases, 1/640; 4 cases, 1/320; 3 cases.1/160 ; 3 cases, 1/80 ; 3 cases, 1/40; 3 cases, 1/20. The resultsof the complement-fixation test were negative in 28 outof 29 cases. In the remaining case a positive reaction

’

(P++++) was obtained. Even in this case the significanceof the reaction was doubtful, as this serum exhibited distinct

. anticomplementary tendency in the serum control tubes.Controls.-Forty-three sera from other cases (not typhus)

were examined. Of these, 40 were negative and 3 yieldedpseudo-positive reactions (one case of rheumatic fever,positive reaction, P++++; 2 cases of syphilis positivereaction, P ++).The formation of agglutinin for this proteus-like organism

in the blood of typhus cases is not accompanied by theformation of immune body. as indicated by the complement-fixation reaction using as antigen a fresh saline suspension ofthis proteus-like organism.

Animal Experimentation.,

Ten monkeys were experimented on-7 grivet (Ceroo-patheeus) and 3 Macacus rhesus.

Of these animals four were used as controls. The six0 others were injected subcutaneously with one dose of

1 to 2 c.cm. of a 24 hours’ broth culture of the proteus-like0 organism. In one monkey a second injection was given0 ten days after the first. Prior to inoculation all had negative

complement-fixation reactions and all yielded negativeagglutination reactions in dilution of 1/10 against this

s proteus-like organism with one exception only. In thise case a positive agglutination of monkey’s serum in a titrel, of 1/10 was obtained, but the reaction vanished in a higher

dilution of 1/20. None of the monkeys showed signs ofn sickness or abnormal temperature after inoculation.d In all cases a positive complement-fixation reaction wasn obtained; in five cases within 12 days of inoculation. In the

386

sixth case the complement fixation, negative on the ninth m:

day, became positive on the eighteenth day. The agglutinin agresponse in this case was not as intense as usual. In all withese cases, accompanying the appearance of agglutinin in an

the blood serum of the inoculated monkeys, immune bodies thwere produced also, which had the property of fixingcomplement in the presence of its specific antigen- ,,

namely, the proteus-like organism. The following gives R.the serological reactions in one of the experimental T(animals (Monkey Z):&mdash; A

(A) Days after inoculation. (B) Agglutination. (C) Complement ca

fixation. w

(A)... 4 6 9 12 18 35 53

(B)... 1’20 1/160 1640 1.1280 11280 1640 1,40 of(C) ... - P++++ &mdash; P++++ P++++ P++++ Neg. W

Significance of Agglutination Reactions in Typh2cs Cases. At the present time Austrian and German observers are 14

agreed that this proteus-like organism must of necessity be 8Ea constant secondary invader in patients infected with 5.

typhus fever, in much the same fashion as certain authori-ties regard a secondary invasion with the Streptococcuspyogenes as occurring in scarlet fever. Such a conception,however, though explaining the appearance of agglutinin inthe blood of typhus cases, is incompatible with the followingfacts..1. Systematic cultural investigations of the urine and blood in cases’of typhus fever over all stages of the diseasedo not yield, except very exceptionally, positive results forthis proteus-like organism, and this despite the fact that itgrows easily on all ordinary laboratory media-i.e., nutrientbroth, agar, &c.

2. Serological investigations of typhus cases, while show-ing the presence of agglutinin for this proteus-like organism, 1failed to demonstrate the presence of any immune body ; whereas serological investigations of monkeys inoculated E

subcutaneously yield both positive agglutination and com- Iplement-fixation reactions. If the "secondary invasion" theory were correct, then inevitably the immune body would (be demonstrable in the sera of typhus cases.

This agglutination phenomenon, then, though yielding (results of the greatest clinical value, is to be regarded as a (

pseudo-agglutination. Possibly it is of the nature of a group agglutination, and, if so, then the virus causing typhus mayprove to be an organism allied to this proteus-like bacillus,but one which is resistant to cultivation on ordinary media.The fact that the Wassermann reaction is not a specific

test for syphilis in the strict immunological sense has notdiminished its practical application as a diagnostic test forthat disease. In similar fashion the appearance of an

agglutinin in the sera of typhus patients for an organismwhich has no apparent relationship to the disease need notbias the student against its great diagnostic value in typhusfever.

Conchusions.

1. This agglutination test is an invaluable aid in the

diagnosis of typhus fever.2. Frequently the reaction appears only in the disease,

and as a rule rapidly increases in titre.3. A rapid agglutination in a dilution of 1/10 on

Garrow’s agglutinometer, while not absolutely reliable, is

sufficiently suspicious to justify the isolation of the case.This is especially true of individuals uninoculated withT.A.B. or cholera. The serum of 50 uninoculated nativesfailed to give a reaction even in a dilution of I,/10. If thetest be repeated at intervals of two days confirmatoryevidence will, in the great majority of cases, be obtained.

4. This test is particularly valuable in the case of nativesin whom it may be difficult to distinguish clinically betweensevere relapsing fever and typhus.Note.-The agglutination was carried out in all cases on

Garrow’s agglutinometer; the results were read after fiveminutes’ rotation and controls were used in every instance.We have found Garrow’s5 agglutinometer of invaluable

service and absolutely reliable. With this instrument,allowing five minutes for rotation, we find a dilution of1/25 diagnostic. It is essential in order to obtain scientificand uniform results from a test of this description that astandardised and uniform technique should be adopted.The method we have described has four great advantages :

1. The rapidity with which results can be obtained is neces-sarily of great value to the clinician, so as to enable speedyisolation. 2. The elimination of the feeble agglutinatingpower of normal serum in low dilutions upon this organism.This seems to take effect more readily if placed in tubes in

mass contact with the organism than with the Garrowagglutinometer. 3. Economy in material, and the facilitywith which the technique can be performed in other than inan elaborately equipped laboratory. 4. The ease with whichthe agglutination can be read and its limits ascertained.Our thanks are due to Captain P. H. Bahr, D S.O.,

R.A.M.C., for assistance during the present investigation.To Colonel R. Fowler, A.A.M.C., commanding officer of -Australian General Hospical, we are indebted for facilities incarrying out the experimental work on the section dealingwith animal inoculation and complement-fixation.

References&mdash;1. Brit. Med. Jour., Jan. 12th, 1918. 2. Wilson: Journalof Hygiene, 1910, x,, 155; 1909, ix., 306. 3. Weil, E., and Felix, A.:Wien. klin Woch., 1916, xxix. (No. 2), 33-35. Vitecek, V.: Ibid, 1917,xxx. (No. 31), 967-972. Sterling, S., and Sterling, K.: Ibid., 972-974.Ballner, F., and Finger, A.: Ibid., 966-967. Dadej, K., andKrahelska, M.: M&uuml;nch, med. Woeh., lxiv. (No. 42), 1379-1381 (No. 43),1408-1409. Schuerer, T., and Stern, W.: Ibid,, 1917, lxiv. (No. 27),886-887. 4. Bordet and Gengou: Ann. de l’Inst. Pasteur, 1901, xv.5. Garrow, R. P., Capt. R A.M.C.: THE LANCET, 1917, i., 262.

NOTES ON

THE INTESTINAL PROTOZOALINFECTIONS

OF 1680 MEN EXAMINED AT THE UNIVERSITY WAR

HOSPITAL, SOUTHAMPTON.

BY DORIS L. MACKINNON, D.Sc.,LECTURER IN ZOOLOGY, UNIVERSITY COLLEGE, DUNDER.

’WHILE acting as protozoologist at the University War’. Hospital, Southampton, from May, 1917, to March, 1918, 1

examined in the course of routine work the stools of 1680- patients. The results of my observations may be of interest" to other workers at dysentery hospitals for three reasons:

(1) the cases were examined as often as six times each ;(2) while 914 of them were admitted as convalescent from

g dysentery (including colitis and enteritis) and severe

a diarrhcea, the other 766 were non-dysenterics, diagnosedp typhoid, paratyphoid, trench fever, P.U.O., and so on;

and (3) with less than 75 exceptions this large body of menhad been invalided from France or had not been out of

’

Britain.1. Number of Examinations.

)t The attempt was made, and, thanks to the excellent)r arrangements at this hospital, very successfully made, to6n examine each case at least six times before pronouncing

him uninfected by Entam&oelig;ba histolytica. Since the demandon accommodation for in-coming patients was always great,

1s and since many of the men left the hospital after but abrief stay, the six examinations were made in the shortestworkable period-i.e., a week-the first and sixth examina-tions corresponding with the two examinations required bythe Army Council Instruction No. 205 of 1917.

es It will be seen from the above that the six-examination

he standard was well maintained. This is the first time, so faras I know, when a large number of cases has been systemati-cally examined so often. 2 That the attempt to make somany examinations was justified is indicated by the follow-en ing table of the findings of E. histelytica. The percentagesof the total findings for which each examination is responsible

on will be seen to be closely in accord with those in the tablesve arrived at by calculation (Carter, Mackinnon, Matthews, and

Smith 1917).

1 The men were all questioned as to their previous travels, and wefound that 447 had been at one time or other in tropical or subtropicalregions.

2 Dobell (1917) first pointed out the advisability of aiming at makingsix examinations per case. Carter, Mackinnon, Matthews, and Smith(1917) showed that it was well worth while to attempt this largenumber, since they calculated that three examinations reveal only65 per cent. of the E. histolytica discoverable by six. The work doneby Jepps and Meakins (1917) at the Hampstead Military Hospital is theonly instance as yet on record in which the six-examination standardwas consistently maintained throughout the investigation, but only65 cases were examined.