Analysis of forty-six cases of actinomycosis with special reference to its etiology

ANALYSIS OF FORTY3IX

CASES OF ACTINOMYCOSIS

WITH SPECIAL REFERENCE

TO ITS ETIOLOGY*

M. I. JEFF DAVIS,M.D. Assistant

in Clinica

DermatoIogy

and Visiting

DermatoIogist,

NASHVILLE,

T

Definition

* Studies

SchooI of Medicine

In a ten year period, 1926 to 1936, there were forty-six cases of actinomycosis in the University HospitaJ. Thirty-two of these cases were confirmed either by finding suIfur granuIes in exudate from the lesions and the demonstration microscopicaJJy of the ray fungus, or by simiIar findings of the cIub-bearing fiIaments in biopsy specimens or at autopsy. A foIIow-up study was made and twenty-two patients were heard from. GEOGRAPHIC DISTRIBUTION In Michigan, as shown by Sanford’s” compIete survey, there is not the greatest incidence of the disease, but it is Iocated in that portion of the Northwest where there are a greater number of cases seen than in any other part of the country, except the states of New York and Massachusetts. However, it is probabIe that this greater prevaIence of the disease in any one Iocality is onIy apparent and may be due to greater interest in and famiIiarity with the disease in that particuIar IocaIity. * McKenty” noted that a11 but severaJ of his thirty-seven cases came from within a radius of one hundred miIes of MontreaI. EarIier writers have suggested this endemic nature in reports of actinomycosis in cattIe where certain herds of cattIe fed in particular salt marshes. This endemic character of the disease has aIso been noted where the infected animaIs were known to be feeding on a certain batch of grain (Jensen); but Johne” in 1888 fed cattIe on food infected with cuItures of the organism and none became infected. * F. author.

given the author by F. T. Lord.

and contributions

University

TENNESSEE

HE Actinomyces bovis has been estabIished by Wright1 since 1905 as organism of actinothe causative mycosis. However, there &II remain many points unsoIved in its pathogenesis. BioIogicaI research has continued to make new contributions2 in an attempt to answer some of these debatable points. Yet, in the light of this newer knowIedge there has been no comprehensive cIinica1 study of any Iarge group of cases of antinomycosis in this country for the past ten years. In view of these facts it was considered an appropriate time to review the cases of actinomycosis occurring in the University HospitaI in the past ten years. It wiII be the purpose of this paper to anaIyze these cases with specia1 reference to: (I) their bearing on the mode of infection of the disease, (2) the resuIts of treatment especiaIIy in the cervica1 cases and (3) to bring up to date our knowIedge of the pathogenesis of actinomycosis. Actinomycosis is an infection characterized by the usua1 presence of Iesions composed of abundant granuIation, or connective tissue surrounding muItipIe areas of abscess formation and by the presence within the abscesses of branching, Gramstaining, fiIamentous, anaerobic organisms with radiaIIy-disposed, cIubbearing fi1aments.t The disease has been fuIIy described by many observers. Ruhrah3 coIIected fifty-eight American cases in 1899. This number was brought up to a hundred by Erving3” in 1902. In 1922 Sanford4 was abIe to coIIect aImost 700 unreported cases in this country. t

Vanderbilt

T.

Lord:

Persona1

from the Department of Dermatology and SyphiloIogy, University of Michigan Medical School. 447

communication

Service

to

the

of Dr. U. J. Wile,

448

American

Journd

Davis-Actinomycosis

of Surgery

Our group of cases was fairly we11 distributed over the state with a few cases from nearby states. There was no evidence of endemicity. An interesting point was the fairly even distribution between rura1 and urban popuIation, a point which has not been mentioned in other reports. AGE

AND

SEX

IN

DISTRIBUTION OTHER

WITH

THAT

GIVEN

REPORTS

I 1

TABLE THAT

TABLE I AGE

Among the cervicofacia1 cases were two primary infections of the tongue, previousIy reported from this department.6 In two instances actinomycotic-like organisms were found on pathologic examination of tissue, occurring as a secondary invaderat the edges of a wound. COMPARISON

Most writers agree in describing actinomycosis as occurring in middIe age. A comparison of age distribution is made in Table I with that reported in other groups. Eighteen, or thirty-nine per cent occurred in persons past midlife which seems to be an unusuaIIy high percentage of occurrence in people in the higher age bracket. COMPARISON OF

JUNE, ,941

OF

OF OTHER

New and Figi.. McKenty., Erving. University HospitaI..

” 70

IOO

64.

46

*One hundred nineteen six Mayo Clinic cases.

CASES

RESPECT

TO

WITH

ANATOMIC

DISTRIBUTION I

I

No. Cases

Name

Sanford and Magath*.....

I

Head and Neck

IIg 1 62

CoIIer . New and Figi, McKenty Erving. University HospitaI.

96 1:; 37 Ioo 46

* One hundred nineteen six Mayo CIinic cases.

. coIIected

60

” 60

.

39

cases and ninety-

As in other reports there was a marked preponderance of maIes over femaIes, a we11 accepted characteristic of the disease. ANATOMIC

IN

I

Thorax

Abdo, men

16

9

4

3

64 %I% I9 53

;% 2 20

23

9

I4

23

Skin

iH.3%

Age, Per Cent

I Ig 96 t 107 37

HOSPITAL

REPORTS

MODE

Sanford and Magath*

II

UNIVERSITY

DISTRIBUTION

It is generaIIy conceded that the majority of actinomycotic infections occur in the cervicofacia1 region. This is we11 demonstrated in the accompanying comparative TabIe II, showing the sites of eIection for the disease as found by other observers. In our group there were equaIIy as many cervicofacia1 cases as there were thoracic and abdomina1 cases together.

OF

colIected

cases and ninety-

INFECTION

For thirty years two wideIy different views have been prevaIent concerning the mode of infection obtaining in actinomycosis. On the one side, there are those who beIieve that the organism gains entrance to the body through grasses, grains, etc., upon which the organism Iives in the outside worId and by means of which the infection is carried to the tissue. Contact with “ Iumpy jaw ” in animaIs is aIso considered a source of infection. More recentIy vegetabIes packed in straw have been suggested as another possibiIity.7s8 This theory is upheId by the foIIowers of Boestrom, who in 1890 isoIated a streptothrix, which was apparentIy identical with an aerobic organism found widespread in nature, upon grasses and grains. This theory is further supported by the fact that many cases are reported in which barbs of barIey or grain have been found in the Iesions of men and animals, and many instances in which there is a history of swalIowing a bIade of grass

NEW SERIES VOL. LII, No 3

Davis-Actinomycosis

The foIIowers of this theory aIso take the reports of instances of endemic actinomycosis among cattIe fed on certain fieIds of grain as further evidence in support of their theory. There is an opposing theory, based on the we11 known and accepted work of IsraeI and W01ff,~ Iater Wright,1 and others, demonstrating that the organism causing true lesions never grows on actinomycotic grasses and grains and has quite different cuItura1 characteristics from that of Bostroem. l1 Actinomvces bovis is a facuItative anaerobe, diffIcuIt to grow, grown onIy at body temperature. These facts have given rise to the view upheId by WoIff and Israel,!’ and Homer Wright,’ as we11 as by Lord,lO and Naeslund,” that the Actinomyces bovis does not have its usua1 habitat in the outside worId on grain and grasses, but normaIIy inhabits the digestive tract, particuIarIy the mouth, where it remains as a saphrophyte. Given the proper conditions of a Iower resistance and porta of entry, the organism becomes pathogenic with the production of a Iesion.

American

Journal

of Surgery

449

two gave history of contact with infected animaIs. Mattson7 is of the opinion that infection is acquired from animaIs in a goodIy number of instances. In twenty-six answers to questionnaires he found fifteen patients gave a history of treating animals with “Iumpy jaw.” New and Figi13 Iikewise through eighty answers to a questionnaire reported that whiIe thirty-five or 43.7 per cent had been associated with the disease in animals, forty-five or 56. I per cent had not come in contact with diseased cattIe. However, the time eIapsing from the time of of onset of exposure, to the recognition symptoms, averaged five and one-haIf years, varying from six months to twentyfive years. They concIuded that the probabiIity of direct infection from cattle was questionabIe. Rather, they feIt that the presence of “Iumpy jaw” in cattIe in the neighborhood from which patients with the infection came, probabIy indicated that the organisms were abundant on the vegetation of the Iocality, rather than that the diseased animaIs were the primary cause of the infection in man. Contagion from man to man has never OCCUPATION been definiteIy estabIished. l4 Ochsner’s14 It is commonIy stated that the disease is two we11 known cases are usuaIIy cited as more common in those handIing hay or exampIes of this occurrence. In one instance those who come in contact with animals the man drove a horse with “Iumpy jaw,” affIicted with “lumpy jaw.” However, in in the other the patient had handIed inreviewing the reports with this particuIar fected cattIe. There is one case reported in point in mind, it wouId seem that the actua1 which infection is presumed to be transfacts do not substantiate this generally ferred from man to man. This is McKenty’s’ accepted impression. case of a nurse who was infected by adSanford and Magath12 reported on I 19 ministering to a patient with actinomycocoIIected cases, and ninety-six cases from sis. But McKenty himseIf states that the the Mayo CIinic; of their I 19 coIIected proof was not convincing. cases only sixteen were farmers (oneThough Mattson7 Ieans toward the frequarter not noted). In Sanford’s4 678 coI- quency of the transmission of the disease Iected cases, 50 per cent onIy were in from animaI to man, he cites SaImon’s contact with cattIe or in occupation which experiment of pIacing twenty-two infected predisposed them to infection. FurthercattIe with healthy cattIe for four months, more, there was no definite reIationship without any of the weI1 cattIe becoming noted between the distribution of the disinfected, which is certainly convincing eviease in the entire country and the distribudence against infection by contact. As tion of the prevalence of diseased cattIe. In CahiII15 has stated, the disease is not Mckenty’s5 thirty-seven cases, nineteen being exceedingly diffIcuIt to contagious, were farmers, one a grain merchant and

430

American

Journal

of Surgery

Davis-Actinomycosis

produce the disease by injecting infected pus into animaIs. OnIy fifteen of our forty-six cases were farmers, or Iess than 35 per cent. The remainder were divided among laborers, professiona men, mechanics and cIerks. EIeven were cIassified as unempIoyed. A questionnaire was sent to twenty-six of our patients in which they were asked the question. Had they at any time come in contact with animaIs who had “Iumpy jaw” or simiIar infections? To this question sixteen answers were received, al1 in the negative. HABIT

OF

CHEWING

GRASS

That actinomycosis is contracted through the introduction into the ora cavity of straw, grain and grasses which are infected, has been handed down to medica students through cIassroom and textbooks for the past fifty years. There are numerous instances in the Iiterature in which infection is said to have foIIowed the chewing of grass, or the swaIIowing of a bIade of grass, or again where infection has occurred through injury of the oral mucous membrane by a barb of barIey or beard of grain. Such is Cope’s8 case of a chiId of ten, who deveIoped an actinomycosis of the jaw after swaIIowing a bIade of grass which stuck in the tonsi1. Soltman’ had a simiIar case in which a youngster deveIoped actinomycotic abscesses in the chest after swaIIowing a barIey awn, which was Iater found in the pus from the abscess. New and Figi13 reported the history of a foreign body in five of their cervica1 cases, and there have been simiIar reports by others. In a11 these reports, however, it must be noted that such instances onIy make up a very smaI1 percentage of any one group of cases. CoIebrooklB and Lord2 have both expressed skepticism, regarding any view which considers the infection as being introduced through such a mechanism as these cases wouId indicate, as the organism, which is commonly found on vegetabIe sources, is an aerobic fungus, not conform-

JUNE,

rgjr

ing bacterioIogicaIIy to the type of organism used to describe this group. The anaerobic organism of the type of Actinomyces bovis has never been demonstrated unquestionabIy in natura1 sources outside the human body; they postuIate, therefore, that anyone who adheres to the idea that the organism of actinomycosis is introduced into the body by vegetabIe matter, must as to presuppose his own hypothesis how this organism existed before entering the human body. There is no bioIogica1 evidence to substantiate any of these hypotheses, such as existence of a spore formation in the Iife cycIe of the organism. Mattson’ is the only one who gives definite figures on patients concerning their habit of chewing grass or grain. He states that of twenty-six patients asked this question in a circuIar Ietter, eIeven answered in the affirmative. Of our forty-six cases there were only three who gave a history of chewing grass or straw. CircuIar letters were Iikewise sent to twenty-six patients, and in answer to the question, “Have you ever been in the habit of chewing on straw or grass? ” there were none who admitted this habit. Epsteinl’ has recentIy stated that the theory that the disease is contracted by chewing straws, grasses, etc., is not we11 founded, and from a cIinica1 standpoint the hypothesis in many cases does not fit the cIinica1 history. Our findings bear substantiation of this cIinica1 impression which has aIso been voiced by others.2*10~12Thus it may be that a very old idea handed down through the textbooks might need revision. DENTAL

ORIGIN

Homer Wright1 showed that the true actinomycosis described by IsraeI and W01ff,~ in 1878, is not found outside of the anima1 body, and the etioIogica1 relationship of the organism of the WoIff-IsraeI type to actinomycosis is now regarded as estabrished. For this reason WoIff and Israel, as we11 as Homer Wright,’ Lord,*” and NaesIund, l1 have at various times expressed the belief that the Actinomyces

Nliw Smr~s

VOL. LII, No. 3

Davis-Actinomycosis

bovis is a natura1 inhabitant of the gastrointestinal tract, especiaIIy of the mouth. Origin of the disease from organisms normaIIy harbored in the buccat cavity is further suggested by finding in carious teeth and tonsiIIar cysts of norma peopIe, smears, and serea1 sections showing organisms with the morphoIogy and staining reaction of actinomyces. In examination of 10,000 tonsiIs WiIkson@ found bodies which resembIed those found in cases of actinomycosis in 177 cases. Actinomycosis occurred second in order of frequency among the pathoIogic conditions found in the tonsiIs examined. These were believed to be of the non-pathogenic variety, however. WelIerlg in examination of 1,000 tonsils found forty-seven out of eighty positive cases of tonsiIIar concretions of the crypts to be actinomyceslike coIonies of mixed mouth organisms. LordlszO noted in the sputum from patients without cIinica1 evidence of actinomycotic infection, the occasiona presence of organisms with the morphoIogy and staining reaction of actinomyces. He Iikewise observed that patients with actinomycosis often gave a history of previous troubIe with their teeth. Knowing the failure of actinomycosis to grow at other than body temperature and the Iack of evidence of its contagiousness, the fact suggested itself to him that the disease arises from within rather than from without the individua1. Thus in 1910 LordzO inocuIated guinea pigs intraperitoneaIIy with the contents of carious teeth of patients without actinomycosis and produced omenta1 tumors in 60 per cent, histoIogicaIIy identica1 with actinomycotic tissue, and containing typical club-bearing actinomyces granuIes. LordzO Iater noted that simiIar Iesions were produced in animaIs inocuIated with the scrappings from sound teeth. An attempt to isoIate the organism in pure cuIture, however, faiIed. These findings furnished evidence in favor of the bucca1 cavity as a normal habitat of Actinomyces bovis. To prove the identity, however, of the organ-

isms found in the mouth with Actinomyces bovis, it wouId be necessary that pure cultures be obtained, as certain strains of streptothrix are capabIe of forming similar appearing coIonies when inocuIated into animals. It was this point in the faiIure of the experimenta work, to which Mattson’ pointed in 1922 as showing the unIikIihood of true actinomycosis being present in the mouths of norma persons. However, in 1925 NaesIund” confirmed Lord’s findings, and produced experimentaIIy in animaIs, Iesions histoIogicaIIy identica1 with actinomycotic tissue and containing cIub-bearing a&nomyces granuIes, by the inocuIation of animaIs with materia1 obtained from the norma mouth. He was further successfu1 in isoIating from the norma mouth anaerobic organisms morphoIogicaIIy and cuIturaIly Iike the actinomyces of the Wolff-IsraeI type. He was Iater abIe to produce with these cuItures Iesions in animaIs characteristic of actinomycosis. This establishes the fact that antinomyces of the WoIff-IsraeI type is present in norma mouths and is a Iong step forward toward substantiating Wright’s theory that the organism is a norma inhabitant of the 0raI cavity. That the disease may arise from organisms normaIIy found in the ora cavity, has been suggested by a number of writers and investigative workers. Many observers have noted and commented upon the greater frequency, with which the disease occurs in the head and neck than in any other portion of the body. Likewise there has been noted the unusuaIIy Iarge number of cases in which carious teeth are found present in the mouth, or in which there was a previous history of some diffIcuIty with the teeth. Sanford and Magath13 in ninetysix cases reported from Mayo’s, noted in aImost a11 cases, a history of preceding tooth or tonsil troubIe. They concIuded that the aIimentary tract pIayed a very important part in the source of infection. Ruhrah3 noted that carious teeth, in which the Iower jaw is invoIved, seemed to pIay an important rBIe. In McKenty’sj thirty-seven

452

American Journal of Surgery

Davis-Actinomycosis

cases, of those in which a history was given, there were eight following extractions or infected teeth. New and Figi13 concIuded that the organism of actinomycosis is probabIy a common inhabitant of the normaI mouth. CoIIeP Iikewise noted the finding of carious teeth, and that the infection was not uncommon folIowing tooth extractions. Warwick22 describes the case of a woman who a few months foIIowing aspiration of a tooth deveIoped a thoracic actinomycosis. There are aIso two cases, one of Cope* in EngIand, and one of McWiIIiams’ in this country, in which actinomycosis of the finger deveIoped in each instance, foIlowing a human bite by a person without actinomycosis. In seven of our cases, infection was preceded by onIy a few days to a few weeks by the extraction of teeth. In one case there was a history of an abscessed tooth, and in another instance actinomycosis was found in the pathoIogica1 examination of the tonsiIs without evidence of infection elsewhere in the body. It is usua1 to consider that some break in the tissue precedes the infection and acts as a porta of entry. There was, however, an unusuaIIy Iarge number of cases in which it was impossibIe to point to a break in the skin, or mucous membrane preceding infection. Eighteen cases, or 39 per cent of the group may be incIuded in such a cIassif?cation. In four abdomina1 and thoracic cases, and three cervica1 cases, there was a history of onset of infection folIowing a severe bIow to the body but without causing an actuaI break in the continuity of the tissues. Six abdomina1 and thoracic cases were preceded by appendectomy, a we11 known occurrence. It wouId seem that either a microscopic break in the tissue is necessary or that the tooth in many instances acts as an area of Iowered resistance. RESULTS

OF TREATMENT CERVICAL

IN THE

CASES

It is the genera1 experience of most observers that the prognosis in the cervica1

JUNE,rg‘+r

cases is far more favorabIe than in the abdominal and thoracic infections. McKenty5 treated nineteen cervica1 cases; of these four were not traced, two showed no improvement and one was chronic over an eight-year period, dying of a chronic nephritis with freedom of nine months from actinomycosis. Of twenty-one cervica1 cases treated by Mattson’ tweIve were entirely cured, seven were improving when heard from. His mortaIity in appendicea1 and puImonary cases was IOO per cent in six months to a year. New and Figi13 traced eighty-five of their patients of whom thirty-six were we11 for one to five years. Seven had died, and in twenty-four the condition was not stated. For numerous reasons, the interva1 between the time the patient was aware of symptoms and the time that he presented himseIf for treatment was in most instances quite Iong. Only one patient was seen in which the infection had been present Iess than a month. Eight were not examined unti1 the Iesion had been present six months or Ionger. The interva1 varied in the remaining seven patients from one to six months. A ten year foIIow-up study was made on twenty-two of our cases. This consisted of thorough re-examination in the cIinic and when this was impossible by foIIow-up letters. The resuIts are striking. Of the tota twenty-three abdomina1 and thoracic cases originaIIy seen, a11 but one had died at the end of fourteen months, and the one stiI1 Iiving had active Iesions in the chest at the end of one and one-haIf years. Of those suffIcientIy we11 to Ieave the hospital, seven months was the average Iength of time before death. ResuIts in the cervica1 cases present a brighter picture. Of the tota twenty-three cervica1 cases treated at the University HospitaI during the past ten years fifteen were folIowed. Of these a11 but one was cured on an average of two and one-half months after discharge from the hospital. The Iength of time required for heaIing varied from ten days, foIlowing discharge

Davis-Actinomycosis

NEW SERIESVOL.LII, No. 3

from the hospita1, to six months. One patient died of a condition unreIated to his infection, which was heaIed at the time of death. One patient had a Iesion present on the cheek for ten years and finaIIy died with generaIized metastases.

ON 22

OF TREATMENT CASES.

OF 46

PATIENTS

DATA ON RESULTS

Name

McKenty ............................ Mattson. ............................ Figi., ............................... University HospitaI. ..................

WlTH

Journal

of Surgery

453

For these reasons the bioIogist beIieves the organism of actinomycosis to be a natura1 inhabitant of the digestive tract, especiaIIy the mouth, and beIieves that the infection is introduced into the body from outside sources. There is no bioIogica1 evi-

TABLE RESULTS

American

III

ACTINOMYCOSIS

OF TREATMENT

IN OTHER

WITH TEN YEAR LARGE

GROUPS

FOLLOW-UP

OF CASES

STUDY

IS GIVEN

Total

19 21

85 46 Abdominal thoracic

.

I

,

22

* Actinomycosis of cheek present for ten years. Died of generatized spread. There were two cases of several years’ cure, in which there was death from entireIy unretated causes. These cases were considered as cured.

Treatment in the University HospitaI both on the dermatoIogica1 and surgica1 services consists of LugoI’s wet dressings, iodides to toIerance, pIus x-ray and surgica1 drainage when indicated. In reviewing the resuIts of these cases the maIignant nature, the frequency of metastases and the high mortaIity in abdomina1 and thoracic cases as compared with cervica1 infection showed the same disproportion as given in other reports. SUMMARY

The theory that actinomycosis is contracted by chewing straws, grasses, etc., is at variance with the present day bioIogica1 and bacteroIogica1 concept of the mode of infection of this disease. An attempt has been made to correIate cIinica1 findings with the facts now we11 established by the laboratory. It is now a definiteIy estabIished and accepted fact that the true causa1 organism of actinomycosis is an anaerobe, never found growing in the outside worId. Furthermore, it has been cuItured from the mouths of norma individuaIs with the subsequent production of typica actinomycotic infection in the tissues of Iaboratory animaIs.

dence to support the hypothesis that actinomycosis is introduced into the body by vegetabIe matter. Findings in our own cases and in that of coIIected materia1 wouId tend to show that the habit of eating grass, the proximity to infected animaIs, and specia1 types of occupation, things which we have heretofore associated with the mode of infection of actinomycosis, are probabIy present considerabIy Iess than 50 per cent of the time. AIthough suficient materia1 has not been presented upon which to make any definite concIusions, clinica findings have been given which wouId at Ieast question whether there is need for a revision of our ideas of the mode of infection of actinomycosis. Actinomycosis is stiI1 a reIativeIy rare infection, there being onIy forty-six cases seen in a period of ten years, or Iess than five cases per year, and this in a portion of the country in which actinomycosis is purported to be reIativeIy prevaIent. The fairIy even distribution of actinomycosis between the rura1 and urban popuIation is noted for the first time. A reIativeIy good prognosis is indicated in the cervicofacia1 cases as evidenced from the resuIts obtained in treatment of such

Davis-Actinomycosis cases whereas the outlook in thoracic abdomina1 cases is extremeIy grave.

and

REFERENCES I. WRIGHT, J. H. The biology of the microorganism of actinomycosis. J. M.-Research, 13: 349: 1905. 2. LORD. F. T. and TREVITT. L. D. Pathonenesis of actinomycosis; recovery of actinomycosis Iike organisms from normal mouth. J. Infect. Dis., 58: 115-120, 1936. 3. HEKTOEN, LUDWIG. Internat. Clin., 4: I 10-121, 1900. 3a. ERVING, W. G. Actinomycosis hominis in America with report of six cases. Bull. Johns Hopkins Hosp., 13: 261-268, 1902. 4. SANFORD, A. H. Distribution of actinomycosis in U. S. J. A. M. A., 81: 655-659, 1923. 5. MCKENTY, F. E. Study of cases of actinomycosis. Am. J. Med. SC., 145: 835-836, 1913. 6. CAMERON, 0. J. Primary actinomycosis of tongue. J. A. M. A., gg: 1146-1150, 1932. 7. MATTSON, W. W. Human actinomycosis with speciaI reference to source and mode of infection. Surg., Gynec. Ed Obst., 34: 482-494, 1922. 8. COPE, V. Z. A CIinicaI study of actinomycosis with illustrative cases. Brit. J. Surg., 3: 55-81, 1915. Q. WOLFF, M. and ISRAEL, J. Vircbows Arch. f. Path. Anat., 126: II, 1891. IO. LORD, F. T. EtioIogy, pathogenesis, and diagnosis of actinomycosis. M. Clin. N. America, 16: 82~844, 1933.

I I. NAESLUND, C. Acta. Path. et hilicrobiol., Scandenar., 5: 334, ‘929. 12. SAYFORD, A. H. and MAGATH, T. B. Etiology and laboratory diagnosis of actinomycosis. Minnesotct M., 5: 71-80, 1922. 13. NEW, G. B. and FIGI, F. A. Actinomycosis of head and neck. Surg., Gynec. CY Obst., 37: 617-625, 1923. 14. Oxford System of Medicine. VoI. v, part I, chap. xrv, 397. 15. CAHILL, J. A., JR. Actinomycosis. Virginiu &I’. Month., 55: 233-237, 1928. 16. COLEBROOK, L. The mvceIia1 and other microorganisms associated &ith human actinomycosis. Hrit. J. Exper. Path., I: 197, 1920. 17. EPSTEIN, N. California Ed West M. J., 30: 395-399, ‘929.

18. WILKIXSON, H. F. PathoIogic changes in ton&; study of 10,000 pairs of tonsils, with special reference to presence of cartiIage bone, tubercuIosis, and bodies suggestive of actinomycosis. Arch. Otolaryngol., 10: 127-151, 1929. rg. WELLER, C. V. Incidence of pathogenesis of tonsilar concretions. Am. Otol., Rhinol. +Y Laryngol., 33: 79-r 19, 1924. 20. LORD, F. T. A contribution to the etiology of actinomycosis. Boston M. c~ Surg. J., 163: 82-65, 1910; J. A. M. A., 57; 1911. 21. COLLER, FREDERICK. J. Michigan M. Sot., 23: 267-372, 1924. 22. WARWICK, W. T. Lancet, 2: 497-501,

Etiology 1923.

of

actinomycosis.