downloaded file 46

8/12/2019 Downloaded File 46

1/11

J. Med. Microb iol. ol. 48 (1999), 789-799999 The Pathological Society of Great Britain and IrelandISSN 0022-26 15

REVIEW ARTICLE

Erysipelothrix rhusiopathiae: bacteriologyepidemiology and clinical manifestations of anoccupational pathogenC. JOSEPHINE BROOKE* and T H O MA S V . R ILEY ?Department of Microbiology, University of We ste rn Australia and Division of Microbiology and InfectiousDiseases, We stern Australian Centre for Pathology and Medical Research, Nedlands 6 9 WA , Australia

Erysipelothrix rhusiupathiae has been recognised as a cause of infection in animals andman since the late 1880s. It is the aetiological agent of swine erysipelas, and also causeseconomically important diseases in turkeys, chickens, ducks and emus, and other farmedanimals such as sheep. The organism has the ability to persist for long periods in theenvironment and survive in marine locations. Infection in man is occupationally related,occurring principally as a result of contact with animals, their products or wastes.Human infection can take one of three forms: a mild cutaneous infection known aserysipeloid, a diffuse cutaneous form and a serious although rare systemic complicationwith septicaemia and endocarditis. While it has been suggested that the incidence ofhuman infection could be declining because of technological advances in animalindustries, infection still occurs in specific environments. Furthermore, infection by theorganism may be under-diagnosed because of the resemblance it bears to otherinfections and the problems that may be encountered in isolation and identification.Diagnosis of erysipeloid can be difficult if not recognised clinically, as culture is lengthyand the organism resides deep in the skin. There have been recent advances inmolecular approaches to diagnosis and in understanding of Erysipelothrix taxonomyand pathogenesis. l h o PCR assays have been described for the diagnosis of swineerysipelas, one of which has been applied successfully to human samples. Treatment byoral and intramuscular penicillin is effective. However, containment and controlprocedures are far more effective ways to

IntroductionThe genus Erysipelothrix consists of two namedspecies, E. rhusiopathiae and E. tonsillarum, and anas yet unnamed third species. All are gram-positive,non-sporing rods. E. rhusiopathiae is a pathogen or acommensal or saprophyte of a wide variety of wild anddomestic animals, birds and fish. Diseases of economicimportance in animals include swine erysipelas,erysipelas of farmed turkeys, chickens and emus, andpolyarthritis in sheep and lambs. Man can be infected;the organism is an occupational pathogen, prevalent inthose working in association with animals and animalproducts. Three forms of human disease are recognised,the mildest and most common of these is the skininfection known as erysipeloid.

Received 19 Jan. 1999; accepted 15 March 1999.Corresponding author: Professor T. V Riley (e-mail:triley @cyllene. wa. du.au)

reduce infection in both man and animals.

A review of E. rhusiopathiae and human infection has notbeen published for 10 years [l]. While few additionalcases have been reported in this time, much has beenachieved in the area of Erysipelothrix taxonomy and inmolecu lar approaches to the diagnosis and pathogenesis ofinfection. It has been suggested that the frequency ofhuman infection is declining, due to changes in technologyby industry which have reduced the use of animal productsand occupational exposure to the organism [11. However,in specific environments, exposure to E. rhusiopathiaecontinues, and the threat of complications such asendocarditis is real. This review presents the recentdevelopments against a background of older literature toensure that awa reness of the organism is maintained.

BacteriologyHistory and nomenclatureBergey's Manual [2] classifies E. rhusiopathiae as aregular non-sporing gram-positive rod. Until recently,


2/11

790 C . J. BROOKE AND T. V. RILEYthe genus consisted of only the type species, E.rhusiopathiae, which was known to demonstrate con-siderable serological, biochemical and antigenic varia-tion [3]. Genetic analyses have revealed a new species,E. tonsillarum [4], Erysipelothrix was originally con-sidered a close relative of the genus Listeria, althoughnumerous molecular taxonomic studies have concludednow that the genus is a distinct cluster of organisms,most similar to the streptococci [2,3].E. rhusiopathiae (from the Greek erysipelas adisease, thrix a hair or thread, rhusius reddishand pathus disease [5]), literally erysipelas threadof red disease [l], has a long history and is the resultof many name changes. Koch [6] first isolated a strainof the genus Erysipelothrix in 1876 from a mouse heinoculated with putrefLing blood. He described theorganism as the bacillus of mouse septicaemia andnamed it E. muriseptica. In 1882, Loeffler isolated asimilar organism from cu taneous blood vessels of a pigwhich had died from swine erysipelas, and was the fmtto describe fully the infectious agent and the disease itcaused in swine [6].The first description of human disease, later attributedto Erysipelothrix, was reported in 1870 in the BritishMedical Journal; hrther cases were documented in1873 as erythema serpens [7]. However, it was not until1884, when Rosenbach isolated an organism similar toKochs from a patient with localised cutaneous lesions,that Erysipelothrix was established as a humanpathogen. He coin ed the term erysipeloid to dif-ferentiate between the human streptococcal diseaseerysipelas and the condition he had observed [6].Rosenbach [6] distinguished three separate species ofthe organism, E. muriseptica, E. porci and E. ery-siploides, based on their isolation from mouse, pig andman, respectively. It was later realised that these werethree nearly identical strains of the same species [3],and they were named E. insidiosa, as originallyproposed by Trevisan in 1885. This name, and all 36other documented names for the organism, wererejected in favour of E. rhusiopathiae in 1966, acombination which originated in 1918 [8].Takahashi et al. [9] isolated a cluster of avirulentserotype 7 strains which were later found to begenetically distinct from E. rhusiopathiae by DNAbase composition and DNA-DNA homology studies[4]. These strains formed the basis of a new species, E.tonsillarum (from the Latin of the tonsils). Originally,E. tonsillarum was considered morphologically andbiochemically identical to E. rhusiopathiae, but it wasshown later that E. tonsillarum could ferment sucrose,while E. rhusiopathiae could not [10, 111. E tonsilla-rum was described as avirulent for pigs, mice andchickens, but pathogenic for dogs [12, 131. There havebeen no studies so far into the pathogenicity of thisspecies for man.

Morphology and growth characteristicsE. rhusiopathiae is a non-motile, non-sporulating, non-acid-fast, slender gram-positive rod, which is easilydecolourised [2]. Gram-negative forms are often seen[14], particularly if the culture is old [15]; thus, theorganism has been cited occasionally as a gram-negative bacillus [161. E. rhusiopathiae was longdescribed as non-capsulate [2], until recent studiesshowed the presence of a capsule and suggested a rolefor it in virulence [171.Based on the colonial appearance of the organism,Erysipelothrix morphology is described as smooth (S)or rough R) [2 ,8 , 181. S-form colonies are convex,with a sm ooth su rface and entire edg e [2, 15, 191. R-form colonies are slightly larger with an irregular edgeand a flattened, rough surface [14]. All colonies areclear, circular and very small (0.1-0.5 mm diameterafter 24 h; 0.5-1.5 mm after 48 h [2, 19]), increasing insize and tending towards a pale blue opacity withfurther incubation or age [2, 191. Most strains exhibit anarrow zone of a-haemolysis on blood agar, which caneven show slight clearing after 48 h [2]. R-formcolonies do not cause haemolysis [181.Growth in broth was best described by Smith (cited byJones [2]), who noted that the suspension had a faintopalescence which on shaking was resolved for amom ent into delicate rolling clouds. forms causeslight turbidity and a powdery deposit, whereas Rforms have a tangled hair-like appearance [3].Cell morphology is closely linked to the colonialcharacteristics of each form [19]. S forms are slender,straight or slightly curved rods with rounded ends, 0.8-2.5 p m in length and 0.2-0.4 p m in diameter. The rodsexist in various formations, often as small chains [2].The R form exhibits a predominantly filamentousmorphology, frequently likened to the mycelial forma-tions of fungi, although branching does not occur. Thefilaments can be 4 p m to > 60 p m in length and canhave a beaded appearance with Grams staining [2].Long chains of distinct rods can also exist in this form[14, 181.The origin of R and forms and their role in diseasehave received much emphasis in the Erysipelothrixliterature, possibly because different forms for otherpathogens, such as Streptococcus pneumoniae, havedistinct roles in virulence. For E. rhusiopathiae theseroles are not definite, and there are conflictingobservations on the role of each form. Furthermore,the distinction between and R forms is not alwaysclear. An intermediate (RS) form, which is the mostcommon conformation [15], has been used to describecolonies sharing the characteristics of both types [8].Some investigators have suggested that forms arecommonly isolated from acute pig diseases, such assepticaemia, and R forms from more chronic syn-


3/11

ERYSIPELOTHRIX RHUSIOPATHIAE 79dromes such as arthritis and endocarditis [8]. However,there are conflicting reports relating to the virulence ofR forms. Gorby [20] reported that, in pig disease, Rforms are the more virulent type, while Taylor [21]stated that the R form was generally considered lessvirulent. Similar possible relationships for S and Rforms in human disease have not been documented.Media and conditions of incubation (see below) play amajor role in morphology formation; R forms arefavoured by incubation at 37C in acidic pH, while Sforms predominate in alkaline conditions (PH 7.6-8.2)with incubation at 33C [14]. Changing growth con-ditions has allowed S forms to give rise to R and RSforms, and S forms to originate from R forms [3]. TheS form often dissociates to the R form with age [8, 151.These changes in morphology and cultural character-istics are reported to lead to changes in virulence andantigenic properties [81.Growth conditions and requirementsE. rhusiopathiae is a facultative anaerobe [3]. Newlyisolated strains are micro-aerophilic, but laboratory-adapted cultures grow both aerobically and anaerobi-cally, with some strains being favoured by incubationin C02 5% or 10% [2]. The organism can grow attemperatures between 5 and 44C [2], optimallybetween 30 and 37C [18]. Best growth is favouredby an alkaline pH. The optimum pH range has beendocumented as 7.2-7.6 [2, 5] or 7.4-7.8 [ lS ,18, 19 ,221 and the limits of growth a s 6.7-9.2 [23].Growth is enhanced by the inclusion of serum 5- lo%,blood, g lucose 0.1 -0.5%, protein hydroly sates, orsurfactants such as Tween 80 in med ia [8, 181. Theexact nutritional requirements of the organism are notknown [181, but riboflavin, small amounts of oleic acidand several amino acids [24] particularly tryptophanand arginine [8, 25 ] are needed for growth. Higherconcentrations of glucose and oleic acid are inhibitoryPI.BiochemistryThe genus Erysipelothrix is relatively inactive andgives negative results for catalase, oxidase, methyl red,indole and Voges-Proskauer reactions [151. Carbo-hydrate fermentations produce acid without gas, butreaction patterns are variable and depend on the basalmedium and indicator used [14,26]. Andrade's agarwith horse serum 10% is the recommended m edium forbiochemical tests. The majority of strains produce H2Sgas, but again the extent of this production varies withthe culture medium. The best reaction is demonstratedon triple sugar iron agar [27]. A more detaileddescription of the biochemical characteristics ofErysipelothrix can be found in the reviews of Ewald[8] Jones [2] or Reboli and Farrar [3]. Traditionally,biochemical reactions were used to differentiate be-

tween Erysipelothrix and morphologically similarbacteria, such as Listeria and Corynebacterium spp.Characteristics used for this purpose included ahaemolysis, lack of m otility, lack of catalase productionand resistance to neomycin [141. While traditionalbiochemical testing may still be of value, particularlyin discriminating between E. rhusiopathiae and E.tonsillarum, rapid identification can be achieved withan API Coryne System strip (bioMerieux) [28].Chemical tolerancesE. rhusiopathiae is a remarkably resistant organism forone that does not form spores [181. The survival of theorganism in the environment is an important factor inthe epidemiology of disease. E. rhusiopathiae is alsotolerant to numerous chemicals. It can grow in thepresence of phenol 0.2% and crystal violet 0.001%,and is said to be one of the organisms most resistant tosodium azide, tolerating 0.1% [23,29]. Some of thesechemical tolerances have been utilised in the develop-ment of selective media.Antigenic structureWatts [30] noted that most strains of Erysipelothrix hadtwo kinds of antigen, a species-specific heat-labileprotein antigen and a heat- and acid-stable polysac-charide antigen, which now form the basis forserotyping strains. Dedik (cited by Reboli and Farrar[3]) recognised two major serotypes, A and B. Strainsthat did not react with these specific antisera werenamed group N. The A rabic numeral serotyping systemof K ucsera [3 13 superseded the alphab etical sch eme ,due to variations in previous methods of antigenicextraction. The current standard serotyping method is adouble agar-gel precipitation test with type-specificrabbit antisera and antigen prepared by hot aqueousextraction [3 1,321.At present, strains of Erysipelothrix are classified asserotypes 1-26 [33]. A group N still exists for strainsthat have no type-specific antigen. In swine, 75-80%of isolates are of serotype 1 or 2 (previously group Aor B) and the less common serotypes make up theremaining 20% [18]. Some investigators have noted arelationship between serotype and clinical condition inpigs, with serotype la most commonly isolated fromacute swine illness, and serotype 2a more prevalent inchronic forms of disease [34]. However, other surveyshave provided contradictory results, and all clinicalconditions can be induced experimentally in susceptibleswine with a variety of serotypes [18,3 5]. Sneath et al.[23] reported that strains from human or pig originwere antigenically similar, but there has been noverification of this since the numerical typing systemwas established. There is a deficiency in the literatureregarding serotypes in human infection, and theepidemiological significance of serotyping in humandisease is questionable.


4/11

792 C. J. BROOKE AND T. V. RILEYMolecular studies have cast further doubt on the valueof serotyping as a reliable taxonomic and epidemio-logical method of classification. DNA -DNA hybridisa-tion [101, polyacrylamide gel electrophoresis [36] andmultilocus enzyme electrophoresis studies [111 classi-fied serotypes la, lb, 2, 4-9, 11, 12, 15-17, 19, 21and 25 in the species E. rhusiopathiae, and serotypes3, 7, 10, 14, 20, 22-24 in E. tonsillarum, with differingresults for serotypes 13 and 18 [111. When Erysipelo-thrix strains were analysed by restriction fragmentlength polymorphisms (RFLP), both E. tonsillarum andE. rhusiopathiae contained serotype 2 (virulent) and 7(avirulent) strains. The creation of a third species ofErysipelothrix was suggested to account for a distinctcluster of strains [37].These findings suggest that there is no direct relation-ship between serotype and virulence, supporting earlierinvestigators who noted that, within each serotype,strains of high, low and no virulence existed, and thatfactors other than serotype were important in theinduction of disease in animal models [38,39]. Furtherinvestigation is required to reach a consensus on typingschemes and their role, if any, in studies of pathogeni-city.

Mechanisms o pathogenicityRelatively little is known about the pathogenesis of E.rhusiopathiae infections. Strains of E. rhusiopathiaeare known to vary considerably in virulence. Despitemuch investigation, there has been no conclusiveevidence of a relationship between virulence andmorphology, chemical structure or antigenic structure[181. Various virulence factors have been suggested,although their relative importance is not yet clear. Thepresence of a hyaluronidase and a neuraminidase hasbeen recognised [40,41], but hyaluronidase was de-tected in strains both virulent and avirulent for pigs. Acorrelation between the amount of neuraminidaseproduced and the virulence of strains was noted [41],although later studies demonstrated that avirulentacapsular mutant strains also produced the enzyme~171.Adhesion to porcine kidney cells in vitro was greaterfor virulent strains [42]; however, the role of this factorin disease has not been investigated further. Furtherwork on adhesion has been carried out [43], but thiswas in relation to arthritis in swine as a model forrheumatoid arthritis in man. Recently, the presence of alabile capsule was reported and acapsular mutants wereconstructed by transposon mutagenesis. In contrast tothe parental strain, the mutants failed to resistphagocytosis by murine polymorphonuclear leucocytes,and could not survive within murine macrophages,suggesting that the capsule was an important virulencefactor [17,44]. The mutant has been used in thedevelopment of a diagnostic PCR [45] (see below) and

vaccine studies [46]. Further investigation of thepathogenesis of E. rhusiopathiae infection is required.

EpidemiologyE. rhusiopathiae and infections caused by this organ-ism occ ur wo rld-wide [6, 141. Infections of m an a ndanimals have been documented from Africa, Australia,several countries in the Americas, Japan, China andthroughout Europe [81. Human disease can originatefrom an animal or environmental source.Animal diseaseSwine erysipelas caused by E. rhusiopathiae is thedisease of greatest prevalence and economic impor-tance [18]. There are three clinical forms: a severeacute septicaemic form of sudden mortality; a milder,subacute urticaria1 form characterised by purplediamond-shaped lesions on the skin and a chronicform with endocarditis or arthritis [14]. Swineerysipelas is economically detrimental to the pigindustries of North America, Europe, Asia andAustralia [181.As well as affecting swine, E. rhusiopathiae causesinfections in a wide variety of domestic and wildmammals (including marine mammals), domestic,game and wild birds, and man [3, 8, 19,221. Poly-arthritis of sheep and lambs, and erysipelas in calves,ducks and domestic turkeys are also economicallysignificant diseases caused by E. rhusiopathiae [6, 191.In Australia, the organism is an emerging problem infarmed emus [47].Domestic swine are believed to be the most importantanimal reservoir of E. rhusiopathiae. The organism isshed by diseased animals in faeces, urine, saliva andnasal secretions, which can contaminate food, water,soil and bedding, leading to indirect transmission of theorganism [18]. Furthermo re, an average of 20-40% ofhealthy swine, and in some herds up to 98%, harbourErysipelothrix in the lymphoid tissue of the alimentarytract, particularly in the tonsils [18,48 ,49]. One studydemonstrated that both virulent (serotypes 2, 6, 11, 12and 16) and avirulent (serotype 7) serotypes werefound on the tonsils [9]. Another showed that thefaeces of apparently healthy animals contained virulentorganisms [50]. The maintenance of E. rhusiopathiaein nature appears to result from asymptomatic carriagein animals and subsequent dissemination of theorganism to the environment [6].Mice are susceptible to infection [6], but other rodentsseem to be affected only occasionally [51]. Theseanimals can harbour the organism and are importantreservoirs in som e environments, such as meat packingplants [6]. Insects have been reported to carry Erhusiopathiae, and are occasional vec tors [6 , 19,521;


5/11

ER YSIPEL THRIX RHUSIOPA THIAE 793however, this is not a known route of infection forman.Environmental reservoirsThe environment appears to be secondary in impor-tance to animal reservoirs as a source of E.rhusiopathiae. However, in some circumstances, suchas in marine environments, the organism may survivelong enough to create a significant hazard to man [6].E. rhusiopathiae is a saprophyte associated with somegroups of animals, particularly marine fish, molluscsand crustacean s [6 ,5 3, 541. Freshwater fish and som especies of bird are also hosts [14]. The organismsurvives and grows on the exterior mucoid slime offish, without causing disease in the fish themselves [6].It is likely that E. rhusiopathiae survives by a similarmechanism on the exterior slime layer of other marinecreatures. The slime on fish appears to be an importantsource of infection for man. In early reports, fishcaught under aseptic conditions did not harbourErysipelothrix [ 5 5 ] , so investigators concluded thatthe organism was transmitted via the slime from otherfish [54,55]. Boxes used for transport of fish seemed toplay a vital role in the transmission of E rhusio-pathiae, and many human cases resulted from contactwith these objects [5 , 54, 551.Once in the environment, E. rhusiopathiae can survivefor long periods although it does not form spores [181.The organism is ubiquitous, and can be found wherevernitrogenous matter decomposes, retaining virulence andviability for months in putrid material [56]. Survival inswine faeces for 1-5 months, depending on seasonalconditions [19], and in soil for up to years from thetime of the last diseased pig [57] has been demon-strated. However, this latter report did not consider thepossibility of asymptomatic shedding. E. rhusiopathiaehas been recovered from sewage eMluent fromabattoirs, streams, drains and fertilizer [3, 151, surviv-ing in drinking water for 4-5 days and sewage for 10-14 days [5]. It was long thought that the organismcould live in soil indefinitely, and early reportssuggested that the source of infection was soil [5].However, studies by Wood [58] did not support thiswidely held belief; E. rhusiopathiae survived for amaximum of only 35 days, depending on temperatureand soil condition. Despite this limited endurance, theorganism does survive long enough in soil to be apotential source of infection to animals and man.E. rhusiopathiae persists in animal tissues for longperiods, despite chilling, freezing, or curing [6]. Theorganism is resistant to pickling, smoking and salting[14]. Erysipelothrix can also survive in decaying tissue,and will remain viable in a carcass for 12 days in directsunlight, for 4 months in putrefied flesh, for 9 monthsin a buried carcass and at least 10 months inrefhgerated tissue [191. E. rhusiopathiae has been

isolated from fresh fish, pork and chicken for humanconsumption [59,60]. The widespread distribution ofE. rhusiopathiae can be attributed to the ability of theorganism to survive for long periods in the environ-ment, and the fact that the organism can colonise orinfect a wide variety of animals [6].

Human infectionRisk of humans infection is based on the opportunityfor exposure, and factors such as age, sex, race andsocio-economic status relate only to this opportunity[3 ,6 I]. Individuals involved in occupations or recrea-tions with contact with animals, animal products oranimal wastes are at greatest risk. Thus, E. rhusio-pathiae infection is said to be occupationally related[56]. It follows that those in occupations with mostfrequent animal contact, such as butchers, abattoirworkers, veterinarians, farmers, fishermen, fish-hand-lers and housewives are the most commonly infected[5 ,6 , 19,621. However, cases have been documentedfrom a very wide variety of occupations (see [l] for acomplete list). The comm on names for human infectionreflect this occupational mode of acquisition. Theseinclude seal finger, whale linger, blubber finger, fishhand, fish poisoning, fish handlers disease and porkfinger [3, 6,6 2,6 3].Human infection can occur from contact with infectedanimals, their secretions, wastes or products, or organicmatter contaminated by any of these [6]. Infection isinitiated either by an injury to the skin with infectivematerial or when a previous injury is contaminated.There have been a few documented instances ofpenetration of the skin by the bacteria [61], and ofinfection by ingestion of contaminated food products[62]. There have been no reports of person-to-persontransmission [3]. Modes of infection tend to be veryoccupation-specific, and transmission is generally byvehicles. These include contaminated objects causingwounds, such as knives, needles, dissecting instru-ments, fish teeth and spines, fish hooks, bone splinters,and crab, lobster and crayfish claws [6]. If a wound isalready present, infection can result from contact withany of a very wide variety of contaminated objects[6,621.Infections in both man and animals appear to have aseasonal incidence, with most cases occurring in thesummer months [6]. While it has been suggested thatthe biological activity of E. rhusiopathiae is related totemperature [55], others think that it is likely to be dueto increased contact between people and sources ofinfection during these months [6].

Clinical manifestations in manE rhusiopathiae can cause three forms of humandisease which closely resemble disease in swine. These


6/11

794 C. J. BROOKE AND T. V. RILEYare erysipeloid (a localised cutaneous form), a general-ised cutaneous form and a septic form often associatedwith endocarditis [141. It is possible that the incidenceof human infection is declining. However, infection ispossibly under-diagnosed, because of the resemblanceit bears to other infections, difficulties in isolation andidentification of the causative organism, and the rapidresponse to empiric antimicrobial therapy [64].ErysipeloidErysipeloid is the most common form of humaninfection [14]. It is an acute localised cutaneousinfection, described as a local cellulitis [3]. Erysipeloidusually occurs on the hand or fingers, reflecting theoccupational nature of acquisition of the disease [62];however, lesions have been described on many areas ofthe body [6].The incubation period is usually < 4 days, but can beup to 7 days after exposure [65]. The infection consistsof a distinctive, well-demarcated, slightly elevatedviolaceous lesion [191. The peripheral edge spreadsslowly as the centre fades [66], and while vesicles areoccasionally present, there is no suppuration or pitting[14,56]. There is associated local swelling, and anintense itching or a severe burning or throbbing pain[6], which is inconsistent with the mild look of thelesion [11. Systemic symp toms can occu r; 10% of casesexperience fever and joint ache, and lymphadenitis andlymphadenopathy appear in 33% of patients [67].Arthritis can manifest in an adjacent joint. The diseaseis self-limiting and usu ally resolves in 3 -4 weekswithout therapy [3], although relapses may occur ifuntreated [61.Diffuse cutaneous fo rmThis form is more generalised than erysipeloid, andincludes the rare cases in which lesions progress fromthe initial site to other locations on the body, or inwhich there is development of lesions remote from thesite of inoculation [68]. The lesions are similar to thoseof the localised form, but bullous lesions can alsooccur [69]. Systemic symptoms are more frequent andinclude fever, malaise, joint and muscle pain andsevere headaches [6], and polyarthritis in rare in-stances. The clinical course is more protracted andrecurrences are more frequent than with erysipeloid[56]. Very few cases have been documented; only 1 of100 cases reported by Klauder [68] was generalisedand none of the 500 cases reported by Nelson [67] wasof this form.Septicaemia and endocarditisA more serious manifestation of E. rhusiopathiaeinfection is septicaemia, to which endocarditis hasalmost always been linked. In 49 cases of systemicinfection in 15 years, 90% were associated with

endocarditis [20]. Although septicaemia and endocar-ditis are relatively uncommon, there does appear to bean increase in incidence [14,201, which could eitherreflect increased exposure or improved diagnosis.E. rhusiopathiae endocarditis has a mortality of 38%and presents as an acute or subacute form, the latterbeing more frequent. In their summary of cases ofendocarditis, Gorby and Peacock [20] discussedpredisposing factors, and comparisons with other formsof endocarditis were made. E. rhusiopathiae endocar-ditis had an increased male to female ratio, possibly theresult of occupational exposure, and mortality wasalmost double the rate of endocarditis of otheraetiologies. The majority of patients had normal nativeheart valves and were immunocompetent. A history ofalcohol abuse, believed to be a risk factor for thedevelopment of this com plication, was noted in 33% ofpatients. Only 36% reported a previous erysipeloidlesion, and 89% of patients were in occupationsinvolving contact with animals.Recent reports [16,701 have demonstrated that Erysi-pelothrix bacteraemia without endocarditis is morecommon than w as thought previously, occurring mainlyin immunocompromised patients. This increased ratewas linked to a more thorough identification of bloodculture isolates from these patients.Miscellaneous infectionsThere are reports of other infections associated withErysipelothrix. These have included chronic arthritis[69], cerebral infection, [71,72] and osseous necrosis[68]. Recent case reports have focused on novelpresentations and complications of Erysipelothrixinfection. These have included erysipeloid with co-existing orf [64], persistent bacteraemia in a hospital-ised patient [73], bacteraemia in an HIV-positivepatient [74], endocarditis with acute renal failure[75], septicaemia and lupus nephritis [76], andsepticaemia in a neonate [77]. In reports of systemicinfection, the typical predisposing factors have beeninvolved: either immunocompromised patients withatypical infection without cardiac involvement orimmunocompetent patients with endocarditis. Renalinvolvement and alcoholism were factors noted in thissecond group.E. rhusiopathiae has been implicated recently in asyndrome known as crayfish poisoning, which affectslobster fishermen in Western Australia and bears aclinical resemblance to erysipeloid [78]. A possibleassociation between the organism and infections infishermen was noted in 1947 by Sheard and Dicks [79],but proper identification was hampered by fieldconditions. It was not until 1996 that this infectionwas investigated hrther. The presence of otherpotential pathogens did not allow a direct causalrelationship between E rhusiopathiae and crayfish


7/11

ERYSIPELOTHRIX RHUSIOPATHIAE 79poisoning to be established; however, the results weresuggestive of this [78].

Treatment and preventionSusceptibility data are still limited [3], despite recentreports o n the subject [9, 80, 811. Erysipelothrix ishighly susceptible to penicillin, cephalosporins andclindamy cin [20 , 23, 801. Most strains are resistant toaminoglycosides, trimethoprim-sulphamethoxazolepolymyxins, sulphonamides, streptomycin, novobiocinand vancomycin. The organism is variably susceptibleto chloramphenicol, tetracyclines and erythromycin [31.Erysipeloid can be treated effectively with oralpenicillin [3]. Although infection is usually self-limit-ing, relapses and progression to more serious forms arepossible. Oral penicillin will resolve a case oferysipeloid in around 48 h, while intravenous penicillinis recommended for more serious E. rhusiopathiaeinfections [141. While the mortality rate for endocardi-tis has been reduced from 100% in the pre-antibioticera, there is still a 38% fatality rate despite availabletreatment [20]. This rate could be partly explained bythe use of vancomycin, to which the organism isresistant, in empiric therapy for endocarditis [8 11.Therefore, early diagnosis of all forms of E. rhusio-pathiae infection is essential [141, In those individualsallergic to penicillin, cephalosporins have been de-scribed as the most appropriate alternative, as clin-damycin and erythromycin are only bacteriostatictowards E. rhusiopathiae [3].Before the advent of penicillin therapy, some allevia-tion of symptoms could be achieved with hyperimmuneserum. However, the resulting serum sickness was oftenmore severe than an episode of erysipeloid, and thistreatment was confirmed to be of little value forcutaneous infections [14, 821. Com mercial vaccines inthe form of bacterins, lysates or live attenuated strainsof E. rhusiopathiae serotype 2 offer protection to pigsand turkeys [25]. Vaccination is not a viable option inman, because clinical erysipeloid appears to conveylittle or no immunity [55,66], as evidenced by relapseand/or re-infection.Since Heilman and Herrel [52] first reported thesuccess of penicillin therapy for Erysipelothrix infec-tions, there has been no recorded resistance of theorganism to this antibiotic. One experiment involvingserial passage of 75 strains for 8 months did notproduce resistance [14]. Plasmids were not found in E.rhusiopathiae in early investigations [SO, 8 11, but laterstudies were able to detect them [83]. Plasmids appearto play no critical role in E. rhusiopathiae resistance.Antibiotics contained in animal feed have beenreported to influence the resistance of some strains ofE. rhusiopathiae, although the mechanism remainsuncertain [9].

Containment and control of E. rhusiopathiae are themost effective means of preventing the spread ofinfection in man and animals. An awareness of theinfection is essential for individuals in occupationswhich put them at risk. Suggested preventive measuresinclude: the wearing of gloves or other protective handwear, good hygiene especially frequent hand washingwith disinfectant soap and the prompt treatment ofany small injuries [6]. Good health is considered animportant factor in prevention, as a poor state of health,including alcoholism, may predispose to the seriousforms of infection [20, 841. Control of animal diseaseby sound husbandry, herd m anagement, good sanitationand immunisation is recommended [181.The removal or regular disinfection of contaminatedsources has been shown to be an important method oflimiting the spread of the organism throughout a workenvironment [6]. E. rhusiopathiae can be killed bycommonly available disinfectants [191. However, man yinvestigators have noted that structurally complexequipment is difficult to clean, and because theorganism is able to survive in organic matter, dis-infecting without cleaning is u seless [85, 861. Ifdisinfection is impractical, other control measuresbecome even more significant.Control of reservoir populations of E. rhusiopathiae isimpractical or impossible, because of the widespreaddistribution of the organism, the large variety of animalhosts and its ability to persist in the environment.However, the possibility of human infection can bereduced by awareness, safe work practices and sensibleprecautions.

Isolation and laboratory identificationIt has been reported that medical practitioners who seecases of human erysipeloid regularly find the lesionand other symptoms are so typical that a biopsy andsubsequent isolation is neither necessary nor justifiable[87]. As a result, the majority of identificationprotocols have been developed with swine erysipelasin mind. However, if human cases are declining,doctors could be less likely to recognise the infection,and, therefore, isolation techniques and methods foridentification of E. rhusiopathiae may gain newimportance.Cu tura1 methodsTraditional cultural methods for E. rhusiopathiaeisolation involve the use of selective and enrichmentmedia. Identification is based on Grams stain, culturalmorphology, motility, haemolytic characteristics andbiochemical properties, particularly H2S production [3].For bacteraemia or endocarditis, a blood samplecultured in standard blood culture media is sufficientfor isolation [19], as E. rhusiopathiae is not particu-


8/11

796 C . J. BROOKE AND T. V. RILEYlarly fastidious [3]. The organism is more difficult toisolate from cases of erysipeloid, because it is said tolive deep in the skin [2]. A biopsy at the advancingedge of the lesion and extending the entire thickness ofthe dermis is required. Aspirates of the lesion orassociated bullae and vesicles are usually less reward-ing [19]. Swabbing does not usually detect thepathogen [651.Biopsies and aspirates are incubated in an infusionbroth of glucose 1%, in air or C02 5-10 andsubcultured to blood agar every 24 h. If a sample islikely to be heavily contaminated, such as from soil,faeces, or animal tissue, selective measures are required[2]. A number of selective and enrichment broths havebeen described. The most commonly used is Erysipe-lothrix selective broth (ESB), a liquid medium contain-ing serum, tryptose, neomycin, vancomycin andkanamycin [MI. Packers medium (SACV) makes useof the organisms tolerance of sodium azide and crystalviolet [29], and is frequently used for subculture aftergrowth in ESB [18]. Modified blood azide (MBA) issimilar to SACV, but does not include crystal violet[89]. Bohms medium contains azide, kanamycin,phenol and water blue [8].Incubation of liquid media at 37C for 18-24 h isusually sufficient for growth. Colonies are visible after24 h on most solid media, although incubation for 48 his recommended for ESB and 72 h fo r SAC1 Analternative for biopsy specimens is to refrigerate themat 4-5C for 4-5 weeks in a liquid enrichment mediaand then subculture to SACV [151.Each medium has advantages, but none is ideal. ESB isstill regarded as the best selective medium, despite areport showing that some strains grow poorly due tokanamycin susceptibility [901. MBA requires lessincubation time than SACV, but is not as selectiveand is not suitable for heavily contaminated samples[89]. Bohms medium does not seem to have beenwidely used, although the reasons for this are unclear.

Mouse protection testThis test is traditionally regarded as the best con-firmatory test of E. rhusiopathiae identity, becausemost strains of the organism are highly virulent forlaboratory mice. One group of animals is inoculatedsubcutaneously with a 24-h broth culture and equinehyperimmune E. rhusiopathiae antiserum, and thesecond group receives broth culture but not antiserum.If the organism is E. rhusiopathiae the second group,but not the first, will die within 5-6 days [2]. However,strains must be pathogenic for mice for this method tobe useful. While this test is useful in studies of swineerysipelas, the reliability of this method for humanpathogenic strains remains unknown.

Fluorescent antibody testDirect and indirect assays have been used to confirmthe identity of E. rhusiopathiae in tissues [91,92], inbroth [93] and in human infection [94]. However,Harrington [93] noted that this method was not assensitive as cultural methods, and as a result it has notbeen used widely.API Coryne systemThe API Coryne system (bioMkrieux) is a commercialstrip system for the identification of coryneformbacteria. Soto [28] compared conventional biochemicalreactions with the commercial system for corynebac-teria and related genera, including Erysipelothrix. Thesystem had few misidentifications and all four strainsof E. rhusiopathiae tested were correctly identified byuse of the strip. The investigators concluded that thecommercial strip was a good alternative to traditionalbiochemical methods, permitting reliable and rapididentification of coryneform bacteria.

PCRTwo PCR methods are available for the detection ofErysipelothrix species [45, 951. While they were bothdeveloped for swine erysipelas, a PCR method wasemployed for detection of organisms in human speci-mens in a recent study in Australia [78]. Makino et al.[95] based their primers on a region of the 16s rRNAgene, specific to Erysipelothrix but shared by both E.rhusiopathiae and E. tonsillarum. Shimoji et al. [45]made use of the avirulent transposon mutant createdduring capsule studies to develop an E. rhusiopathiae-specific PCR. The primers were designed fromsequences presumed to be associated with the virulenceof E. rhusiopathiae. This assay would be particularlyuseful for monitoring of swine disease. However, thevalue of this test in a human clinical situation isuncertain, due to the lack of information regarding thepathogenicity of E. tonsillarum for humans. The timesaved by using the PCR is the greatest advantage it hasover all other methods for detecting Erysipelothrix.Erysipelothrix PCR is very sensitive; Makino et al.[95] detected


9/11

PCR is a powerful tool for the detection of Erysipelo-thrix in all kinds of samples and is able to overcomemany problems inherent in other diagnostic methods.However, it is likely that a combination of culture andmolecular techniques will be used for accurate diag-nosis in the future.

ConclusionsAlthough uncommon, it is likely that human infectionswith E. rhusiopathiae are under-diagnosed. Theorganisms slow growth and small colony size meanthat it may be overlooked in the routine diagnosticlaboratory or overgrown with secondary pathogens suchas Staphylococcus aureus and Streptococcus pyogenes.A high index of suspicion and the application ofmodern molecular techniques will no doubt improvethis situation. As a pathogen of animals, particularlyswine, Erysipelothrix is of great economic importanceand good animal husbandry practice is essential toreduce impact. However, the organisms resilience andability to survive are important in both human andveterinary medicine. Most human infections result fromoccupational exposure and this possibility can bereduced through awareness and safe work practices.

References1.2.

3.

4.

5.6.

7.8.

9.

10.

11.

12.

Reboli AC, Farrar WE. Erysipelothrix rhusiopathiae: anoccupational pathogen. Clin Microbiol Rev 1989; 2: 354-359.Jones D. Genus Erysipelothrix R o ~ e n b a c h 1 9 0 9 ~ ~ ~ ~ .n: SneathPHA, Mair NS, Sharpe ME, Holt JG (eds) Bergeys Manual ofsystematic bacteriology, 9th edn, vol 2. Baltimore, Williamsand Wilkins. 1986: 1245-1249.Reboli AC, Farrar WE. The genus Erysipelothrix. In: BalowsA Trupe r HG, Dwor kin M, Harder W, Schle ifer K (eds) Theprokaryotes. A handbook on the biology of bacteria: ecophy-siology, isolation, identification, applications. New York,Springer-Verlag. 1992: 1629 - 1642.Takahashi T, Fujisawa T, Benno Y et al. Erysipelothrixtonsillarum sp. nov. isolated from tonsils of apparently healthypigs. Int J Sst Bacteriol 1987; 37: 166-168.Woodbine M. Erysipelothrix rhusiopathae: bacteriology andchemotherapy. Bacteriol Rev 1950; 14: 161 178.Wood RL. Erysipelothrix infection. In: Hubbert WT, McCul-loch WF, Schnurrenberger PR (eds) Diseases transmitted fromanimals to man, 6th edn. Springfield, IL, Thomas. 1975:Barber M. Discussion on swine erysipelas infection Erysipe-lothrix rhusiopathiae) in man and animals. Proc R Soc MedEwald FW. The genus Erysipelothrix. In: Starr MP, Stolp H,Triiper HG, Balows A, Schlegel HG (eds) The prokaryotes: ahandbook on habitats, isolation, and identification of bacteria,Vol 2. New York, Springer-Verlag. 1981 1688- 1700.Takahashi T, Sawada T, Muramatsu M et al. Serotype,antimicrobial susceptibility, and pathogenicity of Erysipelothrixrhusiopathiae isolates from tonsils of apparently healthyslaughter pigs. J Clin Microbiol 1987; 25: 536-539.Takahashi T, Fujisawa T, Tamura Y et al. DNA relatednessamong Erysipelothrix rhusiopathiae strains representing alltwenty-three serovars and Erysipelothrix tonsillarum. Int J SystBacteriol 1992; 42: 469-473.Chooromoney KN, Hampson DJ, Eamens GJ, Turner MJ.Analysis of Erysipelothrix rhusiopathiae and Erysipelothrixtonsillarum by multilocus enzyme electrophoresis. J ClinMicrobiol 1994; 32: 371-376.Takah ashi T, Tamu ra Y, Yoshimura H et al. Erysipelothrix

271 -281.

1948; 41: 328-330.

13.

14.15.

16.

17.

18.

19.

20.

21.22.

23.24.25.

26.27.

28.

29.

30.31.

32.

33.

34.

35.

36.

37.

ERYSIPELOTHRIX RHUSIOPATHIAE 797tonsillarum isolated from dogs with endocarditis in Belgium.Res Vet Sci 1993; 54: 264-265.Takahashi T, Tamu ra Y, Endoh YS , Hara N. Cellular fatty acidcomposition of Erysipelothrix rhusiopathiae and Erysipelothrixtonsillarum . J Vet Med Sci 1994; 56: 385-387.Grieco MH, Sheldon C. Erysipelothrix rhusiopathiae. Ann NyAcad Sci 1970; 174: 523-532.Cottral GE. Erysipelothrix In: Manual of standardized methodsfor veterinary microbiology. Ithica, NY, Cornell UniversityPress. 1978: 429-436, 671, 672, 679, 687.Garcia-Restoy E, Espejo E, Bella F, Llebot J. Bacteremia dueto Erysipelothrix rhusiopathiae in immunocompromised hostswithout endocarditis. Rev Infect Dis 1991; 13: 1252-1253.Shimoji Y, Yokomizo Y, Sekizaki T, Mori Y, Kubo M.Presence of a capsule in Erysipelothrix rhusiopathiae and itsrelationship to virulence for mice. Infect Immun 1994; 62:Wood RL. Erysipelas. In: Leman AD, Straw BE, MengelingWL, DAllaire S Taylor DJ (eds) Diseases of swine, 7th edn.Ames, IA, Iowa State University Press. 1992: 475-486.Conklin RH, Steele JH. Erysipelothrix infections. In: Steele JH(ed) CRC handbook. Series in zoonoses, vol 1 (section A).Boca Raton, FL, CRC Press. 1979: 327-337.Gorby GL, Peacock JE. Erysipelothrix rhusiopathiae endocar-ditis: microbiologic, epidemiologic, and clinical features of anoccupational disease. Rev Infect Dis 1988; 10: 3 17-325.Taylor DJ. Pig diseases, 5th edn Cambridge, Burlington Press.1990.Shuman RD. Erysipelothrix. In: Davis JW Anderson RC,Karstad LH, Trainer DO (eds) Infectious and parasitic diseasesof wild birds. Ames, IA, Iowa State University Press. 1971:Sneath PHA, Abbott JD, Cunliffe AC. The bacteriology oferysipeloid. BMJ 1951; 2: 1063-1066.Hunter SH. Some growth requirements of Erysipelothrix andListeria. J Bacteriol 1942; 43: 629-640.Groschup MH, Timoney JF. Modified Feist broth as a serum-free alternative for enhanced production of protective antigenof Erysipelothrix rhusiopathiae. J Clin Microbiol 1990; 28:White TG, Shuman RD. Fermentation reactions of Erysipelo-thrix rhusiopathiae. J Bacteriol 1961; 82: 595-599.Vickers CL, Bierer BW. Triple sugar iron agar as an aid in thediagnosis of erysipelas. J Am Med Vet Assoc 1958; 133:543-544.Soto A, Zapardiel J, Soriano F. Evaluation of API Corynesystem for identifying coryneform bacteria. J Clin PatholPacker RA. The use of sodium azide (NaN3) and crystal violetin a selective medium for streptococci and Erysipelothrixrhusiopathiae. J B acteriol 1943; 46: 343-349.Watts PS. Studies on Erysipelothrix rhusiopathiae. J Path01Bacteriol 1940; 50: 355-369.Kucsera G. Proposal for standardization of the designationsused for serotypes of Erysipelothrix rhusiopathiae (Migula)Buchanan. Int J Syst Bacteriol 1973; 23: 184- 188.Kalf GF, White TG. The antigenic components of Erysipelo-thrix rhusiopathiae. 11. Purification and chemical characteriza-tion of a type-specific antigen. Arch Biochem Biophys 1963;102: 39-47.Nonun g Molin G. A new serotype of Erysipelothrixrhusiopathiae isolated from pig slurry. Acta Vet Hung 1991;Kucsera G. Serological typing of Erysipelothrix rhusiopathiaestrains and the epizootiological significance of the typing. ActaVet Acad Sci Hung 1979; 27: 19-28.Wood RL, Hamngton R. Serotypes of Erysipelothrix rhusio-pathiae isolated from swine and from soil and manure ofswine pens in the United States. Am J Vet Res 1978; 39:Tamura Y, Takahashi T, Zarkasie K Nakamura M , YoshimuraH. Differentiation of Erysipelothrix rhusiopathiae and Erysi-pelothrix tonsillarum by sodium dodecyl sulfate-polyacrylamidegel electrophoresis of cell proteins. Int J Syst Bacteriol 1993;Ahrne S Sternstrom I, Jensen NE Pettersson B, Uhlen M,Molin G. Classification of Erysipelothrix strains on the basisof restriction fragment length polymorphisms. Int J SystBacteriol 1995; 45: 382-385.

2806-28 10.

141 45.

2573-2575.

1994; 47: 756-759.

39: 137-138.

1833- 1840.

43: 111-114.


10/11

798 C. J. BROOKE AND T. V. RILEY38. Eamens GJ, Turner MJ, Catt RE. Serotypes of Erysipelothrixrhusiopathiae in Australian pigs, small ruminants, poultry, andcaptive wild birds and animals . Aust Vet 1988; 65: 249-252.39. Takahashi T, Sawada T, Set0 K, Muramatsu M, Maruyama T,Kanzaki M. Pathogenicity of Erysipelothrix rhusiopathiaestrains of serovars la, 3, 5, 6, 8, 11, 21, and type N, isolatedfrom slaughter pigs affected with chronic erysipelas. NipponJuigaku Zasshi 1985; 47: 1-8.40. Norrung V Studies on Erysipelothrix insidiosa S rhusio-pathiae. I. Morphology, cultural features, biochemical reactionsand virulence. Acta Vet Scand 1970; 11: 577-585.41. Krasemann C, Muller HE. [The virulence of Erysipelothrixrhusiopathiae strains and their neuraminidase production.]Zentralbl Bakteriol Orig A 1975; 231: 206-213.42. Takah ashi T, Hirayama N, Sawada T, Tamura Y, Muram atsu M.Correlation between adherence of Erysipelothrix rhusiopathiaestrains of serovar la to tissue culture cells originated fromporcine kidney and their pathogenicity in mice and swine. VetMicrobiol 1987; 13: 57-64.43. Franz B, Davies ME, Homer A. Localization of viable bacteriaand bacterial antigens in arthritic joints of Erysipelothrixrhusiopathiae-infected pigs. FEMS Immunol Med Microbiol44. Shimoji Y, Yokomizo Y, Mori Y. Intracellular survival andreplication of Erysipelothrix rhusiopathiae within murinemacrophages: failure of induction of the oxidative burst of

macrophages. Infect Immun 1996; 64: 1789- 1793.45. Shimo ji Y, Mori Y, Hyaku take K, Se kizaki T, Yokomizo Y.Use of an enrichment broth cultivation-PCR combination assayfor rapid diagnosis of swine erysipelas. Clin Microbiol 1998;46. Shimo ji Y, Mori Y, Sekizaki T, Shibahara T, Yokom izo Y.Construction and vaccine potential of acapsular mutants ofErysipelothrix rhusiopathiae: use of excision of Tn916 toinactivate a target gene. Infect Immun 1998; 66: 3250-3254.47. Griffiths GL, Buller N. Erysipelothrix rhusiopathiae infectionin semi-inte nsively farmed emus. Aust Vet 1991; 68:12 1 22.48. Rowsell HC. The effects of stomach contents and the soil onthe viability of Erysipelothrix rhusiopathiae. Am Vet Med49. Stephenson EH, Berman DT. Isolation of Erysipelothrixrhusiopathiae from tonsils of apparently normal swine bytwo methods. Am Vet Res 1978; 39: 187-188.50. Wood RL. solation of pathogenic Erysipelothrix rhusiopathiaefrom feces of apparently healthy swine. Am Vet Res 1974;51. Shuman RD. Erysipelas. In: Davis JW arstad LH, TrainerDO (eds) Infectious diseases of wild mammals. Ames, IA,Iowa State University Press. 1970: 267-272.52. Heilman FR, Herrell WE. Penicillin in the treatment ofexperimental infections due to Erysipelothrix rhusiopathiae.Proc StagMeet Mayo Clin 1944; 19: 340-345.53. Klauder V Erysipeloid and swine erysipelas in man. Aclinical and bacteriologic review; swine erysipelas in theUnited State s. JAMA 1 926; 86: 536-541.54. Murase N, Suzuki K, Isayama Y, Murata M. Studies on thetyping of Erysipelothrix rhusiopathiae. 111. Serological beha-viour of the strains isolated from the body surface of marine

fishes and their epizootiological significance in swine erysipe-las. Jpn Vet Sci 1959; 21: 215-219.55. Proctor DM, Richardson IM. A report on 235 cases oferysipeloid in Aberdeen. Br Ind Med 1954; 11: 175-179.56. Klauder V Erysipeloid as an occupational disease. JAMA57. Weaver RE. Erysipelothrix. In: Lennette A, Balows A, HauslerWJ, Shadomy HJ (eds) Manual of clinical microbiology, 4thedn. Washington, DC, American Society for Microbiology.58. Wood RL. Survival of Erysipelothrix rhusiopathiae in soilunder variou s environ menta l conditio ns. Cornell Vet 1973 ; 63:59. Ternstrom A, Molin G. Incidence of potential pathogens onraw pork, beef and chicken in Sweden, with special referenceto Erysipelothrix rhusiopathiae. Food Protection 1987; 50:60. Stenstrom I-M, Norrung V, Ternstrom A Molin G. Occurrenceof different serotypes of Erysipelothrix rhusiopathiae in retailpork and fish. Acta Vet Scand 1992; 33: 169-173.

1995; 12: 137-142.

36: 86-89.

ASSOC 1958; 132: 357-361.

35: 41-43.

1938; 111: 1345-1348.

1985: 209-210.

390-410.

141-146.

61. McGinnes GF, Spindle F. Erysipeloid condition among workersin a bone button factory due to the bacillus of swineerysipelas. Am Public Health 1934; 24: 32-35.62. Hunter D. The diseases of occupations, 5th edn. London,English University Press. 1975: 709-7 12.63. Hillenbrand FKM. Whale finger and seal finger: their relationto erysipeloid. Lancet 1953; 1: 680-68 1.64. Connor MP, Green AD. Erysipeloid in a sheep farmer withcoexistin g orf. Infect 1995 ; 30: 161-163.65. McClain JB. Erysipelothrix rhusiopathiae. In: Mandell GL ,Douglas RG, Bennett JE (eds) Principles and practice ofinfectious diseases, 3rd edn. New York, Churchill Livingstone.66. King PF. Erysipeloid. Survey of 115 cases. Lancet 1946; 2:67. Nelson E. 500 cases of erysipeloid. Rocky Mountain Med68. Klauder V Erysipelothrix rhusiopathiae infection in swine andin human beings: comparative study of the cutaneous lesions.Arch Dematol Syphil 1944; 50: 151-159.69. Ehrlich JC. E rysipelothrix rhusiop athiae infection in m an.Report of a case with cutaneous bullae, in which cure wasachieved with penicillin. Arch Intern Med 1946; 78: 565-577.70. Fakoya A, Bendall Rp Churchill DR, Doherty JF RidgwayGL. Erysipelothrix rhusiopathiae bacteraemia in a patientwithou t endoc arditis. Infect 1995; 30: 180- 181.7 1. Torkildsen A. Intracranial erysipeloid (swine erysipelas)abscess. Variety of abscess not hitherto observed. Bull Hyg1943: 1013.72. Silberstein EB. Erysipelothrix endocarditis: report of a casewith cerebral manifestation s. JAMA 1 965; 191: 862-864.73. Schuster MG, Brennan PJ, Edelstein P. Persistent bacteremiawith Erysipelothrix rhusiopathiae in hospitalized patient. ClinInfect Dis 1993; 17: 783-784.74. Hollick GE, Edinger RE. Erysipelothrix bacteremia. ClinMicrobiol News1 1995; 17: 9-16.75. GimCnez M, Femandez P Padilla E, Matas L, Ausina V:Endocarditis and acute renal failure due to Erysipelothrixrhusiopathiae. Eu r Clin Microbiol Infect Dis 1996; 15:76. Totemchokchyakarn K, Janwityanujit S Sathapatayavongs B,Puavilai S. Erysipelothrix rhusiopathiae septicemia in systemiclupus erythematosus. Int Derm atol 1996; 35: 818-820.77. Jones N, Khoosal M. Erysipelothrix rhusiopathiae septicemiain a neonate. Clin Infect Dis 1997; 24: 511.78. Brooke CJ, McLaughlin V, Mee BJ, Riley TV: An investigationof crayfish poisoning in Western Australia. Med J Aust 1999;170: 288.79. Sheard K, D icks HG. Skin lesions am ong fishermen atHoutmans Abrolhos, Western Australia, with an account oferysipeloid of Rosenbach. Med Aust 1949; 2: 352-354.80. Takahashi T, Sawada T, Ohmae K et al. Antibiotic resistanceof Erysipelothrix rhusiopathiae isolated from pigs with chronicswine erysipelas. Antimicrobial Agents Chemother 1984; 25:81. Venditti M, G elfusa V, Tarasi A, Brandimarte C, Serra FAntimicrobial susceptibilities of Erysipelothrix rhusiopathiae.Antimicrob Agents Chemother 1990; 34: 2038-2040.82. Gledhill AW. Discussion o n sw ine erysip elas infe ction

(Erysipelothrix rhusiopathiae) in man and animals. Proc RSoc Med 1948; 41: 330-332.83. Noguchi N, Sasatsu M, Takahashi T, Ohmae K, Terakado N,Kono M. Detection of plasmid DNA in Erysipelothrixrhusiopathiae isolated from pigs with chronic swine erysipelas.Vet Med Sci 1993; 55: 349-350.84. McCarty D, Bornstein S. Erysipelothrix endocarditis. Report ona septicemic form of the erysipeloid of Rosenbach. Am J ClinPathol 1960; 33: 39-42.85. Spencer R. The sanitation of fish boxes. I. The quantitative andqualitative bacteriology of commercial wooden fish boxes. JAppl Bacteriol 1959; 22: 73-84.86. Mutalib AA, King JM cDonough PL. Erysipelas in cagedlaying chickens and suspected erysipeloid in animal caretakers.Vet Diagn Invest 1993; : 198-201.87. Price JEL, Bennett WEJ. The erysipeloid of Rosenbach. B M J88. Wood RL.A selective liquid medium utilizing antibiotics forisolation of Erysipelothrix insidiosa. Am Vet Res 1965; 26:

1990: 1599-1600.196- 198.1955; 52: 40-42.

347-348.

385-386.

1951; 2: 1060-1062.

1303- 1308.


11/11

downloaded file 46

Documents