Aug 14, 1991 - The man was admitted to the ... Case 3. A 39-year-old man was bitten on the right fourth finger while handling a ... d TSI, triple sugar iron agar.
JOURNAL OF CLINICAL MICROBIOLOGY, Nov. 1991, p. 2535-2538
Vol. 29, No. 11
0095-1137/91/112535-04$02.00/0 Copyright © 1991, American Society for Microbiology
Actinobacillus spp. and Related Bacteria in Infected Wounds of Humans Bitten by Horses and Sheep PEEL,'* K. A. HORNIDGE,2 M. LUPPINO,3 A. M. STACPOOLE,4 AND R. E. WEAVERs Microbiological Diagnostic Unit, University of Melbourne, Parkville, Victoria 3052,1 Microbiology Department, M. M.
Preston and Northcote Community Hospital, Preston, Victoria 3072,2 Department of Pathology, the Geelong Hospital, Geelong, Victoria 3220,3 and Department of Pathology, Ballarat Base Hospital, Ballarat, Victoria 3550,4 Australia, and Meningitis and Special Pathogens Branch, Division of Bacterial and Mycotic Diseases, Centers for Disease Control, Atlanta, Georgia 303335 Received 11 April 1991/Accepted 14 August 1991
We describe the isolation of ActinobaciUus lignieresi and an A. equuli-like bacterium from an infected horse-bite wound in a 22-year-old stable foreman and A. suis from a bite injury in a 35-year-old man who had been attacked by a horse. A. lignieresu was also isolated in pure culture from an infected sheep-bite wound in a rural worker. These species of the genus ActinobaciUus are primarily associated with animals and animal diseases and are rarely isolated from humans. The purpose of this report is to raise awareness of the possible occurrence of ActinobaciUus spp. in bite wounds inflicted by farm animals and to discuss the difficulties encountered in the identification of species of Actinobacillus and related bacteria.
Dogs and cats are responsible for most of the bite wounds inflicted by animals on humans. If the wounds become infected, bacteria from the oropharyngeal flora of these animals, especially Pasteurella multocida, are typically implicated. Actinobacillus spp. occur as commensals in the oropharyngeal flora of horses and sheep. Actinobacillus suis has been found in the upper respiratory tract (7) and oral cavity (2) of horses. Indirect evidence for the presence of A. lignieresii in the same site exists, since this species has been previously isolated from horse-bite wounds in humans (4, 5). A. lignieresii may be present in the rumen of healthy sheep (10). A. suis is an important opportunistic pathogen in pigs and horses, and A. lignieresii is an opportunistic pathogen in cattle and sheep (11). We report here the isolation of species of the genus Actinobacillus along with other bacteria, including P. multocida, from horse-bite wounds and A. lignieresii from a sheep-bite wound in humans. We also discuss the identification of Actinobacillus species.
A Gram-stained smear of the wound swab showed abundant pus cells with mixed bacteria. The specimen was cultured on Columbia agar (Oxoid Ltd., Basingstoke, Hampshire, United Kingdom) containing 5% (vol/vol) horse blood (HBA) and MacConkey agar (Oxoid CM7b) under aerobic conditions with 3% carbon dioxide, on HBA containing neomycin (75 ,ug/ml) anaerobically, and in cooked-meat medium. After incubation at 37°C for 48 h, the predominant, abundant growth consisted of two types of small, gramnegative, rod-shaped bacteria which were later identified by the Microbiological Diagnostic Unit at the University of Melbourne as A. lignieresii and an A. equuli-like bacterium. A few colonies of S. aureus grew on the MacConkey agar only, and a light, mixed population of oral Neisseria and Streptococcus species grew on the other media. Colonies of the gram-negative rods were differentiated initially by the strength of their oxidase reactions and the intensity of their coloration on MacConkey agar, which reflected the different rates of acid production from lactose by these two isolates. Case 2. After being attacked and bitten by a horse, a 35-year-old man arrived at a country hospital in Victoria, Australia, presenting with a compound fracture of the left radius and ulna and a macerated wound on the arm associated with moderate muscle loss. The fracture was openly reduced and internally fixed. The wound was debrided 3 days later, by which time it showed signs of inflammation. Antibiotic therapy with penicillin, flucloxacillin, gentamicin, and metronidazole was begun. Debridement was performed again after another 2 days. Further attempts at wound closure proved futile because of the extent of muscle loss, and the patient was transferred to the Victorian Plastic Surgery Unit at the Preston and Northcote Community Hospital in Preston, Victoria, Australia, for bone and skin grafts. Over the course of the next 3 months, the patient underwent a series of debridement and grafting procedures but sustained repeated infections with purulent wound discharge from which mixed cultures of bacteria, including S. aureus, Prevotella melaninogenica (formerly Bacteroides melaninogenicus), Escherichia coli, and P. multocida (also identified by the Microbiological Diagnostic Unit), were isolated. The
CASE REPORTS Case 1. A 22-year-old man arrived at the Casualty Department of the Geelong Hospital, presenting with a painful, swollen left hand. He had been bitten on the hand by a horse about 12 h previously, while working as a stable foreman. On examination, the hand was warm, swollen, erythematous, and tender to the touch, and ajagged cut about 1 cm in length was present on the dorsum. The man was admitted to the hospital, and a swab of the wound was taken for microscopy and culture. Tetanus prophylaxis was given, and the patient was treated empirically with penicillin to cover infection by oral flora, metronidazole to cover infection by anaerobic flora, and flucloxacillin to cover possible infection by Staphylococcus aureus. The symptoms subsided with treatment and, after 3 days, the patient was discharged. He did not return for a follow-up examination. *
Corresponding author. 2535
2536
J. CLIN. MICROBIOL.
PEEL ET AL.
patient was treated with a variety of antibiotics, including penicillin, cloxacillin, flucloxacillin, gentamicin, and metronidazole. More than 3 months after the initial attack by the horse, a bone biopsy specimen from the injured arm was submitted for culture. This specimen was cultured on HBA and MacConkey agar (Oxoid CM7b) under aerobic conditions with 5% carbon dioxide; on HBA with a kanamycin (1,000 ,ug) disc and a neomycin (10 Fug) disc and Schaedler agar (Oxoid CM437) with 5% (vol/vol) horse blood, nalidixic acid (10 ,ug/ml), and 0.1% (vol/vol) Tween 80 anaerobically; and in cooked-meat medium (Oxoid CM81). From this specimen, A. suis was isolated for the first time, along with a strain of E. coli with the same antibiogram as that of the E. coli isolated previously. At this stage, the antibiotic therapy was varied to include the topical application of co-trimoxazole cream with systemic gentamicin and minocycline. Co-trimoxazole was included primarily because the E. coli isolated was susceptible to this antimicrobial agent. However, A. suis was also susceptible to co-trimoxazole, as well as to gentamicin and tetracycline. A bone graft and flap operation was performed 1 month later. This procedure was uncomplicated by bacterial infection, and cultures of subsequent specimens yielded no significant growth. Case 3. A 39-year-old man was bitten on the right fourth finger while handling a sheep. The bite broke the skin over the dorsum of the terminal interphalangeal joint. He was examined by his medical practitioner, presenting with a grossly swollen, erythematous, painful finger. The palm of his hand was also slightly swollen. A probe through the broken skin penetrated freely into the terminal joint of the finger. A wound swab was taken, and the patient was treated with cephalexin (500 mg four times daily). However, the infection was slow to respond, and residual mild crepitus was still present in the terminal interphalangeal joint 6 months later. A Gram-stained smear of the wound swab showed a small number of pus cells and no bacteria. The specimen was cultured on HBA, heated HBA, MacConkey agar (Oxoid CM7b), and HBA containing colistin (10 ,ug/ml) and nalidixic acid (15 ,ug/ml) under aerobic conditions with 5% carbon dioxide added and on HBA and HBA containing colistin and nalidixic acid anaerobically. A pure growth of a gramnegative, rod-shaped bacterium was obtained after incubation for 24 and 48 h at 37°C. This isolate was later identified at the Microbiological Diagnostic Unit as A. lignieresùi. MATERIALS AND METHODS Conventional biochemical reactions and other characteristics were determined as previously described (4). The sensitivity of the isolates to 0/129 (2,4-diamino-6,7-diisopropylpteridine) was tested against commercially available discs containing 10 or 150 ,ug of the agent (Oxoid). Phosphatase activity was determined by growth on phenolphthalein diphosphate agar followed by exposure of the cultures to ammonia vapor (1). Acid production from carbohydrates was also investigated with fermentation medium with 5% inactivated horse serum, 0.05% Oxoid agar no. 1, and bromocresol purple as the indicator. Gas production from glucose (2%) was tested with both Oxoid MRS broth (CM359) and Bacto Lactobacilli MRS broth (Difco Laboratories Inc., Detroit, Mich.). The gas was detected by displacement of petrolatum layers above the broths in small glass test tubes (13 by 100 mm;
TABLE 1. Characteristics shared by the case isolates of Actinobacillus and the A. equuli-like bacterium Characteristic
Positive Aerobic and anaerobic growth Growth on MacConkey agar Sensitivity to 0/129b discs Urease (Christensen's) Nitrate reduction to nitrite ,B-Galactosidase (ONPG)C Phosphatase Acid from slant and butt of TSId H2S with lead acetate paper
Acid from: D-Galactose D-Glucose Maltose D-Mannose Raffinose Sucrose D-XylOse
Negative Motility Gas from nitrate broth Citrate (Simmons') Indole production Nitrite reduction Lysine decarboxylase Arginine dihydrolase Ornithine decarboxylase H2S in butt of TSI No acid from: Adonitol Dulcitol Erythritol Inulin Melezitose L-Rhamnose a Produced gas from glucose in MRS broth. b 0/129, 2,4-diamino-6,7-diisopropylpteridine. C ONPG, o-Nitrophenyl-3-D-galactopyranoside. d TSI, triple sugar iron agar.
Kimble Glass Inc., Vineland, N.J.). Two standard cultures, A. equuli NCTC 8529 and A. equuli NCTC 9435, were also checked for lack of gas production in MRS broths. RESULTS The Gram-stained smears of the Actinobacillus isolates and the A. equuli-like bacterium from cultures on agar media showed gram-negative coccoid forms interspersed among predominantly bacillary forms and occasional long filaments. Table 1 shows the growth characteristics and biochemical reactions which are shared by the case isolates and which collectively indicate their assignment to the genus Actinobacillus. However, these shared characteristics differ from those of A. actinomycetemcomitans, the species of Actinobacillus that occurs in the oral flora and as a pathogen of humans. A. actinomycetemcomitans is more fastidious in its growth characteristics, since it requires incubation in an atmosphere containing 5% carbon dioxide for good growth and does not grow on MacConkey agar. Biochemically, it does not produce urease or ferment raffinose or sucrose (11).
ACTINOBACILLUS SPP. IN ANIMAL-BITE WOUNDS
VOL. 29, 1991
TABLE 2. Differential characteristics for identification to the species level of case isolates and of A. actinomycetemcomitans Result obtained for isolates of: Reaction for A. A A. Characteristic equuli-like A. actinolignieresii
Hemolysis (SBA)b Oxidase reaction Esculin hydrolysis Acid from: Cellobiose Lactose D-Mannitol Melibiose Salicin Trehalose
A.
suis
(cases 1 and 3)
(case 3)
+ -
+ + +
(+)d
+ + + + +
+ -
bacterium (case D)'
mycetemcomitans
-
-
dc
-
-
+ + +
d
+
a Produced gas from glucose in MRS broth. b SBA, sheep blood agar. C d, 11 to 89%o of strains positive. d (+), delayed reaction, acid first detected on day 3.
Furthermore, the fermentation of D-galactose, maltose, D-mannitol, and D-xylose is variable and may be used in the biotyping of strains of this species (4). The characteristics used for identification of the species isolated are given in Table 2 along with those for A. actinomycetemcomitans. Apart from the hemolysis of sheep blood in sheep blood agar plates, the colonies of A. suis were characteristically sticky and adherent to the surfaces of agar media. The carbohydrate reactions of this strain differed from the typical reactions for A. suis (11) in that arabinose and glycerol were not fermented. The production or delayed production of acid from carbohydrates was the same in basal fermentation medium containing serum with a low concentration of agar and bromocresol purple as the indicator as it was in fermentation medium with Andrade's indicator. The A. equuli-like bacterium differed from A. equuli in producing gas from glucose. This was initially detected in the Durham tube in fermentation medium and confirmed by testing for gas production in MRS broths with petrolatum layers. The A. equuli-like bacterium was capable of raising the petrolatum layers more than 1.5 cm above the surface of MRS broth cultures in 48 h, whereas A. equuli NCTC 8529 and A. equuli NCTC 9435 did not produce gas under these conditions. However, this aerogenic ability could not be demonstrated unless the tubes used for the test were made of glass or gas-impermeable plastic. Except for the production of gas from glucose, the biochemical characteristics of the A. equuli-like bacterium conformed to those of A. equuli (4, 11). In addition to the characteristics shown in Tables 1 and 2, the isolate fermented arabinose, fructose, glycerol, sorbitol, and starch and did not ferment adonitol and inositol. DISCUSSION A. lignieresii has been recognized as an opportunistic pathogen of cattle and sheep for many years. It causes chronic granulomatous lesions of the tongue in cattle (wooden tongue) and lesions of the skin, testes, and mammary glands in sheep. A. suis has been implicated as an etiologic agent of septicemia, pneumonia, and arthritis in pigs and can cause similar diseases in horses. Both species may occur as commensals in the alimentary, respiratory, or genital tracts
2537
of these animals (11). In addition, A. lignieresùi has been isolated from the nasopharynx and middle ear of asymptomatic laboratory rats (8). A. actinomycetemcomitans occurs as part of the normal oral flora and as a pathogen in humans in conjunction with Actinomyces spp. in actinomycosis or as the sole causative agent of soft-tissue abscesses and endocarditis (4, 11). It is also implicated in periodontal disease (15). A. lignieresùi has been reported as present in an infected wound in a 13-year-old boy who was bitten on the face by a horse (5). However, our case 3 is the first report of A. lignieresùi in an infected sheep-bite wound in a human. Moreover, the bacterium was clearly playing a pathogenic role, since it was grown in pure culture from the inflamed wound. The isolation of actinobacilli from situations in which mixed bacterial populations exist is known to be difficult (11). Nevertheless, the isolation of A. suis for the first time from patient 2 more than 3 months after the initial attack by the horse is unusual. Presumably, A. suis had persisted in the infected wound for this period despite repeated debridement and multiple antibiotic treatment. This case represents the first report of A. suis from a horse-bite wound in a human. The paucity of reported infections in humans due to these species of Actinobacillus may reflect the difficulties associated with differentiation of the three genera composing the family Pasteurellaceae-Actinobacillus, Pasteurella, and Haemophilus-and of their species. The genera Actinobacillus and Pasteurella are especially closely related (3, 11). Their generic differentiation is not clearly based on a definitive set of biochemical characteristics but is based on the distinctive combinations of biochemical reactions that characterize their species (3, 8, 11). This means that, according to current schemes of classification, a wide range of biochemical tests must be undertaken to accurately identify Actinobacillus and Pasteurella species. When this is not done, Actinobacillus species are likely to be misidentified as Pasteurella species (5, 8). A further complication in the recognition of Actinobacillus species from human clinical specimens is that A. actinomycetemcomitans, the species which is most familiar to clinical laboratory workers, differs from other species in the genus in that it does not grow on MacConkey agar or in the absence of an atmosphere containing increased carbon dioxide, does not produce urease, and is significantly less active in the fermentation of carbo-
hydrates. The biochemical characteristics of the gas-producing A. equuli-like bacterium from case 1 did not agree with those of the recently described Actinobacillus species which may produce gas, A. rossii (14), or with those of the gasproducing Pasteurella species, P. aerogenes (3), P. dagmatis (9), or P. caballi (12). The taxonomy of this A. equuli-like bacterium poses problems. The isolate of A. suis from patient 2 was unusual in that it did not ferment arabinose or glycerol (11). However, failure to ferment arabinose is apparently typical of equine, as opposed to porcine, isolates of this species. None of the 37 strains of A. suis isolated from swabs of the buccal mucosa of horses fermented arabinose
(2), and neither did any of 307 strains of A. suis-like bacteria isolated from equine tracheal washings and clinical specimens (6).
The need for more detailed studies of the taxonomy of genera and species of the family Pasteurellaceae is generally accepted (13, 14), as is the existence of undescribed members of the family (13). In the meantime, the accurate
2538
PEEL ET AL.
identification of Actinobacillus species according to current schemes of classification depends on adequate biochemical characterization and awareness of the likely presence of these bacteria in sites such as bite wounds inflicted by horses and sheep. ACKNOWLEDGMENTS We thank Michael J. Plunkett of Ararat, Victoria, Australia, for clinical information on case 3; Barry Holmes of the Central Public Health Laboratory, London, United Kingdom, for confirmation of identifications; J. E. Phillips of the Royal (Dick) School of Veterinary Studies, Edinburgh, United Kingdom, for information on the natural habitat and human pathogenicity of A. Iignieresii; and Lorna Barclay of the Microbiological Diagnostic Unit, University of Melbourne, for technical assistance.
REFERENCES 1. Barber, M., and S. W. A. Kuper. 1951. Identification of Staphylococcus pyogenes by the phosphatase reaction. J. Pathol. Bacteriol. 63:65-68. 2. Bisgaard, M., K. Piechulla, Y.-T. Ying, W. Frederiksen, and W. Mannheim. 1984. Prevalence of organisms described as Actinobacillus suis or haemolytic Actinobacillus equuli in the oral cavity of horses. Comparative investigations of strains obtained and porcine strains of A. suis sensu stricto. Acta Pathol. Microbiol. Immunol. Scand. Sect. B 92:291-298. 3. Carter, G. R. 1984. Genus I. Pasteurella Trevisan 1887, 94AL, Nom. cons. Opin. 13, Jud. Comm. 1954, 153, p. 552-557. In N. R. Krieg and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 1. The Williams & Wilkins Co., Baltimore. 4. Clark, W. A., D. G. Hollis, R. E. Weaver, and P. Riley. 1984. Identification of unusual pathogenic gram-negative aerobic and facultatively anaerobic bacteria. Centers for Disease Control, Atlanta. 5. Dibb, W. L., A. Digranes, and S. T0njum. 1981. Actinobacillus
J. CLIN. MICROBIOL.
lignieresii infection after a horse bite. Br. Med. J. 283:583-584. 6. Jang, S. S., E. L. Biberstein, and D. C. Hirsh. 1987. Actinobacillus suis-like organisms in horses. Am. J. Vet. Res. 48:10361038. 7. Kim, B. H., and J. E. Phillips. 1976. Actinobacillus suis in the horse. Vet. Rec. 98:239. 8. Lentsch, R. H., and J. E. Wagner. 1980. Isolation of Actinobacillus lignieresii and Actinobacillus equuli from laboratory rodents. J. Clin. Microbiol. 12:351-354. 9. Mutters, R., P. Ihm, S. Pohl, W. Frederiksen, and W. Mannheim. 1985. Reclassification of the genus Pasteurella Trevisan 1887 on the basis of deoxyribonucleic acid homology, with proposals for the new species Pasteurella dagmatis, Pasteurella canis, Pasteurella stomatis, Pasteurella anatis, and Pasteurella langaa. Int. J. Syst. Bacteriol. 35:309-322. 10. Phillips, J. E. 1981. The genus Actinobacillus, p. 1393. In M. P. Starr, H. Stolp, H. G. Truper, A. Balows, and H. G. Schlegel (ed.), The prokaryotes. A handbook on habitats, isolation, and identification of bacteria, vol. Il. Springer-Verlag, Berlin. 11. Phillips, J. E. 1984. Genus III. Actinobacillus Brumpt 1910, 849AL, p. 570-575. In N. R. Krieg and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 1. The Williams & Wilkins Co., Baltimore. 12. Schlater, L. K., D. J. Brenner, A. G. Steigerwalt, C. W. Moss, M. A. Lambert, and R. A. Packer. 1989. Pasteurella caballi, a new species from equine clinical specimens. J. Clin. Microbiol. 27:2169-2174. 13. Sneath, P. H. A., and M. Stevens. 1985. A numerical taxonomic study of Actinobacillus, Pasteurella and Yersinia. J. Gen. Microbiol. 131:2711-2738. 14. Sneath, P. H. A., and M. Stevens. 1990. Actinobacillus rossii sp. nov., Actinobacillus seminis sp. nov., nom. rev., Pasteurella bettii sp. nov., Pasteurella Iymphangitidis sp. nov., Pasteurella mairi sp. nov., and Pasteurella trehalosi sp. nov. Int. J. Syst. Bacteriol. 40:148-153. 15. Socransky, S. S. 1979. Criteria for the infectious agents in dental caries and periodontal disease. J. Clin. Periodontol. 6:16-21.
