Rapidly Growing Members of the Genus Mycobacterium Affecting Dogs and Cats
Rapidly growing members of the genus Mycobacterium were most often associated with chronic (2 to 72 months), nonhealing skin lesions of dogs and cats. Mycobacterium fortuitum (M. fortuitum) was the most commonly isolated mycobacterium obtained from these lesions, although M. chelonae-abscessus and M. flavescens were occasionally encountered. Isolates were tested in vitro to various antimicrobial agents and found to be susceptible to amikacin (100% of the isolates), cefoxitin (93.8%), ciprofloxacin (75%), clarithromycin (71.4%), doxycycline (28.6%), erythromycin (6.2%), gentamicin (68.8%), kanamycin (75%), minocycline (81.3%), streptomycin (14.3%), tobramycin (43.8%), trimethoprim/sulfonamides (57.1%), and vancomycin (15.4%).
Introduction
Genetic analysis of members of the genus Mycobacterium reveals differences that parallel growth rate in vitro.1–3 There are two groups defined by growth rate on artificial media. There are those that grow slowly (i.e., colonies are visible after several weeks of incubation), including the tuberculous species (M. tuberculosis, M. bovis, M. africanum, and M. microti) as well as some nontuberculous species (such as members of the M. avium intracellulare complex). On the other hand, there are those mycobacteria that grow relatively rapidly (i.e., colonies are visible after several days of incubation), which include M. fortuitum, M. chelonae, M. smegmatis, M. phlei, and M. vaccae.
The rapidly growing mycobacteria are environmental microorganisms isolated from diverse habitats, most commonly water and soil.4 Infection, at least in humans, is usually associated with an underlying immunosuppressive condition such as cancer, corticosteroid administration, trauma (surgical or accidental), or chronic renal failure.2–9 There are several case reports involving dogs and cats that associate trauma (e.g., bite wounds, injection-site abscess) with the development of mycobacteriosis due to rapidly growing mycobacteria.10–13
Etiological diagnosis of conditions involving the rapidly growing mycobacteria exists on two levels. On the one hand, the index of suspicion should be high when confronted with patients with chronic, nonhealing wounds that are unresponsive to the usually successful antimicrobial agents. On the other hand, demonstration of the agent by visualization in stained smears, culture, or both, results in a presumptive or definitive assessment, respectively, as to the etiology.
Antimicrobial drugs are used as part of the treatment of conditions involving rapidly growing mycobacteria.813–15 In vitro measurement of antimicrobial susceptibility of this group of microorganisms is difficult, however, and is best performed by laboratories that are equipped and experienced with these techniques. Timely acquisition of susceptibility data can therefore be problematic, resulting in the initial choice of antimicrobial therapy often being guided by retrospectively acquired susceptibility data. Complicating this is the fact that the antimicrobial drugs used to treat the rapidly growing mycobacteria are different from those that are used to treat the slower growing tuberculous species.3
It has been the authors’ impression that difficulty in making an etiological diagnosis is partly responsible for the chronicity of conditions involving the rapidly growing mycobacteria. For these reasons, this retrospective study was conducted to determine whether there were characteristics of the lesions that might lead more easily to the correct etiological diagnosis. In addition, the authors determined the in vitro antimicrobial susceptibility of the encountered mycobacteria.
Materials and Methods
Source, Collection, and Analysis of Isolates
Specimens were obtained from seven dogs and 11 cats between July 1989 and February 2000. Specimens were collected for analysis by aspiration.
Bacteriological media were inoculated with specimens obtained from each of the affected animals within 1 hour of collection. Trypticase soy broth with agara containing 5% sheep’s blood was inoculated and then incubated at 37°C in 5% carbon dioxide. Inoculated media were incubated for 5 days.
Identification of Isolates and Determination of Susceptibility
Colonies of rapidly growing mycobacteria were usually observed after 48 to 72 hours of incubation, but some strains took as long as 5 days to form visible colonies. Preliminary identification included demonstration of the acid-fastness of the isolates following staining with the Ziehl-Neelsen carbolfuchsin method. Mycobacterial isolates were identified, and their susceptibility to antimicrobial drugs was determined by the National Jewish Medical and Research Center in Denver, Colorado. Because they were tested for susceptibility over a span of several years, not all of the isolates were tested with the same drugs, nor was susceptibility testing performed by the same methods. All but three of the isolates were tested using either a disk diffusion test, an agar disk elution modification of the proportion method, or both.16 The three isolates not tested by these methods were tested by a broth dilution method.16 Likewise, because of taxonomic uncertainties pertaining to M. chelonae and M. abscessus that occurred during the period under study, the species designations of isolates identified as either one of these were combined and reported herein as M. chelonae-abscessus.
Results
The most common presenting complaint was chronic (2 to 72 months), nonhealing skin lesions (10/11 cats; 2/7 dogs) [Table 1]. In all cases, cytopathological evaluation of aspirates revealed pyogranulomatous inflammation. Microorganisms stained poorly, if at all, when smears of aspirates were stained with Gram’s or a Romanowsky-type stain (e.g., Wright’s, Giemsa). It was not unusual to observe “speckled” structures or nonstaining “ghosts,” presumed to be poorly stained or nonstaining mycobacteria, respectively. Acid-fast stained smears of aspirates obtained from affected sites exhibited acid-fast bacteria in 50% (4/8) of cats tested but none (0/6) of the affected dogs tested. Exudates obtained from affected cats showed very few acid-fast bacteria (usually <1 per 10 oil-immersion fields). Six of six tested and affected cats were negative for feline leukemia (FeLV) and feline immunodeficiency (FIV) viruses.
Rapidly growing members of the genus Mycobacterium were associated almost exclusively with the skin in affected cats, whereas in dogs there were more varied locations [Table 1]. Mycobacterium fortuitum was most commonly (11/18 cases) isolated from all affected animals, although M. chelonae-abscessus (6/18) and M. flavescens (1/18) were also encountered [Table 1].
The mycobacterial isolates were susceptible to amikacin (100% of the isolates), cefoxitin (93.8%), ciprofloxacin (75%), clarithromycin (71.4%), doxycycline (28.6%), erythromycin (6.2%), gentamicin (68.8%), kanamycin (75%), minocycline (81.3%), streptomycin (14.3%), tobramycin (43.8%), trimethoprim/sulfonamides (57.1%), and vancomycin (15.4%) [Table 2].
Treatment outcome was not tabulated, as the authors believed this would be potentially misleading because each patient was treated with a variety of antimicrobial agents, making it difficult (if not impossible) to accurately ascribe success (or failure) to a particular one.
Discussion
It is difficult to ascribe an underlying condition to all of the dogs and cats described in this report, although in four of the seven canine patients an underlying condition could be found (after surgery/trauma [n=3]; exudate from the skin around a dialysis catheter in a dog with chronic renal failure [n=1]). As has been reported by others, all of the cats that were tested for FeLV and FIV were negative.141517
The authors found that smears of affected material stained with an acid-fast stain were positive in 50% of tested cats and negative in affected dogs tested. These are useful observations, since other mycobacterial conditions affecting dogs and cats (e.g., feline leprosy, canine leproid granulomatosis, tuberculosis) almost always show acid-fast bacilli in affected material in numbers high enough to be visualized.1518 The results of bacteriological culture also differentiate the rapidly growing mycobacteria from the organisms associated with feline leprosy (i.e., M. lepraemurium) and canine leproid granulomatosis (associated with an unnamed member of the genus Mycobacterium), as these are conditions from which an etiological agent has been difficult to grow in vitro.1518–20 Both of these agents group with the slowly growing mycobacteria.1920 The rapidly growing mycobacteria usually form visible colonies on a blood agar plate within 72 hours of incubation at 37°C in room air.
The types of lesions or affected areas differed between dogs and cats. Almost without exception, cats presented with lesions affecting the skin. The types of lesions observed with dogs, on the other hand, were more varied (e.g., lung, joint, intravenous catheter, as well as skin). These observations are similar to those of other studies, but an explanation is not readily apparent.141521–24
Although surgical intervention is sometimes an important part of the treatment of infectious processes involving rapidly growing mycobacteria, antimicrobial therapy is important as well. Since the results of antimicrobial susceptibility testing take time to acquire, the use of the appropriate antimicrobial agent depends upon data acquired retrospectively. The authors’ results show that these bacteria are more likely (>70% of the isolates) to be susceptible in vitro to amikacin, cefoxitin, ciprofloxacin, clarithromycin, kanamycin, and minocycline and less likely to be susceptible to doxycycline, erythromycin, gentamicin, streptomycin, tobramycin, trimethoprim/sulfonamides, and vancomycin. It has been the authors’ experience that resolution of lesions takes several weeks. Therefore, it is encouraging that orally administered agents such as minocycline, ciprofloxacin, and clarithromycin appear to be rational choices.
Difco Laboratories, Inc., Detroit, MI
