OVERVIEW: What every clinician needs to know

Pathogen name and classification

Originally classified as a Corynebacterium, Rhodococcus is now classified as a member of the family Nocardioform, order Actinomycetes. The classification of this genus results from the following properties of rhodococci: mycolic, acid-containing cell walls and production of tuberculostearic acid. Members of this genus lack the ability to ferment carbohydrates and liquefy gelatin, which are characteristic abilities of pathogenic corynebacteria.

Rhodococcus is a facultative intracellular, gram-positive bacteria. It is an aerobic organism. As the name suggests (rhodo = red and coccus = round bacteria) it sometimes appears coccoid in shape. However, the organism is pleomorphic. Depending on nutrient and growth environment, it can also appear in a bacillary form. The bacillary form can appear as a variety of shapes, from curved and long to branched, short filaments.

History and important pathogens

Infection with Rhodococcus is a zoonosis. It was first discovered in 1923, in sheep with pyogranulomatous pneumonia. Since then, it has been documented in many types of animals, predominately herbivores. In 1967, the first human case was reported in a man being treated with steroids and 6-mercaptopurine for autoimmune hepatitis who developed pneumonia; the organism was isolated from his lung tissue. There were few cases reported subsequently until the time of the human immunodeficiency virus (HIV) epidemic in the 1980s. As impaired cell-mediated immunity in patients seems to play a role in the ability of this organism to cause disease, it follows that more cases have been seen since advances in treatment of malignancies and organ transplantation have led to increases in the numbers of immunocompromised hosts. Historically it has rarely been reported in people with intact immune systems; however, this may be changing because of heightened awareness of the organism.

Rhodococcus includes the following human pathogens: R. equi, R. erythropolis, R. gordoniae, R. fascians (R. luteus), and R. rhodochrous (complex). Most human infections caused by Rhodococcus have been identified as R. equi and much of the literature discusses this organism. Likewise, most of the information in this resource focuses on R. equi. Approximately 10 to 15% of reported infections caused by R. equi have occurred in immunocompetent individuals, although a recent article states a rate as high as 40%.

What is the best treatment?

  • Because there is no established standard for treatment, an Infectious Diseases consultation is recommended to assist in the care of patients. An Infectious Diseases consultation is also recommended because timely initiation of comprehensive treatment is essential to a favorable outcome.

  • Glycopeptide antibiotics (vancomycin, teicoplanin), rifampicin, quinolones, aminoglycosides, carbapenems, and in many cases, macrolides are effective. Alternative agents include tetracyclines, clindamycin, and trimethoprim/sulfamethoxazole. However, susceptibility to these agents is variable. More recently, linezolid has been identified as an effective agent. There has been some controversy surrounding the use of cephalosporins and penicillins due to possible acquired or intrinsic resistance. Hence, these agents are not recommended. If they must be used, it is recommended that a β-lactamase inhibitor be utilized with these agents. Vancomycin-based regimens were shown to be effective in mice and in a multicenter study of HIV-infected persons. In general, regimens containing two to three agents are recommended and are effective in avoiding the emergence of resistance. Use of antibiotics that are active intracellularly, such as azithromycin and rifampicin, can be effective in limiting intrahistiocytic survival. Many experts view the combination of vancomycin, rifampicin, and imipenem as the most efficacious.

  • Duration of treatment is dependent on both degree and anticipated duration of immunosuppression, as well as severity of disease. In immunocompetent hosts treatment can be 2 to 8 weeks versus up to 6 months of total treatment, including initial intravenous antibiotic induction, in patients with impaired cellular immunity. Treatment should be continued until the signs and symptoms of infection resolve.

  • Localized, non-central nervous system infection in immunocompetent hosts can be treated with oral agents.

  • Abscesses and empyemas should be drained. Lobectomy should be considered if there is poor clinical response to antibiotic therapy. If surgery is undertaken, all infected tissues should be excised.

  • Susceptibility testing should be performed to guide definitive therapy. As in most organisms, resistance can develop in rhodococci; however, the resistance patterns have not been identified definitively. There are several postulated mechanisms of resistance. These include:

    altered penicillin binding proteins

    increased antibiotic degradation

    β-lactamase production

  • Susceptibility testing can be performed by the method of Kirby–Bauer on cation-adjusted Mueller–Hinton agar. The minimum inhibitory concentration (MIC) can be measured by Etest on these plates. Extrapolation of Clinical Laboratory and Standards Institute criteria for Corynebacterium spp. breakpoints allows for interpretation of resistance patterns.

How do patients contract this infection, and how do I prevent spread to other patients?

  • Epidemiology—with regards to transmission, most information is based on R. equi.

    Transmission is by inhalation, ingestion, or inoculation of the organism, usually from soil. Exposure to farm animals, soil, and manure has been reported in many cases. Apparently, close contact is not required as evidenced by several documented cases in the literature. There are, however, reports of HIV-positive individuals who have become ill, following contact with roommates infected with R. equi. This raises the issue of possible person-to-person transmission. Other patients with documented disease have reported only secondary exposure. In these cases, the patients have had no direct contact with livestock or soil; instead, they have had contact with an individual who has had those exposures. In animals, ingestion of soil is described as another route of exposure. This has not, as yet, been documented in humans. Human-to-human transmission has, occasionally, been suspected in cases of nosocomial pneumonia involving HIV-positive individuals. There has been discussion of seasonal and regional differences in R. equi transmission on farms possibly secondary to precipitation and heat levels, but there is insufficient data to clarify definitive patterns. There is even less information on seasonality with respect to human infection.

    The organism has also been isolated from water. In the veterinary literature, there is mention of crowding as a risk for transmission, as such most stables/farms employ methods to reduce crowding to mitigate risk. It is reasonable to consider similar measures when dealing with human cases of R. equi. There is even less data on infection prevention for other types of rhodococci. Although found in soil, R. erythropolis has also been cultured from the surface of healthy eyes. Likewise, there is documentation of Rhodococcus as a nasal commensal organism in approximately 40% of human adults. The etiologic role of nasal colonization in Rhodococcus infection is unclear.

    Incidence is difficult to gauge. On the one hand, possible infection and mortality due to R. equi in HIV-infected individuals has decreased as the use of highly active antiretroviral therapy (HAART) becomes widespread. On the other hand, as technology improves and becomes more available, there are larger numbers of persons exposed to chemotherapy, immunosuppressive treatments related to solid organ transplantation and rheumatologic disease, and long-term dialysis. R. equi is an opportunistic pathogen.

    The initiation of HAART in HIV-positive individuals, with decreased CD4 counts, has been associated with decreased risk of R. equi infection. Azithromycin used as prophylaxis in many patients with acquired immune deficiency syndrome may also be beneficial in decreasing infection with this organism. Despite this, there is no evidence for long-term directed prophylaxis. Although speculative, there are two possible arguments that support the practice of not using prophylaxis. The first is that there seems to be evidence that immune reconstitution by use of HAART is more efficacious for protection from rhodococcal disease than adjunctive antibiotics, such as azithromycin. The second argument is that in veterinary practice, in which there is significantly more regular and widespread experience with R. equi, prophylaxis is not part of routine practice.

  • Infection control issues

    Inhalation and inoculation are two important routes of infection. There are no specific recommendations or guidelines for infection prevention in patients with rhodococcal infection. However, in reviewing cases involving R. equi, it seems prudent to recommend that the following be considered: if the patient is coughing or has a definitive pneumonia, surgical masks should be considered for staff and exposed patients. Additionally, masks should be considered when caring for a patient who has documented R. equi pneumonia and who is coughing. Gowns and gloves should be employed during contact with patients with rhodococcal wounds or infections; if wounds are likely to produce material that could be aerosolized, masks might also be considered.

    Vaccine technology has not been effective in preventing R. equi in horses or foals, due to immunological complications. At this time, there is no active research on a vaccine for humans.

What host factors protect against this infection?

  • Intact cellular immunity is protective against infection with R. equi. This may also be true of other rhodococci.

  • Patients with impaired cellular immunity are at higher risk for contracting infection secondary to Rhodococcus. This includes HIV-positive patients, particularly if their CD4 count is less than 100cells/mm3, patients who have undergone chemotherapy, and patients who are on immunosuppressants after receiving organ transplantation. In fact, most patients who are considered immunocompromised can be included here. Therefore, diabetics, alcoholics, patients with chronic renal failure, patients with sarcoidosis, patients on corticosteroids or immunosuppressive monoclonal antibodies, and preterm infants are at risk for infection.

  • A necrotizing granulomatous reaction is usually seen on histopathology. It is characterized by dense histiocytic infiltrates with eosinophilic, granular cytoplasm. Intrahistiocytic coccobacilli are present. This is, likely, the result of R. equi’s ability to survive within macrophages. Its ability to inhibit phagosome-lysosome fusion allows for proliferation of organisms intracellularly. The infected macrophages are, subsequently, destroyed. In most patients who manifest rhodococcal infection, cell-mediated immunity is impaired. They cannot mount an adequate response to rapidly eradicate Rhodococcus. Together with the organism’s ability to survive in the macrophage, this accounts for a low-grade but persistent inflammation and granuloma formation and extrapulmonary manifestations. In some cases, Michaelis–Guttman bodies are prominent (Figure 1); these are concentrically layered basophilic inclusions resulting from impaired macrophage lysosomal function. This leads to failure of lysophagosomal fusion and ineffective killing of ingested organisms. This, together with the Michaelis–Guttman bodies, is called malakoplakia.

Figure 1:

Michaelis-Gutmann bodies.

What are the clinical manifestations of infection with this organism?

  • A cavitary and sometimes necrotizing pneumonia is the most common clinical manifestation of R. equi infection (Figure 2). Eighty percent of R. equi cases demonstrate pneumonia. This is a pneumonia of subacute onset. Disease may start with an insidious, dry cough. Initial symptoms may be constitutional, with cachexia, weight loss, and fatigue. Typically, patients demonstrate progressive cough, fever, and pleuritic chest pain. Hemoptysis is also described in 15% of patients with R. equi pneumonia.

  • Chest radiographs often demonstrate upper lobe cavitation. This radiologic finding makes it easy to confuse rhodococcal infection with M. tuberculosis infection. However, in R. equi, the upper lobe cavitation is thick walled and frequently includes an air-fluid level. Other radiographic findings that can be seen in R. equi infections include pleural effusion, infiltrates, and nodules.

  • There is a significant frequency of bacteremia (30–50%) in R. equi infections which increases with degree of immune impairment. Bacteremia has also been reported in R. erythropolis infection.

  • The next most frequent sites of infection are central nervous system and subcutaneous abscesses.

  • Concurrent extrapulmonary infection is seen in approximately 20% of pulmonary cases. Three patterns of extrapulmonary infection have been described:

    traumatic (including wounds, septic arthritis, and endophthalmitis

    isolated bacteremia

    inoculation of the gastrointestinal tract with dissemination to regional lymph nodes, peritonitis, or pelvic abscess

  • Infections of devices, such as intravascular catheters, have been reported.

  • There have been a variety of case reports of infections with non-equi species such as osteomyelitis caused by R. erythropolis and ocular infections secondary to traumatic inoculation of R. gordoniae.

Figure 2:

Cavitary lesions caused by Rhodococcus.

What common complications are associated with infection with this pathogen?

  • Bacteremia frequently complicates R. equi pneumonia.

  • Pneumonia is also frequently complicated by lung abscess, pleural effusion, empyema, and pneumothorax.

  • Microabscesses are also seen, and these can coalesce into larger abscesses. Abscesses can also occur as a result of hematogenous spread of this pathogen to more distant sites, such as the brain.

  • Other complications of R. equi infection include endobronchial lesions, endocarditis, cardiac tamponade, and mediastinitis. Non-equi rhodococcal meningitis has also been documented.

  • Granuloma formation is common in rhodococcal infections. Both endobronchial and pulmonary granulomas are seen.

  • Malakoplakia is a less common complication of rhodococcal infection. Malakoplakia is a chronic, granulomatous inflammatory process associated with an impaired ability to process microorganisms within histiocytes. It is best described as an accumulation of benign macrophages associated with intracellular and extracellular aggregates of periodic acid-Schiff-positive histiocytes. These histiocytes are known as Michaelis–Guttman bodies, and they contain lamellated iron and calcium inclusions.

  • Relapse of infection is also seen; this is another way in which Rhodococcus emulates mycobacteria. The ability of the Rhodococcus to survive and multiply inside the macrophage, as well as the propensity to form abscesses, is the likely mechanism for this complication of infection.

  • Mortality is greatest in HIV-infected individuals who are not on HAART, and is as high as 50 to 55%. The mortality is 20 to 25% in patients with depressed immunity due to other causes, such as chemotherapy or transplant-related medications, and is 11% in immunocompetent patients.

How should I identify the organism?

Rhodococcus life cycle: R. equi inhabits the soil and gains entry into the respiratory tract via inhalation of dust/soil particles containing bacteria. In nonhumans, ingestion of soil may be another method of transmission; this, then, can lead to bloodstream infections. The organism replicates within the respiratory tract, is then coughed up, and then swallowed. Replication continues as rhodococci travel through the intestinal tract. Subsequently, Rhodococcus is eliminated by defecation, reintroducing large numbers of the bacteria back into the soil.

  • Microscopy and culture are the detection methods of choice.

  • The staining techniques which are useful are: Gram stain which shows the organism as gram-positive and an acid fast stain which shows the organism as partially acid fast. Rhodococcus is a pleomorphic organism. Depending on nutrient, environmental, and growth conditions, it can range in morphology from a long-curved rod, to a short rod with branching or filament formation, to a coccus.

  • The colonial appearance of Rhodococcus is quite variable, rendering phenotypic lab testing insufficient. In addition, the characteristic cell-wall lipids in Rhodococcus have more than one critical function; they are the end-product of complex biochemical pathways. The unique lipids produced by these bacteria do not lend themselves to antigen variation in the same manner as surface proteins found in many other bacteria. Therefore, molecular methods must, frequently, be employed to establish species identification. In addition, Rhodococcus possesses the ability to act as a biocatalyst; this is particularly true of R. erythropolis which possesses a large set of enzymes. For this reason, R. erythropolis has attracted attention in the biotechnology world. As Rhodococcus is so difficult to identify, multiple methods of testing are used simultaneously. The organism can grow at a wide range of temperatures, from 10 to 40oC. However, it grows optimally at approximately 30oC. Colonies grow on solid media (see below for preferred media). It is an obligate aerobe which is nonmotile and does not produce spores.

    Rhodococcus grows well on ordinary media but may be overlooked as a contaminating coryneform or misidentified as Nocardia or Micrococcus. Selective media such as colixin-naladixic agar (CNA), phenyl-ethanol agar (PEA), and ceftazidime-novobiocin agar are useful in the isolation of R. equi from contaminated specimens.

    R. equi colonies appear as irregularly round, smooth, mucoid, and salmon-colored (Figure 3). The latter feature occurs after 4 to 7 days. As indicated above, colonial appearance is variable.

    Rhodococcus is biochemically characterized by positivity for catalase, lipase, urease, and phosphatase. Absent are oxidase, deoxyribonuclease, elastase, lecithinase, and protease.

    Colonies form on solid media in approximately 48 hours. However, the characteristic salmon color may take 4 to 7 days of growth to manifest.

    Polymerase chain reaction is helpful in identifying Rhodococcus, but this testing is not commercially available. Sensitivity and specificity appear to be high.

  • Rhodococcus can be difficult to differentiate from Nocardia, Bacillus, and Micrococcus when grown on nonspecialized media. Therefore, CNA, PEA, or ceftazidime-novobiocin agar are recommended culture media. As previously mentioned, Rhodococcus shares many characteristics with mycobacteria. Rhodococcus can also be difficult to differentiate from mycobacteria because Rhodococcus is weakly acid fast. The 14-day arylsulfatase test is negative in Rhodococcus; this helps to distinguish it from some species of mycobacteria. Synergistic hemolysis is another method which helps to identify Rhodococcus. Blood agar cross-streaked with Staphylococcus aureus, Corynebacterium pseudotuberculosis, or Listeria monocytogenes to which Rhodococcus is added creates synergistic hemolysis (Christie Atkins Munch-Petersen test-like), as a result of enzymatic activity. An additional characteristic and defining feature is that in vitro antagonism between imipenem and other β-lactams is common in R. equi isolates, unlike other bacteria.

  • Restriction fragment length polymorphism and ribotyping can be effective adjuncts in identifying the organism.

Figure 3:

Characteristic “salmon”-colored colonies.

How does this organism cause disease?

  • The significance of virulence factors in human infection is not well understood. These seem to vary between strains of R. equi, just as the host range of R. equi strains seems to vary. In R. equi strains infecting livestock, there is fairly consistent expression of a 20kDa antigen called VapA. Expression of the VapA antigen demonstrates measurable differences in R. equi strains in different geographic regions. VapA is considered to impart significant virulence in these strains which infect nonhumans. But VapA has only been found to be expressed in 20 to 25% of strains recovered in human infection, which is the reason that its role in virulence in humans is questioned. Instead, rhodococcal virulence in humans seems to be related to the organism’s ability to evade the immune system. The ability of R.equi to survive within macrophages is its most important virulence factor in humans. Its ability to inhibit phagosome-lysosome fusion allows for proliferation of organisms intracellularly in the macrophage. Subsequently, inflammation and ultimately destruction of infected macrophages occurs. Person-to-person transmission in pleuropulmonary disease, as well as wounds, can be understood as possible through aerosolization of the pathogen.

  • The clinical manifestations and indolent beginning of most rhodococcal infections correlates with the prolonged presence of this relatively low-virulence organism in cells. The clinical illness is indolent and, even in foals in which R. equi has been a recognized pathogen since the 1920s, symptoms can be hard to recognize. The cough may have an insidious onset, as does the corresponding apathy. Later, unexplained fever may be the first objective indication of active infection. By this time, some of the more definitive aspects of illness may be present, such as: cavitation, empyema, pleural effusion, bacteremia, pneumothorax, microabscesses, etc. Because Rhodococcus has the ability to persist and multiply in cells by evasion of the immune response or the lack of cell-mediated immune response in cases of impaired cell-mediated immunity, this organism causes granuloma formation, cavitary lesions, and slow invasion of tissue to the level of involvement of the bloodstream, in some cases.

WHAT’S THE EVIDENCE for specific management and treatment recommendations?

Chen, XY, Xu, F, Xia, JY, Cheng, YS, Yang, Y. “Bacteremia due to : a case report and review of the literature”. J Zhejiang Univ Sci B. vol. 10. 2009. pp. 933-6. (A case report and a good review of the literature.)

Conlin, PA, Willingham, KJ, Oliver, JW. ” bacteremia in an immunocompetent patient”. Lab Med. vol. 31. 2000. pp. 263-5. (A case report which includes discussion of laboratory techniques for isolation and identification and characteristics of Rhodococcus.)

Corti, M, Palmero, D, Eiguchi, K. “Respiratory infections in immunocompromised patients”. Curr Opin Pulmonary Med. vol. 15. 2009. pp. 209-17. (Overview of pneumonias in immunosuppressed patients, including those secondary to Rhodococcus: prophylaxis, clinical manifestations, microbiological and molecular identification, and treatment strategy.)

Finnerty, WR. “The biology and genetics of the genus “. Annu Rev Microbiol. vol. 46. 1992. pp. 193-218. (An overview of the characteristics of the genus.)

Gelfand, MS, Cleveland, KO, Brewer, SC. ” pneumonia in a patient with fludarabine-treated chronic lymphocytic leukemia and CD4-lymphopenia”. Am J Med Sci. vol. 340. 2010. pp. 80-1. (A case report with discussion of specific immunologic facets of Rhodococcus-related illness.)

Harvey, RL, Sunstrum, JC. ” infection in patients with and without human immunodeficiency virus infection”. Rev Infect Dis. vol. 13. 1991. pp. 139-45. (Overview of Rhodococcus infection with comparison of manifestations in immunocompetent versus immunosuppressed individuals.)

Holt, JG, Krieg, NR, Sneath, PH, Staley, JT, Williams, ST. Bergey’s Manual of Determinative Bacteriology. 1994. pp. 611-39. (A review of morphologic and biochemical characterisics of the organism.)

Kedlaya, I, Ing, MB, Wong, SS. ” Infections in immunocompetent hosts: case report and review”. Clin Infect Dis. vol. 32. 2001. pp. e39-45. (Case report and review of literature.)

Logan, NA, Lappin-Scott, HM, Oyston, PCF. “Prokaryotic diversity: mechanisms and significance. 66th Symposium of the Society for General Microbiology”. 2006. (Includes discussion of biological characteristics of Rhodococcus.)

Lora-Tamayo, J, Fernández-Sabé, N, Viladrich, PF, Verdaguer, R, Gabarrós, A. “Ventriculo-peritoneal shunt infection due to non- sp”. Int J Infect Dis. vol. 13. 2009. pp. e193-4. (Case report of a non-equi Rhodococcus infection.)

Nishi, SP, Valentine, VG, Duncan, S. “Emerging bacterial, fungal, and viral respiratory infections in transplantation”. Infect Dis Clin North Am. vol. 24. 2010. pp. 541-55. (Overview including Rhodococcus as an emerging pathogen, specifically in the transplant population.)

Park, SD, Uh, Y, Jang, IH, Yoon, KJ, Kim, HM, Bae, YJ. ” septicaemia in a patient with acute lymphocytic leukemia”. J Med Microbiol. vol. 60. 2011. pp. 252-5. (Case report of Rhodococcus erythropolis infection.)

Reboli, A., Meyer, D., Mandell, GL, Bennett, JE, Dolin, R. “Other Coryneform Bacteria and Rhodococci”. Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases. 2010. (Reference chapter from the major Infectious Diseases textbook.)

Takai, S, Sasaki, Y, Ikeda, T, Uchida, Y, Tsubaki, S, Sekizaki, T. “Virulence of Isolates from patients with and without AIDS”. J Clin Microbiol. vol. 32. 1994. pp. 457-60. (A comparison of the microbiologic characteristics of Rhodococcus isolates in immunocompetent and immunosuppressed individuals.)

Topino, S, Galati, V, Grilli, E, Petrosillo, N. ” Infection in HIV-infected individuals: case reports and review of the literature”. AIDS Patient Care STDS. vol. 24. 2010. pp. 211-22. (Case reports and literature review.)

Weinstock, DM, Brown, AE. “: an emerging pathogen”. Clin Infect Dis. vol. 34. 2002. pp. 1379-85. (Overview of the diverse clinical manifestations of infection and accurate diagnosis.)

Yam, KC, Okamoto, S, Roberts, JN, Eltis, LD. “Adventures in —from steroids to explosives”. Can J Microbiol. vol. 57. 2011. pp. 155-68. (Overview of biochemical and microbiologic characteristics of Rhodococcus.)

Yamshchikov, AV, Schuetz, A, Lyon, GM. ” infection”. Lancet Infect Dis. vol. 10. 2010. pp. 350-9. (Overview of pathogenesis, clinical characteristics of infections, and approaches to treatment.)