(Chest. 2003;123:1970-1976.)
© 2003
American College of Chest Physicians
Prognosis and Clinical Evaluation of Infection Caused by Rhodococcus equi in HIV-Infected Patients*
A Multicenter Study of 67 Cases
Manuel Torres-Tortosa, MD;
Julio Arrizabalaga, MD;
José L. Villanueva, MD;
Juan Gálvez, MD;
María Leyes, MD;
M. Eulalia Valencia, MD;
Juan Flores, MD;
José M. Peña, MD;
Elisa Pérez-Cecilia, MD and
Carmen Quereda, MD; for Grupo Andaluz para el estudio de las Enfermedades Infecciosas, and Grupo de estudio de SIDA of the Sociedad Española de Enfermedades Infecciosas y Microbiología Clínica
* From the hospital Punta de Europa (Dr. Torres-Tortosa), Algeciras; hospital Nuestra Señora de Aranzazu (Dr. Arrizabalaga), San Sebastián; hospital Virgen del Rocío (Dr. Villanueva), Sevilla; hospital Juan Ramón Jiménez (Dr. Gálvez), Huelva; hospital Son Dureta (Dr. Leyes), Palma de Mallorca; Instituto de Salud Carlos III (Dr. Valencia), Madrid; hospital Arnau de Vilanova (Dr. Flores), Valencia; hospital La Paz (Dr. Peña), Madrid; hospital Clínico de San Carlos (Dr. Pérez-Cecilia), Madrid; and hospital Ramón y Cajal (Dr. Quereda), Madrid, Spain.
A complete list of participants is given in the Appendix.
Correspondence to: Manuel Torres-Tortosa, MD, Apartado n° 289. 11280 Algeciras (Cádiz), Spain; e-mail: mtt{at}comcadiz.com
 |
Abstract
|
|---|
Objective: To assess the clinical characteristics and the factors that influenced the prognosis of patients with HIV and infection caused by Rhodococcus equi.
Design: Observational, multicenter study in 29 Spanish general hospitals.
Setting: These hospitals comprised a total of 20,250 beds for acute patients and served a population of 9,716,880 inhabitants.
Patients: All patients with HIV and diagnosed R equi infection until September 1998.
Results: During the study period, 19,374 cases of AIDS were diagnosed. Sixty-seven patients were included (55 male patients; mean ± SD age, 31.7 ± 5.8 years). At the time of diagnosis of R equi infection, the mean CD4+ lymphocyte count was 35/µL (range, 1 to 183/µL) and the stage of HIV infection was A3 in 10.4% of patients, B3 in 31.3%, C3 in 56.7%, and unknown in 1.5%. R equi was most commonly isolated in sputum (52.2%), blood cultures (50.7%), and samples from bronchoscopy (31.3%). Chest radiographic findings were abnormal in 65 patients (97%). Infiltrates were observed in all of them, with cavitations in 45 patients. The most active antibiotics against the strains isolated were vancomycin, amikacin, rifampicin, imipenem, ciprofloxacin, and erythromycin. After a mean follow-up of 10.7 ± 12.8 months, 23 patients (34.3%) died due to causes related to R equi infection and 6 other patients showed evidence of progression of the infection. The absence of highly active antiretroviral therapy (HAART) was independently associated with mortality related to R equi infection (relative risk, 53.4; 95% confidence interval, 1.7 to 1,699). Survival of patients treated with HAART was much higher than that of patients who did not receive this therapy.
Conclusions: Infection by R equi is an infrequent, opportunistic complication of HIV infection and occurs during advanced stages of immunodepression. In these patients, it leads to a severe illness that usually causes a bacteremic, cavitary pneumonia, although HAART can improve the prognosis.
Key Words: AIDS • bacteremia • HIV • pneumonia • Rhodococcus equi
|
Introduction
|
|---|
Rhodococcus equi, previously known as Corynebacterium equi, is a weak, acid-fast, Gram-positive rod. It is currently classified among the nocardiform actinomycetes.1
The ability of R equi to remain inside macrophages and even destroy them is considered to be the basis for its pathogenicity.1
2
3
It produces a necrotizing granulomatous lesion rich in macrophages with periodic acid-Schiff–positive granular cytoplasm. Occasionally, a peculiar histopathologic lesion is observed, known as malakoplakia, which is characterized by histiocytes with cytoplasmic inclusions laminated with iron and calcium (Michaelis-Gutmann bodies2
4
5
6
). It grows well in ordinary culture media and forms salmon-pink colored colonies.1
2
6
R equi is a common pathogen of pneumonia in foals and sometimes produces infections in other mammals.1
2
Although natural exposure to R equi is frequent,7
the first infection by this organism in humans was described in 1967.8
It mainly affects immunocompromised patients, especially those with HIV infection.9
10
Although it could be underdiagnosed in the past,2
11
and despite some reports that have detected an increasing incidence,9
it is a very uncommon illness, both in general population and in patients with HIV infection. It has been reported only in isolated cases or in short series with revision of cases previously published.10
12
13
This study reports the clinical characteristics and factors that influenced the prognosis of 67 patients with HIV and diagnosed R equi infection.
|
Materials and Methods
|
|---|
An observational, multicenter study was performed in 29 general hospitals from several regions in Spain. These hospitals comprised a total of 20,250 beds for acute patients and served a population of 9,716,880 inhabitants. All of the patients with HIV and R equi infection from the beginning of the AIDS epidemic until September 1998 were included. Information from every patient was collected by a previously designed form. The follow-up of each patient was carried out for as long as possible, and it was finished in December 1998. In those cases diagnosed before the beginning of this study, this information was collected retrospectively.
Isolation, identification, and susceptibility tests of the R equi strains were performed in the microbiology laboratory of each hospital. AIDS was diagnosed using the diagnostic criteria in force in Europe at the time of diagnosis of the R equi infection.14
15
16
Definitions
Outcome of R equi Infection: This was assessed at the end of the follow-up period using three categories: cure, disappearance of the initial lesions and absence of manifestations of infection 1 month after withdrawing antimicrobial treatment; regression, reduction of the initial lesions and/or improvement of symptoms attributable to R equi infection during the course of antimicrobial therapy; and progression, increase of the lesions or exacerbation of symptoms attributable to R equi infection during the course of treatment, as well as the recurrence of infection after withdrawing treatment.
Survival: Survival was evaluated using the following categories: (1) death associated with R equi infection (related mortality), when the patient’s death occurred due to causes directly attributable to the infection, to complications of the infection or to the therapy used; (2) death not associated with the infection, when death occurred with evidence of regression or cure of the infection and due to a cause not associated with the infection or its treatment; (3) and alive, when the patient was still alive at the last evaluation after the diagnosis of R equi infection.
Antibiotic Therapy: Antibiotic therapy was defined as appropriate when at least two active drugs were used against the causal agent and both during a period > 90 days. Treatment was considered as inappropriate in any other case.
Highly Active Antiretroviral Therapy (HAART): HAART was defined as the combination of two reverse transcriptase inhibitors with protease inhibitors for a period of > 30 days.
Statistical Analysis
A descriptive analysis of all the clinical characteristics collected from each patient was performed. Related mortality with R equi infection was considered to be the dependent variable for the analysis of the prognosis. The following variables were analyzed to know their possible association with related death: diagnosis of R equi infection before 1997, bacteremia, extrapulmonary location of the infection, multilobar pneumonia, previous or concomitant diagnosis of AIDS, CD4+ lymphocytes count, inappropriate antibiotic therapy, surgery, and absence of HAART. A statistically significant association was considered to exist when the p value was < 0.05. First, a univariate analysis was performed by means of the Cox proportional hazards method. Relative risks (RRs) and 95% confidence intervals (CIs) were calculated. Then, all the variables associated with related death in the univariate analysis were included in a multivariate Cox regression model, by means of a forward stepwise method, and contrasted by the likelihood ratio. Survival of patients with and without HAART was analyzed using the Kaplan-Meier method, and its statistical significance was assessed using the log-rank test.
|
Results
|
|---|
During the study period, 19,374 new patients received a diagnosis of AIDS in the participant hospitals. Sixty-seven patients with HIV received a diagnosis of R equi infection during 9 years (Fig 1
). Previous exposure to R equi was suspected in 10 patients. Mean ± SD age was 31.7 ± 5.8 years (range, 20 to 60 years), and most were male (55 patients, 82%). Risk activities for HIV infection were as follows: IV drug use, 48 patients (71.6%); heterosexual lifestyle, 11 patients (16.4%); bisexual or homosexual lifestyle, 3 patients (4.5%); transfusions, 2 patients (3%); blood derivatives, 1 patient (1.5%); and unknown, 2 patients (3%). Only eight of the patients who used IV drugs were still using IV drugs at the time of diagnosis of R equi infection.
The stage of HIV infection at the time of diagnosis of R equi infection was A3 in 7 patients (10.4%), B3 in 21 patients (31.3%), C3 in 38 patients (56.7%), and not clear in 1 patient (1.5%). The average CD4+ lymphocyte count was 35/µL (range, 1 to 183/µL). The most common symptoms attributable to R equi infection were fever (91%) and respiratory symptoms: cough (88.1%), expectoration (85.1%), and chest pain (44.8%) [Table 1 ]. The average duration of symptoms before hospital admission was 47.7 ± 48.6 days (range, 4 to 223 days). Most samples from which R equi was isolated were sputum (52.2%), blood cultures (50.7%), and samples from bronchoscopy (31.3%) [Table 2
]. Chest radiographic findings were abnormal in 65 patients (97%). Infiltrates were observed in all of them, and cavitations were observed in 45 patients. There was multilobar involvement in 13 patients, pleural effusions in 11 patients, and mediastinal lymphadenopathies in 2 patients. In patients in whom R equi infection was located in the lungs, the area involved was the upper right lobe in 20 patients, the middle lobe in 5 patients, the lower right lobe in 19 patients, the upper left lobe in 21 patients, and the lower left lobe in 18 patients. There was evidence of malakoplakia in tissue samples from five patients.
The mean follow-up period of patients after diagnosis of R equi infection was 10.7 ± 12.8 months. Organs affected by R equi infection are shown in Table 3
. The lung was involved in 95.5% of episodes, and bacteremia was observed in 59.8%.
Table 4
shows the antimicrobial susceptibility of strains that established the diagnosis of R equi infection. The most active antibiotics were vancomycin, amikacin, rifampicin, imipenem, ciprofloxacin, and erythromycin.
All but three patients received antibiotics for R equi infection. Eleven patients underwent surgery: lung resection (n = 4) and drainage techniques (n = 7). Twenty-four patients received HAART (35.8%), 42 patients (62.7%) did not receive HAART, and this information was unknown in 1 patient.
The mean follow-up was 10.7 months. Regarding the outcome of R equi infection, 10 patients (14.9%) fulfilled the criteria for cure, and there was regression in 28 patients (41.8%) and progression in 29 patients (42.3%). Furthermore, after this period, 30 patients (44.8%) were still alive, 23 patients (34.3%) died due to causes related to R equi infection, and 14 patients (20.9%) died due to causes not related to the infection. Thus, 29 patients (43.3%) died in relation to R equi infection or presented evidence of progression.
Table 5
shows the univariate analysis of the variables that were associated with the worst prognosis. These were multilobar involvement, absence of HAART, and inappropriate antimicrobial therapy for R equi infection. However, only the absence of HAART was independently associated with related mortality in the multivariate analysis (RR, 53.4; 95% CI, 1.7 to 1669). Figure 2
shows that the probability of death related to R equi infection among the 42 patients who did not receive HAART was much higher than that of the 24 patients who did. In fact, all the deaths for this reason occurred in patients who did not receive HAART.
|
Discussion
|
|---|
To our knowledge, this study describes the most extensive series published to date of R equi infection in patients with HIV. As this is a retrospective study, the data presented should be interpreted with caution and do not allow conclusions to be drawn on aspects that would require a previous homogeneous definition for all patients (for example, treatment used). However, we believe that this study furnishes information of interest, as it is a multicenter study carried out over a wide geographic area with a large population of patients with HIV infection. During the enrollment period of the 67 patients included in this study, 19,374 new cases of AIDS were diagnosed in the participant hospitals. This shows how uncommon this infection is, as well as the usefulness of a wide collection of cases in order to increase our knowledge about this process.
At the time of diagnosis of R equi infection, the average CD4+ lymphocyte count was very low. This fact shows that R equi induces disease in patients with HIV infection with advanced immunologic impairment and supports the opinion that R equi infection is considered an AIDS-defining event.17
The most frequently involved organ was the lung, and this fact is a constant in the human infection. Thus, in two series of R equi infection with a total of 36 patients with HIV infection, the lung was involved in 91.7%.12
13
Furthermore, in a review of cases of patients without HIV infection, there was respiratory involvement in 27 of the 54 cases reported.10
This justifies the fact that the clinical expression of the disease is mainly respiratory. In the present series, the lobes were involved with a similar frequency. Imaging techniques revealed pulmonary cavitations in two thirds of cases. In a series of 78 patients with HIV infection with pulmonary cavitations evaluated over a 7-year period, R equi was the causal agent in 7.7% of cases.18
Therefore, R equi should form part of the differential diagnosis of cavitary pneumonia in patients with HIV infection.19
As has already been pointed out,12
13
blood and sputum cultures were the samples with the best yield for the diagnosis of this infection. The pattern of antimicrobial susceptibility of the isolated strains is similar to that described in other studies,13
20
although this can vary in specific geographic areas21
or in isolations with previous antibiotic therapy.22
23
Infection caused by R equi was a severe illness given that, after the follow-up period, more than one third of the patients died due to causes attributable to the infection and in a further 9% there were evidences of progression of the infection despite the prescribed treatment. When variables that influenced the prognosis of the patients were evaluated, only the absence of HAART was associated with R equi-related mortality. Moreover, no patients receiving HAART died in relation to the infection caused by R equi. It has been pointed out that the infection caused by R equi has a worse prognosis in patients with HIV than in patients without HIV,24
and that they need a lifetime of antimicrobial therapy.2
24
However, a great change happened since HAART was introduced as standard therapy for HIV infection: there was a decrease in the incidence of opportunistic infections,25
its prognosis improved,26
27
and secondary chemoprophylaxis could be suspended with no recurrences.28
29
It is likely that HAART can allow us to withdraw the antibiotic therapy at some time in patients with HIV and R equi infections, as well as to reduce the incidence of this opportunistic infection and, as could be observed in this study, to improve its prognosis. In fact, the outcome of patients included in this study was more favorable than that observed in other series with similar patients but which were carried out before HAART was available in clinical practice.12
13
The peculiar pathogenicity of R equi infection determines the treatment of this process. Due to the high frequency of bacteremia and high bacterial loads, it may be appropriate to indicate a combination of IV antibiotics with bactericidal effect such as imipenem plus vancomycin or imipenem plus teicoplanin.30
31
Lipophilic antibiotics with good intracellular penetration must then be administered, and combinations based on macrolides and rifampin have proven to be optimal.1
2
13
Azithromycin seems to be an appropriate drug for the treatment of this infection in combination with others, as reaches high levels in tissues.32
Appropriate antibiotic treatment did not influence outcomes in this study. This could be explained as the drugs that the patients received were very different from each other. This important issue could only be clarified by means of prospective studies, where homogeneous regimens are previously established. Drainage of abscesses must be carried out when possible; however, we think that resection surgery should be limited to selected patients who do not respond to medical treatment.
In summary, R equi infection is an uncommon opportunistic complication of HIV infection that occurs in advanced stages of immunodepression. It usually appears as a subacute pneumonia that is usually cavitary and bacteremic. Although the most effective antimicrobial therapy is unknown, taking into account the pathogenic and biological features of this organism, antimicrobial therapy should be based on a combination of drugs with a good intracellular penetration, and administered over a long period. Infection by R equi in patients with HIV is a severe illness although, as this study shows, HAART can improve its prognosis.
|
Appendix
|
|---|
Other members of Grupo Andaluz para el estudio de las Enfermedades Infecciosas and/or Grupo de estudio de SIDA of the Sociedad Española de Enfermedades Infecciosas y Microbiología Clínica who participated in the study were Koldo Aguirrebengoa, MD, hospital de Cruces, Baracaldo; Antonio Bascuñana, MD, hospital Puerta del Mar, Cádiz; Félix Gutiérrez, MD, hospital General, Elche; Manuel Javaloyas, MD, hospital Sant Llorenç, Viladecans; Fernando Lozano, MD, hospital de Valme, Sevilla; Patricia Muñoz, MD, hospital Gregorio Marañón, Madrid; Antonio Payerás, MD, Complex Hospitalri, Palma de Mallorca; Jesús Rodríguez-Baños, MD, hospital Virgen de la Macarena, Sevilla; Jesús Santos, MD, hospital Virgen de la Victoria, Málaga; Josu Baraia, MD, hospital de Basurto, Bilbao; Marino Blanes, MD, hospital La Fe, Valencia; Mercedes González-Serrano, MD, hospital del Servicio Andaluz de Salud, La Linea de la Concepción; Pablo Labarga, MD, hospital San Millán, Logroño; Jaime Locutura, MD, hospital General Yagüe, Burgos;, Rafael Luque, MD, hospital de Motril; Juan Pascuau, MD, hospital Virgen de la Nieves, Granada; Ignacio Suárez, MD, hospital Infanta Elena, Huelva; Mauricio Telenti, MD, hospital General de Asturias, Oviedo; and Antonio Vergara, MD, hospital de Puerto Real; Spain.
|
Acknowledgements
|
|---|
We thank Dr. Moreno Maqueda from the Infectious Diseases Section, and Dr. Caballero-Granado from the Internal Medicine Service of Hospital Punta de Europa (Algeciras, Spain), for collaboration and advice in the statistical analysis. Furthermore, we thank Merck Sharp & Dohme for their financial assistance in the English translation of this manuscript.
|
Footnotes
|
|---|
Abbreviations: CI = confidence interval; HAART = highly active antiretroviral therapy; RR = relative risk
Received for publication June 19, 2002.
Accepted for publication October 8, 2002.
|
References
|
|---|
- Prescott, JF (1991) Rhodococcus equi: an animal and human pathogen. Clin Microbiol Rev 4,20-34[Abstract/Free Full Text]
- Verville, TD, Huycke, MM, Greenfield, RA, et al Rhodococcus equi infections of humans: 12 cases and a review of the literature. Medicine (Baltimore) 1994;73,119-132[Medline]
- Hondalus, MK, Mosser, DM Survival and replication of Rhodococcus equi in macrophages. Infect Immun 1994;62,4167-4175[Abstract/Free Full Text]
- Kwon, KY, Colby, TV Rhodococcus equi pneumonia and pulmonary malakoplakia in acquired immunodeficiency syndrome: pathologic features. Arch Pathol Lab Med 1994;118,744-748[ISI][Medline]
- Guerrero, MF, Ramos, JM, Renedo, G, et al Pulmonary malacoplakia associated with Rhodococcus equi infection in patients with AIDS: case report and review. Clin Infect Dis 1999;28,1334-1336[CrossRef][ISI][Medline]
- Scott, MA, Graham, BS, Verrall, R, et al Rhodococcus equi an increasingly recognized opportunistic pathogen; report of 12 cases and review of 65 cases in the literature. Am J Clin Pathol 1995;103,649-655[ISI][Medline]
- Vullo, V, Mastroianni, CM, Lichtner, M, et al Serologic responses to Rhodococcus equi in individuals with and without human immunodeficiency virus infection. Eur J Clin Microbiol Infect Dis 1996;15,588-594[CrossRef][ISI][Medline]
- Golub, B, Falk, G, Spink, WW Lung abscess due to Corynebacterium equi: report of first human infection. Ann Intern Med 1967;66,1174-1177[ISI][Medline]
- Linder, R Rhodococcus equi and Arcanobacterium haemolyticum: two "coryneform" bacteria increasingly recognized as agents of human infection. Emerg Infect Dis 1997;3,145-153[ISI][Medline]
- Farina, C, Ferruzzi, S, Mamprin, F, et al Rhodococcus equi infection in non-HIV-infected patients: two case reports and review. Clin Microbiol Infect 1997;3,12-18[Medline]
- Doig, C, Gill, MJ, Church, DL Rhodococcus equi: an easily missed opportunistic pathogen. Scand J Infect Dis 1991;23,1-6[ISI][Medline]
- Donisi, A, Suardi, MG, Casari, S, et al Rhodococcus equi infection in HIV-infected patients. AIDS 1996;10,359-362[ISI][Medline]
- Arlotti, M, Zoboli, G, Moscatelli, GL, et al Rhodococcus equi infection in HIV-positive subjects: a retrospective analysis of 24 cases. Scand J Infect Dis 1996;28,463-467[ISI][Medline]
- Centers for Disease Control.. Revision of the CDC surveillance case definition of acquired immunodeficiency syndrome. MMWR Morb Mortal Wkly Rep 1987;36(suppl 1),3s-15s
- European Centre for the Epidemiological Monitoring of AIDS. 1993 revision of the European AIDS surveillance case definition. AIDS Surveillance in Europe Quarterly Report 1993;37,23-28
- Ancelle-Park, R Expanded European AIDS case definition [letter]. Lancet 1993;341,441[Medline]
- Albrecht, H Redefining AIDS: towards a modification of the current AIDS case definition. Clin Infect Dis 1997;24,64-74[ISI][Medline]
- Rodríguez-Arrondo, F, von-Wichmann, MA, Arrizabalaga, J, et al Lesiones pulmona res cavitadas en los pacientes infectados por el virus de la inmunodeficiencia huma na: analisis de una serie de 78 casos. Med Clin (Barc) 1998;111,725-730
- Gallant, JE, Ko, AH Cavitary pulmonary lesions in patients infected with human immunodeficiency virus. Clin Infect Dis 1996;22,671-682[ISI][Medline]
- McNeil, MM, Brown, JM Distribution and antimicrobial susceptibility of Rhodococcus equi from clinical specimens. Eur J Epidemiol 1992;8,437-443[CrossRef][ISI][Medline]
- Hsueh, PR, Hung, CC, Teng, LJ, et al Report of invasive Rhodococcus equi infections in Taiwan, with an emphasis on the emergence of multidrug-resistant strains. Clin Infect Dis 1998;27,370-375[ISI][Medline]
- Nordmann, P, Chavanet, P, Caillon, J, et al Recurrent pneumonia due to rifampicin-resistant Rhodococcus equi in a patient infected with HIV. J Infect 1992;24,104-107[CrossRef][ISI][Medline]
- Van Etta, LL, Filice, GA, Ferguson, RM, et al Corynebacterium equi: a review of 12 cases of human infection. Rev Infect Dis 1983;5,1012-1018[ISI][Medline]
- Harvey, RL, Sunstrum, JC Rhodococcus equi infection in patients with and without human immunodeficiency virus infection. Rev Infect Dis 1991;13,139-145[ISI][Medline]
- Palella, FJ, Jr, Delaney, KM, Moorman, AC Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med 1998;338,853-860[Abstract/Free Full Text]
- Sepkowitz, KA Effect of HAART on natural history of AIDS-related opportunistic disorders. Lancet 1998;351,228-230[CrossRef][ISI][Medline]
- Carr, A, Marriott, D, Field, A, et al Treatment of HIV-1-associated microsporidiosis and cryptosporidiosis with combination antiretroviral therapy. Lancet 1998;351,256-261[CrossRef][ISI][Medline]
- López Bernaldo deq Uiros, JC, Miro, JM, Peña, JM, et al A randomized trial of the discontinuation of primary and secondary prophylaxis against Pneumocystis carinii pneumonia after highly active antiretroviral therapy in patients with HIV infection. N Engl J Med 2001;344,159-167[Abstract/Free Full Text]
- Ledergerber, B, Mocroft, A, Reiss, P, et al Discontinuation of secondary prophylaxis against Pneumocystis carinii pneumonia in patients with HIV infection who have a response to antiretroviral therapy. N Engl J Med 2001;344,168-174[Abstract/Free Full Text]
- Rouquet, RM, Clave, D, Massip, P, et al Imipenem/vancomycin for Rhodococcus equi pulmonary infection in HIV-positive patient [letter]. Lancet 1991;337,375[Medline]
- Chavanet, P, Bonnotte, B, Caillot, D, et al Imipenem/teicoplanin for Rhodococcus equi pulmonary infection in AIDS patients. Lancet 1991;337,794-795[Medline]
- Reese, RE, Betts, RF Antibiotic use: M. Erythromycin, azithromcyn, clarithromycin and dirithromycin. Reese, RE Betts, RF eds. A practical approach to infectious diseases. 4th ed. 1996,1288-1310 Little, Brown and Company. Boston, MA:
TOP
Abstract
Introduction
