- Email: [email protected]
Disseminated Mycobacterium genavense infection in Canadian AIDS patients
Tubercle and Lung Disease (1995) 76, 168-170 © 1995 Pearson Professional Ltd
.......
il lN
LungDisease
Case report
Disseminated Mycobacterium genavense infection in Canadian AIDS patients S. D. Shafran*, G. D. Taylor*, J. A. Talbot* *Department of Medicine, Division of Infectious Diseases, fDepartment of Medical Microbiology and Infectious Diseases, Division of Clinical Microbiology, University of Alberta, Edmonton, Canada S U M M A R Y. Mycobacterium genavense is a recently described mycobacterial species which thus far has been identified only in persons with advanced H I V disease. It appears to be a r a r e pathogen with an undefined reservoir. We describe the first two cases of M. genavense infection in Canadian AIDS patients. The clinical presentation of fever and wasting with extremely low CD4 lymphocyte counts was indistinguishable f r o m disseminated M. avium complex (MAC) infection. However, blood cultures in B A C T E C 13A m e d i u m required a m e a n of 58 days (range 41-87) to detect growth of M. genavense in contrast to a mean of 10 days for M A C in our laboratory. M. genavense infection is underdiagnosed due to the lack of universal use of B A C T E C liquid m e d i u m and the use of relatively short incubation times (only 6 weeks) by some laboratories. The value of antimycobacterial therapy for M. genavense is unknown, but anecdotal data suggest that treatment with a regimen a p p r o p r i a t e for M A C m a y be beneficial. R E S U M E. Mycobacterium genavense est une esp~ce mycobact~rienne r~cemment d6crite qui jusqu'ici n ' a ~t6 identifi6e que chez des personnes pr~sentant un ~tat 6volu~ d'infection p a r le VIH. Il semble ~tre un pathog~ne r a r e avec un r~servoir non d~fini. Sont d6crits ici les deux premiers cas d'infection ~ M. genavense chez des malades canadiens atteints du SIDA. Le tableau clinique de f~brilit~ et anorexie, ajout6 h des taux extr~mement bas de lymphocytes CD4, ~tait impossible h distinguer d ' u n e infection diss~min~e p a r le complexe M. avium (MAC). N6anmoins les h~mocultures dans un milieu B A C T E C 13A ont n~cessit6 en moyenne 58 jours (de 41 h 87 jours) p o u r d~celer la croissance de M. genavense, compar~e h une moyenne de 10 jours p o u r d6celer la croissance de M A C dans le laboratoire. L'infection M. genavense est sous-diagnostiqu~e h cause d ' u n e sous-utilisation du milieu liquide B A C T E C et de l'utilisation de p6riodes d'incubation relativement courtes (seulement 6 semaines) dans certains laboratoires. L'int~r~t d ' u n traitement antimycobact6rien p o u r M. genavense n ' e s t pas prouv6, mais des donn6es incidentes sugg~rent que l'utilisation d ' u n r6gime appropri6 p o u r le traitement de M A C p o u r r a i t ~tre b~n~fique. R E S U M E N. Mycobacterium genavense es una micobacteria recientemente descrita que, hasta aqul, s61o ha sido identificada en personas con enfermedad V I H avanzada. Parece ser un pat6geno r a r o con un reservorio no definido. Se describe los dos primeros casos de infecci6n con M. genavense en pacientes canadienses con SIDA. El cuadro clinico con fiebre y anorexia con un recuento extremadamente bajo de linfocitos CD4 era imposible de distinguir de una infecci6n diseminada con complejo M. avium (MAC). Sin embargo, los hemocultivos en un medio B A C T E C 13A necesitaron un promedio de 58 dias (de 41 a 87 dias) p a r a detectar el crecimiento de M. genavense, en comparaci6n con un promedio de 10 dlas p a r a M A C en nuestro laboratorio. La infecci6n con M. genavense es subdiagnosticada debido a la escasa utilizaci6n del medio liquido B A C T E C y a los periodos de utilizaci6n relativamente cortos (s61o 6 semanas) p o r algunos laboratorios. Se desconoce el valor del tratamiento antimicobacteriano p a r a M. genavense, pero datos anecd6ticos sugieren que un tratamiento adecuado p a r a M A C podria ser ben~fico.
Correspondence to: Dr Stephen D. Shafran, Division of Infectious Diseases, Departmentof Medicine, 2E4.11, Walter C. Mackenzie Health Sciences Centre, 8440- 112 Street, Universityof Alberta, Edmonton T6G 2B7 Canada Paper received31 March 1994. Final version accepted 27 June 1994. 168
Mycobacterium genavense in AIDS
In 1990, Hirschel et al first reported disseminated infection in a patient with advanced AIDS due to a novel mycobacterial species. ~They subsequently demonstrated that this organism was a new mycobacterial species related to Mycobacterium simiae which they named M. genavense after the city of Geneva, the place of residence of the index case. 2 As of the end of 1993, 31 cases of M. genavense infection in patients with advanced HIV disease have been described in the literature 14 and 44 such cases were presented by Hirschel's group at the Ninth International Conference on AIDS in Berlin in June 1993. 9 Cases of 34. genavense infection have been reported from Switzerland, Germany, Austria, Italy, the United States and Australia. 9 In this report, we describe the first two cases of disseminated M. genavense infection in persons with advanced HIV disease who were life-long residents of Canada. CASE 1 A 32-year-old white homosexual with known HIV infection was hospitalized 2 November 1992 with intermittent fever, night sweats and a 14 kg weight loss over the previous 5 months. He had an episode of Pneumocystis carinii pneumonia (PCP) in August 1991 and had not experienced other major HIV-associated complications. At the time of hospitalization, his CD4 lymphocyte count was 30 x 106/L and his hemoglobin was 10.9 g/ dL. Neither hepatomegaly nor splenomegaly was present. A blood culture collected in BACTEC 13A (Becton Dickinson, Towson, MD, USA) medium grew acid-fast bacilli (AFB) after 62 days incubation. A second blood specimen collected later the same day required 75 days to detect mycobacterial growth. The blood isolate was negative for hybridization with a commercial M. avium complex (MAC) DNA probe (Accuprobe, Gene Probe Inc., San Diego, CA, USA). The high pressure liquid chromatographic pattern of the mycolic acids was suggestive of M. simiae. The organism was forwarded to Dr Stefan Emler at the Laboratoire Central de Virologie at H6pital Cantonal, Universitaire de Gen6ve who identified it as M. genavense by sequencing a portion of the amplified 16S rRNA. After identification of positive mycobacterial blood cultures in January 1993, the patient's stool sample was examined and was smear-positive for AFB. On 17 February, the patient was started on treatment for disseminated mycobacterial infection with a combination of clarithromycin, ethambutol and ciprofloxacin. He was unable to tolerate either clarithromycin or ciprofloxacin due to the development of a diffuse erythematous rash which recurred on rechallenge with either of these agents. From March onward, he was treated with a combination of ethambutol and amikacin. Mycobacterial blood cultures collected 17 February, 5 March, and 26 March, 1993 continued to grow M. genavense after 41, 52 and 87 days incubation respectively. 7 mycobacterial blood cultures collected between 13 April and 15 June 1993 were negative after 12 weeks incubation.
169
On 22 April 1993, the patient underwent colonoscopy because of persistent diarrhea. Although the endoscopic appearance was normal, a biopsy of the transverse colon revealed many AFB. Clofazimine was added to ethambutol and amikacin, and the patient remained on all 3 antimycobacterial agents until 9 July 1993 when upper gastrointestinal endoscopy revealed the presence of multiple duodenal nodules which, on biopsy, contained large numbers of AFB. The patient deteriorated, experiencing a 13 kg weight loss since 20 April 1993 and requested palliative care only. The patient's antimycobacterial drugs were withdrawn and he died on 9 August 1993. A post mortem examination revealed disseminated mycobacterial infection with involvement of the small bowel, large bowel, intraabdominal lymph nodes, liver, spleen, both lungs, and the adrenal glands. All post mortem mycobacterial cultures were negative. CASE 2 A 41-year-old white homosexual presented in September 1992 with fatigue, anorexia, 14 kg weight loss, and chronic cough. He tested HIV-positive, was found to have PCP and his CD4 lymphocyte count was 20 x 106/ L. Despite appropriate therapy he developed respiratory failure and required mechanical ventilation. He ultimately recovered from PCP after a 1-month hospital stay. Over the following 5 months he continued to lose weight and stabilized at 30 kg below his usual weight. Symptomatic anemia requiring blood transfusion developed following introduction of zidovudine therapy. Antimycobacterial prophylaxis with rifabutin was offered, but declined by the patient. 5 months after his original presentation, abdominal pain, bloating, and constipation developed. An abdominal ultrasound demonstrated splenomegaly, lymphadenopathy and attenuation of the liver suggestive of fatty infiltration. The patient was re-admitted to hospital because of abdominal pain and cachexia. He gradually deteriorated and died in April 1993, 7 months after his original presentation. An autopsy was not performed. 2 days prior to death, a blood culture collected 44 days earlier into BACTEC 13A medium was reported as growing AFB. This infection was not treated because of the patient's advanced disease state. A second blood culture, collected into BACTEC 13A medium 2 weeks prior to death, also grew AFB after 43 days of incubation. 3 previous blood culture specimens collected 3-6 months prior to death as well as bronchoscopy specimens at the time of PCP diagnosis failed to grow mycobacteria. As in case 1, the isolate in case 2 failed to hybridize with a commercial DNA probe for MAC and was forwarded to Dr Emler who identified it as M. genavense.
DISCUSSION Since the original case report, ~disseminated M. genavense
170
Tubercle and Lung Disease
infection in advanced HIV disease has been increasingly recognized. 3-9 The clinical disease is remarkably similar to disseminated M. avium complex (DMAC) infection, in that both occur in very advanced HIV disease, both can occur in multiple organs, fever and weight loss are prominent symptoms, and bacteremia is the most common site of detection in life? ° In contrast with MAC, M. genavense infection has only been documented in persons with AIDS. Whereas MAC can easily be grown in a variety of mycobacterial media, growth of M. genavense is difficult. For the most part, M. genavense grows well only in BACTEC liquid media (either 13A or 12B) and may require more than 6 weeks incubation to detect growth. Indeed, Wald et al reported a mean time to detection of growth of 42 days in BACTEC 13A medium in 7 patients. 4 In our laboratory, the mean time to detect growth in these two patients was 58 days for all 7 positive blood cultures and 53 days for the 5 specimens collected prior to antimycobacterial treatment, in contrast to a mean of 10 days for MAC (unpublished data). Detection of growth of M. genavense has ranged from 13 days 4 to 87 days (this report). As some laboratories do not use BACTEC methodology and/or incubate cultures longer than 6 weeks, it is very likely that M. genavense infection is underdiagnosed. Our laboratory routinely incubates mycobacterial blood cultures for 12 weeks. The environmental reservoir, if any, for M. genavense has not been identified. However, M. genavense infection has been reported from three continents: Europe, Australia, and North America. The cases reported here are the first cases of M. genavense reported from Canada. Little is known about the in vitro susceptibility profile of this organism. Jackson et aP tested two M. genavense isolates and both were susceptible to amikacin and resistant to isoniazid, ethambutol, and ciprofloxacin with discordant susceptibility to rifampin and rifabutin. Susceptibility to clarithromycin and azithromycin was not tested. The appropriateness of antimycobacterial therapy and the selection of specific agents for M. genavense infection have not been defined. Wald et al4 reported a longer median survival after the detection of mycobacteremia in the 5 patients whom they treated with antimycobacterial drugs as compared with the 2 patients who were not so treated. Bessesen et a l 7 r e ported sterilization of blood cultures in 3 patients receiving clarithromycin and ethambutol together with at least one other agent. Nadal et al8 noted clinical and bacteriological improvement in 2 of 3 treated children. In our first case, blood cultures remained positive for more than 5 weeks into treatment with ethambutol and amikacin, but subsequently became negative despite the continued presence of mycobacterial enteritis. Definitive treatment recommendations cannot be made on the basis of the present data. Nevertheless, we tentatively suggest treatment with a regimen appropriate for the treatment of DMAC infection, including clarithromycin and ethambutol plus at least one other oral agent." Our report brings to 7 the number of countries from which cases of M. genavense infection have been
reported. All cases have occurred in patients with advanced AIDS and have been clinically indistinguishable from DMAC infection. Cases have occurred in both children and adults, as well as in both sexually and parenterally acquired HIV infection. Fever and weight loss have been the most common symptoms, sometimes associated with anemia, hepatomegaly, splenomegaly, lymphadenopathy and enteritis. The diagnosis has usually been established by blood culture in BACTEC liquid medium, but seldom before 3 weeks incubation, a time by which most cases of MAC infection are identified. 12 Definitive speciation is difficult and usually requires referral to a reference laboratory. The value of antimycobacterial therapy is unknown, but anecdotal data suggest that treatment with a regimen appropriate for DMAC may be beneficial.
Acknowledgements We thank Sylvia Chomyc, Donna Breuer and the mycobacteriology staff at the Provincial Laboratory for Public Health of Northern Alberta for their diligence in nurturing our two isolates of M. genavense. Note in proof: Since submission of this manuscript we have identified 3 additional patients with advanced AIDS who had M. genavense infection.
References 1. Hirschel B, Chang H R, Mach N. Fatal infection with a novel, unidentified mycobacterium in a man with AIDS. N Engl J Med 1990; 323: 109-113. 2. BSttger E C, Hirschel B, Coyle M B. Mycobacterium genavense sp. nov. Int J System Bacteriol 1993; 43: 841-843. 3. BSttger E C, Teske A, Kirschner P e t al. Disseminated 'Mycobacterium genavense' infection in patients with AIDS. Lancet 1992; 340: 76-80. 4. Wald A, Coyle M B, Carlson L C, Thompson R I, Hooton T M. Infection with a fastidious mycobacterium resembling Mycobacterium simiae in seven patients with AIDS. Ann Intern Med 1992; 117: 586-589. 5. Jackson K, Sievers A, Ross B C, Dwyer B. Isolation of a fastidious mycobacterium species from two AIDS patients. J Clin Microbiol 1992; 30: 2934-2937. 6. Coyle M B, Carlson L D C, Wallis C K et al. Laboratory aspects of Mycobacterium genavense, a proposed species isolated from AIDS patients. J Clin Microbiol 1992; 30: 3206-3212. 7. Bessesen M T, Shlay T, Stone-Venohr B, Cohn D L, Reves R R. Disseminated Mycobacterium genavense infection: clinical and microbiological features and response to therapy. AIDS 1993; 7: 1357-61. 8. Nadal D, Caduff R, Kraft R et al. Invasive infection with Mycobacterium genavense in three children with the acquired immunodeficiency syndrome. Eur J Clin Microbiol Infect Dis 1993; 12: 37-43. 9. Prch~re M, Emler S, Wald A e t al. Infection with Mycobacterium genavense: clinical features in 44 cases. Ninth International Conference on AIDS. Berlin, June 1993: Abstract WS-BI0-2. 10. Benson C A, Ellner T J. Mycobacterium avium complex infection and AIDS: advances in theory and practice. Clin Infect Dis 1993; 17: 7-20. 11. US Public Health Service Task Force on Prophylaxis and Therapy for Mycobacterium avium complex. Recommendations on prophylaxis and therapy for disseminated Mycobacterium avium complex for adults and adolescents. M M W R 1993; 42 (RR-9): 14-20. 12. Kiehn T E, Cammarata R. Comparative recoveries of Mycobacterium avium-M, intracellulare from isolator lysis-centrifugation and BACTEC 13A blood culture systems. J Clin Microbiol 1988; 26: 760-761.
.......
il lN
LungDisease
Case report
Disseminated Mycobacterium genavense infection in Canadian AIDS patients S. D. Shafran*, G. D. Taylor*, J. A. Talbot* *Department of Medicine, Division of Infectious Diseases, fDepartment of Medical Microbiology and Infectious Diseases, Division of Clinical Microbiology, University of Alberta, Edmonton, Canada S U M M A R Y. Mycobacterium genavense is a recently described mycobacterial species which thus far has been identified only in persons with advanced H I V disease. It appears to be a r a r e pathogen with an undefined reservoir. We describe the first two cases of M. genavense infection in Canadian AIDS patients. The clinical presentation of fever and wasting with extremely low CD4 lymphocyte counts was indistinguishable f r o m disseminated M. avium complex (MAC) infection. However, blood cultures in B A C T E C 13A m e d i u m required a m e a n of 58 days (range 41-87) to detect growth of M. genavense in contrast to a mean of 10 days for M A C in our laboratory. M. genavense infection is underdiagnosed due to the lack of universal use of B A C T E C liquid m e d i u m and the use of relatively short incubation times (only 6 weeks) by some laboratories. The value of antimycobacterial therapy for M. genavense is unknown, but anecdotal data suggest that treatment with a regimen a p p r o p r i a t e for M A C m a y be beneficial. R E S U M E. Mycobacterium genavense est une esp~ce mycobact~rienne r~cemment d6crite qui jusqu'ici n ' a ~t6 identifi6e que chez des personnes pr~sentant un ~tat 6volu~ d'infection p a r le VIH. Il semble ~tre un pathog~ne r a r e avec un r~servoir non d~fini. Sont d6crits ici les deux premiers cas d'infection ~ M. genavense chez des malades canadiens atteints du SIDA. Le tableau clinique de f~brilit~ et anorexie, ajout6 h des taux extr~mement bas de lymphocytes CD4, ~tait impossible h distinguer d ' u n e infection diss~min~e p a r le complexe M. avium (MAC). N6anmoins les h~mocultures dans un milieu B A C T E C 13A ont n~cessit6 en moyenne 58 jours (de 41 h 87 jours) p o u r d~celer la croissance de M. genavense, compar~e h une moyenne de 10 jours p o u r d6celer la croissance de M A C dans le laboratoire. L'infection M. genavense est sous-diagnostiqu~e h cause d ' u n e sous-utilisation du milieu liquide B A C T E C et de l'utilisation de p6riodes d'incubation relativement courtes (seulement 6 semaines) dans certains laboratoires. L'int~r~t d ' u n traitement antimycobact6rien p o u r M. genavense n ' e s t pas prouv6, mais des donn6es incidentes sugg~rent que l'utilisation d ' u n r6gime appropri6 p o u r le traitement de M A C p o u r r a i t ~tre b~n~fique. R E S U M E N. Mycobacterium genavense es una micobacteria recientemente descrita que, hasta aqul, s61o ha sido identificada en personas con enfermedad V I H avanzada. Parece ser un pat6geno r a r o con un reservorio no definido. Se describe los dos primeros casos de infecci6n con M. genavense en pacientes canadienses con SIDA. El cuadro clinico con fiebre y anorexia con un recuento extremadamente bajo de linfocitos CD4 era imposible de distinguir de una infecci6n diseminada con complejo M. avium (MAC). Sin embargo, los hemocultivos en un medio B A C T E C 13A necesitaron un promedio de 58 dias (de 41 a 87 dias) p a r a detectar el crecimiento de M. genavense, en comparaci6n con un promedio de 10 dlas p a r a M A C en nuestro laboratorio. La infecci6n con M. genavense es subdiagnosticada debido a la escasa utilizaci6n del medio liquido B A C T E C y a los periodos de utilizaci6n relativamente cortos (s61o 6 semanas) p o r algunos laboratorios. Se desconoce el valor del tratamiento antimicobacteriano p a r a M. genavense, pero datos anecd6ticos sugieren que un tratamiento adecuado p a r a M A C podria ser ben~fico.
Correspondence to: Dr Stephen D. Shafran, Division of Infectious Diseases, Departmentof Medicine, 2E4.11, Walter C. Mackenzie Health Sciences Centre, 8440- 112 Street, Universityof Alberta, Edmonton T6G 2B7 Canada Paper received31 March 1994. Final version accepted 27 June 1994. 168
Mycobacterium genavense in AIDS
In 1990, Hirschel et al first reported disseminated infection in a patient with advanced AIDS due to a novel mycobacterial species. ~They subsequently demonstrated that this organism was a new mycobacterial species related to Mycobacterium simiae which they named M. genavense after the city of Geneva, the place of residence of the index case. 2 As of the end of 1993, 31 cases of M. genavense infection in patients with advanced HIV disease have been described in the literature 14 and 44 such cases were presented by Hirschel's group at the Ninth International Conference on AIDS in Berlin in June 1993. 9 Cases of 34. genavense infection have been reported from Switzerland, Germany, Austria, Italy, the United States and Australia. 9 In this report, we describe the first two cases of disseminated M. genavense infection in persons with advanced HIV disease who were life-long residents of Canada. CASE 1 A 32-year-old white homosexual with known HIV infection was hospitalized 2 November 1992 with intermittent fever, night sweats and a 14 kg weight loss over the previous 5 months. He had an episode of Pneumocystis carinii pneumonia (PCP) in August 1991 and had not experienced other major HIV-associated complications. At the time of hospitalization, his CD4 lymphocyte count was 30 x 106/L and his hemoglobin was 10.9 g/ dL. Neither hepatomegaly nor splenomegaly was present. A blood culture collected in BACTEC 13A (Becton Dickinson, Towson, MD, USA) medium grew acid-fast bacilli (AFB) after 62 days incubation. A second blood specimen collected later the same day required 75 days to detect mycobacterial growth. The blood isolate was negative for hybridization with a commercial M. avium complex (MAC) DNA probe (Accuprobe, Gene Probe Inc., San Diego, CA, USA). The high pressure liquid chromatographic pattern of the mycolic acids was suggestive of M. simiae. The organism was forwarded to Dr Stefan Emler at the Laboratoire Central de Virologie at H6pital Cantonal, Universitaire de Gen6ve who identified it as M. genavense by sequencing a portion of the amplified 16S rRNA. After identification of positive mycobacterial blood cultures in January 1993, the patient's stool sample was examined and was smear-positive for AFB. On 17 February, the patient was started on treatment for disseminated mycobacterial infection with a combination of clarithromycin, ethambutol and ciprofloxacin. He was unable to tolerate either clarithromycin or ciprofloxacin due to the development of a diffuse erythematous rash which recurred on rechallenge with either of these agents. From March onward, he was treated with a combination of ethambutol and amikacin. Mycobacterial blood cultures collected 17 February, 5 March, and 26 March, 1993 continued to grow M. genavense after 41, 52 and 87 days incubation respectively. 7 mycobacterial blood cultures collected between 13 April and 15 June 1993 were negative after 12 weeks incubation.
169
On 22 April 1993, the patient underwent colonoscopy because of persistent diarrhea. Although the endoscopic appearance was normal, a biopsy of the transverse colon revealed many AFB. Clofazimine was added to ethambutol and amikacin, and the patient remained on all 3 antimycobacterial agents until 9 July 1993 when upper gastrointestinal endoscopy revealed the presence of multiple duodenal nodules which, on biopsy, contained large numbers of AFB. The patient deteriorated, experiencing a 13 kg weight loss since 20 April 1993 and requested palliative care only. The patient's antimycobacterial drugs were withdrawn and he died on 9 August 1993. A post mortem examination revealed disseminated mycobacterial infection with involvement of the small bowel, large bowel, intraabdominal lymph nodes, liver, spleen, both lungs, and the adrenal glands. All post mortem mycobacterial cultures were negative. CASE 2 A 41-year-old white homosexual presented in September 1992 with fatigue, anorexia, 14 kg weight loss, and chronic cough. He tested HIV-positive, was found to have PCP and his CD4 lymphocyte count was 20 x 106/ L. Despite appropriate therapy he developed respiratory failure and required mechanical ventilation. He ultimately recovered from PCP after a 1-month hospital stay. Over the following 5 months he continued to lose weight and stabilized at 30 kg below his usual weight. Symptomatic anemia requiring blood transfusion developed following introduction of zidovudine therapy. Antimycobacterial prophylaxis with rifabutin was offered, but declined by the patient. 5 months after his original presentation, abdominal pain, bloating, and constipation developed. An abdominal ultrasound demonstrated splenomegaly, lymphadenopathy and attenuation of the liver suggestive of fatty infiltration. The patient was re-admitted to hospital because of abdominal pain and cachexia. He gradually deteriorated and died in April 1993, 7 months after his original presentation. An autopsy was not performed. 2 days prior to death, a blood culture collected 44 days earlier into BACTEC 13A medium was reported as growing AFB. This infection was not treated because of the patient's advanced disease state. A second blood culture, collected into BACTEC 13A medium 2 weeks prior to death, also grew AFB after 43 days of incubation. 3 previous blood culture specimens collected 3-6 months prior to death as well as bronchoscopy specimens at the time of PCP diagnosis failed to grow mycobacteria. As in case 1, the isolate in case 2 failed to hybridize with a commercial DNA probe for MAC and was forwarded to Dr Emler who identified it as M. genavense.
DISCUSSION Since the original case report, ~disseminated M. genavense
170
Tubercle and Lung Disease
infection in advanced HIV disease has been increasingly recognized. 3-9 The clinical disease is remarkably similar to disseminated M. avium complex (DMAC) infection, in that both occur in very advanced HIV disease, both can occur in multiple organs, fever and weight loss are prominent symptoms, and bacteremia is the most common site of detection in life? ° In contrast with MAC, M. genavense infection has only been documented in persons with AIDS. Whereas MAC can easily be grown in a variety of mycobacterial media, growth of M. genavense is difficult. For the most part, M. genavense grows well only in BACTEC liquid media (either 13A or 12B) and may require more than 6 weeks incubation to detect growth. Indeed, Wald et al reported a mean time to detection of growth of 42 days in BACTEC 13A medium in 7 patients. 4 In our laboratory, the mean time to detect growth in these two patients was 58 days for all 7 positive blood cultures and 53 days for the 5 specimens collected prior to antimycobacterial treatment, in contrast to a mean of 10 days for MAC (unpublished data). Detection of growth of M. genavense has ranged from 13 days 4 to 87 days (this report). As some laboratories do not use BACTEC methodology and/or incubate cultures longer than 6 weeks, it is very likely that M. genavense infection is underdiagnosed. Our laboratory routinely incubates mycobacterial blood cultures for 12 weeks. The environmental reservoir, if any, for M. genavense has not been identified. However, M. genavense infection has been reported from three continents: Europe, Australia, and North America. The cases reported here are the first cases of M. genavense reported from Canada. Little is known about the in vitro susceptibility profile of this organism. Jackson et aP tested two M. genavense isolates and both were susceptible to amikacin and resistant to isoniazid, ethambutol, and ciprofloxacin with discordant susceptibility to rifampin and rifabutin. Susceptibility to clarithromycin and azithromycin was not tested. The appropriateness of antimycobacterial therapy and the selection of specific agents for M. genavense infection have not been defined. Wald et al4 reported a longer median survival after the detection of mycobacteremia in the 5 patients whom they treated with antimycobacterial drugs as compared with the 2 patients who were not so treated. Bessesen et a l 7 r e ported sterilization of blood cultures in 3 patients receiving clarithromycin and ethambutol together with at least one other agent. Nadal et al8 noted clinical and bacteriological improvement in 2 of 3 treated children. In our first case, blood cultures remained positive for more than 5 weeks into treatment with ethambutol and amikacin, but subsequently became negative despite the continued presence of mycobacterial enteritis. Definitive treatment recommendations cannot be made on the basis of the present data. Nevertheless, we tentatively suggest treatment with a regimen appropriate for the treatment of DMAC infection, including clarithromycin and ethambutol plus at least one other oral agent." Our report brings to 7 the number of countries from which cases of M. genavense infection have been
reported. All cases have occurred in patients with advanced AIDS and have been clinically indistinguishable from DMAC infection. Cases have occurred in both children and adults, as well as in both sexually and parenterally acquired HIV infection. Fever and weight loss have been the most common symptoms, sometimes associated with anemia, hepatomegaly, splenomegaly, lymphadenopathy and enteritis. The diagnosis has usually been established by blood culture in BACTEC liquid medium, but seldom before 3 weeks incubation, a time by which most cases of MAC infection are identified. 12 Definitive speciation is difficult and usually requires referral to a reference laboratory. The value of antimycobacterial therapy is unknown, but anecdotal data suggest that treatment with a regimen appropriate for DMAC may be beneficial.
Acknowledgements We thank Sylvia Chomyc, Donna Breuer and the mycobacteriology staff at the Provincial Laboratory for Public Health of Northern Alberta for their diligence in nurturing our two isolates of M. genavense. Note in proof: Since submission of this manuscript we have identified 3 additional patients with advanced AIDS who had M. genavense infection.
References 1. Hirschel B, Chang H R, Mach N. Fatal infection with a novel, unidentified mycobacterium in a man with AIDS. N Engl J Med 1990; 323: 109-113. 2. BSttger E C, Hirschel B, Coyle M B. Mycobacterium genavense sp. nov. Int J System Bacteriol 1993; 43: 841-843. 3. BSttger E C, Teske A, Kirschner P e t al. Disseminated 'Mycobacterium genavense' infection in patients with AIDS. Lancet 1992; 340: 76-80. 4. Wald A, Coyle M B, Carlson L C, Thompson R I, Hooton T M. Infection with a fastidious mycobacterium resembling Mycobacterium simiae in seven patients with AIDS. Ann Intern Med 1992; 117: 586-589. 5. Jackson K, Sievers A, Ross B C, Dwyer B. Isolation of a fastidious mycobacterium species from two AIDS patients. J Clin Microbiol 1992; 30: 2934-2937. 6. Coyle M B, Carlson L D C, Wallis C K et al. Laboratory aspects of Mycobacterium genavense, a proposed species isolated from AIDS patients. J Clin Microbiol 1992; 30: 3206-3212. 7. Bessesen M T, Shlay T, Stone-Venohr B, Cohn D L, Reves R R. Disseminated Mycobacterium genavense infection: clinical and microbiological features and response to therapy. AIDS 1993; 7: 1357-61. 8. Nadal D, Caduff R, Kraft R et al. Invasive infection with Mycobacterium genavense in three children with the acquired immunodeficiency syndrome. Eur J Clin Microbiol Infect Dis 1993; 12: 37-43. 9. Prch~re M, Emler S, Wald A e t al. Infection with Mycobacterium genavense: clinical features in 44 cases. Ninth International Conference on AIDS. Berlin, June 1993: Abstract WS-BI0-2. 10. Benson C A, Ellner T J. Mycobacterium avium complex infection and AIDS: advances in theory and practice. Clin Infect Dis 1993; 17: 7-20. 11. US Public Health Service Task Force on Prophylaxis and Therapy for Mycobacterium avium complex. Recommendations on prophylaxis and therapy for disseminated Mycobacterium avium complex for adults and adolescents. M M W R 1993; 42 (RR-9): 14-20. 12. Kiehn T E, Cammarata R. Comparative recoveries of Mycobacterium avium-M, intracellulare from isolator lysis-centrifugation and BACTEC 13A blood culture systems. J Clin Microbiol 1988; 26: 760-761.