113
results have been obtained in blind assays of samples with either the elaborate extraction (as used in the paper) or simple methods such as boiling or even 4 freeze/thaw cycles to liberate HPV before PCR. The consistency of these results with different extraction methods indicates that inhibitors are not a difficulty, and that routine testing for high levels of HPV by PCR could soon become practicable. Imperial Cancer Research Fund, P0 Box 123,
London WC2A 3PX, UK
JACK CUZICK
Department of Chemical Pathology, University College London
GEORGE TERRY
Department of Medical Microbiology, University College London
LINDA HO
Department of Obstetrics,
City Hospital, Nottingham
TONY HOLLINGWORTH
Department of Pathology, University Hospital, Nottingham
MALCOLM ANDERSON
Population movements and cholera spread in Cordillera Province, Santa Cruz Department, Bolivia StR,—The Latin-American cholera epidemic started on the coast of Peru in January, 1991, and spread rapidly throughout the continent. The causative agent is Vibrio cholerae 01, biotype El Tor, serotype Inaba (Wklg Epidemiol Rec 1991; 66: 47, 1992; 67:33-39). By July, 1991, cholera had not yet reached Bolivia. We began a surveillance system for episodes of diarrhoeal diseases to detect promptly any cholera case in Cordillera Province, Santa Cruz, Bolivia. Pacific
Cordillera Province occupies 86 245 km2in south-eastern Bolivia. It borders with the Department of Chuquisaca, Tarija, and Paraguay, and is about 150 km from Argentina. 62 231 inhabitants (0 7/km) were registered in 1987. There are a district hospital in Camiri town, nine area hospitals, and several "health places". Camiri is connected with Santa Cruz and the Argentine border by Carretera Panamericana, one of the most important South American routes. A report form for episodes of diarrhoea was delivered to each of the ten hospitals at the end of July, 1991. The form, including the numbers of new cases by age, province, and outcome, was returned fortnightly to Camiri district hospital. From August, 1991, to January, 1992, the number of diarrhoea cases in Camiri district hospital stayed fairly constant, the increase in November to January probably being related to the beginning of the rainy season (figure). The number of cases among children under six years of age was higher than that among adults. A similar trend of diarrhoea cases was seen in the nine area hospitals. No cholera cases were reported in this period. In February, 1992, the total number of diarrhoea cases increased four-fold compared with January. The number in adults was 1 ’6 times that in children. The first cholera case in this area was confirmed on Feb 12, 1992 by the bacteriology unit of
CENETROP, where all fecal samples from suspected cases had been promptly dispatched. 9 further cases of cholera in the Camiri area and 6 cases from four other areas of the district were confirmed during February. The first cholera cases were detected in patients coming back from a two day meeting organised to commemorate the centenary of the last Guarani native indians’ genocide. The first cholera case was reported in Bolivia in August, 1991, in Rio Abajo Region, and the epidemic was limited to the north of this country (Wkly Epidemiol Rec 1991; 66: 263). But it was impossible to postpone the gathering. About 6000 indians and others from Guarani areas of Argentina, Paraguay, and from northern Bolivian cholera-infected areas participated in the meeting in Kuruyuki, 65 km south of Camiri on Jan 28, 1992. Santa Cruz Department was included in the list of newly infected areas on March 20, 1992, 6 months after the first notified cholera case in Bolivia (Wkly Epidemiol Rec 1992; 67: 87-88). The introduction of the cholera epidemic into Cordillera Province seems to have been favoured by the gathering of Guarani indians in Kuruyuki. Our data underline the role of large population movements and the gathering of people in overcrowded places as risk factors in the spread of cholera, and emphasise the importance of surveillance during the early phase of the epidemic to obtain descriptive information and defme intervention priorities for high-risk groups. Istituto di Malattie Infettive, University of Siena, I-53900 Siena, Italy Academic
Department of University of Florence
P. GUGLIELMETTI
Infectious Diseases,
Unidad Sanitaria Santa Cruz, District of Cordillera, Bolivia
H. GAMBOA
Centro Nacional de Enfermedades Tropicales "CENETROP", Santa Cruz de la Sierra
D.
,5-years-old (hatched), and choleraconfirmed cases (black); *first cholera case in Bolivia, Aug 27, 1991.
J. ANTUNEZ
Laboratory of Medical Bacteriology and Mycology, Superiore diu Sanità,
Istituto Rome
I. LUZZI
Department of Epidemiology and Biostatistics, Istituto Rome
Superiore di Sanità,
F. ROSMINI
Academic
Department of Infectious Diseases,
University of
F. PARADISI
Florence
Predisposition and ascaris infection SIR,-In your May 23 editorial (p 1264), you emphasise the need to control intestinal helminth infections in children, and support the
strategy of mass chemotherapy on the basis of the epidemiology and modes of transmission of such infections. There is, however, an important aspect of the editorial that deserves comment. The term predisposition applied to helminth infections implies that individuals with heavy worm burdens are more likely to reacquire heavy infections after successful treatment, whereas lightly infected individuals tend to become lightly reinfected. This relation has been shown in schistosomes, Opisthorchis, and all the major geohehninth infections; it has been detected at all ages although schistosomes in adults seem to be an exception; and it persists for several cycles of treatment.2,3 Predisposition has been noted in family units and in specific age-classes of a community,4 and does not imply that each individual maintains a consistent position in a ranking of susceptibility to helminth infectionsmerely that correlation analyses indicate that a significant proportion of individuals retain the same rank of intensity of infection before and after treatment and reinfection.5 The fact that some people do not reach the same ranked position is believed to be a result of errors of measurement, factors of chance that affect exposure to infective stages, and heterogeneity in factors that determine the establishment of wonns.6,7These consistent observations, supported by much evidence, were confirmed by our study reported in the same issue (p 1253). We therefore disagree with your editorial statement that the reported study "raises questions about the validity of predisposition". Recorded patterns of the intensity of reinfection raise the interesting issue of what factor or combination of factors generates predisposition; this is the subject of much research at present. ...
Newdiarrhoea cases in Camiri area, Cordillera Province, Bolivia.
A. BARTOLONI M. ROSELLI
114
We do agree, however, with the conclusion that biological variability may in practice work against a selective approach to chemotherapy. This is not a novel conclusion, but it has not previously been supported by such extensive epidemiological data. A selective approach may also be unsatisfactory because of the costs of diagnosis in relation to treatment, because of resistance in the community to selective treatment, and because of the ethical issues of diagnosis without treatment. For these reasons the suggestion that an effective means of controlling helminth infections is mass treatment targeted at age groups with high average worm loads8,9 is gaining increasing support. Wellcome Trust Research Centre for Parasitic Infections, West Beit, Imperial College, London SW7 2BB, UK
ANDREW HALL D. A. P. BUNDY R. M. ANDERSON
Centre for International Child Health, Institute for Child Health, London WC1
A. M. TOMKINS
1. Bensted-Smith R, Anderson RM, Butterworth AE, et al. Evidence for predisposition of individual patients to reinfection with Schistosoma haematobium after treatment, Trans R Soc Trop Med Hyg 1987; 81: 651-54. 2. Holland CV, Asaolu SO, Crompton DWT, et al. The epidemiology of Ascaris lumbricoides and other soil-transmitted helminths in primary school from Ife-Ife, Nigeria. Parasitology 1989; 99: 275-85. 3. Chan LS, Kan SP, Bundy DAP. The effect of repeated chemotherapy on age-related predisposition to Ascaris lumbricoides and Trichuris trichiura. Parasitology 1992; 104: 371-77. 4. Forrester JE, Scott ME, Bundy DAP, Golden MHN. Predisposition of individuals and families in Mexico to heavy infection with Ascaris lumbricoides and Trichuris trichiura Trans R Soc Trop Med Hyg 1990; 84: 272-76. 5. Keymer A, Pagel M. Predisposition to helminth infection. In: Schad GA, Warren KS, eds. Hookworm disease: current status and new directions. London: Taylor and Francis, 1990: 177-209. 6. McCallum HI. Covariance in parasite burdens: the effect of predisposition to infection. Parasitology 1990; 100: 153-59 7. Bundy DAP, Medley GF. Immune-epidemiology of human geohelminthiasis: ecological and immunological determinants of worm burden. Parasitology 1992; 104: 105-19. 8. Butterworth AE, Sturrock RF, Ouma JH, et al. Comparison of different chemotherapy strategies against Schistosoma mansoni in Machakos District, Kenya: effects on human infection and morbidity. Parasitology 1991; 103: 339-55. 9. Bundy DAP, Wong MS, Lewis LL, Horton J. Control of geohelminths by delivery of targeted chemotherapy through schools. Trans R Soc Trop Med Hyg 1990; 84: 115-20.
Artificial
breeding grounds for mosquito control
SIR,-Mosquitoes transmit some of mankind’s most troubling diseases but vector control measures are difficult and expensive. Mosquito densities fluctuate over time, and analysis of the causes of such fluctuations may provide new ideas for control. One such observation in a Colombo suburb suggests a promising line. A large municipal drain carries waste water from houses to the sea. The flow is sluggish, and stagnant puddles and decaying organic matter provide good breeding grounds for certain species. When there is no rain the houses in the area teem with mosquitoes. However, the drain also serves as a conduit for rainwater, and whenever it rains the number of mosquitoes falls. If occasional showers persist over several weeks the mosquito density builds up again, this time unaffected by the continuing rain. This indicates that alternative grounds are available for laying eggs, although they were ignored during spells when the municipal drain is a more attractive site. By drawing mosquitoes to lay their eggs in the drain so that all the larvae are washed out to sea with the first rains, this drain offers a lesson worth emulating. Unfortunately, the drain cannot produce a lasting benefit because the rains clear not just the larvae but also the ingredients that attract mosquitoes to lay eggs there. If the drain continued to attract gravid females, while periodically flushing out the larvae, the impact would be much
greater. Artificial breeding grounds more attractive to mosquitoes than sites available naturally could overcome this difficulty. The provision of controlled breeding grounds from which the larvae can regularly be eliminated should not be difficult and could even be undertaken as a community health venture by householders. Faculty of Medicine, Colombo 8, Sri Lanka
DIYANATH SAMARASINGHE
Corynebacterium pseudodiphtheriticum pulmonary infection in AIDS patients SiR,—Infectious pneumonia remains a leading cause of morbidity and mortality among HIV patients. Although most cases are due to opportunistic agents, common bacterial pathogens and less pathogenic bacteria, especially the coryneform bacterium Rhodoccus equi, have been increasingly implicated.i We present two cases of pneumonia due to Corynebacterium pseudodiphtheriticum, a rare human pathogen, in two HIV seropositive men. Case 1-43, AIDS with tuberculosis adenitis, 6 months earlier admitted to our intensive care unit with 15 day history of high temperature, rigors, non-productive cough and progressive dyspnoea; on zidovudine and co-trimoxazole prophylaxis. Temperature was 39°C, regular heart rate 120/min, arterial blood pressure 100/50 mm Hg, respiratory rate 50/min, and he had pulmonary crackles over both lung fields. Pa02 was 51 mm Hg while breathing room air, CD4 count was 172/ [11, and HIV serology was positive. Radiography showed confluent alveolar opacities in the middle and lower left lung. Bronchoalveolar lavage fluid, blood, and cerebrospinal fluid cultures remained sterile, whereas culture of a protected brush specimen (PBS) obtained under fibreoptic bronchoscopy yielded 4 x 103 coryneform bacteria colony-forming units (CFU) per ml, resistant to macrolides, co-trimoxazole, and teicoplanin. Vancomycin was started to ensure activity against a possible R equi. These coryneform bacteria had rapid urease activity (5 min) that strongly suggested C pseudodiphtheriticum, which was confirmed by the Institut Pasteur, Paris. Clinical and radiographic findings rapidly became normal. After 4 weeks, there was no evidence of recurrent pneumonia. Case 2-34, AIDS with 2 week history of progressive dyspnoea, high temperature, productive cough, mucopurulent sputum, vomiting, and rigors; 5 months earlier, admitted for Pneumocystis carinii pneumonia (PCP) treated by co-trimoxazole, changed to dapsone because of adverse reactions; after discharge he received zidovudine and inhaled pentamidine once a month. Temperature was 38°C, heart rate 100/min, arterial blood pressure 120/70 mm Hg, and he had pulmonary crackles over both lung fields. Pa02 was 70 mm Hg while breathing room air, CD4 count was 130/ul, and HIV serology was positive. Radiography showed alveolar opacities in the right lung. Examination of bronchoalveolar lavage fluid revealed P carinii. Bronchial secretions (fibreoptic bronchoscopy) showed numerous polynuclear cells and gram-positive bacilli with morphological features suggestive of coryneform bacteria. Culture grew IOS CFU per ml of an organism that was identified as C pseudodiphtheriticum. Resistance was confined to fluoroquinolones. The patient was treated with dapsone and inhaled pentamidine (300 mg once a day) for PCP, and vancomycin. Clinical and radiographic findings quickly became normal. 4 weeks later, physical findings were unremarkable and the chest radiograph was
normal.
Cpseudodiphtheriticum can cause endocarditis and pneumonia,3-7 especially in apparently immunocompetent patients with previous pulmonary disease. Another case occurred in a young man on steroids for systemic lupus erythematosus.11 There has been only one report in a patient with a disease clinically compatible with HIV infection, although no serological confirmation for HIV was obtained. This patient had a lung abscess from which only C pseudodiphtherriticum was isolated and subsequently required lobectomy.9 Our two cases had documented HIV infection. They are the only cases from about 200 HIV-positive patients who underwent diagnostic fibreoptic bronchoscopy during the past year in our two institutions. Case 2 had relapsing PCP and concomitant infection by C pseudodiphtheriticum. It is difficult to delineate the role of each organism, but C pseudodiphtheritfczan as pure culture and abundant mucopurulent sputum suggest that this organism was at least in part responsible. Infection with C pseudodiphtheriticum is combined with other pathogens in many instances,IO although it was the sole causative agent in some reports.6 By contrast with R equi, which is susceptible to macrolides,’" one of our isolates
was
resistant
to
these
compounds. This
may have
results have been obtained in blind assays of samples with either the elaborate extraction (as used in the paper) or simple methods such as boiling or even 4 freeze/thaw cycles to liberate HPV before PCR. The consistency of these results with different extraction methods indicates that inhibitors are not a difficulty, and that routine testing for high levels of HPV by PCR could soon become practicable. Imperial Cancer Research Fund, P0 Box 123,
London WC2A 3PX, UK
JACK CUZICK
Department of Chemical Pathology, University College London
GEORGE TERRY
Department of Medical Microbiology, University College London
LINDA HO
Department of Obstetrics,
City Hospital, Nottingham
TONY HOLLINGWORTH
Department of Pathology, University Hospital, Nottingham
MALCOLM ANDERSON
Population movements and cholera spread in Cordillera Province, Santa Cruz Department, Bolivia StR,—The Latin-American cholera epidemic started on the coast of Peru in January, 1991, and spread rapidly throughout the continent. The causative agent is Vibrio cholerae 01, biotype El Tor, serotype Inaba (Wklg Epidemiol Rec 1991; 66: 47, 1992; 67:33-39). By July, 1991, cholera had not yet reached Bolivia. We began a surveillance system for episodes of diarrhoeal diseases to detect promptly any cholera case in Cordillera Province, Santa Cruz, Bolivia. Pacific
Cordillera Province occupies 86 245 km2in south-eastern Bolivia. It borders with the Department of Chuquisaca, Tarija, and Paraguay, and is about 150 km from Argentina. 62 231 inhabitants (0 7/km) were registered in 1987. There are a district hospital in Camiri town, nine area hospitals, and several "health places". Camiri is connected with Santa Cruz and the Argentine border by Carretera Panamericana, one of the most important South American routes. A report form for episodes of diarrhoea was delivered to each of the ten hospitals at the end of July, 1991. The form, including the numbers of new cases by age, province, and outcome, was returned fortnightly to Camiri district hospital. From August, 1991, to January, 1992, the number of diarrhoea cases in Camiri district hospital stayed fairly constant, the increase in November to January probably being related to the beginning of the rainy season (figure). The number of cases among children under six years of age was higher than that among adults. A similar trend of diarrhoea cases was seen in the nine area hospitals. No cholera cases were reported in this period. In February, 1992, the total number of diarrhoea cases increased four-fold compared with January. The number in adults was 1 ’6 times that in children. The first cholera case in this area was confirmed on Feb 12, 1992 by the bacteriology unit of
CENETROP, where all fecal samples from suspected cases had been promptly dispatched. 9 further cases of cholera in the Camiri area and 6 cases from four other areas of the district were confirmed during February. The first cholera cases were detected in patients coming back from a two day meeting organised to commemorate the centenary of the last Guarani native indians’ genocide. The first cholera case was reported in Bolivia in August, 1991, in Rio Abajo Region, and the epidemic was limited to the north of this country (Wkly Epidemiol Rec 1991; 66: 263). But it was impossible to postpone the gathering. About 6000 indians and others from Guarani areas of Argentina, Paraguay, and from northern Bolivian cholera-infected areas participated in the meeting in Kuruyuki, 65 km south of Camiri on Jan 28, 1992. Santa Cruz Department was included in the list of newly infected areas on March 20, 1992, 6 months after the first notified cholera case in Bolivia (Wkly Epidemiol Rec 1992; 67: 87-88). The introduction of the cholera epidemic into Cordillera Province seems to have been favoured by the gathering of Guarani indians in Kuruyuki. Our data underline the role of large population movements and the gathering of people in overcrowded places as risk factors in the spread of cholera, and emphasise the importance of surveillance during the early phase of the epidemic to obtain descriptive information and defme intervention priorities for high-risk groups. Istituto di Malattie Infettive, University of Siena, I-53900 Siena, Italy Academic
Department of University of Florence
P. GUGLIELMETTI
Infectious Diseases,
Unidad Sanitaria Santa Cruz, District of Cordillera, Bolivia
H. GAMBOA
Centro Nacional de Enfermedades Tropicales "CENETROP", Santa Cruz de la Sierra
D.
,5-years-old (hatched), and choleraconfirmed cases (black); *first cholera case in Bolivia, Aug 27, 1991.
J. ANTUNEZ
Laboratory of Medical Bacteriology and Mycology, Superiore diu Sanità,
Istituto Rome
I. LUZZI
Department of Epidemiology and Biostatistics, Istituto Rome
Superiore di Sanità,
F. ROSMINI
Academic
Department of Infectious Diseases,
University of
F. PARADISI
Florence
Predisposition and ascaris infection SIR,-In your May 23 editorial (p 1264), you emphasise the need to control intestinal helminth infections in children, and support the
strategy of mass chemotherapy on the basis of the epidemiology and modes of transmission of such infections. There is, however, an important aspect of the editorial that deserves comment. The term predisposition applied to helminth infections implies that individuals with heavy worm burdens are more likely to reacquire heavy infections after successful treatment, whereas lightly infected individuals tend to become lightly reinfected. This relation has been shown in schistosomes, Opisthorchis, and all the major geohehninth infections; it has been detected at all ages although schistosomes in adults seem to be an exception; and it persists for several cycles of treatment.2,3 Predisposition has been noted in family units and in specific age-classes of a community,4 and does not imply that each individual maintains a consistent position in a ranking of susceptibility to helminth infectionsmerely that correlation analyses indicate that a significant proportion of individuals retain the same rank of intensity of infection before and after treatment and reinfection.5 The fact that some people do not reach the same ranked position is believed to be a result of errors of measurement, factors of chance that affect exposure to infective stages, and heterogeneity in factors that determine the establishment of wonns.6,7These consistent observations, supported by much evidence, were confirmed by our study reported in the same issue (p 1253). We therefore disagree with your editorial statement that the reported study "raises questions about the validity of predisposition". Recorded patterns of the intensity of reinfection raise the interesting issue of what factor or combination of factors generates predisposition; this is the subject of much research at present. ...
Newdiarrhoea cases in Camiri area, Cordillera Province, Bolivia.
A. BARTOLONI M. ROSELLI
114
We do agree, however, with the conclusion that biological variability may in practice work against a selective approach to chemotherapy. This is not a novel conclusion, but it has not previously been supported by such extensive epidemiological data. A selective approach may also be unsatisfactory because of the costs of diagnosis in relation to treatment, because of resistance in the community to selective treatment, and because of the ethical issues of diagnosis without treatment. For these reasons the suggestion that an effective means of controlling helminth infections is mass treatment targeted at age groups with high average worm loads8,9 is gaining increasing support. Wellcome Trust Research Centre for Parasitic Infections, West Beit, Imperial College, London SW7 2BB, UK
ANDREW HALL D. A. P. BUNDY R. M. ANDERSON
Centre for International Child Health, Institute for Child Health, London WC1
A. M. TOMKINS
1. Bensted-Smith R, Anderson RM, Butterworth AE, et al. Evidence for predisposition of individual patients to reinfection with Schistosoma haematobium after treatment, Trans R Soc Trop Med Hyg 1987; 81: 651-54. 2. Holland CV, Asaolu SO, Crompton DWT, et al. The epidemiology of Ascaris lumbricoides and other soil-transmitted helminths in primary school from Ife-Ife, Nigeria. Parasitology 1989; 99: 275-85. 3. Chan LS, Kan SP, Bundy DAP. The effect of repeated chemotherapy on age-related predisposition to Ascaris lumbricoides and Trichuris trichiura. Parasitology 1992; 104: 371-77. 4. Forrester JE, Scott ME, Bundy DAP, Golden MHN. Predisposition of individuals and families in Mexico to heavy infection with Ascaris lumbricoides and Trichuris trichiura Trans R Soc Trop Med Hyg 1990; 84: 272-76. 5. Keymer A, Pagel M. Predisposition to helminth infection. In: Schad GA, Warren KS, eds. Hookworm disease: current status and new directions. London: Taylor and Francis, 1990: 177-209. 6. McCallum HI. Covariance in parasite burdens: the effect of predisposition to infection. Parasitology 1990; 100: 153-59 7. Bundy DAP, Medley GF. Immune-epidemiology of human geohelminthiasis: ecological and immunological determinants of worm burden. Parasitology 1992; 104: 105-19. 8. Butterworth AE, Sturrock RF, Ouma JH, et al. Comparison of different chemotherapy strategies against Schistosoma mansoni in Machakos District, Kenya: effects on human infection and morbidity. Parasitology 1991; 103: 339-55. 9. Bundy DAP, Wong MS, Lewis LL, Horton J. Control of geohelminths by delivery of targeted chemotherapy through schools. Trans R Soc Trop Med Hyg 1990; 84: 115-20.
Artificial
breeding grounds for mosquito control
SIR,-Mosquitoes transmit some of mankind’s most troubling diseases but vector control measures are difficult and expensive. Mosquito densities fluctuate over time, and analysis of the causes of such fluctuations may provide new ideas for control. One such observation in a Colombo suburb suggests a promising line. A large municipal drain carries waste water from houses to the sea. The flow is sluggish, and stagnant puddles and decaying organic matter provide good breeding grounds for certain species. When there is no rain the houses in the area teem with mosquitoes. However, the drain also serves as a conduit for rainwater, and whenever it rains the number of mosquitoes falls. If occasional showers persist over several weeks the mosquito density builds up again, this time unaffected by the continuing rain. This indicates that alternative grounds are available for laying eggs, although they were ignored during spells when the municipal drain is a more attractive site. By drawing mosquitoes to lay their eggs in the drain so that all the larvae are washed out to sea with the first rains, this drain offers a lesson worth emulating. Unfortunately, the drain cannot produce a lasting benefit because the rains clear not just the larvae but also the ingredients that attract mosquitoes to lay eggs there. If the drain continued to attract gravid females, while periodically flushing out the larvae, the impact would be much
greater. Artificial breeding grounds more attractive to mosquitoes than sites available naturally could overcome this difficulty. The provision of controlled breeding grounds from which the larvae can regularly be eliminated should not be difficult and could even be undertaken as a community health venture by householders. Faculty of Medicine, Colombo 8, Sri Lanka
DIYANATH SAMARASINGHE
Corynebacterium pseudodiphtheriticum pulmonary infection in AIDS patients SiR,—Infectious pneumonia remains a leading cause of morbidity and mortality among HIV patients. Although most cases are due to opportunistic agents, common bacterial pathogens and less pathogenic bacteria, especially the coryneform bacterium Rhodoccus equi, have been increasingly implicated.i We present two cases of pneumonia due to Corynebacterium pseudodiphtheriticum, a rare human pathogen, in two HIV seropositive men. Case 1-43, AIDS with tuberculosis adenitis, 6 months earlier admitted to our intensive care unit with 15 day history of high temperature, rigors, non-productive cough and progressive dyspnoea; on zidovudine and co-trimoxazole prophylaxis. Temperature was 39°C, regular heart rate 120/min, arterial blood pressure 100/50 mm Hg, respiratory rate 50/min, and he had pulmonary crackles over both lung fields. Pa02 was 51 mm Hg while breathing room air, CD4 count was 172/ [11, and HIV serology was positive. Radiography showed confluent alveolar opacities in the middle and lower left lung. Bronchoalveolar lavage fluid, blood, and cerebrospinal fluid cultures remained sterile, whereas culture of a protected brush specimen (PBS) obtained under fibreoptic bronchoscopy yielded 4 x 103 coryneform bacteria colony-forming units (CFU) per ml, resistant to macrolides, co-trimoxazole, and teicoplanin. Vancomycin was started to ensure activity against a possible R equi. These coryneform bacteria had rapid urease activity (5 min) that strongly suggested C pseudodiphtheriticum, which was confirmed by the Institut Pasteur, Paris. Clinical and radiographic findings rapidly became normal. After 4 weeks, there was no evidence of recurrent pneumonia. Case 2-34, AIDS with 2 week history of progressive dyspnoea, high temperature, productive cough, mucopurulent sputum, vomiting, and rigors; 5 months earlier, admitted for Pneumocystis carinii pneumonia (PCP) treated by co-trimoxazole, changed to dapsone because of adverse reactions; after discharge he received zidovudine and inhaled pentamidine once a month. Temperature was 38°C, heart rate 100/min, arterial blood pressure 120/70 mm Hg, and he had pulmonary crackles over both lung fields. Pa02 was 70 mm Hg while breathing room air, CD4 count was 130/ul, and HIV serology was positive. Radiography showed alveolar opacities in the right lung. Examination of bronchoalveolar lavage fluid revealed P carinii. Bronchial secretions (fibreoptic bronchoscopy) showed numerous polynuclear cells and gram-positive bacilli with morphological features suggestive of coryneform bacteria. Culture grew IOS CFU per ml of an organism that was identified as C pseudodiphtheriticum. Resistance was confined to fluoroquinolones. The patient was treated with dapsone and inhaled pentamidine (300 mg once a day) for PCP, and vancomycin. Clinical and radiographic findings quickly became normal. 4 weeks later, physical findings were unremarkable and the chest radiograph was
normal.
Cpseudodiphtheriticum can cause endocarditis and pneumonia,3-7 especially in apparently immunocompetent patients with previous pulmonary disease. Another case occurred in a young man on steroids for systemic lupus erythematosus.11 There has been only one report in a patient with a disease clinically compatible with HIV infection, although no serological confirmation for HIV was obtained. This patient had a lung abscess from which only C pseudodiphtherriticum was isolated and subsequently required lobectomy.9 Our two cases had documented HIV infection. They are the only cases from about 200 HIV-positive patients who underwent diagnostic fibreoptic bronchoscopy during the past year in our two institutions. Case 2 had relapsing PCP and concomitant infection by C pseudodiphtheriticum. It is difficult to delineate the role of each organism, but C pseudodiphtheritfczan as pure culture and abundant mucopurulent sputum suggest that this organism was at least in part responsible. Infection with C pseudodiphtheriticum is combined with other pathogens in many instances,IO although it was the sole causative agent in some reports.6 By contrast with R equi, which is susceptible to macrolides,’" one of our isolates
was
resistant
to
these
compounds. This
may have
