Epidemiology of tuberculosis on Gran Canaria: a 4 year population study using traditional and molecular approaches

BACKGROUND: In recent years several population based studies using restriction fragment length polymorphism (RFLP) analysis have shown a higher rate of recent transmission of tuberculosis than previously thought. This study was undertaken to determine the transmission patterns of tuberculosis and the potential causes of recent transmission on the island of Gran Canaria (Spain). METHODS: The strains of all patients diagnosed with tuberculosis confirmed by culture between 1 January 1993 and 31 December 1996 were typed by RFLP using the insertion sequence IS6110. A cluster was defined as two or more isolates with an identical RFLP pattern. Epidemiological linkage through contact tracing was investigated. RESULTS: Of the total of 719 patients, 153 (21.3%) were excluded because there was inadequate bacterial DNA for genotyping (n=129) or the isolates of Mycobacterium tuberculosis had less than five copies of IS6110 (n=24). The isolates from 409 patients (72.3%) were grouped into 78 different clusters with an estimated 58.5% of the cases being due to recent transmission. Young age was the only significant predictor of clustering. Only in 147 (35.9%) of the 409 patients belonging to a cluster could an epidemiological link be found. 111 patients (19.6%) were identified as having had previous contact with a tuberculosis patient and 81 of them (72.9%) belonged to a cluster. The three largest clusters included 75, 49 and 20 patients, respectively. CONCLUSION: Recent transmission is frequent among patients with tuberculosis on Gran Canaria and could be associated with certain aspects of control measures. Some of the clusters described in the study could be due to the prevalence of particular strains of M tuberculosis on the island.     

Molecular epidemiology of tuberculosis in London 1995-7 showing low rate of active transmission

BACKGROUND: Tuberculosis notification rates for London have risen dramatically in recent years. Molecular typing of Mycobacterium tuberculosis has contributed to our understanding of the epidemiology of tuberculosis throughout the world. This study aimed to assess the degree of recent transmission of M tuberculosis in London and subpopulations of the community with high rates of recent transmission. METHODS: M tuberculosis isolates from all persons from Greater London diagnosed with culture positive tuberculosis between 1 July 1995 and 31 December 1997 were genetically fingerprinted using IS6110 restriction fragment length polymorphism (RFLP) typing. A structured proforma was used during record review of cases of culture confirmed tuberculosis. Cluster analysis was performed and risk factors for clustering were examined in a univariate analysis followed by a logistic regression analysis with membership of a cluster as the outcome variable. RESULTS: RFLP patterns were obtained for 2042 isolates with more than four copies of IS6110; 463 (22.7%) belonged to 169 molecular clusters, which ranged in size from two (65% of clusters) to 12 persons. The estimated rate of recent transmission was 14.4%. Young age (0-19 years) (odds ratio (OR) 2.65, 95% confidence interval (CI) 1.59 to 4.44), birth in the UK (OR 1.55, 95% CI 1.04 to 2.03), black Caribbean ethnic group (OR 2.19, 95% CI 1.15 to 4.16), alcohol dependence (OR 2.33, 95% CI 1.46 to 3.72), and streptomycin resistance (OR 1.82, 95% CI 1.15 to 2.88) were independently associated with an increased risk of clustering. CONCLUSIONS: Tuberculosis in London is largely caused by reactivation or importation of infection by recent immigrants. Newly acquired infection is also common among people with recognised risk factors. Preventative interventions and early diagnosis of immigrants from areas with a high incidence of tuberculosis, together with thorough contact tracing and monitoring of treatment outcome among all cases of tuberculosis (especially in groups at higher risk of recent infection), remains most important.     

Associate investigations: detection of tuberculosis infections in children resulting in discovery of undiagnosed tuberculosis in adults

The authors present the design and implementation of associate investigations of young children with positive tuberculin skin test results. Case study analysis of an associate investigation was done using epidemiologic surveillance techniques, medical interviewing, sociogram mapping, tuberculin skin testing, radiographic evidence, and bacteriologic analysis. Deoxyribonucleic acid fingerprinting of the Mycobacterium tuberculosis isolates using a standardized IS6110-based restriction fragment length polymorphism analysis and IS6110-independent DNA spoligotyping methods was done to track and identify specific bacterial strains. Deoxyribonucleic acid fingerprinting and spoligotyping done on isolates obtained from family members demonstrated same-strain transmission of M. tuberculosis. Three adults with active pulmonary disease and six individuals with latent tuberculosis (TB) were discovered during this investigation. The arrival of a family member from Mexico who had the same strain suggests that the source case lives in Mexico. A child with positive tuberculin skin test results indicates recent and potentially ongoing transmission of TB in the community. Targeted tuberculin skin testing performed on high-risk groups by primary care physicians allows for detection of TB infections. When TB infections are discovered in children, associate investigations can result in the discovery of undiagnosed adult cases and prevent further transmission within the community.     

Mycobacterial infection in HIV seropositive and seronegative populations, 1987-93.

BACKGROUND--Although the causes of the worldwide resurgence of tuberculosis are multifactorial, the HIV epidemic is believed to have had a central role. Control is further threatened by the emergence of multidrug-resistant tuberculosis. METHODS--A retrospective evaluation was undertaken of trends in pulmonary and extrapulmonary culture positive mycobacterial pathology, and the prevalence of drug-resistant tuberculosis in both HIV seropositive and, presumptively, HIV seronegative patients receiving their clinical care at St Mary's Hospital, London. Five hundred and thirty eight patients (188 of whom were known to be HIV seropositive) with positive mycobacterial isolates between January 1987 and March 1993 were identified from laboratory records. These were cross referenced with drug surveillance records. RESULTS--Overall, between 1987 and 1992 there was a progressive 3.5 fold increase in positive mycobacterial isolates and a 2.5 fold increase in patients with proven mycobacterial infection. This increase was greater within the HIV seropositive population. A total of 663 positive mycobacterial isolates was evaluated; the major pathogen identified was Mycobacterium tuberculosis (379 isolates, 57%). Three hundred and fourteen patients were diagnosed as having M tuberculosis, 49 of whom were HIV seropositive. M tuberculosis was predominantly isolated from the lung. Of 358 positive cultures for M tuberculosis (68 HIV seropositive, 290 presumptively HIV seronegative), only 27 isolates (7.6%), almost exclusively derived from presumed HIV seronegative patients, were resistant to either isoniazid, rifampicin, or both drugs together. No increases in drug-resistant isolates were observed over this period. CONCLUSIONS--There has been a considerable increase in the incidence of tuberculosis in both HIV seronegative and seropositive populations during the study period. The emergence of drug-resistant tuberculosis was not observed.     

Antibiotic resistant tuberculosis in the United Kingdom: 1993-1999

BACKGROUND: The re-emergence of tuberculosis as a global health problem over the past two decades, accompanied by an increase in tuberculosis drug resistance, prompted the development of a comprehensive national surveillance system for tuberculosis drug resistance in 1993. METHODS: The UK Mycobacterial Resistance Network (Mycobnet), which includes all mycobacterial reference and regional laboratories in the UK, collects a minimum dataset on all individuals from whom an initial isolate of Mycobacterium tuberculosis complex has been isolated and submitted by source hospital laboratories. Data sought include susceptibility to first line antibiotics, demographic, geographical, and risk factor information. RESULTS: There were 25 217 reports of initial isolates of M tuberculosis complex in the UK between 1993 and 1999. All were tested for sensitivity to isoniazid, rifampicin, and ethambutol and 12 692 of the isolates were also tested for sensitivity to pyrazinamide and streptomycin. A total of 1523 (6.1%) isolates were resistant to one or more drugs, 1397 isolates (5.6%) were resistant to isoniazid with or without resistance to other drugs, and 299 (1.2%) were multidrug resistant. Although the numbers of drug resistant isolates increased over the period, the proportions remained little changed. Certain groups of people were at a higher risk of acquiring drug resistant tuberculosis including younger men, residents of London, foreign born subjects, patients with a previous history of tuberculosis and those infected with HIV. CONCLUSION: Although the proportion of drug resistant tuberculosis cases appears to be stable in the UK at present, more than one in 20 patients has drug resistant disease at diagnosis and more than one in 100 has multidrug resistant disease. Tuberculosis control measures should be strengthened to minimise the emergence of drug resistance through rapid diagnosis, rapid identification of drug resistance, supervised treatment, and maintenance of comprehensive surveillance.     

Atypical mycobacterial spondylitis in HIV-negative patients identified by genotyping

Non-tuberculous mycobacterial infections pose a significant diagnostic and therapeutic challenge. We report two cases of such infection of the spine in HIV-negative patients who presented with deformity and neurological deficit. The histopathological features in both specimens were diagnostic of tuberculosis. The isolates were identified as Mycobacterium intracellulare and M. fortuitum by genotyping (MicroSeq 16S rDNA Full Gene assay) and as M. tuberculosis and a mycobacterium other than tuberculosis, respectively, by culture. There is a growing need for molecular diagnostic tools that can differentiate accurately between M. tuberculosis and atypical mycobacteria, especially in regions of the developing world which are experiencing an increase in non-tuberculous mycobacterial infections.     

Tuberculous infection in a vascular prosthesis: a case of aortic graft infection resulting from disseminated tuberculosis.

Insertion of an aortic prosthetic bypass graft in a patient with untreated disseminated tuberculosis resulted in graft infection with formation of a pseudoaneurysm. Microbiologic cultures implicated Mycobacterium tuberculosis as the causative agent, and a subsequent review of the literature indicated this report to be the first clinical demonstration of the susceptibility of vascular prosthesis to M tuberculosis infection to our knowledge.     

Comparison of characteristics of patients and treatment outcome for pulmonary non-tuberculous mycobacterial infection and pulmonary tuberculosis.

BACKGROUND: Patients with non-tuberculous mycobacteria are usually started on conventional antituberculous triple therapy once acid fast bacilli are detected, before the exact type of mycobacteria has been identified. The ability to identify the characteristics of patients with tuberculous and non-tuberculous mycobacteria may be helpful in identifying before treatment those patients more likely to have non-tuberculous infection. METHODS: A retrospective study was conducted of all patients in one unit in whom non-tuberculous mycobacteria were identified in sputum or bronchoalveolar washings in the period 1987-93. The pattern of drug resistance was determined from laboratory records, and all case notes and chest radiographs were reviewed to identify the underlying disease and treatment outcome. All cases were compared with a matched control group of patients with culture positive Mycobacterium tuberculosis diagnosed during the same period. RESULTS: In the period studied there were 70 non-tuberculous and 221 tuberculous isolates. The non-tuberculous bacteria were typed as follows: M xenopi 23 (33%), M kansasii 19 (27%), M fortuitum 14 (20%), others 14 (20%). Of those with non-tuberculous mycobacteria, 83% were white subjects compared with 47% for tuberculosis. Patients with non-tuberculous mycobacteria were older than those with tuberculosis. Pre-existing lung disease or AIDS was present in 81% of patients with non-tuberculous mycobacteria and in 17% of patients with tuberculosis. Sensitivity to rifampicin and ethambutol was seen in 95% of M xenopi and 96% of M kansasii isolates. Relapse occurred in 60% of cases infected with M xenopi, 20% infected with M kansasii, and in 7% of cases with tuberculosis. CONCLUSIONS: In the population studied non-tuberculous mycobacteria occurred most frequently in elderly white subjects with pre-existing lung disease. If mycobacteria are detected in this group, consideration should be given to the possibility of non-tuberculous infection before embarking on treatment. A combination containing rifampicin and ethambutol is effective. The relapse rate for infection with M xenopi is high and prospective studies of the effect of the above combination of antituberculosis drugs are needed.     

The microbial spectrum of brain abscess with special reference to anaerobic bacteria

The bacteriological and clinical findings of 18 cases of brain abscess are presented. Identification and antimicrobial susceptibility of anaerobic strains was performed both by conventional methods and the newer RapID ANA II panel and E-test methods, respectively. Characterization of the anaerobic isolates was done by restriction fragment length polymorphism (RFLP). Anaerobic or aerobic organisms alone were recovered in three (16.6%), and mixed aerobic and anaerobic in another three (16.6%) patients. There were nine anaerobic isolates. The predominant anaerobes were Prevotella melaninogenicus (four isolates), Bacteroides preacutus (three isolates), Fusobacterium nucleatum and Peptostreptococcus sp. (one isolate each). A total of six aerobic isolates, all of which were Gram-positive cocci were recovered. All the anaerobic isolates were susceptible to metronidazole, but two isolates of P. melaninogenicus were resistant to penicillin. There was complete agreement between the identification based on biochemical profiles and RFLP patterns. These findings indicate the microbial complexity of brain abscess and the need to target antimicrobial therapy against both the aerobic and anaerobic components of infection.     

Network of Contacts between Cattle Herds in a French Area Affected by Bovine Tuberculosis in 2010

France attained ‘Officially Tuberculosis‐Free’ status in 2000. However, the Côte d’Or department (a French administrative unit) has since seen an increase in bovine tuberculosis (bTB) cases, with 35% of cases attributed to neighbourhood contamination. The aim of this study was to investigate the characteristics of neighbourhood contacts in an area affected by bTB in 2010, through the use of social network methods. We carried out a survey to determine the frequency and distribution of between‐herd contacts in an area containing 22 farms. Contacts were weighted, as not all types of contact carried the same risk of bTB transmission. Cattle movement was considered to be associated with the highest risk, but was not observed within the studied area during the study period. Contact with wild boars was the most frequent type of contact, but was associated with a very low risk. Direct cattle‐to‐cattle contacts in pasture and contacts with badger latrines were less frequent, but entailed a greater risk of M. bovis transmission. Centrality values were heterogeneous in these two networks. This would enable the disease to spread more rapidly at the start of epidemics than in a perfect randomly mixed population. However, this situation should also result in the total number of infected herds being smaller. We attributed 95% of the contacts to direct contact in pasture or contact with wild boars or badger latrines. Other kinds of contact occurred less frequently (equipment sharing, cattle straying) or did not occur at all (attendance at a show). Most of the contact types were correlated, but none was sufficient in itself to account for all contacts between one particular farm and its neighbours. Contacts with neighbours therefore represent a challenge for the implementation or improvement of control measures.     

Knowledge and beliefs about tuberculosis among non-working women in Ravensmead, Cape Town

The results of a community-based survey on knowledge and beliefs about tuberculosis in non-working women are presented. The women in the sample showed a very good knowledge of the important aspects of tuberculosis: 90% were aware that it is a problem in their area; 97% knew that it affects the chest; 94% said that it could be fatal; 85% considered it to be infectious and 88% knew that the local clinic provided treatment. Their knowledge of symptoms was good overall but the study revealed misconceptions about the causes and transmission of tuberculosis; 16% indicated that they would not be keen to associate with people with tuberculosis owing to fear of infection. The design effect of cluster sampling was considered in the analysis. The highest design effects (i.e. the most clustering of responses) were found for responses to questions on the causes of tuberculosis and places where treatment could be obtained, possibly reflecting that these beliefs are influenced by neighbourhood contacts. Future tuberculosis education in this group needs to build on existing knowledge and awareness and should focus on changing attitudes such as misconceptions about transmission and the stigmatisation of the disease. Health workers face the challenge of changing behaviour in this community to ensure that people with symptoms present early for screening and that people diagnosed as having tuberculosis comply with treatment.     

结核分枝杆菌快速培养和常规药敏试验在脊柱结核治疗中的应用

目的 探讨BACT/ALERT 3D系统快速培养和绝对浓度法药敏试验对指导脊柱结核个体化化疗的应用价值,分析研究脊柱结核耐药情况.方法 根据临床表现、影像学表现、病理检查,50例患者诊断为脊柱结核,并接受手术治疗.收集术中所取脓液、干酪样组织.低温避光保存,8 h内送检,常规处理后接种液体培养基,使用BACT/ALERT 3D系统进行分枝杆菌快速培养.培养阳性者接种PNB和TCH培养基进行菌种鉴定,并将细菌接种至含药改良罗氏培养基,按绝对浓度法进行11种常用一线和二线抗结核药物药敏试验.结果 50例标本培养阳性21例(42%),人型结核杆菌19例,牛型结核杆菌2例.结核杆菌培养和药敏试验平均耗时41 d(28～58 d).其中耐药11例(52.4%),异烟肼耐药4例(19.0%),利福平和乙胺丁醇各1例(4.8%),链霉素3例(14.3%),力克肺疾2例(9.5%),左氧氟沙星8例(38.1%).结论 结核分枝杆菌快速培养和常规药敏试验准确度高,费用低,可检测常用一线和二线药物的敏感性,适用于指导脊柱结核个体化化疗方案的制定.异烟肼、利福平、吡嗪酰胺、乙胺丁醇或(和)链霉素联合用药方案对多数初治脊柱结核患者有效.     

Impact of social interactions in the community on the transmission of tuberculosis in a high incidence area

BACKGROUND: Tuberculosis (TB) is transmitted by close contact with an infectious person. It is assumed that close contact occurs amongst household members and that contact outside the house is "causal" and does not play a major role in the transmission of TB. METHODS: This study was conducted in an impoverished area with a high incidence of TB and a low HIV seropositive prevalence. Thirty three households with 84 TB patients were identified between February 1993 and April 1996 and the transmission of TB was studied by combining Mycobacterium tuberculosis fingerprinting with in depth sociological interviews. RESULTS: Forty two strain genotypes were identified in the 84 patients. In 15 households all the patients had identical strains, in nine households all the patients had different strains, and in nine households some patients had identical strains and one had a different strain. In 26 houses at least one patient had a strain which formed part of a larger community cluster and in 12 of these households the patient(s) had contact with a community member who had the identical strain. In 58% of the cases the contact took place while drinking in social groups. CONCLUSION: In high incidence areas contact outside the household may be important for the transmission of TB. This contact often takes place during recreation which, in the case of this study of impoverished people, consisted of drinking in social groups. Social interaction patterns should be studied and understood for effective implementation of control strategies.     

Follow up of an immunocompromised contact group of a case of open pulmonary tuberculosis on a renal unit.

BACKGROUND--The organisation, management, outcome and cost of follow up of a large group of mainly immunocompromised patients and healthcare workers who were exposed to a staff member of a London renal unit with smear positive pulmonary tuberculosis are described. METHODS--Following British Thoracic Society (BTS) guidelines, 576 close contacts were identified and divided into three groups: (1) 303 renal patients including 61 with renal transplants; (2) 90 surgical patients; and (3) 183 staff members. Screened contacts were interviewed, completed a symptoms questionnaire, and were offered a chest radiograph and Heaf or Mantoux test if appropriate with referral to a chest physician if required. RESULTS--Overall, 524 (85%) living contacts have been screened: 243 (97%) renal (first screening), 63 (70%) surgical, and 135 (74%) staff contacts. Thirty one transplant patients were prescribed isoniazid chemoprophylaxis. Fifty two renal patients had died before screening and 11 deaths occurred after first interview. One case of tuberculosis epidemiologically related to the index case was diagnosed on clinical criteria. A review of the case records and/or death certificates and entries on to tuberculosis registers indicated no further cases. The cost of the investigation was estimated to be approximately franc25 000, or franc44 per contact screened, with staff costs comprising 79% of the total. CONCLUSIONS--Undiagnosed tuberculosis in healthcare workers working with immunosuppressed patients can lead to large and expensive follow up studies. The applicability of the 1990 and 1994 BTS guidelines to the investigation of tuberculosis in an immunocompromised nosocomial group, and the role of the infection control doctor and the consultant in Communicable Disease Control in overlapping nosocomial and community incidents, are discussed.     

Nosocomial bloodstream infection in a tertiary care paediatric intensive care unit

Objective: To determine the frequency, causative organisms and susceptibility pattern of nosocomial bloodstream infections in children. Study Design: Observational study. Place and Duration of Study: Paediatric Intensive Care Unit of the Children's Hospital, Lahore, from January to December 2004. Patients and Methods: All children admitted to the unit during the study period were daily evaluated for features suggestive of nosocomial infection. In addition to other investigations, blood cultures were done in all suspected cases for the confirmation of nosocomial bloodstream infection (BSI). Nosocomial infection was defined according to the criteria set by Centre for Disease Control and Prevention. Demographic, microbiological and other variables were carefully studied to analyze frequency, incidence rate, spectrum of isolates and susceptibility pattern. Children with and without nosocomial BSI were compared with regard to age, duration of stay in hospital, need and duration of ventilation and the outcome. Results: Of the total 406 admissions, 134 children were suspected to have nosocomial infection on at least 214 occasions (episodes). Blood cultures yielded growth of pathological organisms in 62 of these episodes, giving the frequency of nosocomial BSI as 15.2 per 100 admissions (62/406 episodes). Children with nosocomial bloodstream infection were found to have younger mean age (2.1 vs. 4.1 years), longer average duration of stay (13.1 vs. 6.6 days), more frequent need for ventilation (64% vs. 34%) and longer duration of ventilation (9.7 vs. 4.8 days). Majority of isolates (77%) were gram-negative bacteria; Klebsiella being the most common isolate (n= 23). Aztreonam, Ceftiazidime, Ceforuxime and Ciprofloxacin showed high resistance pattern (33-50%). Isolates showed good sensitivity to Vancomycin (100%), Imipenem (80%), Meropenem (100%) and Co-amoxiclav (88%). Conclusion: The frequency of nosocomial BSI in the observed setting was quite high, having marked impact on the duration of stay and outcome. Emergence of resistant pathogens is alarming.     

Tuberculosis Caused by Mycobacterium orygis in Dairy Cattle and Captured Monkeys in Bangladesh: a New Scenario of Tuberculosis in South Asia

Mycobacterium orygis, commonly known as the oryx bacillus and a newly proposed Mycobacterium tuberculosis complex subspecies, was isolated from 18 cattle in a dairy farm and two captured rhesus monkeys in a zoo in Bangladesh. All the infected animals had tuberculosis lesions in their lungs, suggesting transmission and infection with M. orygis by an airborne route. The 20 isolates were analysed using a range of conventional and molecular typing methods, and RD‐deletion typing and sequencing of selected genes confirmed the isolates as M. orygis. Multiple‐locus variable‐number tandem repeat analysis (MLVA) allowed the isolates to be divided into three clusters based on the relatedness of their MLVA profiles. The two monkey isolates shared the same MLVA pattern with 15 of the cattle isolates, whereas the remaining three cattle isolates had different patterns, even though the latter animals had been kept in the same dairy farm. The diversity observed among isolates may suggest the bacteria have been established in this area for a long period. This study along with other recent findings that report the detection of M. orygis from animals as well as humans originating from South Asia potentially indicate endemic distribution of M. orygis in South Asia.     

Tuberculous pseudoaneurysm of the femoral artery complicated by cutaneous tuberculosis of the foot: a case report

An infected pseudoaneurysm of the right common femoral artery in a 69-year-old patient receiving methotrexate therapy was confirmed to have been caused by Mycobacterium tuberculosis. After surgical excision of the aneurysm and revascularization using femoral vein, cutaneous manifestations of M tuberculosis infection in the foot complicated the course. We hypothesized that methotrexate may have triggered the reactivation of dormant tuberculosis in this patient. Because extrapulmonary tuberculous pseudoaneurysms are clinically similar to other types of infected pseudoaneurysm, M tuberculosis infection should always be suspected during the initial diagnosis. We propose that mycobacterial cultures should be routine when initial cultures and Gram stain are negative.     

Mycobacterial infection in patients infected with the human immunodeficiency virus.

Of 207 homosexual or bisexual patients with the acquired immune deficiency syndrome (AIDS), 24 with the AIDS related complex, and 39 with asymptomatic HIV infection, 32 patients were found to have mycobacterial infection. Mycobacterium tuberculosis was found in 13 patients with AIDS and in two with the AIDS related complex. M avium-intracellulare was found in 15 patients with AIDS and was disseminated in 12. One patient was infected with M kansasii and one with M ulcerans. Invasive procedures were frequently required to obtain positive bacteriological results. Subclinical carriage of M avium-intracellulare and other mycobacteria thought to be nonpathogenic was common in patients seronegative for the human immunodeficiency virus and at all stages of human immunodeficiency virus infection. All but one isolate of M tuberculosis were fully sensitive to standard antimycobacterial antibiotics. Response to treatment was usually rapid. M avium-intracellulare isolates were all resistant to first line agents in vitro, and antibiotics such as ansamycin and amikacin were required to obtain a clinical response.     

Pulmonary Mycobacterium kansasii infection: comparison of radiological appearances with pulmonary tuberculosis.

BACKGROUND: A study was undertaken to determine if there are differences in the radiological appearances at presentation between pulmonary infections caused by Mycobacterium kansasii and Mycobacterium tuberculosis. Correct recognition of the organism has important implications with regard to initial therapy and contact tracing. METHODS: The initial chest radiographs of 28 patients with pulmonary M kansasii infection were compared with those of 56 age, sex, and race matched patients with M tuberculosis infection. All patients in both groups were culture positive and none was known to be HIV positive. The radiographs were analysed independently by two radiologists who were unaware of the causative organism. RESULTS: Radiographic abnormalities in patients with M kansasii infection were more frequently unilateral and right side predominant, while those with tuberculosis more frequently involved a lower lobe. Air space shadowing involving more than one bronchopulmonary segment and pleural effusions were seen less frequently in M kansasii infection (four of 28 (14%) versus 30 of 56 (54%) and none of 28 versus 15 of 56 (27%)). Cavitation (21 of 28 (75%) versus 34 of 56 (61%) was seen to a similar extent in patients with M kansasii infection and in those with tuberculosis. Cavities tended to be smaller in patients with M kansasii infection (p < 0.01). CONCLUSIONS: Differences are seen in the radiographic appearances of pulmonary infection caused by M kansasii and M tuberculosis. These differences are not sufficient to allow a positive diagnosis on the basis of radiographic findings alone, but the presence of a pleural effusion or lower lobe involvement makes M kansasii infection very unlikely.