Background: Tuberculosis (TB) and HIV rates continue to escalate in Russia, but true rates for drug resistance, especially multidrug resistant tuberculosis (MDR TB), are unknown. A study was conducted with the aims of identifying first line drug resistance, both in the civilian and prison sectors, for new and previously treated cases; and risk factors for the development of drug resistance. Methods: A cross sectional survey was undertaken of 600 patients (309 civilians, 291 prisoners) with bacteriologically confirmed pulmonary TB over a 1 year period during 2001-2 in Samara Oblast, Russia. Results: The prevalence of isoniazid, rifampicin, streptomycin, ethambutol and pyrazinamide resistance in new TB cases (civilian and prison patients) was 38.0%, 25.2%, 34.6%, 14.7%, and 7.2%, respectively. The prevalence of MDR TB was 22.7%, 19.8%, and 37.3% in all new cases, new civilian cases, and new prison cases, respectively, with an overall prevalence of 45.5% and 55.3% in previously treated cases. Factors associated with resistance included previous TB treatment for more than 4 weeks, smoking (for isoniazid resistance), the presence of cavitations on the chest radiograph, and imprisonment. HIV was not associated with resistance in all patients. The rates of resistance were significantly higher in prisoners, with rate ratios (RR) of 1.9 (95% CI 1.1 to 3.2) for MDR TB, 1.9 (95% CI 1.1 to 3.2) for rifampicin, and 1.6 (95% CI 1.0 to 2.6) for isoniazid. Conclusions: Rates of first line drug resistance are high, particularly in prisoners and previously treated cases. TB control programmes should initially focus on standardised treatment to maximise cure, combined with measures to reduce institutional TB spread (particularly in prisons) coupled with early diagnosis of MDR TB to reduce the spread and development of resistance.

Background-A national survey of tuberculosis was conducted in England and Wales in 1998 to obtain detailed information on the occurrence of the disease and recent trends. This survey also piloted the methodology for enhanced tuberculosis surveillance in England and Wales and investigated the prevalence of HIV infection in adults with tuberculosis. Methods-Clinical and demographic data for all cases diagnosed during 1998 were obtained, together with microbiological data where available. Annual incidence rates in the population were estimated by age, sex, ethnic group, and geographical region using denominators from the 1998 Labour Force Survey. Incidence rates in diVerent subgroups of the population were compared with the rates observed in previous surveys. The tuberculosis survey database for 1998 was matched against the Communicable Disease Surveillance Centre HIV/AIDS database to estimate the prevalence of HIV co-infection in adult patients with tuberculosis. Results-A total of 5658 patients with tuberculosis were included in the survey in England and Wales (94% of all formally notified cases during the same period), giving an annual rate of 10.93 per 100 000 population (95% CI 10.87 to 10.99). This represented an increase of 11% in the number of cases since the survey in 1993 and 21% since 1988. In many regions case numbers have remained little changed since 1988, but in London an increase of 71% was observed. The number of children with tuberculosis has decreased by 10% since 1993. Annual rates of tuberculosis per 100 000 population have continued to decline among the white population (4.38) and those from the Indian subcontinent, although the rate for the latter has remained high at 121 per 100 000. Annual rates per 100 000 have increased in all other ethnic groups, especially among those of black African (210) and Chinese (77.3) origin. Over 50% of all patients were born outside the UK. Recent entrants to the UK had higher rates of the disease than those who had been in the country for more than 5 years or who had been born in the UK. An estimated 3.3% of all adults with tuberculosis were co-infected with HIV.Conclusions-The epidemiology of tuberculosis continues to change in England and Wales and the annual number of cases is rising. More than one third of cases now occur in young adults and rates are particularly high in those recently arrived from high prevalence areas of the world. The geographical distribution is uneven with urban centres having the highest rates. The increase in the number of cases in London is particularly large. Tuberculosis in patients co-infected with HIV makes a small but important contribution to the overall increase, particularly in London. To be most eVective and to make the most eYcient use of resources, tuberculosis prevention and control measures must be based on accurate and timely information on the occurrence of disease. A new system of continuous enhanced tuberculosis surveillance was introduced in 1999, based on the methodology developed in this national survey.

No abstract