Does DOTS work in populations with drug-resistant tuberculosis?

Transcription

1 Does DOTS work in populations with drug-resistant tuberculosis? Kathryn DeRiemer, Lourdes García-García, Miriam Bobadilla-del-Valle, Manuel Palacios-Martínez, Areli Martínez-Gamboa, Peter M Small, José Sifuentes-Osornio, Alfredo Ponce-de-León Summary Background Directly observed therapy (DOTS) is the main strategy for prevention and control of tuberculosis worldwide. However, its effect on tuberculosis transmission in populations with moderate rates of drug-resistant disease is not known. Methods This population-based prospective study in southern Mexico between March, 1995, and February, 2000, was based on passive case finding and detection of acid-fast bacilli in sputum samples to diagnose pulmonary tuberculosis. We also used cultures, drug-susceptibility testing, bacterial genotyping, and monitoring of treatment outcomes. Findings We enrolled 436 patients; the HIV seroprevalence rate was 2%. We used three indicators to monitor continuing tuberculosis transmission: the incidence rate of pulmonary tuberculosis, which decreased by 54 4% between 1995 and 2000, from 42 1 to 19 2 per 10 5 population (p= ); the percentage of clustered pulmonary tuberculosis cases, which decreased by 62 6% from 22% to 8% (p=0 02); and the rate of primary drug resistance, which decreased by 84 0% from 9 4 to 1 5 per 10 5 population (p=0 004). Rates of multidrug-resistant (MDR) tuberculosis also decreased (p ). The case-fatality ratio was 12% for MDR tuberculosis (five of 41), 7% for strains resistant to at least one drug after exclusion of MDR (four of 55), and 3% for pansusceptible strains (nine of 272). There were 13 treatment failures (11%) in 1995 and one (2%) in 2000 (p=0 012). Interpretation Even in settings with moderate rates of MDR tuberculosis, DOTS can rapidly reduce the transmission and incidence of both drug-susceptible and drug-resistant tuberculosis. However, further interventions, such as drug-susceptibility testing and standardised or individualised treatment regimens, are needed to reduce mortality rates for MDR tuberculosis. Introduction Coincident with expanded efforts to strengthen tuberculosis prevention and control programmes worldwide, there is growing concern about currently reported and potential future rates of multidrug-resistant (MDR) tuberculosis. 1,2 Although there is consensus about the desirability of optimum treatment for all patients with tuberculosis, including those who have resistant organisms, there are few data to guide policy on how best to control and prevent the disease in areas where drug resistance is prevalent. WHO and the International Union Against Tuberculosis and Lung Disease have adopted DOTS (directly observed therapy, short course) as the main strategy for programmatic tuberculosis control. DOTS consists of five main elements: political commitment; case detection by sputum microscopy; directly observed therapy of a standard short-course regimen; uninterrupted supply of all essential drugs; and a standard recording and reporting system that allows assessment of treatment results and overall programme performance. DOTS does not include specific therapy for patients with drug-resistant tuberculosis, but a surge in drug-resistant tuberculosis in several parts of the world requires effective implementation of the DOTS strategy to prevent the occurrence of new MDR tuberculosis cases and to reduce transmission of Mycobacterium tuberculosis. 3 Since 1995, we have carried out a population-based molecular epidemiological study of tuberculosis in a health jurisdiction in southern Mexico. Although a national tuberculosis control programme was implemented in Mexico before 1996, a programme review by the WHO Global Tuberculosis Programme identified inadequate technical policies and management deficiencies. 4 The main problems were excessive emphasis on case detection to the detriment of case holding, policy differences between the Ministry of Health and social security services, and inadequate information systems to monitor essential programme activities such as use of antituberculosis drugs and treatment outcomes. To bring the programme in line with the WHO-recommended DOTS strategy, changes were initiated in 1996 in pilot areas including the Orizaba Health Jurisdiction, to be gradually extended nationwide. 5 There were major improvements to the five elements of the WHO DOTS strategy for tuberculosis control in Mexico, including sustained government commitment to tuberculosis control by modification and standardisation of the guidelines for the national tuberculosis control programme, standardisation of programme procedures nationwide, and training of health-care workers to implement the DOTS programme. The programme standardised Lancet 2005; 365: See Comment page 1206 Division of Infectious Diseases and Geographic Medicine, Stanford University Medical Center, Stanford, CA, USA (K DeRiemer PhD, PMSmall MD); Tuberculosis Unit, Instituto Nacional de Salud Pública, Avenida Universidad 655, Colonia Sta María Ahuacatitlán, Cuernavaca, Morelos CP 62508, Mexico (L García-García MD, M Palacios-Martínez MD); Department of Infectious Diseases, Instituto Nacional de Ciencias Médicas y de Nutrición Salvador Zubirán, Mexico DF, Mexico (M Bobadilla-del-Valle PhD, A Martínez-Gamboa MSc, J Sifuentes-Osornio MD, APonce-de-León MD); Bill and Melinda Gates Foundation, Seattle, WA, USA (P M Small); and Universidad Panamericana Medical School, Mexico DF, Mexico (A Ponce-de-León) Correspondence to: Dr Lourdes García-García garcigar@correo.insp.mx Vol 365 April 2,

2 diagnosis on the basis of quality-assured sputum-smear microscopy for passive case finding, mainly among symptomatic patients presenting to health services and by use of trained laboratory workers and quality-control procedures. Standard short-course chemotherapy with direct observation of treatment, by physicians and community health-care workers trained in proper case management, was provided to all cases. In addition, the information and referral systems were improved to ensure adequate, uninterrupted supplies of qualityassured drugs. A standard, computerised nominal registry of patients, shared by local, state, and national authorities, and a system for programme monitoring were established to assess the treatment outcomes of all tuberculosis patients. 6 Surveillance studies in three states by the Mexican Ministry of Health identified moderate prevalence rates of drug-resistant tuberculosis. 1 The pilot DOTS programme in Orizaba provided an opportunity to assess the epidemiological effect of DOTS in a region where 20 7% of new cases were resistant to at least one antituberculosis drug and 3 3% were MDR tuberculosis. 7 We prospectively measured the incidence rate of pulmonary tuberculosis, the amount of tuberculosis transmission demonstrable with molecular epidemiological techniques, and the rate of primary resistance to at least one first-line drug over a 5-year period. Methods Study population The study site and enrolment procedures have been described previously. 8 The study site is located in five municipalities of the Orizaba Health Jurisdiction, Veracruz state, Mexico, with a total population in 1995 of ; 61 8% were aged 15 years or older. 9 The rate of pulmonary tuberculosis in the jurisdiction remained stable (42 6 per 10 5 population) during on the basis of routine surveillance data, but the rate was higher than that for the whole country (18 2 per 10 5 population). 10 We carried out passive case finding supported by community-based healthworkers and screened people who reported coughing for longer than 15 days. Patients with positive sputum smears for acid-fast bacilli underwent epidemiological, clinical (standard questionnaire, physical examination, chest radiograph, HIV test), and mycobacteriological assessment. A retreatment case of tuberculosis was defined as a patient who had previously received at least 30 days of antituberculosis treatment according to reported history of previous treatment, review of medical records, or both. According to the guidelines of the national tuberculosis control programme of Mexico, treatment for new cases was 2 months of isoniazid, rifampicin, and pyrazinamide then 4 months of isoniazid and rifampicin (2HRZ/4HR). Retreatment cases were given either ethambutol or streptomycin also. 11 In 1998, when the preliminary data from our study showed high rates of drug resistance, the local health jurisdiction adopted the WHO standard regimen of initiating therapy with four drugs for newly diagnosed patients and five drugs for previously treated patients. 12 Treatment with secondline drugs was not locally available, and treatment could not be modified by the local health centre on the basis of the results of drug-susceptibility testing. Patients with MDR tuberculosis were referred to the national tuberculosis control programme and were assessed by physicians of the National Institute of Respiratory Diseases, the national referral centre. We used the operational definitions of the national tuberculosis control programme for treatment outcomes except that success, defaulting, and death were defined according to international definitions. 13 Each patient was followed up annually for assessment of treatment status, outcome, and survival. 7 The research protocol was approved by the appropriate institutional review boards in Mexico and the USA. We enrolled only eligible patients who provided informed written consent. Mycobacteriology and genotyping Sputum samples were processed for M tuberculosis by standard procedures, and isolates were genotyped and compared by use of IS6110-based restriction-fragment length polymorphisms (RFLP) and spoligotyping if the IS6110 RFLP patterns had fewer than six bands. 14,15 Statistical analysis Three indicators of tuberculosis transmission were estimated for each successive 12-month period: the incidence rate of pulmonary tuberculosis; the percentage of pulmonary tuberculosis cases that were clustered; and the rate of pulmonary tuberculosis with primary resistance to at least one first-line drug. As previously described, 8 after January, 1996, the registry of tuberculosis patients was reviewed periodically to ensure that all patients with pulmonary tuberculosis who were diagnosed by the local tuberculosis programme were offered enrolment, thus enabling estimation of rates. We also calculated the case-fatality ratio: the number of tuberculosis patients who died divided by the number of cases, for new, retreatment, or total cases. The casefatality ratio is expressed as a percentage. Patients were judged to be in a cluster if they were diagnosed within 12 months of each other and their initial isolates had identical genotypes with either six or more identical IS6110 RFLP bands or fewer than six identical IS6110 bands plus identical spoligotypes. We assumed that a unique genotype represented reactivation of latent tuberculosis, that a cluster represented continuing transmission, and that the first case in each cluster was caused by reactivation of a latent infection. We used the n minus 1 method to estimate continuing transmission Vol 365 April 2, 2005

3 We used the incidence-rate data test or the 2 test for trends to detect significant trends. 17 For univariate analyses we used the 2 test or Fisher s exact test for categorical variables. We used multiple logistic regression analyses to estimate the odds ratio and 95% CI and to identify the independent predictors of death from tuberculosis. All analyses used Stata statistical software (version 7.0). Role of the funding sources The funding sources had no role in the study design; collection, analysis, or interpretation of data; or the Period Total p % change March 1995 to March 1996 to March 1997 to March 1998 to March 1999 to (1995 to 2000) February 1996 February 1997 February 1998 February 1999 February 2000 New tuberculosis cases Number Rate per 10 5 per year Retreatment tuberculosis cases Number Rate per 10 5 per year Tuberculosis cases, status unknown Number All tuberculosis cases Number Rate per 10 5 per year New tuberculosis cases resistant to at least one first-line drug* Number Rate per 10 5 per year Retreatment tuberculosis cases resistant to at least one first-line drug* Number Rate per 10 5 per year All tuberculosis cases resistant to at least one first-line drug* Number Rate per 10 5 per year New MDR tuberculosis cases Number Rate per 10 5 per year Retreatment MDR tuberculosis cases Number Rate per 10 5 per year All MDR tuberculosis cases Number Rate per 10 5 per year Clustered tuberculosis cases n Percentage Number in largest cluster Clustered tuberculosis cases resistant to at least one drug Number Percentage of all genotyped tuberculosis cases resistant to at least one drug Clustered pansusceptible tuberculosis cases Number Percentage of all genotyped pansusceptible tuberculosis cases Clustered MDR tuberculosis cases Number Percentage of all genotyped MDR tuberculosis cases Deaths during treatment among new tuberculosis cases Number Case-fatality ratio (%) Deaths during treatment among retreatment tuberculosis cases Number Case- fatality ratio (%) Deaths during treatment among all tuberculosis cases Number Case- fatality ratio (%) Treatment failures Number Percentage *Based on standard drug-susceptibility testing to isoniazid, rifampicin, ethambutol, and streptomycin. Based on the n 1 method. Table 1: Indicators of tuberculosis transmission between March, 1995, and February, Vol 365 April 2,

4 writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit the paper for publication. Results Among 3679 patients who reported coughing for longer than 15 days, 490 (13%) cases of bacteriologically confirmed pulmonary tuberculosis were diagnosed between March, 1995, and February, Of these, 436 provided informed written consent and were enrolled. 94% of the study participants agreed to voluntary HIV counselling and testing, and 2% of them were HIV seropositive. Overall, 323 (74%) of the 436 participants were newly detected cases, 109 (25%) were retreatment cases, and the previous treatment status of four (1%) was unknown. The incidence rate of pulmonary tuberculosis decreased by 54 4% from 42 1 per 10 5 in 1995 to 19 2 per 10 5 population in 2000 (p= ). The reduction in the rate of new tuberculosis cases was gradual and sustained (table 1). The reduction in the rate of retreatment cases was greatest between the first and second 12-month periods, with no further decline (p=0 053). The initial isolate was genotyped in 313 (72%) of the 436 study participants. Information about the initial chest-radiograph interpretation was missing for 31 of the tuberculosis patients, including six for whom information about their ethnicity was also missing. Patients whose isolates could not be genotyped were more likely to be from an indigenous ethnic group (24/64 [38%] vs 83/366 [23%], p=0 011), or to report a history of incarceration (36/106 [34%] vs 74/330 [22%], p=0 017). 25 clusters were identified by IS6110 RFLP and spoligotyping. There were nine clusters during the first study period, the largest with five tuberculosis cases during the initial period and a total of nine cases during the study. The percentage of cases that were clustered decreased by 62 6% from 22% during 1995 to 8% in 2000 (p=0 02). The size of clusters also decreased over time. Results of drug-susceptibility tests were available for 371 (85%) patients, and 96 (26%) had isolates that were resistant to at least one first-line drug. Patients without drug-susceptibility results were more likely to be from an indigenous ethnic group (16/64 [25%] vs 47/366 [13%], p=0 011) and to have no cavitary disease (49/258 [19%] vs 10/147 [7%], p=0 001). Overall, 46 (17%) new cases and 50 (52%) retreatment cases were resistant to at least one first-line antituberculosis drug; retreatment cases were more likely than new cases to be drug resistant (odds ratio 5 3 [95% CI ], p ). The rate of new cases that were resistant to at least one first-line drug decreased by 84 0% from 9 4 per 10 5 population in 1995 to 1 5 per 10 5 population in 2000 (p=0 004). We identified 41 cases of MDR tuberculosis, six (15%) of them among new cases. Among 24 patients with several episodes of tuberculosis during the study, the strain from only one patient became resistant to several antibiotics. The rate of MDR tuberculosis among new cases decreased from 2 8 per 10 5 population in 1995 to zero in 2000 (p=0 127), and the rate of MDR tuberculosis among retreatment cases decreased significantly from 7 2 per 10 5 population in 1995 to 3 0 Treatment outcome Total cases Cases with Cases with strains resistant to at least Cases with MDR Cases without pansusceptible strains one drug excluding MDR drug-susceptibility results New tuberculosis cases Success 272 (84%) 196 (87%) 33 (83%) 3 (50%) 40 (78%) Default 24 (7%) 18 (8%) 4 (10%) 0 2 (4%) Died during therapy 10 (3%) 5 (2%) 2 (5%) 1 (17%) 2 (4%) Treatment failure 4 (1%) 2 (1%) 1 (3%) 1 (17%) 0 Relapse Reinfection Other* 7 (2%) 4 (2%) 1 (2%) 1 (17%) 2 (4%) Unknown 6 (2%) (10%) Total Retreatment cases Success 50 (46%) 28 (61%) 8 (53%) 5 (14%) 9 (69%) Default 14 (13%) 9 (20%) 0 4 (11%) 1 (8%) Died during therapy 12 (55%) 4 (9%) 2 (13%) 4 (11% 2 (4%) Treatment failure 19 (17%) 1 (33%) 4 (27%) 14 (40%) 0 Relapse 2 (2%) 1 (2%) 1 (7%) 0 0 Reinfection 3 (3%) 3(7%) Other* 4 (4%) (11%) 2 (4%) Unknown 5 (5%) (11%) 5 (10%) Total Results of drug-susceptibility tests were available for 371 (85%) of the 436 participants. Information on history of previous treatment was not available for four patients; therefore data are presented for 432 individuals. *Includes patients who discontinued treatment on medical advice and those who transferred or moved out of the study area. Table 2: Treatment outcome of new and retreatment tuberculosis cases according to M tuberculosis drug-susceptibility pattern, March, 1995, to February, Vol 365 April 2, 2005

5 per 10 5 population in 2000 (p ). The proportion of clustered cases was small, and neither the observed change in the proportion of drug-resistant clustered cases nor the change in clustered MDR tuberculosis cases was significant (p=0 518 and p=0 338, respectively). 22 (5%) patients died of tuberculosis during antituberculosis treatment, and a further 17 died after treatment. The case-fatality ratios among new cases, retreatment cases, and all cases did not change significantly over time (p=0 246, p=0 972, and p=0 313, respectively). The case-fatality ratio was highest for patients with MDR tuberculosis (12%; five of 41), and it was lower among patients with resistance to at least one drug after exclusion of MDR cases (7%; four of 55) and patients with pansusceptible strains (3%; nine of 272). In a model controlling for sex and age, the strongest independent risk factors for death during treatment were resistance to at least one drug (odds ratio 2 4 [95% CI ], p=0 002) and previous treatment (3 7 [ ], p ). In addition, the number of treatment failures declined over time from 13 (11%) in 1995 to one (2%) in 2000 (p=0 012); most of the treatment failures were among retreatment cases (19/23 [83%]) and individuals with MDR tuberculosis (15/41 [37%]; table 2). Discussion Our prospective population-based study shows that implementation of DOTS in a health jurisdiction that had a moderate rate of drug-resistant tuberculosis decreased the transmission of both drug-susceptible and drug-resistant tuberculosis, as shown by decreases in the tuberculosis incidence rate, the proportion of cases that were clustered, and the rate of newly diagnosed drug-resistant cases. We did not detect significant transmission of drug-resistant tuberculosis resulting in new secondary cases, as in other studies, 18,19 or further amplification of resistance over a 5-year period. 20 The findings of this study both confirm and extend our understanding of tuberculosis control. Two previous molecular epidemiological studies showed that the transmission of tuberculosis decreased as tuberculosiscontrol interventions were implemented in the USA. 21,22 To date, however, cohort studies from high-burden countries have focused on treatment outcomes, not transmission dynamics. For example, a retrospective study in six countries concluded that standard shortcourse chemotherapy is an inadequate treatment for some cases of resistant disease, 23 and a Peruvian study showed the feasibility of treating patients who have chronic MDR tuberculosis with second-line drugs. 24,25 By contrast, our study shows the efficacy of DOTS in interrupting tuberculosis transmission in a less developed country. The initial disagreement between advocates of the DOTS strategy and those supporting an intensified DOTS-plus approach has evolved to a consensus on the value of treating MDR tuberculosis. 26 The rates of treatment failure and death among patients with MDR disease in our study (37% [15 of 41] and 12% [five of 41]) support this consensus. Clearly, patients with drugresistant tuberculosis have increased risks of treatment failure and death, and additional interventions are needed to improve their treatment outcomes. Several studies have now shown the feasibility and effectiveness of DOTS-plus and other strategies to reduce mortality among patients with MDR tuberculosis. 20,24,27 The challenge to tuberculosis-control programmes now becomes the provision of care to individuals with MDR tuberculosis without negatively affecting treatment outcomes for patients with drug-susceptible infections. 2 There are several methodological limitations to our study. Sputum microscopy is insensitive for detection of acid-fast bacilli in sputum samples; however, use of this method increases the generalisability of our study because it is the most commonly recommended method for diagnosis of pulmonary tuberculosis worldwide. We could not genotype the initial isolate of all cases, so the proportion of isolates belonging to clusters is subject to error, particularly among individuals belonging to indigenous ethnic groups or with a history of incarceration. Furthermore, we could not analyse trends for single resistance to individual drugs because there were too few patients whose isolates showed such resistance patterns. However, the likelihood that the findings of this longitudinal study are due to bias is low because the detection and enrolment procedures, the study team, the data-collection forms, and the definition of new and retreatment cases were unchanged throughout the study. Furthermore, detailed epidemiological investigation of clustered patients showed that the high number of reported MDR cases at the outset of the study was not due to an outbreak of MDR tuberculosis. Finally, the tuberculosis incidence rate did not decrease during and the tuberculosis incidence rate in 1995 is not higher than that in , which rules out the possibility that the subsequent trends are biased by the detection of prevalent cases at the start of this study. Analysis of the trends during gives the same decreasing trend in tuberculosis incidence rates as the analysis for , additional confirmation that the detection of prevalent cases is not driving the statistical significance. Moreover, the concomitant decreasing trends of newly diagnosed and primary drug-resistant cases support our conclusion that implementation of DOTS truly decreased tuberculosis transmission. Although this study was carried out in accordance with prevailing ethical standards, it raises controversial issues about the standard of care to which research participants are entitled. The study was initiated in 1995, before controversy arose about whether research participants in less developed countries are entitled to Vol 365 April 2,

6 the best available local versus worldwide standard of care. The debate in relation to patients with MDR tuberculosis did not become prominent until Furthermore, the Declaration of Helsinki s contention that the best standard of care worldwide must always be used is not widely shared. Guidelines by the Council for International Organizations of Medical Sciences, the Nuffield Council, the National Bioethics Advisory Commission, and others differ from the Declaration of Helsinki and note that exceptions can be made to the worldwide best standard of care. 28 Indeed, the Declaration of Helsinki s position on this issue that the best treatment worldwide must always be used was not clearly affirmed until its revision in Edinburgh in the autumn of 2000, after our study was finished but before publication. The care initially provided to tuberculosis patients in 1995 by the Orizaba health jurisdiction was that recommended for the entire country of Mexico. In 1998, the health jurisdiction adopted the WHO recommendations, which were the treatment standards being used to treat 98% of the world s tuberculosis patients. The study was completed before the reporting of improved outcomes in patients in Peru given individualised therapy. 20 Throughout the study period, the WHO recommendations made no provision for individualised therapy for drug-resistant tuberculosis. 23 Thus, our study protocol complied with the worldwide standard of care consistent with the Declaration of Helsinki and other ethical guidelines for the conduct of research. The ethical debate that may be more relevant to this study is not the standard of care used but rather, what is the appropriate level of ancillary care that can or should be provided in research studies in less developed countries? Ancillary care is defined as care that goes beyond the requirements of scientific validity, safety, keeping promises, or rectifying injuries. 29,30 There is still no agreement about what constitutes the worldwide standard of ancillary care for tuberculosis patients who have drug-resistant strains, and the topic remains highly controversial. This important issue needs reasonable discussion, and international investigators need clear guidance. In conclusion, in a health jurisdiction with a moderate rate of drug-resistant tuberculosis, DOTS can rapidly reduce the transmission of both susceptible and resistant organisms. Importantly, this effect was seen in a local programme even when regional or national programmes were not yet fully implemented. However, rates of treatment failure and mortality among patients with MDR tuberculosis remained unacceptably high. Thus, additional measures, such as drug-susceptibility testing and standard or individualised therapy, are needed to improve clinical outcomes. Our results highlight the urgent need for both DOTS and DOTSplus in confronting the global epidemic of tuberculosis, including drug-resistant disease. 28 Contributors Kathryn DeRiemer and Ma de Lourdes García-García participated in the study design and management and analysis of data and wrote the report. Miriam Bobadilla-del-Valle, Areli Martínez-Gamboa, Jose Sifuentes-Osornio, and Alfredo Ponce-de-León participated in the study design, organised and carried out the laboratory work, and contributed to the analyses. Manuel Palacios-Martínez contributed to the study design and implementation. Peter Small made valuable contributions to the study s design, analysis, and the report. All authors reviewed and approved the final report. Conflict of interest statement We declare that we have no conflict of interest. Acknowledgments We thank the population, patients, and health-care workers of the Orizaba Health Jurisdiction, Mexico, for their generous support and cooperation; Martien Borgdorff and Dennis Osmond for thoughtful comments and suggestions before submission of the paper; and several anonymous reviewers for their valuable contributions. This study was supported by the US National Institutes of Health (FIC K01 TW and NIAID AI 35969), the Wellcome Trust (176W009), and the Howard Hughes Medical Institute (ID ). References 1 WHO/IUATLD Global Project on Anti-Tuberculosis Drug Resistance Surveillance. Anti-tuberculosis drug resistance in the world. Report 2: prevalence and trends. WHO/CDC/TB/ Geneva: WHO, Nachega JB, Chaisson RE. Tuberculosis drug resistance: a global threat. Clin Infect Dis 2003: 36 (suppl 1): S WHO Global Tuberculosis Programme. An expanded DOTS framework for effective tuberculosis control. Publication WHO/CDS/TB/ Geneva. WHO, Pio A, Luelmo F, Kumaresan J, Spinaci S. National tuberculosis programme review: experience over the period Bull World Health Organ 1997; 75: PanAmerican Health Organization. El control de la Tuberculosis en las Americas. Bol Epidemiol Organ Panamericana de la Salud 1998; 19: Yañez-Velasco LB, Quiroz G, Rodríguez J. Use of DOTS in pilot areas in Mexico (abstract 129-PC03). Inter J Tuberc Lung Dis 1997: 1: S68. 7 Garcia Garcia ML, Ponce de León A, Jiménez Corona ME, et al. Clinical consequences and transmissibility of drug resistant tuberculosis in southern México. Arch Intern Med 2000; 160: García García ML, Palacios Martínez M, Ponce de León A, et al. The role of core groups in transmitting Mycobacterium tuberculosis in a high prevalence community in Southern México. Int J Tuberc Lung Dis 1999; 4: Instituto Nacional de Estadística, Geografía e Informática. Anuario Estadístico del Estado de Veracruz. México: Talleres Gráficos del Instituto Nacional de Estadística, Geografía e Informática, García García ML, Small PM, García-Sancho C, et al. Tuberculosis epidemiology and control in Veracruz, Mexico. Int J Epidemiol 1999; 28: Secretaría de Salud. Norma Oficial Mexicana NOM-006-SSA2 1993, para la prevención y control de la tuberculosis en la atención primaria a la salud. Diario Oficial de la Federación Jan 26, 1995: Secretaría de Salud. Modificación a la Norma Oficial Mexicana NOM-006-SSA2 1993, para la prevención y control de la tuberculosis en la atención primaria a la salud. Diario Oficial de la Federación Oct 31, ( nom/m006ssa23.html; accessed Oct 10, 2003.) 13 WHO/IUATLD/KNCV. Revised international definitions in tuberculosis control. Int J Tuberc Lung Dis 2001; 5: van Embden JDA, Cave MD, Crawford T, et al. Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology. J Clin Microbiol 1993; 31: Yang ZH, Ijaz K, Bates JH, Eisenach KD, Cave MD. Spoligotyping and polymorphic GC-rich repetitive sequence fingerprinting of Mycobacterium tuberculosis strains having few copies of IS1610. JClin Microbiol 2000; 38: Vol 365 April 2, 2005

7 16 Murray M, Alland D. Methodological problems in the molecular epidemiology of tuberculosis. Am J Epidemiol 2002; 155: Rosner B. Fundamentals of biostatistics, 5th edn. Pacific Grove, CA: Duxbury Press; 2000: Dye C, Williams BG, Espinal MA, Raviglione MC. Erasing the world s slow stain: strategies to beat multidrug-resistant pulmonary tuberculosis. Science 2002; 295: Burgos M, DeRiemer K, Small PM, Hopewell PC, Daley CL. Impact of drug resistance on the generation of secondary cases of tuberculosis. J Infect Dis 2003, 188: Mitnick C, Bayona J, Palacios E, et al. Community-based therapy for multidrug-resistant tuberculosis in Lima, Peru. N Engl J Med 2003; 348: Jasmer R, Hahn JA, Small PM, et al. A molecular epidemiologic analysis of tuberculosis trends in San Francisco, Ann Intern Med 1999; 130: Geng E, Kreiswirth B, Driver C, et al. Changes in the transmission of tuberculosis in New York City from 1990 to N Engl J Med 2002; 346: Espinal MA, Kim SJ, Suarez PG, et al. Standard short-course chemotherapy for drug-resistant tuberculosis: treatment outcomes in 6 countries. JAMA 2000; 283: Suarez PG, Floyd K, Portocarrero J, et al. Feasibility and costeffectiveness of standardized second-line drug treatment for chronic tuberculosis patients: a national cohort study in Peru. Lancet 2003; 359: Pablos-Mendez A, Gowda DK, Frieden TR. Controlling multidrugresistant tuberculosis and access to expensive drugs: a rational framework. Bull World Health Organ 2002; 80: Kim JY, Mukherjee JS, Rich ML, Mate K, Bayona J, Becerra MC. From multidrug-resistant tuberculosis to DOTS expansion and beyond: making the most of a paradigm shift. Tuberculosis (Edinb) 2003; 83: Park SK, Lee CM, Heu JP, et al. A retrospective study for the outcome of pulmonary resection in 49 patients with multidrugresistant tuberculosis. Int J Tuberc Lung Dis 2002; 6: Elzinga G, Raviglione MC, Maher D. Scale up: meeting targets in global tuberculosis control. Lancet 2004; 363: Lie R, Emanuel E, Grady C, Wendler D. The standard of care debate: the Declaration of Helsinki versus the international consensus opinion. J Med Ethics 2004; 30: Richardson HS, Belsky L. The ancillary-care responsibilities of medical researchers. Hastings Center Report, Jan Feb 2004, Vol 365 April 2,