Rifamycins are critical to the success of current tuberculosis (TB) chemotherapy regimens. The development of resistance to rifampin (RMP) necessitates the extension of treatment, is associated with poor treatment outcomes [1,2], and may lead sequentially to multidrug-resistant disease (MDR-TB) . Thus, the loss of RMP from treatment regimens has serious implications for individual patients and TB control programs.
Spontaneously occurring resistance-conferring mutations to RMP (more than 95% of which occur in an 81-bp region of rpoB) are less common than to all other first-line anti-TB drugs . Global data on RMP monoresistance (RMR) are scarce, although most national drug resistance surveys have found it to occur in fewer than 1% of TB cases . RMR-TB presents substantial challenges for TB programs, however, and is increasingly identified in high TB incidence settings because of scale-up of rapid molecular tests . Further, RMR-TB is associated with HIV comorbidity, especially in the setting of intermittent TB treatment regimens [7–10], and increasingly identified in high HIV-burden settings [11,12]. Yet, the longitudinal relationship of incident drug-resistant TB (including RMR-TB and MDR-TB) and HIV burden has been infrequently examined on a population level.
California has the largest number of TB cases and the second highest number of persons living with HIV/AIDS in the United States. We describe trends, sociodemographic and clinical characteristics, and treatment outcomes of RMR-TB in California in relation to isoniazid (INH) monoresistant TB (IMR-TB), MDR-TB, and drug-susceptible TB during an era of dynamic change in population-level HIV burden.
The study population included TB cases reported to the California Department of Public Health TB registry between 1 January 1993 and 31 December 2008. We excluded culture-negative cases and culture-positive cases without initial drug susceptibility test (DST) results for at least INH, RMP, ethambutol (EMB), and pyrazinamide (PZA). DST was performed at local laboratories or at the California Microbial Diseases Laboratory using the BACTEC 460 (Becton Dickinson Diagnostic Instruments, Sparks, Maryland, USA), MGIT 960 (Becton Dickinson), or the agar proportion method.
TB patients with HIV coinfection were identified through a state-wide registry match with the California Office of AIDS using Registry Plus Link Plus software , a probabilistic record linkage program developed by the US Centers for Disease Control and Prevention (CDC). Registry cross-match criteria included name, sex, race/ethnicity, date, and place of birth. Manual review of all matched cases was performed, with only those matches above a predetermined priority threshold considered to represent a coinfected case .
Overall state population denominators were based on California Department of Finance estimates . California HIV prevalence estimates were obtained through nonparametric backcalculation of HIV incidence rates based on racial/ethnic-group-specific incidence counts of reported AIDS cases and reported AIDS deaths from 1981 to 2008 (see Supplementary text, http://links.lww.com/QAD/A373).
This analysis was determined to be a public health analysis, which used de-identified and routinely collected surveillance and treatment data, and not subject to human participants review (according to the US Code of Federal Regulations, 45 CFR 46.101) by the Tuberculosis Control Branch, Division of Communicable Disease Control.
A RMR-TB patient was a patient with a Mycobacterium tuberculosis isolate from any anatomic site with resistance to RMP, with documented sensitivity to INH, and without documented resistance to EMB or PZA. Similarly, an IMR-TB patient was a patient with a M. tuberculosis isolate from any site with resistance to INH, with documented sensitivity to RIF, and without documented resistance to EMB or PZA. A MDR-TB patient was a patient with an M. tuberculosis isolate from any site with resistance to at least RMP and INH, regardless of additional drug resistance. We defined a drug-susceptible TB patient as a patient with a M. tuberculosis isolate from any site with documented sensitivity to INH and RMP and no documented resistance to PZA or EMB. We defined acquired drug resistance as an initially drug-susceptible isolate that demonstrated drug resistance at the final reported DST within a single TB treatment course. Primary drug resistance was defined as patients with isolates having drug resistance at the reported initial DST. Among retreatment TB cases, lack of genotypic data precluded accurate differentiation between acquired drug resistance and reinfection with a drug-resistant strain. Because expanded drug resistance in a subsequent TB episode is rare in California , patients with drug resistance noted at initial DST were considered to have ‘primary’ drug resistance, regardless of history of prior TB diagnosis. Patients with both primary and acquired drug resistance were combined to assess trends in drug resistance. Timeframes for comparison were chosen to represent a ‘pre-HAART’ (i.e. 1993–1996, prior to wide availability of HAART) and ‘HAART’ era (i.e. 2005–2008, following the widespread availability of HAART).
Categorical data were analyzed by the χ 2 test, or by calculating prevalence ratios, 95% confidence intervals (CIs), and P values for the comparison of RMR-TB, IMR-TB, and MDR-TB with drug-susceptible TB. Differences in prevalence of binary covariates throughout the study period (1993–2008) were determined using logistic regression with robust standard errors. Associations with absolute mortality (death at diagnosis or at any time following diagnosis) adjusted for covariates based on a priori subject-matter knowledge (including age, sex, race/ethnicity, HIV status, foreign birth, self-administered treatment, and year of report) were examined using generalized linear models with a log link and robust standard errors. Differences in distribution of continuous variables were determined using the Wilcoxon rank-sum test. All analyses were performed with Stata 12.1 (StataCorp., College Station, Texas, USA).
Frequency and trends of drug-resistant tuberculosis in California, 1993–2008
A total of 57 525 cases of TB were reported in California between 1993 and 2008, of which 44 307 (77%) were culture-confirmed. Of these, 42 582 (96%) had first-line DST results available. Characteristics of patients were similar for those with and without culture and DST performed (data not shown). Of isolates with available DST, 178 (0.4%) were RMR, 3469 (8.0%) were IMR, and 635 (1.5%) were MDR (Fig. 1). The proportion of acquired drug resistance was higher among individuals with RMR-TB [18%, n = 18/178; 83% (15/18) HIV-infected] than either IMR-TB [1.0%, n = 35/3469; 14% (5/35) HIV-infected] or MDR-TB [2.8%, n = 18/635; 50% (9/18) HIV-infected; P less than 0.001].
Among 3254 (7.5% of total) culture-confirmed TB patients with HIV comorbidity, 74 (2.3%) had RMR-TB/HIV, 172 (5.3%) had IMR-TB/HIV, and 35 (1.1%) had MDR-TB/HIV. There was a greater decline in the incidence of RMR-TB/HIV from the pre-HAART era to the HAART era (12.0 per 100 000 vs. 0.5 per 100 000) than IMR-TB/HIV (18.9 per 100 000 vs. 8.9 per 100 000; P <0.001), MDR-TB/HIV (3.5 per 100 000 vs. 0.5 per 100 000; P <0.01), or RMR-TB among patients without HIV coinfection (0.038 per 100 000 vs. 0.009 per 100 000; P = 0.02).
The proportion of all TB patients with RMP resistance who had RMR-TB rather than MDR-TB decreased from 31% (95% CI 26–38%) in the pre-HAART era to 11% (95% CI 5–19%) in the HAART era (P <0.001). This proportion decreased at a faster rate among HIV-coinfected patients (−13% per year, 95% CI −18 to −7%) than among those without known HIV coinfection (−6% per year, 95% CI −10 to −2%). Seventy-eight percent (n = 18/23) of RMP-resistant isolates concurrently tested for rifabutin susceptibility were also rifabutin-resistant.
Sociodemographic characteristics of the study population are shown in Table 1. In univariate analysis, patients with primary RMR-TB were more likely than patients with drug-susceptible TB to be of Hispanic ethnicity (prevalence ratio 1.39, 95% CI 1.01–1.90), black race (prevalence ratio 1.76, 95% CI 1.18–2.64), incarcerated within the prior year (prevalence ratio 2.32, 95% CI 1.31–4.11), or injection drug users (prevalence ratio 1.94, 95% CI 1.02–3.68). Patients with primary RMR-TB were less likely than patients with drug-susceptible TB to be foreign-born (prevalence ratio 0.50, 95% CI 0.37–0.69), or Asian (prevalence ratio 0.39 95% CI 0.27–0.57). In contrast, patients with primary MDR-TB were less likely than patients with drug-susceptible TB to be men (prevalence ratio 0.75, 95% CI 0.64–0.88), and more likely to be foreign-born (prevalence ratio 2.52, 95% CI 1.99–3.19) or Asian (prevalence ratio 1.97, 95% CI 1.68–2.31). Patients with IMR-TB had similar sociodemographic characteristics as those with MDR-TB (see Table 1). The majority of RMR-TB patients born outside the United States (57% of total) were born in Mexico (42%), Central America (10%), the Philippines (13%), and mainland south-east Asia (12%). Foreign-born MDR-TB patients most often originated from Mexico (25%), the Philippines (21%), and mainland south-east Asia (26%).
Clinical and treatment characteristics
In univariate analysis, patients with primary RMR-TB were seven times more likely than drug-susceptible TB patients to be coinfected with HIV (prevalence ratio 7.23, 95% CI 5.23–9.99), whereas patients with primary MDR-TB (prevalence ratio 0.54, 95% CI 0.37–0.81) or primary IMR-TB (prevalence ratio 0.63, 95% CI 0.54–0.74) were less likely to be coinfected with HIV (Table 1). Of 160 primary RMR-TB patients, 23 (14.5%) reported a previous diagnosis of TB occurring a median of 3.5 years [interquartile range (IQR) 2.4–6.2 years] prior to the index diagnosis. In comparison, prior TB diagnoses were noted among 186 (30.4%) of primary MDR-TB and 279 (8.2%) of primary IMR-TB patients. Among patients with a documented culture conversion, the median time to conversion was longer for RMR-TB and MDR-TB (94 and 80 days, respectively) than for IMR-TB and drug-susceptible TB (48 and 50 days, respectively). Patients with MDR-TB (prevalence ratio 2.02, 95% CI 1.71–2.40) were more likely to have cavitary lung disease than patients with drug-susceptible TB, a finding not demonstrated among those with RMR-TB (prevalence ratio 1.04, 95% CI 0.70–1.53). Patients with MDR-TB (prevalence ratio 0.30, 95% CI 0.23–0.39) and IMR-TB (prevalence ratio 0.83, 95% CI 0.77–0.90) were less likely to have received self-administered therapy (SAT) throughout their treatment course relative to patients with drug-susceptible TB; patients with RMR-TB had a similar probability as those with drug-susceptible TB to have received SAT throughout their treatment course (prevalence ratio 0.92, 95% CI 0.66–1.27; Table 2).
The percentage of RMR-TB patients with HIV coinfection decreased from 1993 to 2008, whereas the percentage of IMR-TB, MDR-TB, and drug-susceptible-TB patients with known HIV coinfection, and RMR-TB without HIV did not substantially change (P <0.05; Fig. 2). RMR-TB/HIV patients were more likely to have had a prior TB diagnosis (prevalence ratio 3.06, 95% CI 1.48–6.33), to be men (prevalence ratio 1.48, 95% CI 1.07–2.05), and to have died during anti-TB therapy (prevalence ratio 3.03, 95% CI 1.90–4.81) relative to patients with drug-susceptible TB/HIV (See Supplementary Table 1, http://links.lww.com/QAD/A373). Among 74 patients with RMR-TB/HIV (including both primary and known acquired cases), 31 (42%) died while on anti-TB therapy, and two (3%) died prior to initiating treatment.
Treatment outcomes were available for 92% (n = 147/160) of primary RMR-TB, 82% (n = 509/617) of primary MDR-TB, 95% (n = 3249/3434) of primary IMR-TB, and 94% (n = 34 814/37 121) of drug-susceptible TB patients. Among those with complete data, RMR-TB patients and MDR-TB patients were significantly less likely to complete treatment than drug-susceptible patients (prevalence ratio 0.40, 95% CI 0.28–0.55 and prevalence ratio 0.47, 95% CI 0.39–0.57, respectively). Unadjusted mortality during TB treatment was higher for RMR-TB (22.5%, n = 33/147) than MDR-TB (14.1%, n = 72/509; P = 0.01), IMR-TB (6.9%, n = 223/3249; P < 0.001), or drug-susceptible TB (9.2%, n = 3196/34814; P < 0.001). Unadjusted RMR-TB mortality was substantially higher in the pre-HAART (36.2%; n = 34/94) relative to HAART era (12.5%; n = 9/72; P <0.001). In multivariate analysis adjusting for age, sex, race/ethnicity, foreign birth, HIV status, acid-fast bacilli smear status, therapy administration, and year of diagnosis, death during treatment was more likely among patients with RMR-TB (RR 1.94, 95% CI 1.40–2.69) and MDR-TB (RR 2.24 95% CI 1.78–2.82) compared with drug-susceptible TB (Table 3). Death was similar among patients with IMR-TB (RR 1.01, 95% CI 0.88–1.15) and drug-susceptible TB.
We analyzed a population of approximately 38 million people over a 16-year period to describe demographic and clinical trends associated with drug-resistant TB and its intersection with California's HIV epidemic. We found a striking decline of RMR-TB among HIV coinfected patients that was more prominent than the decline of drug-susceptible TB and all other forms of drug-resistant TB. Death was a frequent outcome for patients with RMR-TB, especially prior to the availability of HAART.
Decreasing RMR-TB incidence was temporally associated with population-level improvements in HIV care (e.g. availability of HAART, improved HIV testing/awareness)  and improvements in programmatic TB control [18,19]. TB treatment and control interventions may have differentially influenced RMR-TB over other forms of drug-resistant TB because of increased use of directly observed therapy and daily, rather than intermittent, anti-TB treatment regimens among patients with HIV coinfection . Two trials examining rifapentine (600 mg weekly) and rifabutin (300 mg twice weekly) continuation phase treatment [Tuberculosis Trials Consortium (TBTC)/US Public Health Service Study 22 and Study 23, respectively], recruited patients with TB/HIV comorbidity in California, some of whom acquired rifamycin resistance [8,21,22]. Genotyping data prior to 2003 are unavailable, however, and the contribution of these cases to downstream RMR-TB incidence is unknown. Although particular RMP resistance-conferring mutations predict the likelihood of RMP-resistant/rifabutin-sensitive strains  or low-level RMP resistance , the role of rifabutin or high-dose RMP in clinical management remains unclear. Preliminary results from a randomized trial of high-dose rifapentine demonstrated no occurences of rifamycin monoresistance .
As in previous studies [9,10,21,26], we found a strong association of RMR with HIV coinfection in California. The causal determinants of this relationship include suboptimal pharmokinetics due to intermittent treatment regimens, drug interactions, and/or malabsorption, as well as potentially higher disease burden related to immunosuppression [21,27–29]. That the propensity for acquired RMP resistance is high is supported by our data. Due to sociological factors, such acquired resistance would be preferentially transmitted among persons at high risk for early progression to active disease. Relatively less fit organisms may also propagate or develop compensatory mutations in the setting of severe immunocompromise [3,30,31]. Among patients with HIV coinfection in California, those with RMR-TB differed from those with drug-susceptible disease in being more likely to have had a previous diagnosis of TB, less likely to have been foreign-born or female, and more likely to have had poor treatment outcomes.
The ongoing scale-up of rapid molecular drug-susceptibility testing provides an opportunity to better determine the burden of RMR-TB in diverse settings, including sub-Saharan Africa where rates may be increasing due to expansion of continuation-phase rifamycins and high HIV prevalence [11,12]. GeneXpert MTB/RIF (Cepheid Diagnostics, Sunnyvale, California, USA), an automated nucleic acid amplification assay, has been implemented in 83 low and middle-income countries since WHO endorsement in 2010, identifying more than 4500 additional RMP-resistant cases in 2011 alone . The WHO recommendation for immediate initiation of MDR-TB regimens including INH for patients at moderate or high risk of drug resistance detected as RMP-resistant  is supported by our finding of similar adverse treatment outcome among patients with RMR-TB and MDR-TB . Cohort studies examining clinical outcomes of patients with RMP resistance detected by rapid molecular tests are needed.
Initial RMP resistance as identified by rapid molecular tests is a surrogate marker for MDR-TB in 73–90% of individuals, depending on regional prevalence and pretest probability . In California, the proportion of RMP-resistant INH-susceptible isolates decreased substantially during an era when the MDR-TB prevalence remained low, in particular among those with HIV comorbidity. Population-level changes in the proportion of HIV-infected individuals who have AIDS or are being treated with antiretroviral medications may be an additional consideration in determining the predictive value of Xpert MTB/RIF-detected RMP resistance for MDR-TB, and RMR-TB rates may be an indicator of the success of TB and HIV control activities.
Our study has some potential limitations. HIV infection was determined via a match between California Department of Public Health HIV/AIDS and TB surveillance data; misclassification may have occurred, but likely declined in a time-dependent fashion . Level of immunosuppression (e.g. CD4+ T-lymphocyte count), antiretroviral treatment, and intermittent TB regimen data were unavailable. As any death that occurred in a TB patient was reported, an unknown proportion of deaths included in the analysis were unrelated to TB. As treatment outcome data were incomplete, particularly for those with MDR-TB, we may have differentially underestimated the frequency of death for MDR-TB patients on anti-TB therapy.
In conclusion, in a large, ethnically diverse population in a low TB incidence setting, we found a striking decline in RMR-TB temporally associated with improvements in care of TB-HIV patients and declines in incident AIDS cases. Although poor treatment outcomes, particularly the increased risk of death, suggest potential missed prevention opportunities, the risk of death was high regardless of case management techniques (i.e. use of directly observed therapy) and presence of HIV coinfection. Current investigation of TB deaths on a national level may provide further insight into prevention of these deaths . These findings serve as a reminder that rifamycin-resistant TB requires effective treatment with rigorous monitoring.
The authors would like to thank Gisela Schecter for her insightful comments, as well as the public health departments in California which reported the data used in this analysis.
This work was supported in part by the National Institutes of Health (K23 AI094251 to J.Z.M., R24TW008822 to L.M.P., and R01 AI076476 to P.C.H.) and the Robert Wood Johnson Foundation (AMFDP Medical Faculty Development Award to J.Z.M.).
Conflicts of interest
All authors report no conflicts of interest.
1. Mitchison DA, Nunn AJ. Influence of initial drug resistance on the response to short-course chemotherapy of pulmonary tuberculosis
. Am Rev Respir Dis
2. Espinal MA, Kim SJ, Suarez PG, Kam KM, Khomenko AG, Migliori GB, et al. Standard short-course chemotherapy for drug-resistant tuberculosis: treatment outcomes in 6 countries
3. Bifani P, Mathema B, Kurepina N, Shashkina E, Bertout J, Blanchis AS, et al. The evolution of drug resistance in Mycobacterium tuberculosis: from a mono-rifampin-resistant cluster into increasingly multidrug-resistant variants in an HIV-seropositive population
. J Infect Dis
4. David HL. Probability distribution of drug-resistant mutants in unselected populations of Mycobacterium tuberculosis
. Appl Microbiol
5. WHOAntituberculosis drug resistance in the world 4th global report
. 2008; Geneva, Switzerland:WHO, 152.
7. Ridzon R, Whitney CG, McKenna MT, Taylor JP, Ashkar SH, Nitta AT, et al. Risk factors for rifampin mono-resistant tuberculosis
. Am J Respir Crit Care Med
8. Vernon A, Burman W, Benator D, Khan A, Bozeman L. Acquired rifamycin monoresistance in patients with HIV-related tuberculosis treated with once-weekly rifapentine and isoniazid. Tuberculosis Trials Consortium
9. Munsiff SS, Joseph S, Ebrahimzadeh A, Frieden TR. Rifampin-monoresistant tuberculosis in New York City, 1993–1994
. Clin Infect Dis
10. Sandman L, Schluger NW, Davidow AL, Bonk S. Risk factors for rifampin-monoresistant tuberculosis: a case-control study
. Am J Respir Crit Care Med
11. Mukinda FK, Theron D, van der Spuy GD, Jacobson KR, Roscher M, Streicher EM, et al. Rise in rifampicin-monoresistant tuberculosis in Western Cape, South Africa
. Int J Tuberc Lung Dis
12. Dramowski A, Morsheimer MM, Jordaan AM, Victor TC, Donald PR, Schaaf HS. Rifampicin-monoresistant Mycobacterium tuberculosis disease among children in Cape Town, South Africa
. Int J Tuberc Lung Dis
13. Link Plus Version 2.10 probabilistic record linkage software. Atlanta, GA: Centers for Disease Control and Prevention; 2006.
14. Xia Q, Westenhouse JL, Schultz A, Nonoyama A, Elms W, Wu N, et al. Matching AIDS and tuberculosis registry data to identify AIDS/tuberculosis comorbidity cases in California
. Health Inform J
15. State of California, Department of Finance, Race/Ethnic Population with Age and Sex Detail, 1990–1999 and 2000–2050. Sacramento, CA; 2009.
16. Pascopella L, Deriemer K, Watt JP, Flood JM. When tuberculosis comes back: who develops recurrent tuberculosis in California?
. PLoS One
17. Du P, Camacho F, Zurlo J, Lengerich EJ. Human immunodeficiency virus testing behaviors among US adults: the roles of individual factors, legislative status, and public health resources
. Sex Transm Dis
18. Chaulk CP, Moore-Rice K, Rizzo R, Chaisson RE. Eleven years of community-based directly observed therapy for tuberculosis
19. Blumberg HM, Burman WJ, Chaisson RE, Daley CL, Etkind SC, Friedman LN, et al. American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: treatment of tuberculosis
. Am J Respir Crit Care Med
20. Centers for Disease Control and PreventionAcquired rifamycin resistance in persons with advanced HIV disease being treated for active tuberculosis with intermittent rifamycin-based regimens
. MMWR Morb Mortal Wkly Rep
21. Burman W, Benator D, Vernon A, Khan A, Jones B, Silva C, et al. Acquired rifamycin resistance with twice-weekly treatment of HIV-related tuberculosis
. Am J Respir Crit Care Med
22. Benator D, Bhattacharya M, Bozeman L, Burman W, Cantazaro A, Chaisson R, et al. Rifapentine and isoniazid once a week versus rifampicin and isoniazid twice a week for treatment of drug-susceptible pulmonary tuberculosis in HIV-negative patients: a randomised clinical trial
23. Tan Y, Hu Z, Zhao Y, Cai X, Luo C, Zou C, et al. The beginning of the rpoB gene in addition to the rifampin resistance determination region might be needed for identifying rifampin/rifabutin cross-resistance in multidrug-resistant Mycobacterium tuberculosis isolates from Southern China
. J Clin Microbiol
24. van Ingen J, Aarnoutse R, de Vries G, Boeree MJ, van Soolingen D. Low-level rifampicin-resistant Mycobacterium tuberculosis strains raise a new therapeutic challenge
. Int J Tuberc Lung Dis
25. Jindani A, Hatherill M, Charalambous S, Mungofa S, Zizhou S, van Dijk J, et al.A multicentre randomized clinical trial to evaluate high-dose rifapentine with a quinolone for treatment of pulmonary TB: The RIFAQUIN Trial
. [Oral abstract and paper 147LB]. 20th Conference on Retroviruses and Opportunistic Infections, 3–6 March, Atlanta, GA.
26. Swaminathan S, Narendran G, Venkatesan P, Iliayas S, Santhanakrishnan R, Menon PA, et al. Efficacy of a 6-month versus 9-month intermittent treatment regimen in HIV-infected patients with tuberculosis: a randomized clinical trial
. Am J Respir Crit Care Med
27. Weiner M, Benator D, Burman W, Peloquin CA, Khan A, Vernon A, et al. Association between acquired rifamycin resistance and the pharmacokinetics of rifabutin and isoniazid among patients with HIV and tuberculosis
. Clin Infect Dis
28. Chideya S, Winston CA, Peloquin CA, Bradford WZ, Hopewell PC, Wells CD, et al. Isoniazid, rifampin, ethambutol, and pyrazinamide pharmacokinetics and treatment outcomes among a predominantly HIV-infected cohort of adults with tuberculosis from Botswana
. Clin Infect Dis
29. Gurumurthy P, Ramachandran G, Hemanth Kumar AK, Rajasekaran S, Padmapriyadarsini C, Swaminathan S, et al. Decreased bioavailability of rifampin and other antituberculosis drugs in patients with advanced human immunodeficiency virus disease
. Antimicrob Agents Chemother
30. Gagneux S, Long CD, Small PM, Van T, Schoolnik GK, Bohannan BJ. The competitive cost of antibiotic resistance in Mycobacterium tuberculosis
31. Comas I, Borrell S, Roetzer A, Rose G, Malla B, Kato-Maeda M, et al. Whole-genome sequencing of rifampicin-resistant Mycobacterium tuberculosis strains identifies compensatory mutations in RNA polymerase genes
. Nat Genet
32. Weyer K, Mirzayev F, Migliori G, Van Gemert W, D’Ambrosio L, Zignol M, et al. Rapid molecular TB diagnosis: evidence, policy-making and global implementation of Xpert(R)MTB/RIF
. Eur Respir J
33. Van Rie A, Mellet K, John MA, Scott L, Page-Shipp L, Dansey H, et al. False-positive rifampicin resistance on Xpert(R) MTB/RIF: case report and clinical implications
. Int J Tuberc Lung Dis
34. Smith SE, Kurbatova EV, Cavanaugh JS, Cegielski JP. Global isoniazid resistance patterns in rifampin-resistant and rifampin-susceptible tuberculosis
. Int J Tuberc Lung Dis
35. Centers for Disease Control and Prevention (CDC)Vital signs: HIV testing and diagnosis among adults – United States, 2001–2009
. MMWR Morb Mortal Wkly Rep
36. Beavers S, Flood J, Weinfurter P, Davidow A, Hirsch-Moverman Y, Thickstun P, et al.Tuberculosis mortality in the United States: how can it be prevented?
American Thoracic Society 2012 International Conference. San Francisco, California; 2012.