You are on page 1of 12

new england

journal of medicine
established in 1812 august 7, 2008 vol. 359  no. 6

Comprehensive Treatment of Extensively Drug-Resistant
Carole D. Mitnick, Sc.D., Sonya S. Shin, M.D., Kwonjune J. Seung, M.D., Michael L. Rich, M.D.,
Sidney S. Atwood, B.A., Jennifer J. Furin, M.D., Ph.D., Garrett M. Fitzmaurice, Sc.D., Felix A. Alcantara Viru, M.D.,
Sasha C. Appleton, Sc.M., Jaime N. Bayona, M.D., Cesar A. Bonilla, M.D., Katiuska Chalco, R.N.,
Sharon Choi, M.S., Molly F. Franke, B.A., Hamish S.F. Fraser, M.B., Ch.B., Dalia Guerra, Rocio M. Hurtado, M.D.,
Darius Jazayeri, M.S., Keith Joseph, M.D., Karim Llaro, R.N., Lorena Mestanza, R.N., Joia S. Mukherjee, M.D.,
Maribel Muñoz, R.N., Eda Palacios, R.N., Epifanio Sanchez, M.D., Alexander Sloutsky, Ph.D.,
and Mercedes C. Becerra, Sc.D.


Extensively drug-resistant tuberculosis has been reported in 45 countries, including From Harvard Medical School (C.D.M.,
countries with limited resources and a high burden of tuberculosis. We describe the H.S.F.F., M.C.B.), Brigham and Women’s
Hospital (S.S.S., S.S.A., J.J.F., G.M.F.,
management of extensively drug-resistant tuberculosis and treatment outcomes R.M.H.), Partners in Health (K.J.S., M.L.R.,
among patients who were referred for individualized outpatient therapy in Peru. S.C., D.J., K.J., J.S.M.), the Harvard School
of Public Health (S.C.A., M.F.F.), and the
Methods Massachusetts State Laboratory Institute
A total of 810 patients were referred for free individualized therapy, including drug (A.S.) — all in Boston; and Socios en Salud
(F.A.A.V., K.C., D.G., K.L., L.M., M.M., E.P.,
treatment, resective surgery, adverse-event management, and nutritional and psycho- J.N.B.), the Peruvian Ministry of Health
social support. We tested isolates from 651 patients for extensively drug-resistant (C.A.B.), and Hospital Nacional Sergio E.
tuberculosis and developed regimens that included five or more drugs to which the Bernales (E.S.) — all in Lima, Peru.
infecting isolate was not resistant. N Engl J Med 2008;359:563-74.
Copyright © 2008 Massachusetts Medical Society.
Of the 651 patients tested, 48 (7.4%) had extensively drug-resistant tuberculosis; the
remaining 603 patients had multidrug-resistant tuberculosis. The patients with ex-
tensively drug-resistant tuberculosis had undergone more treatment than the other
patients (mean [±SD] number of regimens, 4.2±1.9 vs. 3.2±1.6; P<0.001) and had
isolates that were resistant to more drugs (number of drugs, 8.4±1.1 vs. 5.3±1.5;
P<0.001). None of the patients with extensively drug-resistant tuberculosis were coin-
fected with the human immunodeficiency virus (HIV). Patients with extensively drug-
resistant tuberculosis received daily, supervised therapy with an average of 5.3±1.3
drugs, including cycloserine, an injectable drug, and a fluoroquinolone. Twenty-nine
of these patients (60.4%) completed treatment or were cured, as compared with 400
patients (66.3%) with multidrug-resistant tuberculosis (P = 0.36).
Extensively drug-resistant tuberculosis can be cured in HIV-negative patients through
outpatient treatment, even in those who have received multiple prior courses of ther-
apy for tuberculosis.

n engl j med 359;6  august 7, 2008 563
The New England Journal of Medicine
Downloaded from on November 21, 2012. For personal use only. No other uses without permission.
Copyright © 2008 Massachusetts Medical Society. All rights reserved.

16 Drug-susceptibility testing was ordered for all Outpatient therapy.10-12 sis.20 Care is provided to patients not Drug-susceptibility testing was performed at the cured by first-line tuberculosis regimens. Africa.19. or amikacin). During the study consists of hospital-based regimens that are cus. Peru. however. because they died before individualized treatment vidualized treatment for multidrug-resistant tu.7. susceptibility test panel included isoniazid. For personal use only. ethambutol. This tients with extensively drug-resistant tuberculosis.19 At the we report the baseline prevalence of extensively start of the enrollment period.21 Drug-susceptibility pro- poor settings. the standard drug- drug-resistant tuberculosis and the characteris. infecting isolates that are resistant to drugs in these classes — now defined as extensively drug.2 Isoniazid and rifampin are the 1999. streptomycin. is associated with a area by a consortium led by the National Tuber- decreased probability of cure. Study Population resistant tuberculosis9 — the probability of cure Our study included 810 patients with tuberculo- is often even lower. 564 n engl j med 359. changes in program policy permitted en- tomized according to the treatment history and rollment of patients who had received fewer prior strain-susceptibility profile. Copyright © 2008 Massachusetts Medical Society.4-6 Treatment regi. rifam­ tics of patients receiving individualized tubercu. 2002. Comprehensive outpa- anchors of standard first-line therapy. irrespective of the of untreatable tuberculosis. second-line injectable drug (kanamycin. with super.14. period. The study was approved by the mens for multidrug-resistant tuberculosis rely on Office for Research Subject Protection at Harvard the most active second-line drug classes — fluo. capreo- vised outpatient treatment of resistant tuberculosis mycin. referred for individualized tuberculosis treatment drug-resistant tuberculosis strains collected on in metropolitan Lima. other reports have raised the specter ment and contact history.3 Resis. No other uses without permission. Written informed consent was obtained from all capreomycin. . Eligibility for treat- drug-resistant tuberculosis is rapidly fatal if not ment was determined on the basis of the treat- treated13.6  www.17. Medical School and the Ministry of Health in Peru.10. one fluoroquinolone. delivered to all eligible patients in the catchment drug-resistant tuberculosis. over. which defines multi. ture and insufficient numbers of trained on November 21. tient treatment. berculosis was introduced in a pilot project in northern Lima in 1996 and subsequently offered Drug-Susceptibility Testing and Treatment nationwide. could be initiated. obstacles to appropriate files were not normally known at the time of re- treatment exist.15 The standard severity of the disease and whether or not the of care for patients with resistant tuberculosis patient had been hospitalized. pyrazinamide. decentral. Here. ferral. The n e w e ng l a n d j o u r na l of m e dic i n e E xtensively drug-resistant tubercu. azid and rifampin. Most had been treated unsuccessfully for the A report from KwaZulu–Natal Province. rifampin. seminal survey found extensively drug-resistant tuberculosis — then defined as Mycobacterium Me thods tuberculosis strains with resistance to at least iso- niazid. roquinolones and injectable agents (amikacin. losis  august 7. 2008 The New England Journal of Medicine Downloaded from nejm. and members of three of six We conducted a retrospective study of patients classes of second-line drugs — in 10% of multi. treatment regimens. All rights reserved. extensively drug-resistant tuberculosis. Seven patients who under- delivered under the aegis of a model. was tance to these two drugs. between February 1. provided free of charge to referred patients. suggested that in patients coinfected with a few had had contact with patients with multi- human immunodeficiency virus (HIV). including inadequate infrastruc.18 Indi. and one gram has been established in Peru. and kanamycin). bility test results for at least four drugs: isoni- comes many of these impediments. culosis Program. In many resource. and July 31. went drug-susceptibility testing were not included ized national tuberculosis program. Inclusion in the study was re- patients through a country’s public health system stricted to patients with baseline drug-suscepti- and delivered by community health workers. free of charge to patients. pin. six continents. Massachusetts State Laboratory Institute.8 For patients with par­ticipants. Such a pro. 2012. We also describe the strategy losis has been reported in 45 countries1 used and the outcomes achieved in treating pa- since it was first described in 2006. South disease (defined as treatment failure or relapse).18.

No other uses without permission. Copyright © 2008 Massachusetts Medical Society. or PAS). initiation of clavulanate. for recurrent disease.6  www. clofazimine. demographic characteristics. presence or days. Added drugs included Patients were classified according to the results those that the patient had received previously but of all baseline drug-susceptibility tests.9 Although subsequent resistance testing n engl j med 359. and any second-line injectable tion was available.nejm. tests for amino­ outpatient care once their condition had stabi- salicylic acid (para-aminosalicylic acid. For personal use only. based on the treatment patients were screened. Data categorized as likely to be effective. sively drug-resistant tuberculosis was defined as Comprehensive treatment included other stan. amoxicillin– lected before. when possible.24 Supplementary Appendix for details on dosing)8.8 of the regimen (defined as the addition or sub- Regimens were constructed with a goal of stitution of two agents that were likely to be ef- prescribing at least five antituberculosis agents fective).org  august 7. All rights reserved. add.22 After surgery. reinforcing on the frequency of adverse events and related strategies were implemented. at minimum: isoniazid. including a fluoroquino­ necessary. Definitions of Terms and Outcomes ing other drugs. Patients rifampin were excluded from analysis.25. which ble 1 in the Supplementary Appendix. was started pending the re. Baseline amikacin. or up to 31 days after. ethiona­ requiring hospitalization were transferred to mide. Nutritional support. lected for smear microscopy and culture. The duration of treatment was at least 18 absence of cavitary and bilateral disease on chest months for oral agents and at least 8 months radiography. and ciprofloxacin. abstraction included previous treatment expo- tient had not received that drug for at least 30 sure. fluoroquinolone. clinically and bacterio- and contact history. group therapy. with the full text of this article at www. sis. effective if all baseline drug-susceptibility tests showed susceptibility to that drug. or both. and rifabutin. logically. Adverse events were managed aggres- lone and an injectable agent (see Table 2 in the sively with the use of standard algorithms. regimens were adjusted as transmission and. according to criteria described to both isoniazid and rifampin but not to both an previously. lowing.20 Hospitaliza.26 After the completion of treatment. lized and they had been discharged. regimen changes were not abstracted. a drug Variables collected through standardized chart was considered likely to be effective if the pa. available was performed at local laboratories. as tients with extensively drug-resistant tuberculo. and oppor- there were no special treatment protocols for pa. Empirical therapy. TREATMENT OF EXTENSIVELY DRUG-RESISTANT TUBERCULOSIS kanamycin. 2008 565 The New England Journal of Medicine Downloaded from nejm. if possible. Data Collection susceptibility test results were available. with HIV (HIV testing was routinely offered at If a regimen did not contain five medications baseline). drug-suscepti- bility testing for patients as required to develop a bility testing was performed. rifampin. Patients whose iso- have questionable activity against multidrug-resis. Patients which to which the patient’s isolate had confirmed in were those performed on sputum specimens col- vitro susceptibility. Exten- continued. cycloserine. (Concentrations and methods used elsewhere. if medically indicated. and limited hospitalization data. .org on November 21. any vised daily ambulatory treatment. capreomycin. A drug was considered likely to be vidualized regimens. Once test detected. 2012. these were extend. This practice was repeated as likely to be effective. needed. and levofloxacin were also routinely screening and ongoing monitoring are detailed performed. often for more than 18 months. ing the duration of treatment with the injectable agent or the duration of the whole regimen. with reinforcement satisfactory regimen. Community health workers super. Multidrug- with localized disease were referred for resective resistant tuberculosis was defined as resistance thoracic surgery. could also be added.19 Monthly sputum samples were col- for drug-susceptibility testing are detailed in Ta. lates were not resistant to at least isoniazid and tant tuberculosis.nejm. If disease was sults of the drug-susceptibility tests. presence or absence of coinfection after culture conversion for the injectable drug. tunities for income generation were provided. By 2001.) In patients without conversion of sputum cul- Clinicians could request additional drug-suscepti. medical therapy was injectable agent and a fluoroquinolone. to restart indi- necessary. laboratory-confirmed resistance to all of the fol- dard elements. efforts were made to reduce household results were available. which the individualized regimen. if no drug. Clarithromycin. ture after 4 months of treatment.

001). 3. 5 to 54). All P values — calculated with the chi-square tary Appendix for details of the regimen for each test for dichotomous variables. and one fluoroqui- teriologic results. was performed in 7 patients with extensive­ ly drug-resistant tuberculosis (14. of the 12 agents or classes for which drug-suscep. The effect of extensively drug-resistant tubercu. sis (90 days vs. at least one injectable agent. had extensively drug-resistant tuberculosis.6%). 2007. Forty-six patients with extensively treatment completion or cure. 90.6  www.nejm.001). Coinfection with HIV was with extensively drug-resistant tuberculosis than rare. Copyright © 2008 Massachusetts Medical Society. multidrug-resistant tuberculosis was iden. drug-resistant tuberculosis whose sputum cul- berculosis. (7. 5. Extensively drug-resistant tu. either previously ad- hazard ratios and with Student’s or Satterthwaite’s ministered or with inconsistent results on drug- t-test for continuous values — were two-sided. hazard ratio. rifabutin (Table 2. for patients with multidrug-resistant tuberculo- resistant tuberculosis (1. or clofazimine). A median of 99. Pa. All rights reserved. In two of these patients. the drugs at baseline.00) (Table 1). 1). were often included.6%) and in 87 R e sult s patients with multidrug-resistant tuberculosis (14. None of the tures were performed (interquartile range. clarithromycin. Three patients with extensively drug- Of 810 patients receiving individualized regimens. including tient follow-up ended (and data were censored) on hospitalization for surgical resection (Table 3). September  august 7. and patient). tures remained positive after 4 months of treat- Patients with extensively drug-resistant tuber.27 bacterial agents for which either susceptibility had Treatment default was a physician-defined end been documented or the duration of prior expo- point assigned upon the failure of attempts to sure had not exceeded 1 month and susceptibility return to therapy those patients who had not been had not been tested. both had poor outcomes. the me- product-limit method. clarithromycin.4%). treatment failure.12. as an adjunctive treatment for tubercu- treatment.2±1. and the addition of regimens.1 tive to a negative culture was longer for patients vs. odd ratios. Adjustments in the regimen included before the study than had patients with multi. susceptibility tests.9 on November 21. 0. All patients with extensively adhering to their treatment regimen. losis. 2012. The median time to conversion from a posi- tibility testing was routinely performed (8.6. resistant tuberculosis had positive cultures at the 651 had isolates that were tested for the requisite time of surgery. disease was defined as two or more positive bac. for data anal. Recurrent drug-resistant tuberculosis received cycloserine.8%). for patients still receiving Surgery.6% of expected monthly cul- tified in the remaining 603 (92. The n e w e ng l a n d j o u r na l of m e dic i n e was performed in cases of nonresponse to treat.4±1. or reinitiation of therapy. days (interquartile range. The of one or more other agents (amoxicillin–clavu- former group also had isolates resistant to more lanate. For personal use only. 61 days.28 Data were censored when dian cumulative duration of hospital care was 14 an outcome other than death was recorded.5. and death were used. 2008 The New England Journal of Medicine Downloaded from nejm. and Among 16 patients (33. after nolone (Fig. occurring in nine patients with multidrug. see Table 3 in the Supplemen- ysis.4%).2±1.3%) apy and outcomes at the end of treatment. patients were not reclassified according to sis (P = 1.63. Eleven patients with extensively drug-resistant losis on time to an end point (culture conversion. and We used SAS software. these results. or death) was estimated with the use of the points during individualized treatment.3%) with extensively all the others (85. . culosis received an average of 5. 95% 566 n engl j med 359. No other uses without permission. P<0.3 antimyco- pletion.3±1. Additional drugs. 4. ment. tuberculosis were hospitalized at one or more cure.7%) had multidrug-resistant tu. version 9. sputum culture had remained positive during in- berculosis was identified in 48 of these patients dividualized treatment.9%) received two or more of the following agents: amoxicillin– Statistical Analysis clavulanate.5%) and none of the pa. P<0. one of these patients (14. clofazimine. a drug-resistant tuberculosis (mean [±SD] number change in the injectable agent.3±1. drug-resistant tubercu­lo­sis (97.9 to seven patients who died before treatment were 100). Patients with extensively drug-resistant tuber- Standard definitions for cure. regimen reinforcement was possible in only culosis had undergone more treatment regimens 3 (18. treatment com. tients with extensively drug-resistant tuberculo- ment. a change to a later-generation fluoroquinolone. permitting evaluation of response to ther- infected with HIV.

levofloxacin. one in each group. Distribution of Baseline Patient Characteristics According to Resistance Profile./total no.3) 26/603 (4. never exposed No. PAS.47 Demographic data Female sex — no.54 Age — yr 32. All rights reserved.0) <0.2±1.4 0. of agents to which baseline isolate was resistant. Asterisks indicate that some testing was performed for these agents. as well as the lack of standardization or confirmed clinical relevance of tests for these drugs.8) 315/573 (55. 42/48 (87.nejm. † P values were calculated with Student’s or Satterthwaite’s t-test for continuous variables.2±1. clinicians relied less on these results than on those for other drugs. (%) 26/45 (57. Downloaded from nejm. of previous treatment regimens‡ 4. ethambutol. and later-generation fluoroquinolones (gatifloxacin. kanamycin or amikacin. with Fisher’s exact test for HIV infection and hospitalization at treatment initiation. of Regimens 30 20 10 0 zin ol Am e in in in in cin cin cin in in e* e id * n* e* tin id id te t ac ac yc yc yc ac Cy xac rin in i bu bu xa xa xa am m na yc om m m ik ox im ic a se m fa lo flo flo flo m la na eo n l ifl az icy of Ri ha ro clo vu pt Am thio O vo ar ox pr Ka pr of ra ith re Et al cla Sp Le Ca M Cl Py Ci St E os ar n– in Cl lli ici ox Am Figure 1. ‡ Only two patients. first-generation fluoroquinolones (cipro- floxacin.5 <0. exposed >1 mo No DST available. Drug-­ Susceptibility Testing.001 Previous treatment (>1 mo) with a fluoroquinolone and an inject. had not previously received treatment for tuberculosis. ofloxacin).4) 241/603 (40.001 Cumulative months of previous treatment 34. (%) 17/48 (35.7 25. of possible 12§ 8. 2012.6  www.0) 0. pyrazinamide. iso- niazid. moxifloxacin).7±23.3) 0. TREATMENT OF EXTENSIVELY DRUG-RESISTANT TUBERCULOSIS Table 1. ICM AUTHOR: Mitnick RETAKE 1st FIGURE: 1 of 1 2nd REG F 3rd n engl CASE j med 359.9 3. Copyright © 2008 Massachusetts AUTHOR.72 Hospitalized at treatment initiation — no.5) 1.76 * Plus–minus values are means ±SD.6 <0.0±9.0) 0./total no. cavitary findings — no.001 No./total no. 50 Resistant No DST available. Figure has been redrawn and type has been Combo No other uses without permission. cycloserine. (%) 3/48 (6. However. 2008 Revised 567 EMail Line 4-C SIZE The New ARTIST: ts England H/T Journal H/T of Medicine 33p9 Enonon November 21. Some susceptibility testing was performed for these agents.* XDR Tuberculosis MDR Tuberculosis Characteristic (N = 48) (N = 603) P Value† Resistance and prior exposure No. . PLEASEMedical NOTE:Society.4±1. For personal use only. and Prior Exposure to a Particular Agent.5) 378/600 (63.1 5./total no. Use of Antituberculosis Agents in 48 Individualized Treatment Regimens for MDR Tuberculosis. because of the relative infrequency of testing. and with the chi-square test for all other baseline characteristics.3±1. rifampicin. HIV human immunodeficiency virus. (%) 0/48 9/587 (  august 7.001 able agent (other than streptomycin) — no. (%) Clinical data HIV infection — no. MDR multidrug-resistant tuberculosis. § Resistance to the following 12 drugs or drug classes was tested: capreomycin. DST denotes drug-susceptibility test.1±16. streptomycin.00 Bilateral. ethionamide.5±12. and XDR extensively drug-resistant tuberculosis. exposed >1 mo Susceptible.9 31./total no.6 <0. never exposed 40 Susceptible.

5 (8. No other uses without permission. The n e w e ng l a n d j o u r na l of m e dic i n e Table 2.0 Duration of treatment with injectable agents — mo Median 15.2) Duration of treatment (mo) — median (interquartile range) 12.9 Interquartile range 13.6  www.7 No. of patients — % 6 (12. For personal use only. — % 16 (34.4) First-line injectable agent Streptomycin No.9) Duration of treatment (mo) — median (interquartile range) 24. of patients — % 10 (21.8–18.2–9. of patients — % 7 (14. of drugs in regimen (no.0–28.4 Interquartile range 11.7 (10. 2012.0–25. All rights reserved.6–13.7) 568 n engl j med 359.0) Duration of treatment (mo) — median (interquartile range) 24.3–29.1 (13.nejm.0 (4.4 (14. of patients — % 8 (  august 7.3) Duration of treatment (mo) — median (interquartile range) 19.1) Second-line injectable agents Amikacin No.1) Capreomycin No.9) Ofloxacin No.0–26.5) Kanamycin No. without documented resistance or prior exposure for >1 mo) Among all available agents 5. of patients — % 9 (19.0) First-generation fluoroquinolones Ciprofloxacin No.6) Pyrazinamide No.5 (8.8–18.4–19.2±1.1) Duration of treatment (mo) — median (interquartile range) 14. of patients — % 18 (38.3) Duration of treatment (mo) — median (interquartile range) 8.8) Duration of treatment (mo) — median (interquartile range) 9.2 Drugs included First-line oral agents Ethambutol No.0 (3.3 Among 12 agents or classes for which routine DST was performed† 3. Individualized Regimens Prescribed for 47 Patients with XDR Tuberculosis.* Regimen Value Characteristics Duration of treatment — mo Median 24. of patients — % 25 (53.3–13.3±1.9 (6. Copyright © 2008 Massachusetts Medical on November 21. .0) Duration of treatment (mo) — median (interquartile range) 9. 2008 The New England Journal of Medicine Downloaded from nejm.

of patients — % 45 (95.8 (8.9) Duration of treatment (mo) — median (interquartile range) 24.6–18.5 (11. All rights reserved. moxifloxacin).6  www. kanamycin or amikacin. 2012.nejm. first-generation fluoroquinolones (cip- rofloxacin.1 (  august 7.3) Duration of treatment (mo) — median (interquartile range) 21.5–27. n engl j med 359. of patients — % 47 (100) Duration of treatment (mo) — median (interquartile range) 24. No other uses without permission.9) Other second-line oral agents Cycloserine No.2 (8. For personal use only. levofloxacin. Copyright © 2008 Massachusetts Medical Society. of patients — % 21 (44. ofloxacin). cycloserine. of patients — % 31 (66.1) Duration of treatment (mo) 12.7) * One patient died less than 1 month after starting treatment. DST denotes drug-susceptibility test.9) PAS No. isoniazid.7) Duration of treatment (mo) — median (interquartile range) on November 21.5–14.7–27.1 Moxifloxacin No.0) Sparfloxacin No. of patients — % 20 (42. ethambutol. of patients — % 46 (97. Plus–minus values are means ±SD. 2008 569 The New England Journal of Medicine Downloaded from nejm.) Characteristic Value Later-generation fluoroquinolones Levofloxacin No. streptomycin. TREATMENT OF EXTENSIVELY DRUG-RESISTANT TUBERCULOSIS Table 2.6) Duration of treatment (mo) — median (interquartile range) 12. and later-generation fluoroquinolones (gatifloxacin. before regimen data were recorded.4–23.8–27. pyrazinamide.0) Duration of treatment (mo) — median (interquartile range) 12.9) Clofazimine No. of patients — % 8 (17.9) Rifabutin No. of patients — % 1 (2. (Continued. † Resistance to the following 12 drugs or drug classes was tested: capreomycin.3) Other agents Amoxicillin–clavulanate No. rifampicin.9 (12.7–28.1) Ethionamide No. of patients — % 34 (72.0 (10.7 (12.1 (7. .7) Duration of treatment (mo) — median (interquartile range) 24.0) Duration of treatment (mo) — median (interquartile range) 24. and XDR extensively drug-resistant tuberculosis.1) Clarithromycin No.7–28. of patients — % 47 (100) Duration of treatment (mo) — median (interquartile range) 24.7–28. PAS. ethionamide.0 (11.

No. 2008.4) age.7 to 27. mycin. were inconsistent.2 (25.36). The hazard ratio for death among pa.09 (95% CI. at the highest toler- Patients with positive sputum culture at surgery — no. of 10 µg per milliliter was documented.02. 2012./total no. yet the frequency of cure or Patients hospitalized during individualized treatment regi.6) parison population — patients with multidrug- ized treatment regimen — no. as compared with those with multidrug-resis.29.6  www. and rifabu- 570 n engl j med 359.9%) with extensively drug-resistant tuberculo. The com- Patients undergoing surgical resection during individual. The n e w e ng l a n d j o u r na l of m e dic i n e delayed by nearly 1 month in the study patients Table 3. median of less than 6 months in other studies Positive bacteriologic results were reported af. whenever possible. Type of surgery — no. was  august 7. in which the duration of use of injectables was ter cure or completion of treatment in 2 patients reported.1) resistant disease were concurrently applied to all No. The Patients undergoing surgery who subsequently died or 2 (28. of days of hospitalization during individualized treat. study the injectable agent was delivered for a P = 0.45 to 0. — median (interquartile range) lasting more than 2 years in most patients. Treatment was protracted. Moxifloxacin and levofloxacin were com- sis and 15 patients (3. For personal use only. Twenty-nine patients (60. In this tant tuberculosis.3%) with multidrug-resistant tuberculosis drug-resistant tuberculosis (52%). This practice reflects the 10. Causes of death and reasons for default median of nearly 15 months as compared with a were not available. which may explain the attenuated difference.6 (7.1–57.32-34 This study shows that an aggressive.59 to 2.nejm. The median duration — even in patients with isolates that were resis- of follow up was 19. aggressive regi- median (interquartile range) mens — with many drugs.8%) with multidrug-resis.1) patients in this program. of months of treatment for patients undergoing surgery 31. . when necessary.9) ated doses — were used to maximize the chemo- No. despite resistance to ciprofloxacin. PAS. tant to ciprofloxacin. and with amoxicillin–clavu- tuberculosis treatments. of months from treatment initiation to surgery — 11.1–24. capreomycin. as compared with 400 patients was used in 25 of the patients with extensively (66. Culture conversion was lanate.4%) with extensively Since resistance to more than one aminogly- drug-resistant tuberculosis completed treatment coside was common. 14 (5–54) The sample size afforded 80% power to detect ment regimen — median (interquartile range) differences in risk of more than on November 21. Regimens relied heavily on three agents with little prior use in Peru: capreo- confidence interval [CI].30 (6. Copyright © 2008 Massachusetts Medical Society. Hospitalization and Resective Thoracic Surgery among Patients with XDR Tuberculosis.8. was another alternative.0). or were cured. and cycloserine. No other uses without permission. (%) prior treatment exposure. monly included in the individualized regimens tant tuberculosis (P = 0. ment of multidrug-resistant tuberculosis7. 11/48 (22.1) Several principles of management of highly Cavitary resection 1 (2. 0./total no. First. clofazimine.4 months (interquartile range. 0.31 and is supported by evidence that the efficacy of Discussion later-generation fluoroquinolones may be pre- served. All rights reserved. comprehen. 2008 The New England Journal of Medicine Downloaded from nejm. (%) to design (and adjust) regimens containing at least five drugs that were likely to be effective * XDR denotes extensively drug-resistant. a polypeptide. (%) nificantly between the two groups of patients. Data on treatment reinitiation were importance of using fluoroquinolones in the treat- unavailable as of March 17.79). clarithromycin. Pneumonectomy 1 (2. Finally.9) therapeutic benefit. (%) 3 (42. Streptomycin (P = 0. susceptibility to streptomycin at a concentration sis.40). clinicians reinforced sive management program can cure more than regimens with pyrazinamide and ethambutol 60% of patients with extensively drug-resistant (despite extensive prior exposure to these drugs) tuberculosis who are not infected with HIV but with drugs for which susceptibility test results who have received numerous unsuccessful anti. it was prescribed when tients with extensively drug-resistant tuberculo.89]) (Table 4). (%) resistant tuberculosis — had extensive resistance.* who had extensively drug-resistant tuberculosis as compared with those who had multidrug-re- Characteristic Value sistant tuberculosis.9) relapse and the risk of death did not differ sig- men — no. and parenchymal dam- Lobectomy 5 (10. 7/48 (14.6) results of drug-susceptibility testing were used whose treatment failed — no.

fewer than five cultures were grown in the final 12 months of therapy). XDR Tuberculosis MDR Tuberculosis Outcome (N = 48) (N = 603) Effect Estimate and P Value* Response at end of treatment Good outcome — on November 21. These ment default. either by increasing Third. to the strategy. Monthly monitoring permitted calized disease.4) 400 (66. was severely restrict­ tuberculosis (39. and Korea. which were managed only three patients with extensively drug-resis. the United States.42. 1. P = 0. however. however. was frequent. The odds ratio (OR) and the hazard ratios (HR) are unadjusted. aggressively by nurses and clinicians. Since patients for whom ed by extensive resistance and prior exposure. 2012. of baseline drug-susceptibility testing. and lack of initial response. relatively lo.6%) than did not have a response to treatment.89. P values for the OR and the HRs were calculated with the use of the chi-square test. Workers ensured a high level of sibility that additional toxic effects induced by treatment adherence and promptly identified cir- this strategy increased the probability of treat. Poor outcomes among patients tion. cumstances requiring additional attention. four transferred out of the program and one remained in treatment.2) 62 (10.2) HR.3) Cured 29 (60.5–25. 0. 95% CI. repeated drug-susceptibility tests were per- with extensively drug-resistant tuberculosis who formed and regimens adjusted for patients who had surgery were less frequent (28. No other uses without permission.4) 395 (65. of months to cure 26.4) 13 (2. even early identification of patients with no response.6%). ‡ Treatment default was defined as the failure of attempts to return to therapy patients who were not adhering to their treatment regimens.45–0. we cannot junctive therapies were rejected for lack of evi­ exclude the possibility that one or more contrib. at health centers.1) Died 11 (22. frequent contact with health care work­ the regimen’s activity or by providing protection ers afforded many benefits.35-38 Although evidence of the efficacy of although not recorded. dence. The ability among all patients with extensively drug-resistant to adjust regimens.72–2. supervised against the emergence of resistance to more treatment was delivered in patients’ homes and active agents. 0.0) 5 (0.9) 123 (20. all of whom had MDR tuberculosis. n engl j med 359. in patients with restricted lung volume.6) Completed† 0 (0. All rights reserved. For personal use only.3) Treatment failed§ 5 (10. additional toxicity is social needs were also assessed continuously and an unlikely influence. 95% CI.008 No. With the de­fault from treatment of included adverse events.33 * Effect estimates are for the group of patients with extensively drug-resistant (XDR) tuberculosis as compared with the group that had multidrug-­resistant (MDR) tuberculosis. 0.8) 24.24 Other ad- these approaches is limited at best. we cannot draw a causal were better than most outcomes reported from inference about its effect..39 We also cannot exclude the pos.26 Second. Psycho- tant tuberculosis. tin. TREATMENT OF EXTENSIVELY DRUG-RESISTANT TUBERCULOSIS Table 4. § Treatment was considered to fail for those patients who had two or more positive cultures among the five cultures recorded in the final 12 months of the study or for whom any one of the final three cultures was positive.22 Medi. In addi- ceiving surgery. Outcomes were not available for five patients.21.4) Time to interim response and to response at end of treatment — median (95% CI) No. however.0 (24. P = 0. .56– Defaulted‡ 3 (6.6  www.8) OR.8) Poor outcome — no. Individualized regimen design relied on the results cal treatment was prolonged among patients re.nejm.32. treatment failed or who died were not eligible It is noteworthy that the outcomes in our study for surgery. (%) 19 (39. 0. † Patients who completed treatment are defined as those who finished treatment according to protocol but who did not meet the definition for cure or treatment failure owing to lack of bacteriologic results (i. Copyright © 2008 Massachusetts Medical Society.6–27.8 (24. resective surgery was indicated for Fourth. Corticosteroid use. 0. hospitals in  august 7. bacteriologic assessment was integral patients with high-grade resistance. of days to culture conversion 90 (57–115) 61 (59–67) HR. Daily. 95% CI.40-44 uted to treatment success. 2008 571 The New England Journal of Medicine Downloaded from nejm.e. addressed with a range of interventions. Response and Time to Response According to Type of Resistance at Beginning of Individualized Treatment Regimen. (%) 29 (60.6) 198 (32. P = 0.

and cin as part of a 12-month retreatment regimen. produced better results than the regimen intro- Unlike most of the patients in the KwaZulu– duced in 1997. Subsequently. revealed that severe adverse events were mutations during sequential exposure to inade. for tuberculosis. The hospitalized patients in KwaZulu– included in the standardized five-drug regimen Natal13 had advanced acquired immunodeficiency introduced by the program in October 1997 for syndrome. ed progressive acquisition of drug-resistance however. This resistance prob.45. kanamycin and ciprofloxacin were line drugs. and their effect on completion sively drug-resistant tuberculosis probably reflect­ of the regimen. the National tuberculosis was defined as M. Indeed. ever. vidualized regimens. treatment outcomes in the absence of a compari- 572 n engl j med 359. private practitioners prescribed aminogly- patients with extensively drug-resistant tubercu­ cosides and fluoroquinolones for tuberculosis losis. standardized regimen for multidrug-resistant tu- verely ill patients.47 In Peru. of adverse events. For personal use  august 7. how- agents in uncontrolled situations or as part of in. The study had long-standing tuberculosis. be mediated through tuberculosis had never received prior treatment careful use of these agents. exten. prior work. members of three of the six classes of second. extensively drug-resistant tuberculosis after hav- tion for spontaneously occurring resistant mu­ ing received multiple unsuccessful treatments. without controls) and the ted to a hospital. For se. extensively drug-resistant tant tuberculosis. disease. They were therefore This regimen also included ethambutol and py­ neither expected to have nor treated for resistant razinamide. berculosis is inevitable when injectable agents The patients in our study received therapy for and fluoroquinolones are used. more than half had never received the treatment of multidrug-resistant tuberculosis.6  www. in the patients in Peru. to assess the effect of a survivor bias on adequate regimens. Less than 20% design (an observational study involving indi- of the study patients had previously been admit. only one patient with extensively drug-resistant tance emerges can. In the mid-1990s. hospitalization may be berculosis that contained at least four drugs required. however. almost 90% of the patients with exten- of the KwaZulu–Natal patients highlights the sively drug-resistant tuberculosis in our study had benefit of avoiding nosocomial transmission by prior exposure to these agents. in these situations.49 Natal study. No other uses without permission. All rights reserved. however. treatment for tuberculosis. part of treatment regimens for multidrug-resis- In this study. The large number of patients who died Of 466 patients treated with that regimen — after a short interval in this study underscores which consequently contained only two or three the importance of timely diagnosis of resistant new drugs — 48% were cured. Future received prior treatment.18 The use of ka- tuberculosis and initiation of effective antituber. nents of the approach on outcomes. rifampin. which patients had already received. in before these agents were formally introduced as which cure was achieved in 61% of such patients.10-12 One exception was a study in Latvia. These patients may therefore ably reflects the danger of using second-line represent a survival cohort. were not available. It is impossible. namycin and ciprofloxacin in subcurative regi- culosis and antiretroviral therapy in patients who mens probably contributed substantially to the are also coinfected with HIV. adequate infection categorized as likely to be effective may have control is essential to protect staff and patients. tuberculosis isolates Tuberculosis Program in Peru included kanamy- that were resistant to isoniazid. The detection of a development of extensively drug-resistant tuber- common tuberculosis strain in a large proportion culosis.51 Also unavailable were tensively drug-resistant tuberculosis had received data on other management approaches for ex- even more prior therapy than had the patients tensively drug-resistant tuberculosis in Peru with with multidrug-resistant tuberculosis who had which our approach could be compared. most of the patients in our study Our study has several on November 21. Copyright © 2008 Massachusetts Medical Society. Data on the frequency this group. .50.48. The n e w e ng l a n d j o u r na l of m e dic i n e where cure was achieved in fewer than half of tries. 2008 The New England Journal of Medicine Downloaded from nejm.46 The speed and extent to which resis. tants. uncommon and rarely resulted in discontinua- quate treatment regimens. It is therefore improbable that small number of patients preclude identification nosocomial transmission of highly resistant of patient or treatment characteristics that had a strains of tuberculosis accounted for disease in causal effect on outcomes. because of selec. however. the patients with ex. work will estimate the effect of various compo- Development of extensively drug-resistant tu. tion of the regimen. Prescription of a delivering treatment in the community. 2012. as in many coun.nejm.

our results tensively drug-resistant tuberculosis. No other uses without permission. Fox W. Kim comes in 6 countries. not decreased. Chauhan LS. Padayatchi N. Masjedi MR.) 2007. the Francis Family Foundation. Espinal MA. treatment for extensively drug-resistant tubercu. Sotomayor A. Revised definition of extensively drug. standardised second-line drug treatment undertaken by the British Medical Re. or complacency. to Dr. Sorooch S. Med 2004. et XDR-TB in South Africa: no time for denial World Health Organization. References 1. Iademarco MF. Lancet 2002. (Report al. normas y procedimientos 6. Since it is common under program gram. As part ment awards from the National Institute of Allergy and Infec- of future efforts. et al. Dis 2006.52 Nevertheless. reported. ond-line drugs — worldwide.) ment of multidrug-resistant tuberculosis.5:575-8. 2 years of surveillance in Iran. aggressive regimens. tal Wkly Rep 2006. Geneva: World Health Organi- Worldwide emergence of extensively drug. of South Africa. Shah NS. Soc Sci al. Feasibility and cost-effectiveness of Studies on the treatment of tuberculosis culosis with extensive resistance to sec. Martin Hirsch and Megan Murray for com- ments on an earlier version of the manuscript. nity health care workers who participated in this effort.276: Assoc 2007. 2001. MMWR Morb Mortal Wkly Rep 2006. the David Rockefeller Center for Latin American ceptions include patients with HIV coinfection Studies at Harvard University. Thomas J.43:841-7. patients with extensively prehensive outpatient therapeutic approach. and those who are initially infected with a strain the Pittsfield Anti-tuberculosis Association. ceiving community-based treatment in A multi-institutional outbreak of highly 15. . and commu- in Peru. Moll A. 1946. Emergence of Mycobacterium tuber. Ex. Girardi E. Surgery for patients with drug-resis- Tuberc Lung Dis 2001. Socci AR. First-line tuberculosis therapy and culosis. 10. Migliori GB. of Health. Singh JA. (Report no. Emerg Infect Dis resistant tuberculosis.6  www.55: al cohort study in Peru. with relevant subsequent publi­ 301-5. our study underscores the conditions that therapy for multidrug-resistant importance of developing other interventions tuberculosis or extensively drug-resistant tuber. The WHO/IUATLD Global Project on drug-resistant tuberculosis: epidemiology ahead to future challenges. as culosis is reserved for patients who have received well as stem the development and spread of ex- repeated treatments for tuberculosis. 2001. Mitnick) and the ment what will probably be the superior effect of National Heart. Somocurcio JG. TREATMENT OF EXTENSIVELY DRUG-RESISTANT TUBERCULOSIS son group — that is. 20. Global tuberculosis control: WHO re- 2. a cause of death in patients co-infected para el control de la tuberculosis en el Tung on November 21. 12. Frieden TR. Farmer P. 2000–2004. et al. J Indian Med Anti-Tuberculosis Drug Resistance Surveil. N Engl Standard short-course chemotherapy for definition. Suppl 2:S231-S279. and all the patients. zation. et al. We thank Drs. For personal use only. Int J Extensively drug-resistant tuberculosis: et al.363:474-81. with the WHO standard retreatment regi. Upshur R. PLoS Med 2007. no. Clinical and operational value of the Community-based therapy for multidrug- 4. and the Hatch Family Foundation and by career develop- of extensively drug-resistant tuberculosis. the Eli Lilly Founda- tion. Bayona J. the world: fourth global report. drug-resistant tuberculosis after treatment 80. treatment in Peru’s National Tuberculosis Pro- risk of death. RNTCP 2007: looking Lima. Mukherjee JS. et al.394. as part of a com. 2001. Int J Tuberc Lung Dis 1999. Mathieu C. Kim HJ. to Dr. J Med 2003. Palacios E. Suárez PG. 18. 13. Shin S. Shin S. Eur Respir J 2007. 14. Wright A. cured drug-resistant tuberculosis who had not received more than half of patients with extensively drug- prior treatment. Hwang SS. that will further improve treatment outcomes. Lung. Peru. Lancet 2006. port 2001. MMWR Morb Mor. 16. Peru.53-58 can be generalized to many patient populations Supported by grants from the Bill and Melinda Gates Founda- tion. One relevant comparison can be resistant tuberculosis who had previously been made between those with more and those with treated for tuberculosis. A case series: with tuberculosis and HIV in a rural area Perú. despite the limited resources research assistance. Quy HT. Mitnick C. 21. Sherman LF. 196. Lima. Copyright © 2008 Massachusetts Medical Society. Salud de las Personas. Ministerio de Salud. Lancet 1999. 2008 573 The New England Journal of Medicine Downloaded from nejm. No potential conflict of interest relevant to this article was losis in patients who have not had prior treat. lance. families. Kim HR. Binkin NJ. extensively drug-resistant tuberculosis resistant tuberculosis in Lima.287. Rich ML. Programmes and principles in treat. 194.3: 11. White. All rights reserved. Coninx R.30:623-6. 9. Eva Tomczyk for In conclusion.13:380-7. Mate K. Becerra). Perú: Dirección General de initial outcome of persons with multi. Debacker M.45:1290-5. et al.368:1575. Suarez PG. 3. men in Ho Chi Minh City. Actualización de sis in prisons.4(1):e50. Peru: 7 years of experience. Extensively drug-resistant tuberculosis as la doctrina. Ellard GA.105:192. Community-based treatment 2537-45. This strategy has now less exposure to prior treatment: more exposure been integrated into the routine approach to is associated with an increased. Anti-tuberculosis drug resistance in 1229-35. et al. Vietnam.283: Impact of extensive drug resistance on  august 7. ment for tuberculosis.359: 1986. al. Farnia P. WHO/CDS/TB/ resistant tuberculosis. Portocarrero J. Clin Infect Dis 2007. cations. et al.nejm.353:969-73. Lancet middle-income countries. and clinical outcomes. Bai GH. JAMA 1996. Sturm AW. 17. Clin Infect tant tuberculosis: report of 121 cases re- 7. 2012. Geneva: 8. Besozzi G. Cĵ NV. Lima. the Peruvian Ministry treated in low. treatment outcomes in non-HIV-infected of multidrug-resistant tuberculosis in 5. Lan NTN. 1980-9. Kim SJ. et patients with multidrug-resistant tuber. for chronic tuberculosis patients: a nation- search Council tuberculosis units. drug-resistant tuberculosis: treatment out. drug-resistant Mycobacterium tuberculo. tious Diseases (5 K01 A1065836. Floyd K. Maw KL.62:416-21. Thorax 2007. Ministerio de Salud. et 19. n engl j med 359. JAMA 2000. Furin J. et al.348:119-28. WHO/HTM/TB/2008. Gandhi NR. Bayona J. it will be important to docu. and Blood Institute (5 K01 HL080939. 22. 2008. Partners in Health. Mitchison DA.59:1529-39.

monary disease caused by multidrug-­ tant Mycobacterium tuberculosis: epide- bial exposure.45:1415-6. RS. drug-resistant and extensively drug-resis- mutations and relationship to antimicro. Bishai WR. et al. Antimicrob Agents Chemother 50. et al. Speaking the same language: treat. Int J Tuberc Lung vage therapy for cases of chronic multi.117:744-51. extensively drug-resistant F15/LAM4/KZN A DOTS-Plus handbook: guide to the monary tuberculosis. The PIH guide to the hara S. Psy. Antimicrobial activities of levo. Madsen LA. Iseman MD.45:1409-14.9:175-80.53:253-60. et al. tals: an epidemiological modelling study. Dis 2005. Mitchison DA. Cirillo D. drug-resistant pulmonary tuberculosis in 51. Lee CN. Int J to patients with MDR-TB. Prevention of nosocomial transmis- sistance. Outcome recommended treatment regimens.45: study. 2:877-84. Grosset JH. Chan CK. apy for multidrug-resistant tuberculosis. Int J 49. tant enzymes carrying mutations at posi. Leimane V. Shin SS. Girard E. Raviglione MC. Antimicrob 46. 52. Peru. ed. J Antimicrob Chemother resistant mycobacteria. Preliminary results of Centis R. Newell JD. Sofer C. Jouveshomme S. Am J Respir Crit miology and control.370:1500-7. 55. culture. Extensively drug-resistant treatment of multidrug-resistant tubercu. 32. 45. TREATMENT OF EXTENSIVELY DRUG-RESISTANT TUBERCULOSIS 23. patients with sputum smear-positive pul. Veziris N. medical management of multidrug-resis. Venter A. Poolman EM. Holtz TH. sis drug development pipeline: progress Outcomes of patients with multidrug. Farnia P. Multi- tuberculosis isolates: associated genetic collapse therapy with plombage for pul. vaccae immunotherapy for treating tuber. 27. win would understand. et al. 41. Mestanza L. Dis 2007. Chalco K. Taiwan.252:91-106.6  www. Leimane V. Colorimetric detection of multidrug- sistant to fluoroquinolones. et al. et al. .8:749-59. Int Nurs Rev bacterium tuberculosis in whole-blood Tuberc Lung Dis 2005. Smith IM. Chest 2000. 53. J Chemother 1996. resistant pulmonary tuberculosis treated Treatment of multidrug-resistant pulmo. Mycobacterium avium complex replicating regimens for MDR-TB. Song SD. Cochrane Database Syst Rev 2003. roquinolone resistance in Mycobacterium dach H. KwaZulu-Natal. de Bruyn G. Early bactericidal activity of amoxicillin 2007. Wu DY. Dautzenberg B. Sturm AW. Expert Rev Anti 2007. Adverse events in the treatment of tistical analysis of failure time data.11:434-40. Ten-year experience al. Wang JY. Lee IH. Ginsburg AS. et al. of chemotherapy in 107 patients with pul. and hope. All rights reserved.33:466-9. Derenne JP. Gillespie S. al. Condos R. Fluo. Drug resistance profiles of Mycobacteri- sis. 56. Akaki T. Raviglione M. chiatric issues in the management of pa. 2002.16:219-22. 38.59:860-5. Antimicrob Agents 2000. Basu S.10:290-6. Mycobacterium tuberculosis: Charles Dar. 57. Schluger NW. Dolfi L. Kalbfleisch JD. resistant tuberculosis. 58. Bak.356:656-9. floxacin. Socci AR. et al. culosis. Lack of activity of orally administered rience and insight into treatment strate- Nurses as providers of emotional support clofazimine against intracellular Myco. 48. 54. Park SK. Mitnick CD.349:1513-5. Norma técnica monary tuberculosis resistant to isoniazid 40. Huamani P. Rich ML. Lee LN. J In. Int J Tuberc Lung Dis 2001. 31. 1:CD001166. Riekstina V. 2001. and rifampin. The sta. Donald PR. MV. Lima. Masjedi MR.367:945-7.nejm. et 42. Copyright © 2008 Massachusetts Medical Society. 2007. 37. Chang MF. Lai HC. XDR tuber- with ofloxacin/levofloxacin-containing nary tuberculosis with interferon-gamma culosis — implications for global public regimens. Pillay M. Harvard Medical in combination with clavulanic acid in 47.8:137-43. Yew WW. et 28. 2008 The New England Journal of Medicine Downloaded from nejm. Nathanson E. Acha J. Park SK. Int J Infect Dis or repeating it? Clin Infect Dis 2007. and re. Nitta AT. and KRM-1648 Representative drug susceptibility pat- tant tuberculosis: international edition. Matrat S. Tomioka H. Int J Tuberc Lung 25. Normark BH. 1980. health. Timperi R.3:432-42. Semin berc Lung Dis 2004. Partners In Health. Int J Tuberc Lung Dis 1998. Migliori GB. Kim CT. Clin Infect Boston: Partners In Health. Spigelman M. Normark S. Han LL.5:648-55. tuberculosis. Blum- al. tern Med 2002. Andrews JR. Thorpe LE. Chau CH. Rifabutin as sal. et Fluoroquinolones. Mayer C. Gupta R. against Mycobacterium tuberculosis and terns for guiding design of retreatment Boston: Partners In Health. et al. sion of extensively drug-resistant tubercu- 34. Clin Infect Dis cator dye. stage. Rom WN. Sterling TR. Kawa. Furin JJ. Care Med 1998. Sato K. Lancet 2006. drug-resistant pulmonary  august 7. Bill & Melinda Gates Foundation.16:25-9.46:796-9. Laserson KF. Sirgel FA. within Mono Mac 6 human macrophage Dis 2006. 2003. Matteelli A. 44. clarithromycin. Int J Tuberc Lung Dis 2004. on November 21. with artificial pneumoperitoneum for end. Lancet Infect Dis 2003. Tuberculo- 30. berculosis by use of malachite green indi- 35. of tuberculosis: justification for currently from the DOTS-Plus initiative.157:1609-15.50:4170-3. Wallis um tuberculosis isolates: five years’ expe- 26. Clin Infect Dis 1993. For personal use only. Lancet 2005.25:307-15. et al. 2006. losis in rural South African district hospi- Functional analysis of DNA gyrase mu. Jimenez patients receiving community-based ther- ment outcome definitions for multidrug. via aerosol. 338-42. Vega P. Iseman MD. Prentice RL. tients with multidrug-resistant tuberculo. and A-549 type II alveolar cell lines. Goble M. No other uses without permission. Scand J Infect Dis strain of Mycobacterium tuberculosis in community-based treatment of MDR-TB. N Engl J Med 2007. Clinical outcome of individualised neously administered interferon-gamma berg HM. New 39. Song HY. South Africa. Lancet 1997. 2006.9:640-5. Janulionis E. Sloutsky A. Ministerio de Salud.5:857-71. Antimicrobial therapy multidrug-resistant tuberculosis: results York: John Wiley. strains of Mycobacterium tuberculosis re. Evolution Lancet 2007.365:318-26. Perú: Ministerio de Salud. Evolution of the School.48:3133-5. Copyright © 2008 Massachusetts Medical Society.8:1382-4. 2004. 574 n engl j med 359. Cho S. 33. Sweetland A. Lin TP. tion 88 in the A subunit found in clinical and spread of antibiotic resistance. Rich ML. Mohammadi F. gies for MDR-TB in Lima. 2012. et al. losis. J Clin Microbiol 2008. Mitnick CD. et al. 36. Int J Tu- 29. for the treatment of multidrug-resistant tuberculosis: are we learning from history losis in Latvia: a retrospective cohort pulmonary tuberculosis. Mycobacterium de salud para el control de la tuberculosis. 43. Garner P. Olliaro P. Extensively drug-resistant resistant or extensively drug-resistant tu- Agents Chemother 2006. Respir Crit Care Med 2004. 24. Occurrence of serious adverse effects in et al. Infect Ther 2007. Dukes Hamilton C.