Our high prevalence (25%) of isoniazid-resistant tubeculosis highlights the rapid spread of a drug-resistant Mycobacterium tuberculosis strain among high-risk populations. Patients with isoniazid-resistant tuberculosis received prolonged treatment regimens, and isoniazid resistance had no impact on sputum-culture conversion or initial treatment outcomes.
Keywords: tuberculosis, isoniazid resistance, fluoroquinolones, time to sputum culture conversion
Abstract
Background
Although isoniazid-resistant tuberculosis is more common than multidrug-resistant tuberculosis, it has been much less studied. We examined the impact of isoniazid resistance and treatment regimen, including use of a fluoroquinolone, on clinical outcomes.
Methods
A retrospective cohort study among patients with sputum culture-positive tuberculosis was performed. Early fluoroquinolone (FQ) use was defined as receiving ≥5 doses during the first month of treatment. The primary outcome was time to sputum culture conversion (tSCC). A multivariate proportional hazards model was used to determine the association of isoniazid resistance with tSCC.
Results
Among 236 patients with pulmonary tuberculosis, 59 (25%) had isoniazid resistance. The median tSCC was similar for isoniazid-resistant and -susceptible cases (35 vs 29 days; P = .39), and isoniazid resistance was not associated with tSCC in multivariate analysis (adjusted hazard ratio = 0.83; 95% confidence interval [CI], .59–1.17). Early FQ use was higher in isoniazid-resistant than -susceptible cases (20% vs 10%; P = .05); however, it was not significantly associated with tSCC in univariate analysis (hazard ratio = 1.48; 95% CI, .95–2.28). Patients with isoniazid-resistant tuberculosis were treated with regimens containing rifampin, pyrazinamide, and ethambutol +/− a FQ for a median of 9.7 months. Overall, 191 (83%) patients were cured. There was no difference in initial treatment outcomes; however, all cases of acquired-drug resistance (n = 1) and recurrence (n = 3) occurred among patients with isoniazid-resistant tuberculosis.
Conclusions
There was no significant association with isoniazid resistance and tSCC or initial treatment outcomes. Although patients with isoniazid-resistant tuberculosis had a high cure rate, the cases of recurrence and acquired drug resistance are concerning and highlight the need for longer-term follow-up studies.
Although isoniazid-resistant rifampin-susceptible tuberculosis (isoniazid-resistant tuberculosis) accounts for 9.5% of tuberculosis cases worldwide and is considered a major risk factor for multidrug-resistant tuberculosis, it has received much less attention than other forms of drug resistance [1–3]. In a recent meta-analysis, standardized therapy for isoniazid-resistant tuberculosis with first-line drugs was associated with poor outcomes [4]. However, the role of fluoroquinolones (FQs) and isoniazid for the treatment of isoniazid-resistant tuberculosis and the impact of isoniazid resistance level on treatment outcomes remains unclear [1].
The 2003 American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America guidelines (the 2016 guidelines focus on drug-susceptible tuberculosis) recommend a 6-month regimen of rifampin (RIF), pyrazinamide (PZA), and ethambutol (EMB) for the treatment of isoniazid-resistant tuberculosis with the option of adding an FQ in cases of extensive disease [5, 6]. In addition, these guidelines recommend discontinuing isoniazid once resistance is detected. These recommendations are based on a review of trials conducted by the British Medical Research Council [7, 8]. Importantly, the benefit of isoniazid for the treatment of isoniazid-resistant tuberculosis cannot be ascertained from these trials because all regimens contained isoniazid for the full length of therapy, irrespective of baseline resistance. It has been suggested that standard-dose isoniazid may retain activity against some isoniazid-resistant strains, particularly when low-level resistance is present [9–12]. FQs successfully replaced isoniazid for the treatment of drug-susceptible tuberculosis in the RIFAQUIN trial and have been proposed to replace isoniazid for the treatment of isoniazid-resistant tuberculosis given comparable bactericidal activity, but this has not been evaluated in trials [13–15]. These uncertainties are reflected in the substantial variation in treatment regimens, including the use and duration of isoniazid and FQs for the treatment of isoniazid-resistant tuberculosis [16–18].
The primary aim of this study was to compare time to sputum culture conversion (tSCC) between patients with isoniazid-resistant and patients with isoniazid-susceptible tuberculosis. To our knowledge, there are no published studies comparing tSCC between patients with isoniazid-resistant and patients with isoniazid-susceptible tuberculosis. tSCC accounts for the effect of time on treatment and is a US Food and Drug Administration (FDA) recommended endpoint for phase 2 clinical trials [19, 20]. In addition, we aimed to evaluate the treatment outcomes for isoniazid-resistant tuberculosis and to describe the outcomes and use of FQs in this population. This study was conducted in the setting of an outbreak of isoniazid-resistant tuberculosis among persons experiencing homelessness in Atlanta [21]. Our overall goal was to provide data that would improve the care of patients with isoniazid-resistant tuberculosis.
METHODS
A retrospective cohort study design was used. Adults (aged ≥18 years) with culture-confirmed pulmonary tuberculosis, including those with concomitant extrapulmonary disease, treated at Grady Memorial Hospital between January 2008 and October 2015 were eligible for inclusion. Exclusion criteria included the presence of rifampin-resistant tuberculosis and incomplete information regarding treatment course or sputum culture results. Grady Memorial Hospital is a 1000-bed safety-net hospital in Atlanta. Following hospital discharge, all patients were referred to a county public health tuberculosis clinic for further management.
Laboratory Methods
All included patients had sputum samples submitted for acid-fast bacilli (AFB) smear microscopy and cultures incubated in solid Middlebrook 7H11 and liquid media using the BacT/ALERT 3D mycobacterial detect system. Between 2008 and 2014, direct Mycobacterium tuberculosis detection test using nuclear acid amplification method was performed for all smear-positive samples; subsequently the Xpert MTB/RIF assay was performed on all smear-positive samples and upon request. Baseline M. tuberculosis culture isolates were sent to the Georgia Public Health Laboratory where drug-susceptibility testing for rifampin, isoniazid, and ethambutol was performed using mycobacterial growth indicator tube (MGIT) 960. Pyrazinamide was added to this panel in 2015. All isolates with isoniazid resistance (growth at 0.1 ug/mL of isoniazid) from the Georgia Public Health Laboratory were sent to the Centers for Disease Control and Prevention (CDC) where drug-susceptibility testing for first-line drugs was repeated and an extended panel that included PZA and ofloxacin was performed. The level of isoniazid resistance was ascertained by indirect proportion method on agar at isoniazid concentrations of 0.2, 1.0, and 5.0 µg/mL. Follow-up sputum samples from county health clinics were sent to the Georgia Public Health Laboratory, where smear microscopy and culture in both solid Lowenstein-Jensen media and liquid Middlebrook 7H9 media in MGIT was performed. Genotyping for baseline samples was performed by the CDC [22].
Definitions
Low-level isoniazid resistance was defined as M. tuberculosis isolates with growth at an isoniazid concentration of ≥0.2 µg/mL, and high-level resistance was defined as growth at an isoniazid concentration of ≥1.0 µg/mL [5]. Disseminated tuberculosis disease was defined as having a blood culture positive for M. tuberculosis and/or the presence of miliary tuberculosis based on radiology report. Early FQ use was defined as receiving ≥5 doses within the first month of treatment. All treatment was recommended to be given by directly observed therapy (DOT). Early poor DOT adherence was defined as receiving ≤40 doses during the first 3 months, and late poor DOT adherence was defined as missing ≥2 weeks of therapy consecutively afterwards. tSCC was defined as the number days from beginning of tuberculosis treatment to the first of 2 consecutive negative sputum culture results that were at least 30 days apart [19].
Data Collection
A review of inpatient, local health department, and Georgia State Electronic Notifiable Surveillance System records was performed. Georgia State Electronic Notifiable Surveillance System records were last examined in December 2016. Case report forms were used for data abstraction, and data were entered into an online REDCap database [23].
Data Analysis
Data analyses were performed using R version 3.3.2. For univariate comparisons, differences in nominal variables were tested using either a Fisher’s exact or χ2 test and for continuous variables, either a Mann-Whitney or 2-sample t test was used as appropriate. A two-sided P value < .05 was considered significant. Cox proportional hazard models were used to estimate the association of isoniazid resistance with tSCC. Covariates included in the model were based on previous literature and bivariate analysis, and the model was built using the purposeful selection strategy [24]. Patients were censored at the tSCC or at the time of treatment completion and at loss to follow-up or death for those who did not achieve culture conversion. The study was approved by the Emory University and the Georgia Department of Public Health Institutional Review Boards and the Grady Memorial Hospital Research Oversight Committee.
RESULTS
Among 361 patients treated for active tuberculosis during the study period, 270 had culture-confirmed disease, including 249 with pulmonary tuberculosis. After excluding patients with incomplete data and multidrug-resistant tuberculosis, 236 patients were included for further analysis (Figure 1).
Figure 1.
Study population flow diagram. Abbreviation: MDR, multidrug-resistant.
A total of 59 (25%) patients had isoniazid resistance at baseline, with most (n = 54) having low-level isoniazid-resistance (Table 1). There was a bimodal temporal distribution of isoniazid-resistant cases, with an initial peak in 2008–2009 and a second peak in 2014 (Figure 2), and most cases (n = 46) were genotypically linked. Two patients had PZA resistance detected; no resistance to ethambutol was detected. Among 57 patients who had drug-susceptibility testing for ofloxacin performed, no resistance was detected.
Table 1.
Baseline Patient Cohort Characteristics by Isoniazid Resistance Status
| Characteristic | Total n = 236 | Isoniazid-resistant n = 59 | Isoniazid-susceptible n = 177 | P value |
|---|---|---|---|---|
| Age, median (IQR), y | 47.6 (38.6–54.7) | 49.6 (41.8–55.8) | 47.1 (35.9–54.4) | .10 |
| Male | 182 (77) | 57 (97) | 125 (71) | <.01 |
| Black | 193 (82) | 53 (90) | 140 (79) | .09 |
| Index BMI, median (IQR) | 21.0 (19.0–24.4) | 20.3 (18.3–22.6) | 21.2 (19.0–24.8) | .03 |
| US born | 173 (73) | 53 (90) | 120 (68) | <.01 |
| Homeless | 108 (46) | 51 (86) | 57 (32) | <.01 |
| History of incarceration | 99 (42) | 30 (51) | 69 (39) | .14 |
| Medical history | ||||
| Diabetes mellitus | 37 (16) | 9 (16) | 28 (16) | 1.00 |
| HCV Ab seropositive (n = 190) | 30 (16) | 11 (20) | 19 (14) | .27 |
| HBSAg positive (n = 190) | 9 (5) | 4 (8) | 5 (4) | .25 |
| Current malignancy | 4 (2) | 0 | 4 (2) | .57 |
| HIV positive | 87 (37) | 29 (49) | 58 (33) | .03 |
| Known HIV prior to TB diagnosis (n=87) | 66/87 (76) | 27/29 (93) | 39/58 (67) | .01 |
| Receiving ART at TB diagnosis (n=66) | 10/66 (15) | 5/27 (19) | 5/39 (13) | .72 |
| Baseline CD4, median (IQR) | 83 (28–190) | 110 (30–322) | 75 (28–148) | .42 |
| Drug use | ||||
| Tobacco use | 135 (57) | 44 (75) | 91 (51) | <.01 |
| Alcohol use | 128 (54) | 40 (68) | 88 (50) | .02 |
| Illicit drug use | 62 (26) | 23 (39) | 39 (22) | .01 |
| Tuberculosis history | ||||
| Previous active tuberculosis | 24 (10) | 6 (10) | 18 (10) | 1.00 |
| History of LTBI | 36 (15) | 7 (12) | 29 (16) | .53 |
| LTBI treatment | 221 (9) | 5 (9) | 17 (9) | 1.00 |
| Baseline laboratory values | ||||
| Albumin ≤2.5 gm/dL | 55 (23) | 15 (26) | 40 (23) | .74 |
| Hemoglobin ≤10 gm/dL | 58 (24) | 14 (24) | 44 (24) | 1.00 |
| GFR ≤60 mL/min | 39 (17) | 13 (22) | 26 (15) | .24 |
| Tuberculosis presentation | ||||
| Isoniazid low-level resistance | 54 (23) | 54 (92) | … | … |
| Positive TST and/or IGRA (n = 162) | 103 (64) | 25 (57) | 78 (66) | .27 |
| AFB smear result | ||||
| Smear-negative | 65 (28) | 15 (25) | 50 (28) | .53 |
| 1–2 + | 48 (20) | 15 (25) | 33 (19) | |
| 3–4 + | 123 (52) | 29 (49) | 94 (53) | |
| Extrapulmonary involvement | ||||
| Any non-pulmonary site | 65 (28) | 15 (25) | 50 (28) | .80 |
| CNS | 10 (4) | 4 (7) | 6 (3) | .27 |
| Disseminated | 28 (12) | 6 (10) | 22 (12) | .81 |
| Radiologic features | ||||
| Abnormal CXRa | 203 (87) | 49 (83) | 154 (89) | .39 |
| Cavitary diseaseb | 45 (19) | 7 (12) | 38 (22) | .13 |
Data are no. (%) unless otherwise specified.
Abbreviations: AFB, acid-fast bacilli; ART, antiretroviral therapy; BMI, body mass index; CNS, central nervous system; CXR, chest X-ray’ GFR, glomerular filtration rate; HBSAg, hepatitis B surface antigen; HCV Ab, hepatitis C virus antibody; HIV, human immunodeficiency virus; IGRA, interferon-gamma release assay; IQR, interquartile range; LTBI, latent tuberculosis infection; TB, tuberculosis; TST, tuberculin skin test.
aCXR missing (n = 3).
bCavitary disease by CXR.
Figure 2.
Culture-confirmed pulmonary tuberculosis cases per year. Numbers above graph represent percentage of isoniazid-resistant cases per year.
Cohort Characteristics
Most patients were male (77%), black (82%), and born in the United States (73%). Almost half of the patients (46%) had a history of homelessness, and there were high rates of tobacco (57%), alcohol (54%) and illicit drug (26%) use. All patients were tested for human immunodeficiency virus (HIV), and 87 (37%) were positive. Although most coinfected patients (76%) were known to have HIV for many years (median, 6.5) prior to diagnosis with tuberculosis, only 15% were receiving antiretroviral therapy (ART) at the time of tuberculosis diagnosis. In regards to clinical presentation, 123 (52%) patients had an AFB sputum smear microscopy result ≥3+, and 19% had cavitary disease on chest X-ray. The median total treatment duration was 9.3 months (Table 2). Sixty-seven (28%) patients received an FQ during tuberculosis therapy, including 29 (12%) with early FQ use. Among 81 (93%) HIV-positive patients who survived ≥8 weeks after tuberculosis diagnosis, 70 (86%) received ART during tuberculosis therapy, and 30 (37%) were started on ART within 8 weeks of initiating tuberculosis therapy
Table 2.
Treatment Characteristics and Outcomes by Isoniazid Resistance Status-
| Characteristic | Total n = 236 |
Isoniazid resistant n = 59
|
Isoniazid sensitive n = 177 | P value |
|---|---|---|---|---|
| Median days of index hospitalization (IQR) | 8 (6–14) | 10 (6–16) | 8 (5–14) | .15 |
| ICU admission during index hospitalization | 33 (14) | 6 (10) | 27 (15) | .43 |
| Subsequent hospitalization during treatment | 73 (31) | 21 (36) | 52 (30) | .49 |
| Follow-up AFB sputum cultures | ||||
| Median number of follow-up cultures (IQR) | 10 (7–15) | 10 (7–18) | 10 (7–15) | .79 |
| Median days between last positive sample prior and culture conversion (IQR) | 8 (4–19) | 9 (6–19) | 8 (3–18) | .25 |
| Patients without a follow-up culturea | 14 (6) | 3 (6) | 11 (6) | 1.00 |
| Follow-up DST performed | 80 (34) | 31 (53) | 49 (28) | <.01 |
| Treatment | ||||
| Received an FQb | 67 (28) | 31 (53) | 36 (20) | <.01 |
| Levofloxacin | 39 (17) | 11 (19) | 28 (16) | .76 |
| Moxifloxacin | 36 (15) | 23 (39) | 13 (7) | <.01 |
| Early FQ usec | 29 (12) | 12 (20) | 17 (10) | .05 |
| Median days on FQ (IQR) | 86 (13–284) | 202 (67–284) | 53 (8–225) | .03 |
| Median days of PZA Use (IQR) | 81 (63–157) | 259 (178–313) | 74 (61–92) | <.01 |
| Median days of ethambutol Use (IQR) | 65 (44–183) | 251 (177–321) | 56 (39–81) | <.01 |
| Median treatment duration (IQR), mod | 9.3 (7.2–10.6) | 9.7 (8.6–11.2) | 9.2 (7.1–10.5) | .20 |
| ART use (n = 81)e | ||||
| ART during tuberculosis therapy | 70 (86) | 27 (96) | 43 (81) | .11 |
| ART use within 8 weeks of tuberculosis therapy | 30 (37) | 11 (39) | 19 (36) | .97 |
| Medication adverse effects requiring drug interruption | ||||
| Rifamycin (n = 231) | 33 (14) | 8 (14) | 25 (14) | 1.00 |
| Isoniazid (n = 231) | 37 (16) | 7 (12) | 30 (17) | .45 |
| PZA (n = 230) | 29 (13) | 12 (20) | 17 (10) | .06 |
| Ethambutol (n = 227) | 19 (8) | 8 (14) | 11 (6) | .09 |
| Levofloxacin (n = 39) | 2 (5) | 1 (9) | 1 (4) | .48 |
| Moxifloxacin (n = 36) | 4 (11) | 4 (17) | 0 | .27 |
| Poor DOT adherence | ||||
| First 3 months of therapy (n = 187)f | 14 (7) | 1 (2) | 13 (9) | .19 |
| Late treatment interruption (n = 200)g | 28 (14) | 10 (19) | 18 (14) | .33 |
| Clinical outcomes | ||||
| Achieved culture conversionh | 214 (91) | 54 (91) | 160 (91) | 1.00 |
| Median days to culture conversion (IQR) | 30 (11–54) | 35 (14–53) | 29 (11–54) | .39 |
| Treatment outcomes | ||||
| Acquired drug resistance | 1 (1) | 1 (2) | 0 | .25 |
| Death during index admission | 16 (7) | 5 (8) | 11 (6) | .55 |
| Transferred out | 6 (2) | 1 (2) | 5 (2) | 1.00 |
| Final outcomesi | .10 | |||
| Cured | 191(83) | 47 (81) | 144 (84) | |
| Lost to follow-up | 7 (3) | 0 | 7 (4) | |
| Censored on therapy | 1 (1) | 0 | 1 (1) | |
| Recurrence | 3 (1) | 3 (5) | 0 | |
| Death | 28 (12) | 8 (14) | 20 (11) | |
Data are no. (%) unless otherwise specified.
Abbreviations: AFB, acid-fast bacilli; ART, antiretroviral therapy; DOT, directly observed therapy; DST, drug-susceptibility test; FQ, fluoroquinolone; ICU, intensive care unit; IQR, interquartile range; PZA, pyrazinamide.
aThirteen of 14 patients without a follow-up culture died.
bMoxifloxacin and/or levofloxacin.
cDefined as starting FQ within 28 days of tuberculosis therapy and receiving at least 5 doses under DOT.
dAmong patients who completed therapy n = 194
eAmong HIV-positive patients who survived ≥8 weeks
fDefined as ≤ 40 doses. Excluded patients with missing DOT sheet and whho had <90 days of treatment.
gDefined as treatment interruption of ≥2 weeks after first 3 months of therapy. Excluded patients with missing DOT sheets.
hEight patients without culture conversion had ≥1 follow-up culture. None had 2 consecutive negative cultures.
iExcluded patients that were transferred out during tuberculosis therapy.
Comparison of Cohort According to Isoniazid Resistance
Patients with isoniazid-resistant tuberculosis were more likely to be male (97% vs 71%), born in the United States (90% vs 68%), homeless (86% vs 32%), tobacco users (75% vs 51%), alcohol users (68% vs 50%), illicit drug users (39% vs 22%), and have HIV co-infection (49% vs 33%) (all with P < .05) than patients without isoniazid resistance. Three (5%) patients were diagnosed with isoniazid-resistant tuberculosis by contact investigation [21]. Acid-fast bacilli smear grade, proportion with extrapulmonary involvement, and cavitary disease were similar between groups.
Although there was no significant difference in treatment duration when stratified by isoniazid resistance, differences in regimens were observed (Table 3). Patients with isoniazid-resistant tuberculosis were more likely to receive an FQ early in therapy (12 [20%] vs 17 [10%]; P = .05) and at any time during treatment (31 [53%] vs 36 [20%]; P < .01]. The median duration in days of PZA (259 vs 74), ethambutol (251 vs 56), and FQ (202 vs 53) use were higher for patients with isoniazid-resistant tuberculosis as compared with patients with isoniazid-susceptible tuberculosis (all with P < .05). Fifty (85%) patients with isoniazid-resistant tuberculosis completed therapy, and all received ≥6 months of a rifamycin. Forty-one (82%) of these received a rifamycin and PZA combination for ≥6 months. The remaining 9 patients with isoniazid-resistant tuberculosis who completed therapy received an FQ as part of their treatment regimen. Four of these 9 patients suffered adverse events related to PZA, and 5 were changed to an FQ due to physician preference. Among these 9 patients, 1 patient had adverse effects while on FQ with drug-induced hepatitis while on RIF/moxifloxacin. This patient completed therapy on the same regimen with no further adverse events. All but 1 patient with isoniazid-resistant tuberculosis received isoniazid, and patients with isoniazid-resistance received a median of 35 (interquartile range, 25–46) isoniazid doses during the first 3 months of chemotherapy. All patients received standard dose isoniazid, except 1 patient with isoniazid resistance and meningeal disease for which the dose was increased after culture conversion. The proportion of patients with HIV who received ART at any time during and within 8 weeks of initiating tuberculosis therapy was similar between patients with isoniazid-resistant and patients with isoniazid-susceptible tuberculosis.
Table 3.
Treatment Regimens by Duration and Isoniazid Resistance Among Patients Who Completed Treatment (n = 194)a
| Treatment regimens | Isoniazid resistant (n = 50) | Isoniazid susceptible (n = 144) |
|---|---|---|
| 6 months | 9 (18%) | 18 (13%) |
| HREZ (2), HR (4) | … | 16 |
| HREZ (1–2), REZ (4–5) | 7b, d | … |
| REZQ (3), REZ (3) | 1 | … |
| REZQ (6) | 1 | … |
| Other | … | 2 |
| 7–12 months | 34 (68%) | 114 (79%) |
| HREZ (2–4), HR (2–8) | … | 102e,f,f,g |
| HREZQ (2–4), HR (4–6) | … | 10c |
| HREZ (1), REZ (5–8) | 18 b,b | 2f |
| HREZQ (2), REZ (4–7) | 2 c | … |
| HREZQ (1–2), REZQ (4–10) | 14 e,e | … |
| >12 months | 7 (14%) | 12 (8%) h |
| HREZ (1), HRZ (4), HR (5) | … | 1 |
| HREZQ (2), HREQ (4), RE (6) | … | 1 |
| HREZQ (1–6), RQ (8) | … | 3 |
| HREZQ (2–3), HR (10) | … | 3c,c,f |
| HREZQ (2), REQ (10) | … | 1 |
| HREZQ (1), HZEQ (15) | … | 1 |
| HREZQ Linezolid (2), REZ (10) | 1 | … |
| HREZ (1), REZ (27) | 1 | … |
| HREZQ(4), REQ (5–11) | 3 f | … |
| HREZQ (2–4), REZQ (6–15) |
2 c,c,e |
… |
| HZEQ(3), EQ (15) | … | 1 |
Medications included in the table were given for ≥14 days. Numbers in parentheses denote months on therapy.
Abbreviations: E, ethambutol; H, isoniazid; Q, fluoroquinolone (levofloxacin and/or moxifloxacin); R, rifampin; Z, pyrazinamide.
a Includes all patients that completed therapy (cure and recurrence).
bPatients with tuberculosis recurrence (n = 3).
cReceived injectable for 1–2 months (n = 6).
dAcquired drug resistance (n = 1).
eMeningeal tuberculosis (n = 4).
fBone tuberculosis (n = 5).
gOne patient had a diagnosis of bone and meningeal tuberculosis.
hOne patient received >12 months of therapy; stop dates for individual drugs unknown.
Among the 227 patients who survived beyond 2 weeks, 222 (98%) had ≥1 follow-up sputum culture. Thirteen of the 14 patients without a follow-up sputum culture died on index admission. A median of 10 follow-up sputum cultures were obtained per patient, with no significant differences when stratified by isoniazid susceptibility (Table 2). Among 214 patients with sputum culture conversion (SCC), 211 (99%) had ≥1 sputum culture performed in MGIT. There was no significant difference in the median days between last positive sputum culture and date of SCC.
Ninety-one percent of patients in both groups achieved SCC, and although patients with isoniazid-resistant tuberculosis had a trend toward longer tSCC (35 vs 29 days; P = .39), the difference was not significant (Figure 3). Isoniazid resistance was not associated with SCC at 28 and 56 days by bivariate analysis (Table 4). Isoniazid resistance did not impact tSCC in the multivariate model (Table 5). Age, body mass index, HIV status, and AFB smear grade were the only covariates that had a significant association with tSCC (Table 5).
Figure 3.
Cumulative sputum culture conversion by isoniazid resistance status. Abbreviations: CI, confidence interval; HR, hazard ratio.
Table 4.
Cumulative Sputum Culture Conversion
| 28 days (n = 218) | 56 days (n = 216) | |||||
|---|---|---|---|---|---|---|
| Covariate | n/N | % | P value | n/N | % | P value |
| Overall | 98/218 | 45 | … | 166/216 | 77 | … |
| Group | .31 | .45 | ||||
| Isoniazid resistant | 21/55 | 38 | 44/54 | 81 | ||
| Isoniazid susceptible | 77/163 | 47 | 122/162 | 73 | ||
| Age, y, median | <.01 | .11 | ||||
| >48 | 31/99 | 31 | 70/98 | 71 | ||
| ≤48 | 67/119 | 56 | 96/118 | 81 | ||
| Sex | .31 | .38 | ||||
| Male | 71/166 | 43 | 124/166 | 75 | ||
| Female | 27/52 | 54 | 42/51 | 82 | ||
| BMI, kg/m2a | .30 | <.01 | ||||
| <=18.5 | 16/42 | 38 | 25/42 | 60 | ||
| >18.5 | 81/167 | 49 | 134/165 | 81 | ||
| HIV status | <.01 | <.01 | ||||
| Positive | 48/81 | 60 | 70/80 | 87 | ||
| Negative | 50/137 | 36 | 96/136 | 71 | ||
| Diabetes | .29 | .54 | ||||
| Yes | 12/34 | 35 | 28/34 | 82 | ||
| No | 86/184 | 47 | 138/182 | 76 | ||
| Cavitary diseaseb | <.01 | .23 | ||||
| Yes | 7/41 | 17 | 28/41 | 68 | ||
| No | 89/174 | 51 | 135/172 | 78 | ||
| High-grade smearc | <.01 | <.01 | ||||
| Yes | 32/112 | 29 | 71/110 | 65 | ||
| No | 66/106 | 62 | 95/106 | 90 | ||
| Homeless | <.01 | .79 | ||||
| Yes | 34/99 | 34 | 74/98 | 76 | ||
| No | 64/119 | 54 | 92/118 | 78 | ||
| Tobacco use | .21 | .23 | ||||
| Yes | 53/129 | 41 | 95/129 | 74 | ||
| No | 45/89 | 51 | 71/87 | 82 | ||
| Alcohol use | .17 | .19 | ||||
| Yes | 48/119 | 40 | 87/119 | 73 | ||
| No | 50/99 | 51 | 79/97 | 81 | ||
| Illicit drug use | .53 | .43 | ||||
| Yes | 24/59 | 41 | 48/59 | 81 | ||
| No | 74/159 | 47 | 118/157 | 75 | ||
| Early FQ used | .91 | .03 | ||||
| Yes | 12/25 | 48 | 23/24 | 96 | ||
| No | 86/193 | 46 | 143/192 | 74 | ||
| Albumin, gm/dLe | .09 | .13 | ||||
| <=2.5 | 14/43 | 33 | 29/43 | 69 | ||
| > 2.5 | 84/178 | 47 | 137/172 | 80 | ||
| Previous active tuberulosis | .60 | .92 | ||||
| Yes | 12/23 | 55 | 17/23 | 74 | ||
| No | 86/195 | 44 | 149/193 | 77 | ||
| TST/IGRAf | .15 | .94 | ||||
| Positive | 51/101 | 50 | 74/100 | 74 | ||
| Negative | 20 /54 | 37 | 41/54 | 76 | ||
| DOT doses first 3 months of therapyg | .72 | 1.00 | ||||
| <40 | 5/14 | 36 | 11/14 | 78 | ||
| > 40 | 77/173 | 44 | 134/173 | 78 | ||
Abbreviations: BMI, body mass index; DOT, directly observed therapy; FQ, fluoroquinolone; HIV, human immunodeficiency virus; IGRA, interferon-gamma release assay; TST, tuberculin skin test.
aBMI missing for n = 9.
bBy chest X-ray (missing for n = 3).
cDefined as 3–4+.
dDefined as starting FQ within 28 days of tuberculosis therapy and receiving at least 5 doses under DOT.
eData missing for n = 1.
fTST and/or IGRA result available for n = 163.
gExcluded patients with missing DOT sheet and who had <90 days of treatment.
Table 5.
Univariate and Multivariate Model for Time to Sputum Culture Conversion
| Predictor | Univariate analysis | Multivariate analysis |
|---|---|---|
| HR (95% CI) | aHR (95% CI) | |
| Isoniazid resistance | 0.97 (.71–1.32) | 0.83 (.59–1.17) |
| Age, per y | 0.98 (.97–.99) | 0.98 (.97–.99) |
| Male | 0.87 (.63–1.20) | … |
| Black | 0.84 (.59–1.18) | … |
| US born | 0.89 (.66–1.20) | … |
| BMI | 1.00 | 1.00 |
| BMI ≤ 18.5 kg/m2 | 0.60 (.42–.85) | 0.71 (.49–1.02) |
| Missing | 0.60 (.30–1.19) | 0.43 (.20–.89) |
| Previous active tuberculosis | 0.85 (.54–1.35) | … |
| Diabetes | 1.25 (.86–1.81) | … |
| HIV | 1.53 (1.15–2.03) | 1.39 (1.03–1.89) |
| Homeless | 0.86 (.66–1.14) | … |
| Tobacco use | 0.73 (.55–.97) | … |
| Alcohol use | 0.77 (.59–1.02) | … |
| Illicit drug use | 0.97 (.72–1.32) | … |
| Cavitary disease | 0.69 (.48–.97) | … |
| High-grade sputum AFB smear | 0.45 (.34–.60) | 0.47 (.35–.63) |
| Baseline albumin ≤2.5 gm/dL | 0.73 (.52–1.02) | … |
| Intensive care unit admission | 0.96 (.59–1.57) | … |
| Early fluoroquinolone | 1.48 (.95–2.28) | … |
Abbreviations: AFB, acid-fast bacilli; aHR, adjusted hazard ratio; BMI, body mass index; CI, confidence interval; HIV, human immunodeficiency virus.
Among 214 patients with SCC, 24 (11%) received an early FQ. Although early FQ use was associated with SCC at 56 days by bivariate analysis (P = .03), it was not significantly associated with tSCC in univariate (hazard ratio [HR], 1.48; 95% confidence interval [CI], .95–2.28) or multivariate analysis. Among the 54 patients with isoniazid-resistant tuberculosis that achieved SCC, 9 (17%) received an FQ prior to SCC, and 5 of these were also receiving isoniazid at SCC. All 45 (83%) remaining patients were receiving RIF/PZA/EMB at SCC; 31(69%) of these were receiving isoniazid at SCC. There was no difference in median tSCC among those with isoniazid resistance that did and did not receive early FQ (29 vs 36 days, respectively; P = .24).
Treatment Outcomes
Overall, 191 (83%) patients were cured, 28 (12%) died, 3 (1%) suffered recurrent disease, 7 (3%) were lost to follow-up, and 1 (1%) was censored on therapy. The overall cure rate was 89%, after excluding the 16 patients who died at index hospital admission. The final outcomes were similar when stratified by isoniazid resistance, with the exception of acquired-drug resistance (P = .25) and recurrence (P = .01), which occurred exclusively among patients with low-level isoniazid resistance.
The episodes of recurrence occurred between 238 and 1000 days after completion of tuberculosis therapy. All recurrences occurred among patients with isoniazid-resistant tuberculosis and with a M. tuberculosis strain of the same genotype as their initial infection. Two recurrences occurred in HIV-positive patients with CD4 count <50 cells/mm3 who were started on ART during tuberculosis therapy but were not retained in care upon completion of tuberculosis therapy, and 1 of these patients had acquired-resistance to rifampin and thus developed multidrug resistance. Although all patients with recurrence received ≥6 months of PZA and a rifamycin, none received an FQ (Table 3). Among isoniazid-resistant cases, those who suffered recurrence received shorter treatment duration than those who did not (median of 7.8 vs 9.6 months; P = .17).
DISCUSSION
In an inner-city setting characterized by high rates of HIV and during an ongoing outbreak of isoniazid-resistant tuberculosis, we found a very high prevalence (25%) of isoniazid resistance among patients with pulmonary tuberculosis. This rate of isoniazid resistance is much higher than published worldwide and US rates and is a stark reminder of how a drug-resistant M. tuberculosis strain can take hold and rapidly spread among high-risk populations [2, 25]. We found that the tSCC and overall rate of favorable treatment outcomes were similar regardless of isoniazid resistance status. American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America guidelines recommend 6 months of RIF/PZA/EMB, a regimen with good outcomes in a recent meta-analysis [4, 5]. Almost all patients with isoniazid resistance in our cohort received RIF/PZA/EMB, but the median duration of therapy was 9.7 months. Our data on the effectiveness and tolerability of regimens containing a long course of an FQ are encouraging given the high rates of pyrazinamide and ethambutol drug interruption with prolonged use.
Although isoniazid resistance did not significantly impact tSCC, episodes of recurrence and acquired drug resistance occurred exclusively among isoniazid-resistant patients. Our findings suggest that low-level isoniazid resistance may not impact tSCC in patients receiving standard dose isoniazid and first-line treatment regimens for tuberculosis. However, the data suggest isoniazid resistance may have had significant impact on long-term outcomes in this cohort.
As a result of an outbreak of isoniazid-resistant tuberculosis among persons experiencing homelessness in Atlanta, there was a high rate of isoniazid resistance in this cohort compared with the global (9.5%) and US (8%) rates in 2015 [2, 21, 25]. The importance of isoniazid resistance has been debated, and research efforts have been limited despite the role of isoniazid resistance in the genesis of multidrug-resistant tuberculosis [1, 3]. In addition, pharmacovigilance for isoniazid resistance is lacking, and the diversity of resistance mutations is a challenge for rapid molecular testing [1]. Grady Memorial Hospital uses the Xpert MTB/RIF, which exclusively detects rifampin resistance. Rapid isoniazid resistance detection could improve outcomes by allowing earlier initiation of second-line drugs such as FQs [26]. Although assays for rapid diagnosis of isoniazid are available (eg, Hain MTBDRplus), none are FDA approved, and availability is limited to specialized laboratories [27].
To our knowledge, this is the first published study evaluating tSCC (as opposed to 2-month SCC) for isoniazid-resistant tuberculosis [7, 16–18, 28, 29]. Most patients in the cohort achieved SCC before receiving an FQ. Thus, these data suggest that either isoniazid retains some activity or RIF/PZA/EMB can compensate for loss of isoniazid activity in the early phase of therapy for isoniazid-resistant tuberculosis.
Isoniazid has the highest early bactericidal activity among first-line drugs (RIF/INH/PZA/EMB) and kills a significant portion of replicating bacilli within 48 hours, after which RIF and PZA have most bactericidal activity for the treatment of drug-susceptible tuberculosis [30, 31]. The potent early bactericidal activity of isoniazid translates into important public health benefits by decreasing transmission and patient-level benefits by prevention of acquired drug resistance and early mortality from severe forms of tuberculosis [30, 32]. The similar tSCC and deaths on index inpatient admission in this cohort when stratified by isoniazid resistance suggest that RIF/isoniazid/PZA/EMB is a safe initial regimen for the early phase of chemotherapy for isoniazid-resistant tuberculosis, particularly in settings where low-level resistance is predominant. However, for rapidly fatal forms of tuberculosis such as meningitis, a different initial regimen should be considered [32, 33]. Standard-dose isoniazid may retain activity against isoniazid-resistant strains, particularly for low-level resistance [9–12]. Interestingly, lower breakpoints for isoniazid resistance have been proposed based on in vitro models [34]. The ongoing ACTG 5312 study is evaluating the early bactericidal activity of escalating isoniazid doses for low-level isoniazid-resistant strains and should elucidate some of these issues. FQs have potent early bactericidal activity and, as in trials for drug-susceptible tuberculosis, early FQ use was associated with SCC at 56 days in our cohort [14, 35]. Two retrospective cohort studies evaluated FQ use for isoniazid-resistant tuberculosis [17, 18]. One evaluated 2-month SCC rates and found FQ use had no effect on SCC in contrast with isoniazid use, which was associated with increased rates of 2-month SCC [17].
In a recent meta-analysis, patients with isoniazid resistance who received standardized therapy with first-line drugs had poor outcomes compared with patients with drug-susceptible tuberculosis [4]. However, patients receiving at least 6 months of RIF/PZA/EMB, as recommended by American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America guidelines, had outcomes comparable with those of patients with drug-susceptible tuberculosis [5]. Thus, the similar outcomes of patients with isoniazid-resistant and patients with isoniazid-susceptible tuberculosis are in line with this meta-analysis and with other cohorts receiving individualized regimens [16]. Although all recurrences in our cohort occurred among patients with isoniazid resistance who did not receive FQ, the sample size precludes conclusions regarding FQ impact on final outcomes. FQ use improved unadjusted outcomes for isoniazid-resistant tuberculosis in both previously mentioned studies [17, 18]. Finally, likely due to the high rates of comorbidities and late presentation, the death rate in this cohort among isoniazid-resistant (14%) and isoniazid-susceptible (11%) cases was higher than the overall tuberculosis death rate in the United States (7%) [36].
Our study has limitations. Sputum culture collection was not standardized, but sampling frequency was similar between subjects with isoniazid-resistant and patients with isoniazid-susceptible tuberculosis. Drug concentrations, an important factor for tuberculosis outcomes, and acetylator phenotypes, a key determinant for isoniazid concentrations, were not performed [37]. We were only able to detect recurrent tuberculosis episodes that were reported to the Georgia Department of Public Health Program and, in the absence of whole-genome sequencing, could not ascertain whether recurrent cases were due to relapse or reinfection. Finally, the predominance of low-level isoniazid resistance and outbreak-associated cases limits generalizability.
In conclusion, we found no significant association with isoniazid resistance and tSCC or initial treatment outcomes among patients who were mainly started on standard first-line drug therapy. The use of a long course of RIF/PZA/EMB led to a high rate of successful outcomes among patients with isoniazid-resistant tuberculosis. Additionally, patients receiving an FQ tolerated the drug well and had high rates of successful treatment outcomes. The cases of recurrence and acquired drug resistance among patients with isoniazid-resistant tuberculosis are concerning and highlight the need for longer-term follow-up studies.
Notes
Acknowledgments. We are indebted to Dr Aliya Yamin, Dr Omar Mohamed, and Sadie Sellers at the Fulton County Health Department; Bella Siangonya and Raquel Heidelburg at the Dekalb County Board of Health; Dr Rose-Marie F. Sales and Robin Connelly at the Georgia Department of Public Health; and nurses providing care for patients with tuberculosis in Georgia for their assistance with data collection.
Financial support. The study was funded in part by the National Institutes of Health National Institute of Allergy and Infectious Diseases (K23AI103044 to R. R. K.), and the Atlanta Clinical and Translational Science Institute (UL1TR000454).
Potential conflicts of interest. All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
References
- 1. Stagg HR, Lipman MC, McHugh TD, Jenkins HE. Isoniazid-resistant tuberculosis: a cause for concern? Int J Tuberc Lung Dis 2017; 21:129–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. World Health Organization. Global Tuberculosis Report 2014. Geneva, Switzerland: World Health Organization; 2014. [Google Scholar]
- 3. Manson AL, Cohen KA, Abeel T et al. TBResist Global Genome Consortium Genomic analysis of globally diverse Mycobacterium tuberculosis strains provides insights into the emergence and spread of multidrug resistance. Nat Genet 2017; 49:395–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Gegia M, Winters N, Benedetti A, van Soolingen D, Menzies D. Treatment of isoniazid-resistant tuberculosis with first-line drugs: a systematic review and meta-analysis. Lancet Infect Dis 2017; 17:223–34. [DOI] [PubMed] [Google Scholar]
- 5. Blumberg HM, Burman WJ, Chaisson RE et al. ; American Thoracic Society, Centers for Disease Control and Prevention, and the Infectious Diseases Society of America American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: treatment of tuberculosis. Am J Respir Crit Care Med 2003; 167:603–62. [DOI] [PubMed] [Google Scholar]
- 6. Nahid P, Dorman SE, Alipanah N et al. Official American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America clinical practice guidelines: treatment of drug-susceptible tuberculosis. Clin Infect Dis 2016; 63:e147–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Mitchison DA, Nunn AJ. Influence of initial drug resistance on the response to short-course chemotherapy of pulmonary tuberculosis. Am Rev Respir Dis 1986; 133:423–30. [DOI] [PubMed] [Google Scholar]
- 8. Five-year follow-up of a controlled trial of five 6-month regimens of chemotherapy for pulmonary tuberculosis. Hong Kong Chest Service/British Medical Research Council. Am Rev Respir Dis 1987; 136:1339–42. https://www.ncbi.nlm.nih.gov/pubmed/?term=Five-year+follow-up+of+a+controlled+trial+of+five+6-month+regimens+of+chemotherapy+for+pulmonary+tuberculosis.+Hong+Kong+Chest+Service%2FBritish+Medical+Research+Council.+Am+Rev+Respir+Dis+1987%3B+136%3A1339%E2%80%9342. [DOI] [PubMed] [Google Scholar]
- 9. Dooley KE, Mitnick CD, Ann DeGroote M et al. ; Efficacy Subgroup, RESIST-TB Old drugs, new purpose: retooling existing drugs for optimized treatment of resistant tuberculosis. Clin Infect Dis 2012; 55:572–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Donald PR. Chemotherapy for tuberculous meningitis. N Engl J Med 2016; 374:179–81. [DOI] [PubMed] [Google Scholar]
- 11. Escalante P, McKean-Cowdin R, Ramaswamy SV et al. Can mycobacterial katG genetic changes in isoniazid-resistant tuberculosis influence human disease features? Int J Tuberc Lung Dis 2013; 17:644–51. [DOI] [PubMed] [Google Scholar]
- 12. Otto-Knapp R, Vesenbeckh S, Schönfeld N et al. Isoniazid minimal inhibitory concentrations of tuberculosis strains with katG mutation. Int J Tuberc Lung Dis 2016; 20:1275–6. [DOI] [PubMed] [Google Scholar]
- 13. World Health Organization. WHO Treatment Guidelines for Drug-Resistant Tuberculosis—2016 Update. Geneva, Switzerland: World Health Organization; 2016. [PubMed] [Google Scholar]
- 14. Gosling RD, Uiso LO, Sam NE et al. The bactericidal activity of moxifloxacin in patients with pulmonary tuberculosis. Am J Respir Crit Care Med 2003; 168:1342–5. [DOI] [PubMed] [Google Scholar]
- 15. Jindani A, Harrison TS, Nunn AJ et al. ; RIFAQUIN Trial Team High-dose rifapentine with moxifloxacin for pulmonary tuberculosis. N Engl J Med 2014; 371:1599–608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Cattamanchi A, Dantes RB, Metcalfe JZ et al. Clinical characteristics and treatment outcomes of patients with isoniazid-monoresistant tuberculosis. Clin Infect Dis 2009; 48:179–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Chien JY, Chen YT, Wu SG, Lee JJ, Wang JY, Yu CJ. Treatment outcome of patients with isoniazid mono-resistant tuberculosis. Clin Microbiol Infect 2015; 21:59–68. [DOI] [PubMed] [Google Scholar]
- 18. Lee H, Jeong BH, Park HY et al. Treatment outcomes with fluoroquinolone-containing regimens for isoniazid-resistant pulmonary tuberculosis. Antimicrob Agents Chemother 2015; 60:471–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. US Department of Health and Human Services, Food and Drug Administration. Pulmonary Tuberculosis: Developing Drugs for Treatment. Rockville, MD: Center for Drug Evaluation and Research; 2013. [Google Scholar]
- 20. Kurbatova EV, Cegielski JP, Lienhardt C et al. Sputum culture conversion as a prognostic marker for end-of-treatment outcome in patients with multidrug-resistant tuberculosis: a secondary analysis of data from two observational cohort studies. Lancet Respir Med 2015; 3:201–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Powell KM, VanderEnde DS, Holland DP et al. Outbreak of drug-resistant Mycobacterium tuberculosis among homeless people in Atlanta, Georgia, 2008–2015. Public Health Rep 2017; 132:231–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Centers for Disease Control and Prevention (CDC). Tuberculosis genotyping—United States, 2004–2010. MMWR Morb Mortal Wkly Rep 2012; 61:723–5. [PubMed] [Google Scholar]
- 23. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009; 42:377–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Hosmer DW, Lemeshow S, May S. Applied Survival Analysis Regression Modeling of Time-to-Event Data. Hoboken, NJ: Wiley-Interscience, 2008. [Google Scholar]
- 25. Centers for Disease Control and Prevention (CDC). Reported Tuberculosis in the United States, 2015. Atlanta, GA: US Department of Health and Human Services; 2016. [Google Scholar]
- 26. World Health Organization. The Use of Molecular Line Probe Assay for the Detection of Resistance to Isoniazid and Rifampicin: Policy Update. Geneva, Switzerland: World Health Organization; 2016. [Google Scholar]
- 27. Curry International Tuberculosis Center, California Department of Public Health. Drug-Resistant Tuberculosis: A Survival Guide for Clinicians, 3rd ed. 2016. http://www.currytbcenter.ucsf.edu/sites/default/files/tb_sg3_front_matter.pdf#ch1intro [Google Scholar]
- 28. Escalante P, Graviss EA, Griffith DE, Musser JM, Awe RJ. Treatment of isoniazid-resistant tuberculosis in southeastern Texas. Chest 2001; 119:1730–6. [DOI] [PubMed] [Google Scholar]
- 29. Wang TY, Lin SM, Shie SS et al. Clinical characteristics and treatment outcomes of patients with low- and high-concentration isoniazid-monoresistant tuberculosis. PLoS One 2014; 9:e86316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Mitchison DA. Role of individual drugs in the chemotherapy of tuberculosis. Int J Tuberc Lung Dis 2000; 4:796–806. [PubMed] [Google Scholar]
- 31. Jindani A, Dore CJ, Mitchison DA. Bactericidal and sterilizing activities of antituberculosis drugs during the first 14 days. Am J Respir Crit Care Med 2003; 167:1348–54. [DOI] [PubMed] [Google Scholar]
- 32. Donald PR. The chemotherapy of tuberculous meningitis in children and adults. Tuberculosis (Edinb) 2010: 90:375–92. [DOI] [PubMed] [Google Scholar]
- 33. Vinnard C, King L, Munsiff S et al. The long-term mortality of tuberculosis meningitis patients in New York City: a cohort study. Clin Infect Dis 2017; 64:401–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Gumbo T. New susceptibility breakpoints for first-line antituberculosis drugs based on antimicrobial pharmacokinetic/pharmacodynamic science and population pharmacokinetic variability. Antimicrob Agents Chemother 2010; 54:1484–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Lanoix JP, Chaisson RE, Nuermberger EL. Shortening tuberculosis treatment with fluoroquinolones: lost in translation? Clin Infect Dis 2016; 62:484–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Hannah HA, Miramontes R, Gandhi NR. Sociodemographic and clinical risk factors associated with tuberculosis mortality in the United States, 2009–2013. Public Health Rep 2017; 132:366–375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Pasipanodya JG, McIlleron H, Burger A, Wash PA, Smith P, Gumbo T. Serum drug concentrations predictive of pulmonary tuberculosis outcomes. J Infect Dis 2013; 208:1464–73. [DOI] [PMC free article] [PubMed] [Google Scholar]