SUMMARY
SETTING
Nine months of daily isoniazid (9H) and 3 months of once-weekly rifapentine plus isoniazid (3HP) are recommended treatments for latent tuberculous infection (LTBI). The risk profile for 3HP and the contribution of hepatitis C virus (HCV) infection to hepatotoxicity are unclear.
OBJECTIVES
To evaluate the hepatotoxicity risk associated with 3HP compared to 9H, and factors associated with hepatotoxicity
DESIGN
Hepatotoxicity was defined as aspartate aminotransferase (AST) >3 times the upper limit of normal (ULN) with symptoms (nausea, vomiting, jaundice, or fatigue), or AST >5 × ULN. We analyzed risk factors among adults who took at least 1 dose of their assigned treatment. A nested case-control study assessed the role of HCV.
RESULTS
Of 6862 participants, 77 (1.1%) developed hepatotoxicity; 52 (0.8%) were symptomatic; 1.8% (61/3317) were on 9H and 0.4% (15/3545) were on 3HP (P < 0.0001). Risk factors for hepatotoxicity were age, female sex, white race, non-Hispanic ethnicity, decreased body mass index, elevated baseline AST, and 9H. In the case-control study, HCV infection was associated with hepatotoxicity when controlling for other factors.
CONCLUSION
The risk of hepatotoxicity during LTBI treatment with 3HP was lower than the risk with 9H. HCV and elevated baseline AST were risk factors for hepatotoxicity. For persons with these risk factors, 3HP may be preferred.
Keywords: isoniazid, hepatitis C, aspartate aminotransferases
APPROXIMATELY ONE THIRD of the world’s population is infected with latent Mycobacterium tuberculosis.1 One of the main strategies for tuberculosis (TB) control and elimination is the treatment of latent tuberculous infection (LTBI) among persons at highest risk for progression to TB disease.2–5 While LTBI treatment is generally safe and well-tolerated, hepatotoxicity occurs in 0.1–4% of persons after initiating treatment.6 Documented risk factors for hepatotoxicity include increasing age, elevated baseline transaminases, underlying liver disease, alcohol consumption, malnutrition, and being pregnant or in the immediate post-partum period.6 Hepatotoxicity may occur with all LTBI treatment regimens.6 Previous studies have found no consistent association between co-infection with hepatitis B or C virus and increased hepatotoxicity incidence during LTBI treatment.7 The incidence of hepatotoxicity associated with 3 months of rifapentine (P, RPT) +isoniazid (H, INH) in PREVENT TB was 0.4%;8 however, risk factors for hepatotoxicity among persons receiving this regimen have not been previously reported.
To provide a more complete picture of the hepatotoxicity risk associated with LTBI treatment, including the hepatotoxicity-specific safety profile of 3 months of directly observed once-weekly RPT + INH (3HP), we examined treatment interruptions or discontinuations associated with LTBI treatment among adults enrolled in the Tuberculosis Trials Consortium’s PREVENT TB study. We also conducted a nested case-control study to evaluate the role of co-infection with chronic hepatitis C virus (HCV) in treatment-limiting hepatotoxicity.
METHODS
Definition of hepatotoxicity and clinical monitoring
Methods and primary findings for the PREVENT TB study have been published previously.8 Analytic decisions made in determining risk factors (e.g., differentiating race from ethnicity) were similar between the main PREVENT TB study publication and this report to ensure comparability of published results. Study participants were enrolled from clinical trial sites in Brazil, Canada, Spain, and the United States. Participants from one PREVENT TB study site were excluded from this analysis due to discrepancies regarding receipt of study drug and directly observed therapy.
Monitoring for hepatotoxicity in the PREVENT TB study was primarily symptom-driven. Baseline and routine liver chemistries were drawn at the discretion of individual investigators per local guidelines and for participants at risk for hepatotoxicity.6 Participants were queried monthly about symptoms of nausea, vomiting, fatigue, or jaundice during study treatment. If hepatotoxicity was suspected, investigators were encouraged to query participants about alcohol use, and testing for viral hepatitis was suggested. Suspected hepatotoxicity events meeting the criteria of AST >3 times the upper limit of normal (ULN) with symptoms of nausea, vomiting, fatigue or jaundice, or AST >5 × ULN were systematically reported to the PREVENT TB study team at the US Centers for Disease Control and Prevention (CDC) via standardized adverse event case report forms. Toxicities were graded by local clinicians according to the Common Toxicity Criteria Version 2.9 Events of AST >3 × ULN with symptoms were considered grade 3 toxicities. Treatment-limiting hepatotoxicity was defined as the interruption of any study drug(s) for any length of time, including permanent discontinuation, as a result of suspected hepatotoxicity.
Nested case-control study methods
For the nested case-control study, eligible case participants were those in the PREVENT TB study aged >18 years who experienced treatment-limiting hepatotoxicity. The primary objective of the nested study was to assess the effect of HCV co-infection on the discontinuation of LTBI treatment due to hepatotoxicity.
Study enrollment for the nested case-control study was open from March 2002 to November 2009; 5.5% of adults enrolled in the PREVENT TB study were systematically selected as potential controls on enrollment into the PREVENT TB study. Age of potential control participants was evaluated for selection such that the mean age for all controls would be 45 years, as isoniazid (INH) associated hepatotoxicity increases with age. If a selected control developed treatment-limiting hepatotoxicity, the participant was no longer considered a control and could instead be enrolled as a case.
All cases and controls provided written informed consent. The study was approved by the institutional review boards (IRBs) at all study sites and the CDC, Atlanta, GA, USA. One site was not IRB approved for the nested study and contributed no cases or controls.
Both case and control participants in the nested study were queried regarding alcohol use and concomitant medications in the month prior to enrollment; blood samples of all participants were collected for viral hepatitis serological testing at CDC. Blood specimens were tested for immunoglobulin (Ig) M antibody to hepatitis A virus (IgM anti-HAV), hepatitis B virus (HBV) surface antigen (HBsAg), total and IgM antibody to hepatitis B core (anti-HBc), and antibody to hepatitis C virus (anti-HCV). All positive anti-HCV specimens were subsequently tested using a nucleic acid test for HCV RNA (AMPLICOR® Hepatitis C Virus Test, v2.0; Roche Molecular Systems Inc, Branchburg, NJ, USA) to ascertain current unresolved HCV infection. HBV infection was defined as a positive HBsAg and negative IgM anti-HBc. Current (acute or chronic) HCV infection was defined as a positive anti-HCV antibody and HCV RNA, if available.
Statistical methods
For the analysis of incidence and risk factors for treatment-limiting hepatotoxicity, all participants aged ≥18 years enrolled in the PREVENT TB study between June 2001 and February 2008 who took at least one dose of the study drug were included. Statistical analyses were performed using SAS version 9.2 (Statistical Analysis System Institute, Cary, NC, USA). Race was reported by all trial participants, whereas ethnicity was reported only by participants in the United States or Canada. For descriptive analyses, all races were included; however, for modeling analyses, Asian/Pacific Islander, Black, and White were separate and North American Indian and other races were combined due to small numbers of North American Indian participants. Alcohol use was defined as self-reported current or past drinking of beer, wine or other alcoholic beverage in any amount. Alcohol abuse was defined as any affirmative response to the CAGE alcohol assessment questionnaire.10 History of chronic liver disease was defined as any self-reported hepatitis B or C, hepatitis due to alcohol use, or hepatitis or cirrhosis of unknown cause.
We calculated hepatotoxicity incidence proportions and 95% confidence intervals (CIs) among adults in the PREVENT TB study. Incidence proportions were statistically compared using CIs. Using a Wilcoxon-Gehan test, Kaplan-Meier survival analysis was performed to compare days from treatment start to hepatotoxicity event. For comparisons of persons with and without hepatotoxicity, dichotomous variables were compared using Pearson’s χ2 or Fisher’s exact test. Multiplicative interactions of treatment regimen and participant factors were explored; none were statistically significant. Continuous variables were compared using Student’s t-test. For the analysis of risk factors for hepatotoxicity among adults in PREVENT TB, risk ratios (RRs) and 95%CIs were estimated using log-binomial regression. For the nested study, odds ratios (ORs) and 95%CIs were estimated using logistic regression. For multivariable modeling, all variables with P < 0.20 were entered into an initial model, and only those with P ≤ 0.05, through backwards selection, were retained in the final model. The multivariable model for the PREVENT TB study cohort omitted HCV co-infection, as these data were routinely collected only for nested study participants. The multivariable model for the case-control study excluded age because control participants were weighted by age and elevated baseline AST. Baseline AST was missing for 24% of the study population. We tested for co-linearity of risk factors included in multivariable models, and found that factors describing liver health (history of chronic liver disease, elevated baseline AST, and HCV co-infection) were co-linear. Final multivariable models included only the predictors that contributed to the best model fit as determined by the minimized Akaike information criterion. We estimated attributable risk percentages for statistically significant predictors in our multivariable logistic regression models.
RESULTS
A total of 6862 adult participants in the PREVENT TB study took at least one dose of the study treatment: respectively 3317 (48%) and 3545 (52%) were prescribed 9H (9 months of daily INH) and 3HP. Baseline AST evaluation was conducted for 76% of all study participants, regardless of study regimen, with baseline AST results reported for the majority of participants from all trial sites except Brazil and California, USA (data not shown). Repeated AST monitoring (i.e., at least two AST results reported) during the study phase was conducted for 37% of participants (2558/6862), 40% of those on 9H (1312/3317) and 35% of those on 3HP (1246/3545, RR 1.25, 95%CI 1.06–1.20; χ2 P = 0.0002).
Treatment-limiting hepatotoxicity occurred four times more frequently in those receiving 9H than 3HP among all participants (9H 1.9% vs. 3HP 0.4%, RR 4.42, 95%CI 2.52–7.75, χ2 P < 0.001), as well as among those with symptomatic (9H 1.3% vs. 3HP 0.3%, RR 4.51, 95%CI 2.27–8.97, χ2 P < 0.001) and asymptomatic (9H 0.6% vs. 3HP 0.1%, RR 4.10, 95%CI 1.53–11.0, χ2 P = 0.002) hepatotoxicity. In the nested study, there were 49 cases, of whom 35 were symptomatic (27 on 9H, 8 on 3HP) and 14 were asymptomatic (all on 9H), and 243 controls. No hospitalizations or deaths were associated with hepatotoxicity.
The clinical and demographic characteristics of the 6862 persons in the study population and the 292 participants enrolled in the nested study are shown in Appendix Table A.1.* Among the PREVENT TB study participants, those who developed any hepatotoxicity or symptomatic hepatotoxicity tended to be older, female, white, and non-Hispanic. In addition, most had elevated baseline AST, a history of chronic liver disease or alcohol abuse, reported current injection drug use or cigarette smoking, and were receiving 9H (Appendix Table A.1). Both cases and controls in the nested study were similar to non-enrolled cases and controls on all medical and social factors assessed (data not shown).
Within various clinical or demographic strata, persons who received 3HP consistently had a lower cumulative incidence of hepatotoxicity than those who received 9H (Appendix Table A.2). Persons administered 9H who developed hepatotoxicity received a median 4.5 mg/kg dose of INH compared to a 4.0 mg/kg dose received by those without hepatotoxicity (Wilcoxon signed-rank test P = 0.001). There were no significant differences in median mg/kg weekly dose of INH or RPT stratified by hepatotoxicity among persons on 3HP (hepatotoxicity, 10.9 mg/kg each of INH and RPT; no hepatotoxicity, 12.1 mg/kg each of INH and RPT, Wilcoxon signed-rank test P = 0.59).
The median number of days to the onset of treatment-limiting hepatotoxicity differed according to treatment regimen (9H median 101 days, inter-quartile range [IQR] 62–161; 3HP median 22 days, IQR 19–49; Wilcoxon-Gehan P < 0.001). Similarly, days to the onset of treatment-limiting symptomatic or asymptomatic hepatotoxicity also differed by treatment regimen (symptomatic 9H 97 days, IQR 62–155; 3HP 23 days, IQR 19–40, Wilcoxon-Gehan P < 0.001; asymptomatic 9H 105 days, IQR 59–183; 3HP 21 days, IQR 20–49, Wilcoxon-Gehan P < 0.001).
Permanent discontinuation of study treatment due to hepatotoxicity occurred more than five times more frequently among those taking 9H than 3HP (9H 1.7%, 3HP 0.3%, RR 5.54, 95%CI 2.91–10.5, χ2 P < 0.001). Among those who restarted treatment, the median length of treatment interruption in days did not vary by treatment regimen (9H 36 days, IQR 21–44; 3HP 15 days, IQR 13–30; Student’s t-test P = 0.48).
In the nested study, most cases (29/49, 59%) developed an AST 5–10 × ULN, and all events but one were grade 3 toxicities. The one serious adverse event (i.e., grade 4 toxicity) was in a Hispanic female participant assigned to 9H. She developed elevated AST (1025 U/l, 26× ULN), alanine aminotransferase (1468 U/l, 37× ULN) and bilirubin (2.9 mg/dl, 2.4× ULN) levels, with nausea, dark urine and pruritus after 89 doses of 9H; LTBI treatment was permanently discontinued. The majority of the cases presented with at least one symptom (37/49, 76%), of which fatigue was the most common (29/37, 76%).
In both the PREVENT TB and the nested study, female sex, history of chronic liver disease, elevated AST at LTBI treatment start, current or past alcohol use, and treatment with 9H were significantly associated with hepatotoxicity or symptomatic hepatotoxicity (Appendix Tables A.3 and A.4) on univariable analyses. Furthermore, increasing age and current or past injection drug use were associated with hepatotoxicity and symptomatic hepatotoxicity, ethnicity was associated with hepatotoxicity, and race with symptomatic hepatotoxicity in the PREVENT TB study, and with HCV co-infection with treatment-limiting hepatotoxicity in the nested study (Appendix Tables A.3 and A.4).
In a multivariable analysis of risk factors for hepatotoxicity in the PREVENT TB study, increasing age, female sex, White race, non-Hispanic ethnicity, low body mass index (BMI), history of chronic liver disease, elevated AST at LTBI treatment start, and receipt of 9H were independently associated with an increased risk of hepatotoxicity (Appendix Table A.2). The combined attributable risk for the presence of any of these factors was 56% (95%CI 50–61). In an analysis of symptomatic hepatotoxicity, the same risk factors were associated with hepatotoxicity, except low BMI (Table), with a combined attributable risk of 61% (95%CI 54–66). In a multivariable analysis of risk factors in the nested study, when controlling for factors significant in the parent study except for age, HCV co-infection was an independent risk factor for hepatotoxicity, with an attributable risk of 51% (Appendix Table A.4).
Table.
Multivariable analyses of risk factors for treatment-limiting hepatotoxicity* in the PREVENT TB study
| All hepatotoxicity
|
Symptomatic hepatotoxicity
|
|||||
|---|---|---|---|---|---|---|
| RR (95%CI)† | P value† | Attributable risk % | RR (95%CI)† | P value† | Attributable risk % | |
| Demographic factors | ||||||
| Age, per year increase | 1.03 (1.02–1.05) | <0.0001 | — | 1.03 (1.01–1.05) | 0.002 | |
| Female sex | 2.70 (1.65–4.42) | 0.0001 | 52 | 3.34 (1.70–6.28) | 0.0003 | 64 |
| Race | <0.0001 | 0.008 | ||||
| White | 1.00 (reference) | 38 | 1.00 (reference) | 57 | ||
| Black | 0.32 (0.17–0.62) | 0.25 (0.11–0.55) | ||||
| Asian | 0.13 (0.04–0.43) | Not estimated | ||||
| Other | 1.32 (0.60–3.22) | 0.87 (0.27–2.87) | ||||
| Ethnicity | 0.01 | 0.001 | ||||
| Hispanic | 1.00 (reference) | 1.00 (reference) | ||||
| Non-Hispanic | 2.22 (1.28–3.85) | 42 | 3.57 (1.86–6.88) | 40 | ||
| Not applicable (not US or Canada) | 0.87 (0.11–6.76) | Not estimated | ||||
| Health factors | ||||||
| BMI, per km/m2 increase | 0.94 (0.91–0.99) | 0.008 | — | — | — | — |
| Elevated baseline AST | 5.57 (3.31–9.37) | <0.0001 | 80 | 3.19 (1.53–6.63) | 0.002 | 66 |
| Anti-tuberculosis treatment factors | ||||||
| Treatment with 9H | 4.55 (2.53–8.18) | <0.0001 | 77 | 4.22 (2.10–8.47) | <0.0001 | 78 |
Hepatotoxicity was defined as serum AST >3 × ULN with symptoms of nausea, vomiting, jaundice or fatigue or AST >5 × ULN regardless of symptoms.
P values are from the Wald χ2 test. RRs are estimated through log-binomial regression. Only participant factors with a univariable P < 0.20 were entered into the multivariable model, and only factors with a P ≤ 0.05 were retained in the final multivariable model. For the multivariable model, P values without a RR indicate terms that were entered into the model but not retained in the final model.
RR =risk ratio; CI =confidence interval; BMI =body mass index; AST =aspartate aminotransferase; 9H =9 months of self-administered daily isoniazid at 5–15 mg/kg rounded to the nearest 50 mg, with a maximum dose of 300 mg; ULN = upper limit of normal.
DISCUSSION
Our study provides contemporary estimates of the incidence of and risk factors for hepatotoxicity during treatment for LTBI with 9H or 3HP under clinical trial conditions, and suggests that 3HP is less hepatotoxic than 9H and may be preferred in persons at increased risk of hepatotoxicity. The incidence of hepatotoxicity in persons treated with 9H was comparable to rates previously reported for daily INH.11,12 In contrast, the incidence of any hepatotoxicity or symptomatic hepatotoxicity among persons treated with 3HP in this study was significantly lower than that among those who received 9H. Although rare, severe liver injury and death are well-documented adverse effects of anti-tuberculosis drugs used for LTBI treatment, particularly INH.12,13 The participants in the PREVENT TB clinical trial had frequent interactions with health care providers, which may explain why more severe toxicity was not observed.14
The data from the PREVENT TB study allowed us to characterize the incidence of treatment-limiting hepatotoxicity and associated risk factors. The case-control study allowed us to focus on the specific role of HCV co-infection in treatment-limiting hepatotoxicity. Data from the PREVENT TB study demonstrated that participants with a history of chronic liver disease or elevated AST upon initiating LTBI treatment have an increased incidence of hepatotoxicity. This incidence was substantial for patients receiving 9H, but lower among those receiving 3HP. Also consistent with other studies, we found hepatotoxicity to be associated with increasing age, elevated AST at LTBI treatment start, and 9H.2,6 The nested study demonstrated that HCV co-infection was independently associated with hepatotoxicity. This association had been previously documented only among injection drug users, but even among this population the association has not been a consistent finding.15,16 Furthermore, female sex, white race, and non-Hispanic ethnicity were identified as risk factors for hepatotoxicity, findings that have not been commonly described.17
Our results, combined with this prior work, suggest that close clinical monitoring is required, not only of persons of increased age and those with a history of chronic liver disease, but also of women and persons of White race and non-Hispanic ethnicity. Although lower BMI was found to be a risk factor, the median BMI of 26 kg/m2 among persons with hepatotoxicity was not consistent with malnutrition.
Our study had two main limitations. First, AST monitoring differed by treatment regimen, with those on 9H being more likely to have two or more AST evaluations. There was thus potential for undetected asymptomatic hepatotoxicity, particularly among those on 3HP. Our RR estimates for 9H with any hepatotoxicity may therefore have been an overestimation. However, the RR estimates for 9H with any or symptomatic hepatotoxicity are very similar, suggesting minimal overestimation of the RR in our study. Second, this study was conducted in the context of a clinical trial. As patient selection (and self-selection) and monitoring may not reflect populations treated in operational settings, our observed incidence of hepatotoxicity might not accurately represent findings in program conditions.
LTBI treatment in all patients should include a discussion and evaluation of the risk of drug-induced liver injury including hepatotoxicity, including chronic ethanol consumption, known viral hepatitis, pre-existing liver disease, post-partum within 3 months, concomitant hepatotoxic medication and previous abnormal liver enzymes.6 Furthermore, evaluation of AST at treatment initiation may help identify persons at increased risk of hepatotoxicity.
The possibility of severe hepatitis and liver dysfunction should be considered when deciding whether a patient with HCV is a candidate for LTBI treatment. If treatment is initiated, an adverse event clinical monitoring plan should be put in place. Given that 3HP is as effective as 9H,8,18 these results add to the growing body of evidence that 3HP may be a preferred choice for the treatment of LTBI, particularly among those patients who are at high risk of hepatotoxicity.
Acknowledgments
The authors thank all enrolled persons for participation in the study; W Bower, L Bozeman, W el Sadr, S Goldberg, W Kuhnert, L Podewils, K Robergeau-Hunt, and P Spradling for their contributions to the protocol and study management; A Borisov and N Scott for assistance with data management; and N Khudyakov and M M Khin who conducted serologic and molecular testing.
This work was supported by the US Centers for Disease Control and Prevention (CDC). Sanofi donated the RPT used in this study, and since 2007 has donated US$2.3 million to the CDC Foundation to supplement available US federal funding for RPT research. Sanofi did not participate in the design of the study; in the collection, analysis, or interpretation of data; in the writing of this manuscript; or in the decision to submit this manuscript for publication.
APPENDIX
Table A.1.
Demographic and clinical characteristics of adults enrolled in PREVENT TB and the nested case-control study
| PREVENT TB
|
Case-control study
|
|||||
|---|---|---|---|---|---|---|
| Hepatotoxicity* (n = 77)†
n (%) |
Symptomatic hepatotoxicity* (n = 52) n (%) |
Asymptomatic hepatotoxicity* (n = 24) n (%) |
No hepatotoxicity* (n = 6785) n (%) |
Cases (n = 49) n (%) |
Controls (n = 243) n (%) |
|
| Demographic factors | ||||||
| Age, years, median [IQR] | 44 [34–52] | 44 [33–52] | 47 [37–50] | 37 [28–48] | 44 [33–52] | 46 [32–53] |
| Female sex | 49 (64) | 36 (69) | 12 (50) | 3065 (45) | 33 (67) | 107 (44) |
| Race | ||||||
| White | 52 (68) | 39 (75) | 13 (54) | 3799 (56) | 35 (67) | 121 (60) |
| Black | 13 (17) | 8 (15) | 5 (21) | 1780 (26) | 10 (20) | 76 (31) |
| Asian/Pacific Islander | 6 (7.8) | 2 (3.9) | 4 (17) | 887 (13) | 4 (8.2) | 32 (13) |
| North American Indian | 2 (2.6) | 1 (1.9) | 0 (0) | 109 (1.6) | 0 (0) | 7 (2.9) |
| Other | 4 (5.2) | 2 (3.9) | 2 (8.3) | 210 (3.1) | 2 (4.1) | 7 (2.9) |
| Ethnicity | ||||||
| Hispanic | 25 (32) | 16 (31) | 9 (37) | 2666 (39) | 11 (22) | 62 (25) |
| Non-Hispanic | 51 (66) | 36 (69) | 14 (58) | 3591 (53) | 37 (76) | 144 (59) |
| Not applicable (not in the United States or Canada) | 1 (1.3) | 0 (0) | 1 (4.2) | 528 (7.8) | 1 (2.0) | 37 (15) |
| Health factors | ||||||
| BMI, kg/m2, median [IQR] | 26 [23–28] | 26 [23–28] | 25 [23–28] | 27 [24–31] | 26 [23–30] | 28 [24–31] |
| HIV status | ||||||
| Non-infected | 41 (53) | 25 (48) | 15 (63) | 3366 (50) | 31 (63) | 132 (54) |
| Infected | 2 (2.6) | 1 (1.9) | 1 (4.2) | 149 (2.2) | 2 (4.1) | 9 (3.7) |
| Unknown | 34 (44) | 26 (50) | 8 (33) | 3270 (48) | 16 (33) | 102 (42) |
| History of chronic liver disease | 15 (19) | 10 (19) | 5 (21) | 322 (4.8) | 11 (20) | 16 (6.6) |
| Elevated baseline AST | 21/72 (29) | 9/48 (19) | 12/23 (52) | 369/5151 (7.2) | 11/46 (24) | 15/168 (8.9) |
| Hepatitis serology results‡ | ||||||
| Hepatitis B virus co-infection | — | — | — | — | 0 (0) | 3 (1.2) |
| Hepatitis C virus co-infection | — | — | — | — | 12 (24) | 23 (9.5) |
| Self-reported social factors | ||||||
| Homeless >6 months | 6 (7.8) | 5 (9.6) | 1 (4.2) | 495 (7.3) | 4 (8.2) | 24 (9.9) |
| Unemployed >12 months | 10 (13) | 7 (13) | 3 (13) | 769 (11) | 7 (14) | 25 (10) |
| Correctional institute >1 month | 7 (9.1) | 3 (5.8) | 4 (17) | 377 (5.6) | 3 (6.1) | 18 (7.4) |
| Current or past alcohol use | ||||||
| None | 28 (36) | 22 (42) | 6 (25) | 3107/6777 (46) | 17 (35) | 119 (49) |
| Use | 36 (47) | 21 (40) | 14 (58) | 3147/6777 (47) | 24 (49) | 110 (45) |
| Abuse | 13 (17) | 9 (17) | 4 (17) | 469/6777 (6.9) | 8 (16) | 14 (5.8) |
| Current or past IDU | 10 (13) | 8 (15) | 2 (8.3) | 254/6776 (3.8) | 5 (10) | 15 (6.2) |
| Current cigarette smoker | 28 (36) | 18 (35) | 10 (42) | 1977 (29) | 19 (39) | 89 (35) |
| Anti-tuberculosis treatment factors | ||||||
| Indication for TLTBI§ | ||||||
| Close contact | 50 (65) | 31 (60) | 19 (79) | 4711 (69) | 29 (59) | 181 (74) |
| Recent converter | 25 (32) | 19 (37) | 5 (21) | 1798 (27) | 18 (37) | 50 (21) |
| HIV-infected | 0 (0) | 0 (0) | — | 107 (1.6) | 0 (0) | 8 (3.3) |
| Fibrosis on chest X-ray | 2 (2.6) | 2 (3.9) | — | 169 (2.5) | 2 (4.1) | 4 (1.7) |
| 9H | 62 (81) | 42 (81) | 19 (80) | 3353 (48) | 41 (84) | 95 (39) |
Hepatotoxicity was defined as serum AST >3 × ULN with symptoms of nausea, vomiting, jaundice or fatigue, or AST >5 × ULN regardless of symptoms.
Symptoms were not assessed for one participant.
Hepatitis B and C serological and NAT tests were performed systematically only for participants in the nested case-control study.
Subjects were counted only once in the order presented. The total number of HIV-infected persons who were enrolled is listed separately in this table.
BMI = body mass index; HIV = human immunodeficiency virus; AST = aspartate aminotransferase; IDU = injection drug user; TLTBI = treatment for latent tuberculous infection; 9H =9 months of self-administered daily isoniazid at 5–15 mg/kg rounded to the nearest 50 mg, with a maximum dose of 300 mg; ULN = upper limit of normal; NAT = nucleic acid testing.
Table A.2.
Treatment-limiting hepatotoxicity cumulative incidence (95%CI) by treatment arm among adults enrolled in the Tuberculosis Trials Consortium PREVENT TB study
| All
|
9H
|
3HP
|
||||
|---|---|---|---|---|---|---|
| n/N | % (95%CI) | n/N | % (95%CI) | n/N | % (95%CI) | |
| Overall* | 77/6862 | 1.1 (0.9–1.4) | 61/3317 | 1.8 (1.4–2.3) | 15/3545 | 0.4 (0.2–0.6) |
| Treatment permanently discontinued | 68/6862 | 1.0 (0.8–1.2) | 57/3317 | 1.7 (1.3–2.2) | 11/3545 | 0.3 (0.1–0.5) |
| Symptomatic | 52/6862 | 0.8 (0.6–1.0) | 42/3317 | 1.3 (0.9–1.6) | 10/3545 | 0.3 (0.1–0.5) |
| Asymptomatic | 24/6862 | 0.3 (0.2–0.5) | 19/3317 | 0.6 (0.3–0.8) | 5/3545 | 0.1 (0.02–0.3) |
| Demographic factors | ||||||
| Female sex | 49/3114 | 1.6 (1.1–2.0) | 40/1528 | 2.6 (1.8–3.4) | 9/1586 | 0.6 (0.2–0.9) |
| Male sex | 28/3748 | 0.7 (0.5–1.0) | 22/1789 | 1.2 (0.7–1.7) | 6/1959 | 0.3 (0.1–0.6) |
| Race | ||||||
| White | 52/3851 | 1.4 (1.0–1.7) | 44/1873 | 2.3 (1.7–3.0) | 8/1978 | 0.4 (0.1–0.7) |
| Black | 13/1793 | 0.7 (0.3–1.1) | 8/873 | 0.9 (0.3–1.5) | 5/920 | 0.5 (0.1–1.0) |
| Asian/Pacific Islander | 6/893 | 0.7 (0.1–1.2) | 5/443 | 1.1 (0.1–2.1) | 1/450 | 0.2 (0–0.7) |
| North American Indian | 2/111 | 1.8 (0–4.3) | 2/31 | 6.5 (0–15) | 0/80 | 0 |
| Other | 4/214 | 1.9 (0–3.7) | 3/97 | 3.1 (0–6.5) | 1/117 | 0.9 (0–2.5) |
| Ethnicity | ||||||
| Hispanic | 25/2691 | 0.9 (0.6–1.3) | 21/1283 | 1.6 (0.9–3.0) | 4/1408 | 0.3 (0–0.6) |
| Not Hispanic | 51/3642 | 1.4 (1.0–1.8) | 40/1770 | 2.3 (1.6–3.0) | 11/1872 | 0.6 (0.2–0.9) |
| Not applicable (not US or Canada) | 1/529 | 0.2 (0–0.6) | 1/264 | 0.4 (0–1.1) | 0/265 | 0 |
| Health factors | ||||||
| HIV status | ||||||
| Non-infected | 41/3407 | 1.2 (0.8–1.5) | 32/1688 | 1.9 (1.2–2.5) | 9/1719 | 0.5 (0.2–0.9) |
| Infected | 2/151 | 1.3 (0–3.1) | 1/78 | 1.3 (0–3.8) | 1/73 | 1.4 (0–4.0) |
| Unknown | 34/3304 | 1.0 (0.7–1.4) | 29/1551 | 1.9 (1.2–2.5) | 5/1753 | 0.3 (0–0.5) |
| History of chronic liver disease | 15/337 | 4.5 (2.2–6.7) | 13/171 | 7.6 (3.6–12) | 2/166 | 1.2 (0–2.9) |
| Elevated baseline AST | 21/390 | 5.4 (3.1–7.6) | 17/188 | 9.0 (4.9–13.0) | 4/202 | 2.0 (0.1–3.9) |
| Self-reported social factors | ||||||
| Homeless >6 months | 6/501 | 1.2 (0.2–2.2) | 6/214 | 2.8 (0.6–5.0) | 0/287 | 0 |
| Unemployed >12 months | 10/779 | 1.3 (0.5–2.1) | 9/369 | 2.4 (0.9–4.0) | 1/410 | 0.2 (0–0.7) |
| Correctional institute >1 month | 7/384 | 1.8 (0.5–3.2) | 6/169 | 3.6 (0.8–6.3) | 1/215 | 0.5 (0–1.4) |
| Current or past alcohol use | ||||||
| None | 28/3135 | 0.9 (0.6–1.2) | 21/1475 | 1.4 (0.8–2.0) | 7/1660 | 0.4 (0.1–0.7) |
| Use | 36/3210 | 1.1 (0.8–1.5) | 30/1616 | 1.9 (1.2–2.5) | 6/1594 | 0.4 (0.1–0.7) |
| Abuse | 13/509 | 2.6 (1.2–3.9) | 11/222 | 5.0 (2.1–7.8) | 2/287 | 0.7 (0–1.7) |
| Current or past IDU | 10/264 | 4.0 (1.7–6.3) | 8/130 | 6.2 (2.0–10) | 2/134 | 1.5 (0–3.5) |
| Current cigarette smoker | 28/2005 | 1.4 (0.9–1.9) | 24/959 | 2.5 (1.5–3.5) | 4/1046 | 0.4 (0–0.8) |
| TB treatment factors | ||||||
| Indication for TLTBI† | ||||||
| Close contact | 50/4761 | 1.1 (0.8–1.3) | 42/2252 | 1.9 (1.3–2.4) | 8/2506 | 0.3 (0.1–0.5) |
| Recent converter | 25/1823 | 1.4 (0.8–1.9) | 18/924 | 1.9 (1.1–2.8) | 7/899 | 0.8 (0.2–1.4) |
| HIV-infected | 0/107 | 0 | 0/52 | 0 | 0/55 | 0 |
| Fibrosis on chest X-ray | 2/171 | 1.2 (0–2.8) | 2/89 | 2.2 (0–5.3) | 0/82 | 0 |
Symptoms were not assessed for one participant.
Subjects were counted only once in the order presented. The total number of HIV-infected persons who were enrolled is listed separately in this table.
CI=confidence interval; 9H=9 months of self-administered daily isoniazid at 5–15 mg/kg rounded to the nearest 50 mg, with a maximum dose of 300 mg; 3HP= 3 months of directly-observed once weekly 900 mg rifapentine plus isoniazid at 15–25 mg/kg rounded to the nearest 50 mg, with a maximum dose of 900 mg; HIV = human immunodeficiency virus; AST =aspartate aminotransferase; IDU = injection drug user; TLTBI = treatment for latent tuberculous infection.
Table A.3.
Univariable analyses of risk factors for treatment-limiting hepatotoxicity in the PREVENT TB study
| All hepatotoxicity*
|
Symptomatic hepatotoxicity*
|
|||
|---|---|---|---|---|
| RR (95%CI)† | P value† | RR (95%CI)† | P value† | |
| Demographic factors | ||||
| Age, per year increase | 1.03 (1.01–1.04) | <0.0001 | 1.03 (1.01–1.04) | 0.002 |
| Female sex | 2.11 (1.33–3.34) | 0.002 | 2.71 (1.51–4.88) | 0.009 |
| Race | 0.07 | 0.04 | ||
| White | 1.00 (reference) | 1.00 (reference) | ||
| Black | 0.54 (0.30–0.98) | 0.44 (0.21–0.94) | ||
| Asian | 0.50 (0.21–1.15) | 0.22 (0.05–0.92) | ||
| Other | 1.37 (0.59–3.16) | 0.92 (0.28–2.95) | ||
| Ethnicity | 0.04 | 0.09 | ||
| Hispanic | 1.00 (reference) | 1.00 (reference) | ||
| Non-Hispanic | 1.51 (0.94–2.42) | 1.66 (0.92–3.00) | ||
| Not applicable (not US or Canada) | 0.20 (0.03–1.50) | |||
| Health factors | ||||
| BMI, per kg/m2 increase | 0.97 (0.93–1.01) | 0.10 | 0.97 (0.93–1.02) | 0.24 |
| HIV status | 0.77 | 0.96 | ||
| Non-infected | 1.00 (reference) | 1.00 (reference) | ||
| Infected | 1.10 (0.27–4.51) | 0.90 (0.12–6.63) | ||
| Unknown | 0.86 (0.54–1.34) | 1.07 (0.62–1.85) | ||
| History of chronic liver disease | 4.68 (2.69–8.15) | <0.0001 | 4.67 (2.36–9.22) | <0.0001 |
| Elevated baseline AST | 5.10 (3.10–8.39) | <0.0001 | 2.94 (1.44–6.03) | 0.003 |
| Social factors | ||||
| Homeless >6 months | 1.07 (0.47–2.46) | 0.87 | 1.34 (0.54–3.37) | 0.52 |
| Unemployed >12 months | 1.17 (0.60–2.26) | 0.65 | 1.22 (0.54–2.68) | 0.63 |
| Correctional institute >1 month | 1.69 (0.78–3.64) | 0.18 | 1.04 (0.33–3.32) | 0.95 |
| Current or past alcohol use | 0.006 | 0.03 | ||
| None | 1.00 (reference) | 1.00 (reference) | ||
| Use | 1.26 (0.77–2.05) | 0.93 (0.52–1.70) | ||
| Abuse | 2.86 (1.49–5.48) | 2.53 (1.17–5.47) | ||
| Current or past IDU | 3.73 (1.94–7.16) | <0.0001 | 4.56 (2.17–9.58) | <0.0001 |
| Current cigarette smoker | 1.38 (0.87–2.20) | 0.17 | 1.28 (0.72–2.27) | 0.39 |
| TB treatment factors | ||||
| Indication for TLTBI‡ | 0.55 | 0.23 | ||
| Close contact | 1.00 (reference) | 1.00 (reference) | ||
| Recent converter | 1.31 (0.81–2.10) | 1.60 (0.91–2.82) | ||
| Fibrosis on chest X-ray | 1.11 (0.27–4.54) | 1.79 (0.43–7.41) | ||
| Treatment with 9H | 4.42 (2.52–7.75) | <0.0001 | 4.51 (2.27–8.97) | <0.0001 |
Hepatotoxicity was defined as serum AST >3 × ULN with symptoms of nausea, vomiting, jaundice or fatigue or AST >5 × ULN regardless of symptoms.
P values are from the Wald χ2 test. RRs are estimated through log-binomial regression.
Subjects were counted only once in the order presented. The total number of HIV-infected persons who were enrolled is listed separately in this table.
RR =risk ratio; CI =confidence interval; BMI =body mass index; HIV =human immunodeficiency virus; AST =aspartate aminotransferase; IDU =injection drug user; TLTBI =treatment for latent tuberculous infection; 9H =9 months of self-administered daily isoniazid at 5–15 mg/kg rounded to the nearest 50 mg, with a maximum dose of 300 mg; ULN = upper limit of normal.
Table A.4.
Univariable and multivariable analyses of risk factors for treatment-limiting hepatotoxicity* in the nested case-control study
| Univariable analysis
|
Multivariable analysis
|
||||
|---|---|---|---|---|---|
| OR (95%CI)† | P value† | OR (95%CI)† | P value† | Attributable risk‡ % | |
| Demographic factors | |||||
| Age, per year increase | 1.00 (0.98–1.02) | 0.97 | |||
| Female sex | 2.62 (1.37–5.01) | 0.003 | 2.75 (1.28–5.91) | 0.001 | 50 |
| Race | 0.18 | 0.01 | |||
| White | 1.00 (reference) | 1.00 (reference) | 41 | ||
| Black | 0.48 (0.23–1.04) | 0.23 (0.08–0.63) | |||
| Asian | 0.46 (0.15–1.39) | 0.19 (0.05–0.72) | |||
| Other | 0.52 (0.11–2.42) | 0.97 (0.15–6.47) | |||
| Ethnicity | 0.07 | 0.005 | |||
| Hispanic | 1.00 (reference) | 1.00 (reference) | 43 | ||
| Non-Hispanic | 1.45 (0.69–3.02) | 2.97 (1.13–7.86) | |||
| Not applicable (not US or Canada) | 0.15 (0.02–1.23) | 0.14 (0.02–1.30) | |||
| Health factors | |||||
| BMI, per kg/m2 increase | 0.95 (0.90–1.01) | 0.08 | 0.92 (0.86–0.99) | 0.02 | |
| HIV status | 0.48 | ||||
| Non-infected | 1.00 (reference) | ||||
| Infected | 0.95 (0.20–4.60) | ||||
| Unknown | 0.68 (0.35–1.29) | ||||
| History of chronic liver disease | 3.64 (1.54–8.60) | 0.005 | — | 0.79 | |
| Elevated baseline AST | 3.21 (1.36–7.58) | 0.006 | |||
| Hepatitis B virus co-infection | 0.69 (0.04–13.7) | 1.00 | |||
| Hepatitis C virus co-infection | 3.10 (1.42–6.77) | 0.003 | 3.24 (1.12–9.34) | 0.03 | 51 |
| Social factors | |||||
| Homeless >6 months | 0.81 (0.27–2.45) | 0.71 | |||
| Unemployed >12 months | 1.45 (0.59–3.58) | 0.42 | |||
| Correctional institute >1 month | 0.82 (0.25–2.88) | 0.75 | |||
| Current or past alcohol use | 0.03 | — | 0.06 | ||
| None | 1.00 (reference) | — | |||
| Use | 1.57 (0.78–2.99) | — | |||
| Abuse | 4.00 (1.46–10.9) | — | |||
| Current or past IDU | 1.72 (0.59–4.98) | 0.31 | — | 0.10 | |
| Current cigarette smoker | 1.16 (0.61–2.18) | 0.65 | |||
| Anti-tuberculosis treatment factors | |||||
| Indication for TLTBI§ | 0.04 | — | 0.94 | ||
| Close contact | 1.00 (reference) | — | |||
| Recent converter | 2.25 (1.15–4.38) | — | |||
| Fibrosis on chest X-ray | 3.12 (0.55–17.8) | — | |||
| Treatment with 9H | 7.98 (3.59–17.8) | <0.0001 | 9.20 (3.79–22.4) | <0.0001 | 79 |
Hepatotoxicity was defined as serum AST >3 × ULN with symptoms of nausea, vomiting, jaundice or fatigue or AST >5 × ULN regardless of symptoms.
P values are from the Wald χ2 test. ORs are estimated through logistic regression. Only participant factors with a univariable P < 0.20 were entered into the multivariable model, and only factors with a P ≤0.05 were retained in the final multivariable model. For the multivariable model, P values without an OR indicate terms that were entered into the model but not retained in the final model. Elevated baseline ASTwas not included in the multivariable model as data were missing in 27% of case-control participants.
Calculated for White race and non-Hispanic ethnicity only, not all races and ethnicities.
Subjects were counted only once in the order presented. The total number of HIV-infected persons who were enrolled is listed separately in this table.
OR =odds ratio; CI =confidence interval; BMI =body mass index; HIV =human immunodeficiency virus; AST =aspartate aminotransferase; IDU =injection drug user; TLTBI =treatment for latent tuberculous infection; 9H =9 months of self-administered daily isoniazid at 5–15 mg/kg rounded to the nearest 50 mg, with a maximum dose of 300 mg; ULN = upper limit of normal.
Footnotes
The appendix is available in the online version of this article, at http://www.ingentaconnect.com/content/iuatld/ijtld/2015/00000019/00000009/art00008
Conflicts of interest: Potential conflicts of interest and financial disclosure: TRS: Sanofi consultation for data presented to the Food and Drug Administration (Washington DC, USA). Otsuka (Tokyo, Japan): member of data safety monitoring board for tuberculosis clinical trial. All other authors report no conflicts of interest apart from those already disclosed above.
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC.
Study sites (number of patients enrolled), principal investigators and study coordinators:
University of North Texas Health Science Center (UNTHSC), Fort Worth, TX, USA (1171 in PREVENT TB, 51 in substudy): S E Weis, M Fernandez, B King, L Turk, N Shafer, G Stevenson, G Bayona, R Dean, J Helal, G Burgess; Hospital Universitario Clementino Fraga Filho, Rio de Janeiro, RJ, Brazil and Johns Hopkins University, Baltimore, MD, USA (529 in PREVENT TB, 38 in substudy): M B Conde, F C Q Mello, A Efron, C Loredo, M B S Fortuna, M Cailleaux-Cezar, R L Guerra, G Mota, C Felix, C dos Santos Sacramento; South Texas Department of State Health Services (Austin, TX) Region 11 Clinics (516 in PREVENT TB, not IRB-approved for substudy): R Wing, D Wing, D Valenzuela, J Gonzalez, J Uribe, B R Smith; Audie L Murphy Veterans Affairs Hospital, San Antonio, TX (493 in PREVENT TB, 14 in substudy): M Weiner, M Engle, J A Jimenez, H Pavon, V Rodriguez, K B. West, D Dooley, D Hospenthal; Vanderbilt University Medical Center and Nashville Metro Public Health Department, Nashville, TN, (361 in PREVENT TB, 16 in substudy): T Sterling, L R Hammock, A Kerrigan, B Redd, I Montgomery, K Miller, G-S McKee, D Freeman; University of Southern California, Los Angeles, CA (319 in PREVENT TB, 14 in substudy): B E Jones, P Escalante, P Molina, C Silva, A Grbic, M Brown, B Oamar, E Rayos, C Luken; Columbia University College of Physicians and Surgeons, New York, NY (310 in PREVENT TB, 5 in substudy): N W Schluger, J Burzynski, V Lozano, M Wolk; University of Medicine and Dentistry New Jersey, Newark, NJ (339 in PREVENT TB, 18 in substudy): B T Mangura, L B Reichman, G McSherry, A Lardizabal, M C Leus, M Owens, E Napolitano, L Kellert, V Anokute; Denver Public Health Department, Denver, CO (304 in PREVENT TB, 21 in substudy): W Burman, R Reves, R Belknap, D Cohn, J Tapy, G Sanchez, L Luna; University of California, San Diego Medical Center, San Diego, CA, USA (269 in PREVENT TB, 11 in substudy); A Catanzaro, P LoBue, K Moser, M Tracy, P Francisco, J Davis; Montreal Chest Institute, Montreal, QC, Canada (267 in PREVENT TB, 9 in substudy): R I Menzies, K Schwartzman, C Greenaway, L Lands, S Mannix, P Brassard, B Mortezai, B Rabinovitch, M Pelletier, C Valiquette, J Tremblay, P A Plaisir, R Binet; Public Health – Seattle and King County Public Health, Seattle, WA, USA (227 in PREVENT TB, 11 in substudy): M Narita, C M Nolan, S Goldberg, D Schwartz, L Deretsky, M Stone, C Friedly; Agencia de Salut Publica, Barcelona, Spain and UNTHSC (246 in PREVENT TB, 14 in substudy); J A Cayla, J M. Miró, M Antonia Sambeat, J L López Colomés, J A Martinez, X Martinez-Lacasa, A Orcau, P Sanchez, C Tortajada, I Ocana, J P Millet, A Moreno, J Nelson, O Sued, L de Souza, M A Jiménez, L del Baño, L Fina, L Roldan, A Romero; University of California, San Francisco, San Francisco, CA (199 in PREVENT TB, 4 in substudy): P Nahid, P Hopewell, C Daley, R Jasmer, C Merrifield, W Stanton, I Rudoy, J Israel; Johns Hopkins University, Baltimore, MD (210 in PREVENT TB, 20 in substudy): R E Chaisson, S E Dorman, J Hackman, G Maltas, J Fisher; Duke University/FHI 360, Durham, NC (183 in PREVENT TB, 12 in substudy): C D Hamilton, J Stout, A Mosher, E J Hecker, B Ho, E Rich; Boston University Medical Center, Boston, MA (171 in PREVENT TB, 0 in substudy): J Bernardo, J Saukkonen, C Murphy, D Brett-Curran; Edward Hines Jr VA Medical Center, Chicago, IL (174 in PREVENT TB, 11 in substudy): C T Pachucki, A Lee, S Marantz, M P Samuel, A Zulaga; Harlem Hospital Center, New York, NY (156 in PREVENT TB; 4 in substudy): W M El-Sadr, M Klein, C Badshah, J S Schicchi, Y Hirsh-Moverman; Emory University Department of Medicine, Atlanta, GA (114 in PREVENT TB, 4 in substudy): S M Ray, D P Holland, D Dixon, O Mohamed, K Folami, J Bush, C D Simpson, G Barika, W N Favors, N Snow; Jesse Brown VA Medical Center, Chicago, IL, USA (119 in PREVENT TB, 3 in substudy): M Bhattacharya, S Lippold, W Clapp, J Fabre; The University of British Columbia, Vancouver, BC, Canada (114 in PREVENT TB, 8 in substudy): J M Fitzgerald, K Elwood, E Hernandez, B Peyvandi, K Alasaly; VA Little Rock, Arkansas – Arkansas Department of Health, Little Rock, AR, USA (95 in PREVENT TB; 3 in substudy): I Bakhtawar, F Wilson, P Wassler, A Arnold, K Haden, J Owen; University of Manitoba, Winnipeg, MB, Canada (84 in PREVENT TB, 7 in substudy): W Kepron, E Hershfield, M Roth, G A Izon; Michael E DeBakey VA Medical Center –Baylor College of Medicine, Houston, TX (60 in PREVENT TB, 4 in substudy): E Guy, C Lahart, T Scott, R Nickson; Washington DC Veterans Affairs Medical Center, Washington DC (31 in PREVENT TB, 3 in substudy): F Gordin, D Benator, D S Conwell; New York University Bellevue Hospital Center, New York, NY (33 in PREVENT TB, 1 in substudy): R Condos, W Rom, L Sandman; Walter Reed National Military Medical Center, Washington DC (13 in PREVENT TB): K Petersen, T Whitman; Prince George’s County Health Department, Cheverly, MD, USA (2 in PREVENT TB): T E Walsh, W Karney, A Adelakun.
A preliminary version of these results was presented at the 2011 American Thoracic Society Annual Meeting in Denver, CO, USA (abstract number 18812, ‘Hepatitis C virus infection and female sex are risk factors for treatment limiting hepatotoxicity in a large clinical trial of treatment of latent tuberculosis infection: results of a nested case-control study’ at Session C16: New tuberculosis treatment options for active and latent disease).
References
- 1.Dye C, Scheele S, Dolin P, Pathania V, Raviglione MC. Global burden of tuberculosis. Estimated incidence, prevalence, and mortality by country. JAMA. 1999;282:677–686. doi: 10.1001/jama.282.7.677. [DOI] [PubMed] [Google Scholar]
- 2.American Thoracic Society, Centers for Disease Control and Prevention. Targeted tuberculin testing and treatment of latent tuberculosis infection. Am J Respir Crit Care Med. 2000;161(4 Pt 2):S221–S247. doi: 10.1164/ajrccm.161.supplement_3.ats600. [DOI] [PubMed] [Google Scholar]
- 3.Institute of Medicine. Ending neglect: the elimination of tuberculosis in the United States. Washington DC, USA: National Academy of Sciences; 2000. [Google Scholar]
- 4.World Health Organization. Guidelines for intensified tuberculosis case-finding and isoniazid preventive therapy for people living with HIV in resource-constrained settings. Geneva, Switzerland: WHO; 2011. [Accessed June 2015]. http://whqlibdoc.who.int/publications/2011/9789241500708_eng.pdf. [Google Scholar]
- 5.Sterling TR, Bethel J, Goldberg S, Weinfurter P, Yun L, Horsburgh CR. The scope and impact of treatment of latent tuberculosis infection in the United States and Canada. Am J Respir Crit Care Med. 2006;173:927–931. doi: 10.1164/rccm.200510-1563OC. [DOI] [PubMed] [Google Scholar]
- 6.Saukkonen JJ, Cohn DL, Jasmer RM, et al. An official ATS statement: hepatotoxicity of antituberculosis therapy. Am J Respir Crit Care Med. 2006;174:935–952. doi: 10.1164/rccm.200510-1666ST. [DOI] [PubMed] [Google Scholar]
- 7.Bliven EE, Podewils LJ. The role of chronic hepatitis in isoniazid hepatotoxicity during treatment for latent tuberculosis infection. Int J Tuberc Lung Dis. 2009;13:1054–1060. [PubMed] [Google Scholar]
- 8.Sterling TR, Villarino ME, Borisov AS, et al. Three months of rifapentine and isoniazid for latent tuberculosis infection. N Engl J Med. 2011;365:2155–2166. doi: 10.1056/NEJMoa1104875. [DOI] [PubMed] [Google Scholar]
- 9.Cancer Therapy Evaluation Program. The revised common toxicity criteria: version 2.0. Bethesda, MD, USA: CTEP; 1999. [Access June 2015]. http://ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/ctcmanual_v4_10-4-99.pdf. [Google Scholar]
- 10.Ewing JA. Detecting alcoholism. The CAGE questionnaire. JAMA. 1984;252:1905–1907. doi: 10.1001/jama.252.14.1905. [DOI] [PubMed] [Google Scholar]
- 11.Nolan CM, Goldberg SV, Buskin SE. Hepatoxicity associated with isoniazid preventive therapy: a 7-year survey from a public health tuberculosis clinic. JAMA. 1999;281:1014–1018. doi: 10.1001/jama.281.11.1014. [DOI] [PubMed] [Google Scholar]
- 12.Kopanoff DE, Snider DE, Jr, Caras GJ. Isoniazid-related hepatitis: a US Public Health Service Cooperative Surveillance Study. Am Rev Respir Dis. 1978;117:991–1001. doi: 10.1164/arrd.1978.117.6.991. [DOI] [PubMed] [Google Scholar]
- 13.Snider DE, Jr, Caras GJ. Isoniazid-associated hepatitis deaths: a review of available information. Am Rev Respir Dis. 1992;145:494–497. doi: 10.1164/ajrccm/145.2_Pt_1.494. [DOI] [PubMed] [Google Scholar]
- 14.Moulding T. Toxicity associated with isoniazid preventative therapy. JAMA. 1999;282:2207. doi: 10.1001/jama.282.23.2207. [DOI] [PubMed] [Google Scholar]
- 15.Sadaphal P, Astemborski J, Graham NM, et al. Isoniazid preventive therapy, hepatitis C virus infection, and hepatotoxicity among injection drug users infected with Mycobacterium tuberculosis. Clin Infect Dis. 2001;33:1687–1691. doi: 10.1086/323896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Fernandez-Villar A, Sopena B, Vazquez R, et al. Isoniazid hepatotoxicity among drug users: the role of hepatitis C. Clin Infect Dis. 2003;36:293–298. doi: 10.1086/345906. [DOI] [PubMed] [Google Scholar]
- 17.Pettit AC, Bethel J, Hirsch-Moverman Y, Colson PW, Sterling TR. Female sex and discontinuation of isoniazid due to adverse effects during the treatment of latent tuberculosis. J Infect. 2013;67:424–432. doi: 10.1016/j.jinf.2013.07.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Martinson NA, Barnes GL, Moulton LH, et al. New regimens to prevent tuberculosis in adults with HIV infection. N Engl J Med. 2011;365:11–20. doi: 10.1056/NEJMoa1005136. [DOI] [PMC free article] [PubMed] [Google Scholar]