0
Original Research: Chest Infections |

Isoniazid-Associated Hepatitis in Adults Infected With HIV Receiving 36 Months of Isoniazid Prophylaxis in BotswanaIsoniazid-Associated Hepatitis in HIV FREE TO VIEW

Zegabriel Tedla, MD; Minh-Ly Nguyen, MD; Thabisa Sibanda, MBChB; Samba Nyirenda, MD; Tefera B. Agizew, MPhil, MD; Sonali Girde, MD; Charles E. Rose, PhD; Taraz Samandari, MD, PhD
Author and Funding Information

From the CDC Botswana (Drs Tedla, Sibanda, Agizew, and Samandari), Gaborone, Botswana; the Centers for Disease Control and Prevention, Division of Tuberculosis Elimination (Drs Nguyen and Samandari) and Division of HIV/AIDS Prevention (Drs Girde and Rose), Atlanta, GA; the CDC Botswana (Dr Nyirenda), Francistown, Botswana; and ICF International, Inc (Dr Girde), Atlanta, GA.

CORRESPONDENCE TO: Taraz Samandari, MD, PhD, Centers for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop E-45, Atlanta, GA 30329; e-mail: tts0@cdc.gov


FUNDING/SUPPORT: This study was funded by the US Centers for Disease Control and Prevention and the US Agency for International Development.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2015;147(5):1376-1384. doi:10.1378/chest.14-0215
Text Size: A A A
Published online

BACKGROUND:  The World Health Organization recommends 36 months of isoniazid preventive therapy (36IPT) for adults infected with HIV living in TB-endemic countries. We determined the rates and risk factors for isoniazid-associated hepatitis with the use of 36IPT.

METHODS:  One thousand six adults infected with HIV received 36IPT during a pragmatic randomized trial set in Botswana public health clinics providing HIV care. Enrollment exclusion criteria included jaundice or elevations of serum transaminases (ESTs) > 2.5-fold the upper limit of normal (ULN). Participants with any CD4+ lymphocyte count were eligible and received antiretroviral therapy (ART) when CD4+ < 200 cells/μL. 36IPT was stopped for severe hepatitis (more than fivefold ULN EST) but not for moderate hepatitis (2.5-fold to fivefold ULN EST).

RESULTS:  Pharmacy refill records showed 2,237 person-years of isoniazid receipt; 48% of participants initiated ART by 36 months. A total of 1.9% (19 of 1,006) of participants were diagnosed with severe hepatitis; three had jaundice and two of these developed hepatic encephalopathy. Another 3.1% (31 of 1,006) of participants experienced moderate hepatitis. Thirty-eight percent (19 of 50) of participants with moderate to severe hepatitis concomitantly received ART. Forty percent (20 of 50) of moderate to severe cases occurred within the first 2 months of IPT and during this period were not associated with receipt of ART at baseline (hazard ratio, 1.49; 95% CI, 0.20-11.1; P = .70).

CONCLUSIONS:  Adults infected with HIV receiving 36IPT did not have an increased incidence of moderate to severe hepatitis or hepatic encephalopathy compared with published reports among people infected with HIV, people not infected with HIV in trials or public health programs. Compared with participants not receiving ART, the risk of moderate to severe hepatitis was not increased by ART.

TRIAL REGISTRY:  ClinicalTrials.gov; No.: NCT00164281; URL: www.clinicaltrials.gov

Figures in this Article

Although recommended by the Word Health Organization (WHO), 6 months of isoniazid preventive therapy (IPT) for the treatment of latent TB infection in people infected with HIV does not have a durable effect in TB-endemic settings. Prior to the availability of antiretroviral therapy (ART), researchers in Uganda and Zambia conducted follow-on studies after 6-month IPT trials among adults infected with HIV. They reported that within 6 to 18 months after IPT cessation, TB incidence among participants in IPT arms reached levels observed in their respective placebo arms.1,2 Since the advent of ART provision to people infected with HIV living in TB-endemic settings, it has become increasingly evident that ART alone inadequately prevents TB.3,4

A two-arm, randomized, double-blind, placebo-controlled clinical trial conducted in a public health setting in Botswana, a country where ART is freely available to people living with HIV, showed that compared with receipt of 6 months of IPT, adults infected with HIV receiving 36 months of IPT experienced a 43% reduction in their hazard of TB.5 Moreover, the trial found that participants in the 36-month arm who had a positive tuberculin skin test at enrollment experienced a 73% decline in TB incidence compared with their 6-month counterparts. In 2011, the WHO recommended up to 36 months of IPT in TB-endemic settings even when ART is provided.6 Hepatic toxicities associated with ART are not negligible, and the addition of isoniazid to triple therapy regimens requires closer monitoring, particularly in such settings where chronic viral hepatitis is common.

Surveillance of US programs for the treatment of latent TB infection has shown that for every 1,000 people initiating IPT there was one hepatitis death or the need for liver transplantation.7,8 We previously reported that the rates of severe hepatitis in the Botswana IPT cohort, whether biochemical (1.5%), symptomatic (0.4%), or resulting in hepatic encephalopathy (0.1%), were similar to published reports about people infected with HIV and people not infected with HIV.5,911 Rates of severe hepatitis in the 6-month and 36-month arms were 1.6% and 1.9%, respectively.

It is well established that isoniazid may cause transient elevations in serum transaminases, which typically resolve with the continuation of the drug.12,13 For this reason, our protocol permitted the continuation of isoniazid when participants experienced nonsevere hepatitis. To better inform clinicians and public health practitioners about the risks of 36-months of IPT, we describe outcomes of moderate and severe hepatitis among participants receiving 36 months of IPT and explore potential risk factors for isoniazid-associated hepatitis such as viral hepatitis and ART during the first 2 months of IPT when the risk of hepatitis was highest.

The enrollment process is described in detail elsewhere.14 Between November 26, 2004, and July 20, 2006, people ≥ 18 years of age infected with HIV and attending eight government clinics were invited to participate in the study. Enrollment exclusion criteria included not being infected with HIV, jaundice, active TB by symptoms or chest radiograph, a history of hepatitis, or elevations in serum transaminases > 2.5-fold the upper limit of normal (ULN). Participants with any CD4+ lymphocyte count were eligible for enrollment. Consistent with both the Government of Botswana and the WHO’s recommendations, there was no requirement for tuberculin skin testing prior to IPT initiation. However, we performed tuberculin skin tests; induration ≥ 5 mm was regarded as positive.

On a monthly basis, participants received open-labeled isoniazid for 6 months to be taken daily at a dose of 300 mg for weight 30 to 49 kg and 400 mg for weight ≥ 50 kg, supplemented with 25 mg of vitamin B6. In late 2005, the national guidelines changed, and beginning January 1, 2006, all study participants were provided 300 mg daily.5 On three occasions approximately 1 year apart and without prior notification, urine samples were collected from 200 randomly selected participants in the 36-month arm and 50 participants in the 6-month arm among those who continued to return to the clinic. These samples were analyzed for isoniazid metabolites using the potassium cyanide-chloramine-T assay of Eidus and Hamilton.15 ART was offered according to Botswana national guidelines: CD4+ count < 200 cells/μL or WHO clinical stage 3 or 4. Also according to these guidelines, cotrimoxazole prophylaxis was provided for CD4+ count < 200 cells/μL; the dosage was two tablets (trimethoprim/sulfamethoxazole 80/400 mg) taken po once daily or one tablet three times weekly.

Blood draws for serum transaminases and total bilirubin were routinely taken at baseline and 2 weeks after IPT was initiated. Given the pragmatic nature of the trial, regular blood draws for liver function tests were not subsequently scheduled. However blood was drawn for pre-ART screening, if a participant had symptoms concerning for hepatitis, and during a liver safety campaign in 2007 requested by the trial’s Data Safety Monitoring Board.

We defined isoniazid-associated hepatitis as elevations in transaminases with or without symptoms that were at least possibly attributed to isoniazid using an attribution scale.9 Using serum transaminase values, hepatitis was graded from 1 to 5.16 Elevations in either aspartate or alanine transaminase were used to grade hepatitis: grade 1 (mild) > 1 to 2.5 times the ULN; grade 2 (moderate) ≥ 2.5 to 5.0 times ULN; and ≥ grade 3 (severe) if > 5 times ULN. Participants with mild or moderate hepatitis continued to receive isoniazid; however isoniazid was stopped with no rechallenge if a participant developed severe hepatitis. Attribution of hepatitis to the study drug was decided by study clinicians who were blinded to treatment arm; additionally, severe cases were reviewed by an independent committee of four experienced clinicians who were also masked to treatment arm. The attribution of hepatitis to the study drug was graded as follows: definite (“clearly related”), probable (“likely related”), possible (“may be related”), unlikely, and unrelated.

If participants developed moderate to severe transaminase elevations they were tested for serologic evidence of both hepatitis B virus (HBV) and hepatitis C virus (HCV) infection. To detect HCV infection, we used either MEIA (AxSYM HCV 3.0; Abbott Laboratories) or Anti-HCV v. 4.0 VK 47/48 (Murex Biotech). To test for HBV surface antigen (HBsAg), anti-HBV surface antibody, and anti-HBV core antibody (HBcAb), we used HBsAg Determine strip (Abbott Diagnostics) or Murex version 3.0 GE 34/36 (Abbott); anti-HBV surface antibody and HBcAb Architect Anti-HBs reagent Kit 7C18 (Abbott Diagnostic Division).

The analysis presented in this report is limited to the 1,006 participants who were randomized to the 36-month arm of the trial. Participants who had no evidence of grade > 1 hepatitis were compared with participants with moderate to severe hepatitis. Using SAS (version 9.2; SAS Institute Inc), univariable hazard ratios using a Cox proportional hazards model were determined for the following risk factors: sex, age > 35 years at enrollment, low BMI at enrollment, elevated transaminase level (> 1 to < 2.5 ULN) at enrollment, self-reported alcohol use at any time, evidence of alcohol dependence using CAGE criteria,17 and CD4+ count < 200 cells/μL at enrollment. To assess the risk of hepatitis from concomitant ART use, we restricted our analysis to the period within the first 2 months of IPT and censored participants who initiated ART after first receipt of IPT; a multivariable analysis of this same timeframe, including age > 35 years, baseline aspartate and alanine transaminase levels (> 1 to < 2.5 ULN), baseline CD4+ count < 200 cells/μL, and ART, was also conducted using SAS. Serologic tests for HBV coinfection were conducted among participants with isoniazid-associated hepatitis and 99 randomly selected participants without evidence of biochemical hepatitis (ie, abnormal liver function tests with or without symptoms); the odds of isoniazid-associated hepatitis were tested in a case-control analysis. P values < .05 were considered statistically significant. The study was approved by the Botswana Ministry of Health’s Health Research and Development Committee, PPME 13/1/1 Vol V (368) and the CDC Institutional Review Board, protocol #3441. This trial is registered at clinicaltrials.gov number NCT00164281.

Seventy-one percent of the 1,006 participants infected with HIV enrolled in the 36-month arm of the study were women (Table 1). Their median age was 32 years, their median CD4+ count was 307/μL, and 48% initiated ART before the end of the trial. Of 482 participants starting ART, 209 (43%) received zidovudine, lamivudine, and nevirapine; 206 (43%) received zidovudine, lamivudine, and efavirenz; 31 (6%) received stavudine, lamivudine, and nevirapine or efavirenz; and 14 (3%) received tenofovir, emtricitabine or lamivudine, and nevirapine or efavirenz. For those initiating ART after IPT (n = 449), ART was begun a median of 174 days (interquartile range [IQR], 50-525) after IPT initiation, and they received it for a median of 875 days (IQR, 452-1,030) concomitantly with IPT. Based upon pharmacy records, participants received a median of 1,050 days of IPT (IQR, 450-1,080). IPT adherence was 78% by pharmacy refill and 74% to 80% by isoniazid metabolite testing.18 Among the 1,006 participants, 98% had at least one and 76% had at least two follow-up liver function tests.

Table Graphic Jump Location
TABLE 1 ]  Characteristics of Study Participants Infected With HIV Who Received 36 Months of Isoniazid (N = 1,006)

Data are No. (%), median (interquartile range), or mean (SD) unless otherwise stated. BCG = bacille Calmette-Guérin; CD = cluster of differentiation; CPT = cotrimoxazole preventive therapy.

a 

Data were missing for 54 participants.

b 

Data were missing for 46 participants.

c 

Data were missing for 13 in the continued isoniazid group.

d 

Data were missing for 32 participants.

Mild, moderate, and severe hepatitis at least possibly related to study drug was detected in 12.8% (129 of 1,006), 3.1% (31 of 1,006), and 1.9% (19 of 1,006) of participants, respectively (Fig 1). Forty percent (20 of 50) of moderate to severe cases occurred during the 2 months after IPT initiation, and 84% (16 of 19) of severe cases of hepatitis occurred within the first 9 months (Fig 2). Three of the 19 participants with severe hepatitis were initially identified as having moderate hepatitis but within 22 to 36 days progressed to severe hepatitis, although they remained asymptomatic (Fig 3E).

Figure Jump LinkFigure 1 –  Outcomes of isoniazid-associated hepatitis in adults infected with HIV initiating 36 mo of IPT. G1 = EST > upper limit of normal to 2.5-fold upper limit of normal; G2 = EST > 2.5- to 5.0-fold upper limit of normal; G3 = EST > 5.0- to 20.0-fold upper limit of normal; G4 = EST > 20.0-fold upper limit of normal; G5 = death with G4. ART = antiretroviral therapy; EST = elevated serum transaminase; IPT = isoniazid preventive therapy.Grahic Jump Location
Figure Jump LinkFigure 2 –  Moderate and severe hepatitis among participants receiving 36 mo of isoniazid by study month. Solid portions of the bars represent cases of severe (grade 3-5) hepatitis, and hatched portions of the bars represent moderate (grade 2) hepatitis at least possibly associated with isoniazid. The line graph represents participants continuing to receive IPT. From the 1,006 initiating 36-mo IPT, three participants were lost to follow-up, and 91 withdrew before the end of the trial. See Figure 1 legend for expansion of abbreviation.Grahic Jump Location
Figure Jump LinkFigure 3 –  Outcomes among representative participants with moderate (grade 2) hepatitis among adults infected with HIV receiving isoniazid by month. Color-filled rows for INH and ART indicate months when INH or ART were received by the participant. A, Elevated serum transaminase resolved with INH continued, no ART initiated. B, Elevated serum transaminase resolved with INH continued, ART initiated. C, Elevated serum transaminase persisted with INH continued. D, Elevated serum transaminase persisted despite INH discontinuation. E, Elevated serum transaminase worsened resulting in discontinuation of INH. INH = isoniazid; PMTCT = single antiretroviral drug for the prevention of mother-to-child transmission of HIV. See Figure 1 legend for expansion of other abbreviations.Grahic Jump Location

Two participants with severe hepatitis died of hepatic encephalopathy in the sixth and ninth months of therapy with probable and possible attributions to isoniazid, respectively.5 The first was a 41-year-old woman with baseline CD4+ count of 384 cells/μL. A nonstudy clinician advised her to stop IPT when she presented with dark urine and icteric sclera and had a total bilirubin level of 572 mg/L. She declined hospital admission. She may not have stopped IPT until 10 days later when she presented to study clinicians. Four days later she was seen at her local clinic with a complaint of cough and vomiting. She received a diagnosis of pneumonia and was prescribed phenoxymethylpenicillin and acetaminophen. That night she presented to the ED encephalopathic and was admitted. Her liver function test results and coagulation parameters were highly abnormal; she had serologically negative tests for HCV antibody and HBsAg. She died 4 days later. The second was a 32-year-old woman with baseline CD4+ count of 396 cells/μL. She complained of icteric sclera and a right-sided T2 to T5 zosteriform rash at a local clinic and was told to stop her medication and referred to a hospital, but she did not follow this advice. Two days later she saw another nonstudy clinician and stopped her medication. She returned to another local clinic on the same day to have her liver functions checked but was prescribed acetaminophen. Two days later she was admitted to a hospital for confusion, restlessness, and irritability. Admission tests showed a total bilirubin level of 320 mg/dL and were negative for HBsAg and HCV antibody. A lumbar puncture showed elevated protein, normal glucose, and WBC count of 122 (95% lymphocytes). She died 4 days after admission despite administration of IV acyclovir and supportive therapies. Neither of these patients was receiving any other medication, including traditional medications, and neither drank alcohol.

Of the other 17 participants with severe hepatitis, four were symptomatic: two had nausea and vomiting; one had abdominal pain with malaise; and one had jaundice, nausea, and anorexia. Twenty-one percent (four of 19) of participants with severe hepatitis reported alcohol use; three had evidence of alcohol dependence. Seven were receiving zidovudine/lamivudine plus either nevirapine or efavirenz (one was also receiving cotrimoxazole), one was receiving ketoconazole, and a ninth participant had received a traditional medication. None of 12 of these 19 participants who were tested for viral hepatitis had a positive HCV antibody or HBsAg test.

Beside the three participants who progressed from moderate to severe hepatitis, moderate hepatitis was detected in an additional 31 participants. Four had symptoms at the time that hepatitis was detected: One was receiving chemotherapy for Kaposi sarcoma, one had a cough, one was dehydrated, and one complained of jaundice but was not hyperbilirubinemic. The remainder were asymptomatic and had blood draws at the routine 2-week blood draw (n = 12), the liver safety campaign (n = 11), or during pre-ART screening (n = 4). Fifty-two percent (16 of 31) of participants with moderate hepatitis reported alcohol use; among these, nine had evidence of alcohol dependence. Thirty-eight percent (12 of 31) were receiving zidovudine/lamivudine plus either nevirapine or efavirenz or lamivudine/stavudine plus nevirapine. Ten percent (three of 31) were receiving cotrimoxazole. HBsAg tests for two of 25 (8%) of these 31 participants were positive, and a HCV test for one of 16 tested was positive. In all, 40 of 50 participants (80%) with moderate to severe hepatitis were asymptomatic.

After detection of moderate hepatitis, the 31 participants continued to receive isoniazid for a median of 18 and a maximum of 36 additional months. Twenty-four experienced a reduction in transaminases below the moderate level either with (n = 15) or without (n = 9) ART, two had persistent moderate hepatitis, three had no symptoms but had no additional blood test for transaminases, and one had persistent moderate hepatitis after IPT was stopped (Fig 3D). One developed active TB 8 months after resolution of hepatitis and then stopped IPT. One participant died 2 days after having had the blood draw showing moderate hepatitis. This participant died of uncertain causes; her family reported that she had a few hours of weakness, nausea, vomiting, and shortness of breath but no cough or jaundice. She had not initiated ART despite a CD4+ count of 70 cells/μL 4 months earlier.

Univariable analysis showed statistically significant associations between moderate to severe hepatitis and CD4 < 200 cells/μL (P = .001) as well as baseline aspartate transaminase greater than ULN at enrollment (P < .0001) (Table 2). There was a statistically significant trend in the association between hepatitis and lower CD4+ count category (χ2 for trend, P = .0024; data not shown). Restricted to the first 2 months of IPT, 34 participants had initiated ART at baseline. There was one case of moderate to severe hepatitis among participants initiating ART prior to or on the same day as IPT and 19 cases among participants not receiving concomitant ART. ART was not associated with hepatitis (hazard ratio, 1.49; 95% CI, 0.20-11.1; P = .70); the multivariable analysis showed only aspartate transaminase greater than ULN to be significantly associated with hepatitis (adjusted hazard ratio, 4.1; 95% CI , 1.4-11.7; P = .009). There were no cases of hepatitis among participants concomitantly receiving cotrimoxazole in the first 2 months of IPT receipt.

Table Graphic Jump Location
TABLE 2 ]  Risk Factors for Moderate to Severe Hepatitis Among Adults Infected With HIV Receiving 36 Months of Isoniazid Preventive Therapy

ALT = alanine transaminase; AST = aspartate transaminase; CD = cluster of differentiation; ULN = upper limit of normal.

a 

Heights were not available for all participants to calculate BMI.

b 

Fisher exact test used for small numbers; otherwise χ2 test.

c 

OR used, as this was a case-control analysis in which randomly selected participants without hepatitis were selected as control subjects.

The case-control analysis showed that having a serologically positive HBV surface antigen test did not increase the odds of moderate to severe hepatitis (P = .31), whereas having serologically positive HBV core antibody test was associated with hepatitis 2.8 (95% CI, 1.1-7.0; P = .025) (Table 2). Multivariable analysis limited to the group tested for HBV core antibody and including the variables HBV core antibody, age > 35 years, baseline aspartate or alanine transaminase levels (> 1 to < 2.5 ULN), CD4 count < 200 cells/μL, and ART showed a significant association only between CD4 < 200 cells/μL and hepatitis (P = .0057).

In this article, we extend safety observations from the 36-month arm of the Botswana IPT trial. A total of 1.9% and 3.1% of participants infected with HIV initiating 36 months of IPT were diagnosed with severe and moderate hepatitis, respectively. Jaundice associated with isoniazid was uncommon (0.3%), and the two cases of hepatic encephalopathy may have been avoided through better education of patients to immediately stop IPT when icteric and better education of health providers that acetaminophen is contraindicated in jaundiced patients. No hepatitis symptom was reported by 80% of participants who developed moderate to severe hepatitis. Among those with moderate hepatitis, 87% either experienced resolution of transaminase elevations or had no symptoms despite continuing IPT for a median of 24 additional months (range, 7-36) with or without ART. We observed an increased hazard of moderate to severe hepatitis in participants with CD4+ counts < 200 cells/μL.

In addition to the Botswana trial, two other clinical trials have reported adverse events among people infected with HIV receiving 36 months of IPT: one in Soweto and one in Chennai.19,20 In comparison with the Soweto trial, the Botswana trial had far fewer severe (grade ≥ 3) hepatitis events: 28% (46 of 164) vs 2% (19 of 1,006). There were 0.9% (three of 339) cases of jaundice in the Chennai trial compared with 0.3% (three of 1,006) in the Botswana trial; in the former trial, all three patients resumed IPT. In neither the Soweto nor the Chennai trial was any death attributed to study drug; in both trials hepatic enzymes were checked every 6 months, and in both trials ART use was limited or was a protocol-required reason for study exit. As hepatitis typically occurs in the first few months of IPT, it is helpful to consider the incidence of hepatitis in people infected with HIV receiving shorter durations of IPT in the context of ART provision: among 544 South African miners receiving ART and 9 months of IPT there were no cases of clinical hepatotoxicity21; in a Thai ART clinic, the incidence of severe elevations of aspartate transaminase was 1.2% among 799 patients infected with HIV who received 9 months of IPT, 23% of whom were also receiving ART22; and in a cluster-randomized trial in Brazilian HIV clinics, there were 0.25% (three of 1,186) liver toxicities among people infected with HIV receiving 6 months’ IPT, 60% of whom also received ART.23 Thus, the rates of severe hepatitis we observed among 36-month IPT recipients appear typical. As for the two deaths, we recall that there were none among participants in the 6-month arm of the trial, resulting in an overall rate of 0.1% (two of 1,998) for participants initiating IPT, which is in keeping with US program statistics.7,8

It is not known why people infected with HIV with lower CD4+ lymphocyte counts tended to have a higher risk of drug-induced hepatitis. People infected with HIV with lower nadir CD4+ counts are more likely to have abnormal liver tests.24 We speculate that as the histologic picture of isoniazid-related hepatitis is virtually indistinguishable from that of viral hepatitis,25 and since the host immune response is a key element of the response to both viral and drug-induced hepatitis,26,27 the association with lower CD4+ counts may represent a dysregulation of the immune system reacting against isoniazid metabolites, which may be mitigated by immune reconsitution.28 An additional explanation is suggested in the case-control analysis: Participants testing positive for core HBV antibody and largely negative for HBsAg were 2.8 times more likely to develop moderate to severe hepatitis. As people infected with HIV testing negative for HBsAg and positive for HBcAb are more likely to have occult HBV infection, particularly if they have lower CD4 counts or are ART-naive, it is possible that occult HBV infection was a factor in isoniazid-associated hepatitis.29,30 A retrospective study of 73 children infected with HIV receiving both ART and IPT found a decreased risk of hepatitis in participants receiving both ART with IPT and a higher risk of hepatitis in children with lower CD4+ counts.31

Because of the pragmatic nature of our study, a limitation of this paper is that routine blood monitoring of hepatic enzymes throughout the 36 months of IPT was not performed. Had we done so, the overall rate of biochemical hepatitis (ie, asymptomatic elevations in liver function test results) might have increased, but the rate of symptomatic hepatitis would not have changed, since we asked about hepatitis symptoms at each monthly visit. Our finding that grade 1 to 2 baseline aspartate transaminase elevation was associated with subsequent moderate to severe hepatitis is interesting but impractical to implement, given that only 19 of 1,006 people with such elevations progressed to hepatitis and only five developed severe hepatitis. A second limitation is that if many participants did not take their pills, the rate of hepatitis may be an underestimate. Our adherence rate was adequate for a clinical trial and is likely higher than adherence rates expected in programs; therefore, our findings are helpful to program managers as the upper limits of what could be expected.32

Routine blood monitoring of hepatic enzymes during IPT is not recommended for people being treated for latent TB infection, whether infected with HIV or otherwise.33 However, if the baseline serum alanine aminotransferase is above the ULN, blood monitoring is recommended.34 Most people who experience these abnormalities even with mild hyperbilirubinemia resolve them while continuing isoniazid as the liver adapts.12,13 It is difficult to predict which individuals will progress to severe clinical disease, which is why it is important to educate patients about the symptoms of hepatitis and to advise them to immediately stop IPT should they experience any symptom. During the first 2 months of IPT use, when the risk of hepatitis was highest, concomitant use of efavirenz or nevirapine-based ART did not significantly increase the risk of moderate to severe hepatitis. The overall rate of symptomatic isoniazid-associated hepatitis was very low during 36 months of IPT.

Author contributions: T. Samandari is the guarantor of the article. Z. T. contributed to the design of the study, abstraction of study data, and analysis of the data, and wrote the first draft of the manuscript; M.-L. N. contributed to the analysis of the data and the writing of the manuscript; T. Sibanda, S. N., T. B. A., and T. Samandari contributed to the design of the study, abstraction of study data, analysis of the data, and the writing of the manuscript; and S. G. and C. E. R. contributed to analyzing the data and writing the manuscript.

Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Role of sponsors: The sponsors had no role in the design of the study, the collection and analysis of the data, or the preparation of the manuscript.

Other contributions: We thank the participants of the trial, colleagues in the Ministry of Health, and the CDC Botswana nursing staff. The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the CDC.

ART

antiretroviral therapy

HBcAb

anti-hepatitis B virus core antibody

HBsAg

hepatitis B virus surface antigen

HBV

hepatitis B virus

HCV

hepatitis C virus

IPT

isoniazid preventive therapy

IQR

interquartile range

ULN

upper limit of normal

WHO

World Health Organization

Johnson JL, Okwera A, Hom DL, et al; Uganda-Case Western Reserve University Research Collaboration. Duration of efficacy of treatment of latent tuberculosis infection in HIV-infected adults. AIDS. 2001;15(16):2137-2147. [CrossRef] [PubMed]
 
Quigley MA, Mwinga A, Hosp M, et al. Long-term effect of preventive therapy for tuberculosis in a cohort of HIV-infected Zambian adults. AIDS. 2001;15(2):215-222. [CrossRef] [PubMed]
 
Nicholas S, Sabapathy K, Ferreyra C, Varaine F, Pujades-Rodríguez M; AIDS Working Group of Médecins Sans Frontières. Incidence of tuberculosis in HIV-infected patients before and after starting combined antiretroviral therapy in 8 sub-Saharan African HIV programs. J Acquir Immune Defic Syndr. 2011;57(4):311-318. [CrossRef] [PubMed]
 
Lawn SD, Harries AD, Williams BG, et al. Antiretroviral therapy and the control of HIV-associated tuberculosis. Will ART do it? Int J Tuberc Lung Dis. 2011;15(5):571-581. [CrossRef] [PubMed]
 
Samandari T, Agizew TB, Nyirenda S, et al. 6-month versus 36-month isoniazid preventive treatment for tuberculosis in adults with HIV infection in Botswana: a randomised, double-blind, placebo-controlled trial. Lancet. 2011;377(9777):1588-1598. [CrossRef] [PubMed]
 
 Guidelines for Intensified Tuberculosis Case-Finding and Isoniazid Preventive Therapy for People Living With HIV in Resource Constrained Settings. Geneva, Switzerland: World Health Organization; 2011.
 
Kopanoff DE, Snider DE Jr, Caras GJ. Isoniazid-related hepatitis: a US Public Health Service cooperative surveillance study. Am Rev Respir Dis. 1978;117(6):991-1001. [PubMed]
 
Centers for Disease Control and Prevention (CDC). Severe isoniazid-associated liver injuries among people being treated for latent tuberculosis infection - United States, 2004-2008. MMWR Morb Mortal Wkly Rep. 2010;59(8):224-229. [PubMed]
 
Tedla Z, Nyirenda S, Peeler C, et al. Isoniazid-associated hepatitis and antiretroviral drugs during tuberculosis prophylaxis in HIV-infected adults in Botswana. Am J Respir Crit Care Med. 2010;182(2):278-285. [CrossRef] [PubMed]
 
Whalen CC, Johnson JL, Okwera A, et al; Uganda-Case Western Reserve University Research Collaboration. A trial of three regimens to prevent tuberculosis in Ugandan adults infected with the human immunodeficiency virus. N Engl J Med. 1997;337(12):801-808. [CrossRef] [PubMed]
 
Gordin FM, Matts JP, Miller C, et al; Terry Beirn Community Programs for Clinical Research on AIDS. A controlled trial of isoniazid in people with anergy and human immunodeficiency virus infection who are at high risk for tuberculosis. N Engl J Med. 1997;337(5):315-320. [CrossRef] [PubMed]
 
Mitchell JR, Zimmerman HJ, Ishak KG, et al. Isoniazid liver injury: clinical spectrum, pathology, and probable pathogenesis. Ann Intern Med. 1976;84(2):181-192. [CrossRef] [PubMed]
 
Scharer L, Smith JP. Serum transaminase elevations and other hepatic abnormalities in patients receiving isoniazid. Ann Intern Med. 1969;71(6):1113-1120. [CrossRef] [PubMed]
 
Mosimaneotsile B, Mathoma A, Chengeta B, et al. Isoniazid tuberculosis preventive therapy in HIV-infected adults accessing antiretroviral therapy: a Botswana Experience, 2004-2006. J Acquir Immune Defic Syndr. 2010;54(1):71-77. [PubMed]
 
Eidus L, Hamilton EJ. A new method for the determination of N-acetyl isoniazid in urine of ambulatory patients. Am Rev Respir Dis. 1964;89:587-588. [PubMed]
 
Cancer Therapy Evaluation Program. Common terminology criteria for adverse events v3.0 (CTCAE). National Institutes of Health website. http://ctep.cancer.gov/protocolDevelopment/electronic_applications/ctc.htm. Accessed October 28, 2014.
 
Dhalla S, Kopec JA. The CAGE questionnaire for alcohol misuse: a review of reliability and validity studies. Clin Invest Med. 2007;30(1):33-41. [PubMed]
 
Gust DA, Mosimaneotsile B, Mathebula U, et al. Risk factors for non-adherence and loss to follow-up in a three-year clinical trial in Botswana. PLoS ONE. 2011;6(4):e18435. [CrossRef] [PubMed]
 
Swaminathan S, Menon PA, Gopalan N, et al. Efficacy of a six-month versus a 36-month regimen for prevention of tuberculosis in HIV-infected people in India: a randomized clinical trial. PLoS ONE. 2012;7(12):e47400. [CrossRef] [PubMed]
 
Martinson NA, Barnes GL, Moulton LH, et al. New regimens to prevent tuberculosis in adults with HIV infection. N Engl J Med. 2011;365(1):11-20. [CrossRef] [PubMed]
 
Grant AD, Mngadi KT, van Halsema CL, Luttig MM, Fielding KL, Churchyard GJ. Adverse events with isoniazid preventive therapy: experience from a large trial. AIDS. 2010;24(suppl 5):S29-S36. [CrossRef] [PubMed]
 
Khongphatthanayothin M, Avihingsanon A, Teeratakulpisarn N, et al. Feasibility and efficacy of isoniazid prophylaxis for latent tuberculosis in HIV-infected clients patients in Thailand. AIDS Res Hum Retroviruses. 2012;28(3):270-275. [CrossRef] [PubMed]
 
Durovni B, Saraceni V, Moulton LH, et al. Effect of improved tuberculosis screening and isoniazid preventive therapy on incidence of tuberculosis and death in patients with HIV in clinics in Rio de Janeiro, Brazil: a stepped wedge, cluster-randomised trial. Lancet Infect Dis. 2013;13(10):852-858. [CrossRef] [PubMed]
 
Crum-Cianflone N, Collins G, Medina S, et al. Prevalence and factors associated with liver test abnormalities among human immunodeficiency virus-infected people. Clin Gastroenterol Hepatol. 2010;8(2):183-191. [CrossRef] [PubMed]
 
Lee WM. Drug-induced hepatotoxicity. N Engl J Med. 1995;333(17):1118-1127. [CrossRef] [PubMed]
 
Guidotti LG, Chisari FV. Immunobiology and pathogenesis of viral hepatitis. Annu Rev Pathol. 2006;1:23-61. [CrossRef] [PubMed]
 
Liu ZX, Kaplowitz N. Immune-mediated drug-induced liver disease. Clin Liver Dis. 2002;6(3):755-774. [CrossRef] [PubMed]
 
Boasso A, Shearer GM, Chougnet C. Immune dysregulation in human immunodeficiency virus infection: know it, fix it, prevent it? J Intern Med. 2009;265(1):78-96. [CrossRef] [PubMed]
 
Cohen Stuart JWT, Velema M, Schuurman R, Boucher CA, Hoepelman AI. Occult hepatitis B in people infected with HIV is associated with low CD4 counts and resolves during antiretroviral therapy. J Med Virol. 2009;81(3):441-445. [CrossRef] [PubMed]
 
Hamers RL, Zaaijer HL, Wallis CL, et al; PharmAccess African Studies to Evaluate Resistance (PASER). HIV-HBV coinfection in Southern Africa and the effect of lamivudine-versus tenofovir-containing cART on HBV outcomes. J Acquir Immune Defic Syndr. 2013;64(2):174-182. [CrossRef] [PubMed]
 
Gray D, Nuttall J, Lombard C, et al. Low rates of hepatotoxicity in HIV-infected children on anti-retroviral therapy with and without isoniazid prophylaxis. J Trop Pediatr. 2010;56(3):159-165. [CrossRef] [PubMed]
 
Pekovic V, Mayanja H, Vjecha M, et al; Uganda-Case Western Reserve University Research Collaboration. Comparison of three composite compliance indices in a trial of self-administered preventive therapy for tuberculosis in HIV-infected Ugandan adults. J Clin Epidemiol. 1998;51(7):597-607. [CrossRef] [PubMed]
 
Centers for Disease Control and Prevention (CDC). Treatment of tuberculosis. MMWR Morb Mortal Wkly Rep. 2003;52(RR11):1-77.
 
Saukkonen JJ, Cohn DL, Jasmer RM, et al; ATS (American Thoracic Society) Hepatotoxicity of Antituberculosis Therapy Subcommittee. An official ATS statement: hepatotoxicity of antituberculosis therapy. Am J Respir Crit Care Med. 2006;174(8):935-952. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1 –  Outcomes of isoniazid-associated hepatitis in adults infected with HIV initiating 36 mo of IPT. G1 = EST > upper limit of normal to 2.5-fold upper limit of normal; G2 = EST > 2.5- to 5.0-fold upper limit of normal; G3 = EST > 5.0- to 20.0-fold upper limit of normal; G4 = EST > 20.0-fold upper limit of normal; G5 = death with G4. ART = antiretroviral therapy; EST = elevated serum transaminase; IPT = isoniazid preventive therapy.Grahic Jump Location
Figure Jump LinkFigure 2 –  Moderate and severe hepatitis among participants receiving 36 mo of isoniazid by study month. Solid portions of the bars represent cases of severe (grade 3-5) hepatitis, and hatched portions of the bars represent moderate (grade 2) hepatitis at least possibly associated with isoniazid. The line graph represents participants continuing to receive IPT. From the 1,006 initiating 36-mo IPT, three participants were lost to follow-up, and 91 withdrew before the end of the trial. See Figure 1 legend for expansion of abbreviation.Grahic Jump Location
Figure Jump LinkFigure 3 –  Outcomes among representative participants with moderate (grade 2) hepatitis among adults infected with HIV receiving isoniazid by month. Color-filled rows for INH and ART indicate months when INH or ART were received by the participant. A, Elevated serum transaminase resolved with INH continued, no ART initiated. B, Elevated serum transaminase resolved with INH continued, ART initiated. C, Elevated serum transaminase persisted with INH continued. D, Elevated serum transaminase persisted despite INH discontinuation. E, Elevated serum transaminase worsened resulting in discontinuation of INH. INH = isoniazid; PMTCT = single antiretroviral drug for the prevention of mother-to-child transmission of HIV. See Figure 1 legend for expansion of other abbreviations.Grahic Jump Location

Tables

Table Graphic Jump Location
TABLE 1 ]  Characteristics of Study Participants Infected With HIV Who Received 36 Months of Isoniazid (N = 1,006)

Data are No. (%), median (interquartile range), or mean (SD) unless otherwise stated. BCG = bacille Calmette-Guérin; CD = cluster of differentiation; CPT = cotrimoxazole preventive therapy.

a 

Data were missing for 54 participants.

b 

Data were missing for 46 participants.

c 

Data were missing for 13 in the continued isoniazid group.

d 

Data were missing for 32 participants.

Table Graphic Jump Location
TABLE 2 ]  Risk Factors for Moderate to Severe Hepatitis Among Adults Infected With HIV Receiving 36 Months of Isoniazid Preventive Therapy

ALT = alanine transaminase; AST = aspartate transaminase; CD = cluster of differentiation; ULN = upper limit of normal.

a 

Heights were not available for all participants to calculate BMI.

b 

Fisher exact test used for small numbers; otherwise χ2 test.

c 

OR used, as this was a case-control analysis in which randomly selected participants without hepatitis were selected as control subjects.

References

Johnson JL, Okwera A, Hom DL, et al; Uganda-Case Western Reserve University Research Collaboration. Duration of efficacy of treatment of latent tuberculosis infection in HIV-infected adults. AIDS. 2001;15(16):2137-2147. [CrossRef] [PubMed]
 
Quigley MA, Mwinga A, Hosp M, et al. Long-term effect of preventive therapy for tuberculosis in a cohort of HIV-infected Zambian adults. AIDS. 2001;15(2):215-222. [CrossRef] [PubMed]
 
Nicholas S, Sabapathy K, Ferreyra C, Varaine F, Pujades-Rodríguez M; AIDS Working Group of Médecins Sans Frontières. Incidence of tuberculosis in HIV-infected patients before and after starting combined antiretroviral therapy in 8 sub-Saharan African HIV programs. J Acquir Immune Defic Syndr. 2011;57(4):311-318. [CrossRef] [PubMed]
 
Lawn SD, Harries AD, Williams BG, et al. Antiretroviral therapy and the control of HIV-associated tuberculosis. Will ART do it? Int J Tuberc Lung Dis. 2011;15(5):571-581. [CrossRef] [PubMed]
 
Samandari T, Agizew TB, Nyirenda S, et al. 6-month versus 36-month isoniazid preventive treatment for tuberculosis in adults with HIV infection in Botswana: a randomised, double-blind, placebo-controlled trial. Lancet. 2011;377(9777):1588-1598. [CrossRef] [PubMed]
 
 Guidelines for Intensified Tuberculosis Case-Finding and Isoniazid Preventive Therapy for People Living With HIV in Resource Constrained Settings. Geneva, Switzerland: World Health Organization; 2011.
 
Kopanoff DE, Snider DE Jr, Caras GJ. Isoniazid-related hepatitis: a US Public Health Service cooperative surveillance study. Am Rev Respir Dis. 1978;117(6):991-1001. [PubMed]
 
Centers for Disease Control and Prevention (CDC). Severe isoniazid-associated liver injuries among people being treated for latent tuberculosis infection - United States, 2004-2008. MMWR Morb Mortal Wkly Rep. 2010;59(8):224-229. [PubMed]
 
Tedla Z, Nyirenda S, Peeler C, et al. Isoniazid-associated hepatitis and antiretroviral drugs during tuberculosis prophylaxis in HIV-infected adults in Botswana. Am J Respir Crit Care Med. 2010;182(2):278-285. [CrossRef] [PubMed]
 
Whalen CC, Johnson JL, Okwera A, et al; Uganda-Case Western Reserve University Research Collaboration. A trial of three regimens to prevent tuberculosis in Ugandan adults infected with the human immunodeficiency virus. N Engl J Med. 1997;337(12):801-808. [CrossRef] [PubMed]
 
Gordin FM, Matts JP, Miller C, et al; Terry Beirn Community Programs for Clinical Research on AIDS. A controlled trial of isoniazid in people with anergy and human immunodeficiency virus infection who are at high risk for tuberculosis. N Engl J Med. 1997;337(5):315-320. [CrossRef] [PubMed]
 
Mitchell JR, Zimmerman HJ, Ishak KG, et al. Isoniazid liver injury: clinical spectrum, pathology, and probable pathogenesis. Ann Intern Med. 1976;84(2):181-192. [CrossRef] [PubMed]
 
Scharer L, Smith JP. Serum transaminase elevations and other hepatic abnormalities in patients receiving isoniazid. Ann Intern Med. 1969;71(6):1113-1120. [CrossRef] [PubMed]
 
Mosimaneotsile B, Mathoma A, Chengeta B, et al. Isoniazid tuberculosis preventive therapy in HIV-infected adults accessing antiretroviral therapy: a Botswana Experience, 2004-2006. J Acquir Immune Defic Syndr. 2010;54(1):71-77. [PubMed]
 
Eidus L, Hamilton EJ. A new method for the determination of N-acetyl isoniazid in urine of ambulatory patients. Am Rev Respir Dis. 1964;89:587-588. [PubMed]
 
Cancer Therapy Evaluation Program. Common terminology criteria for adverse events v3.0 (CTCAE). National Institutes of Health website. http://ctep.cancer.gov/protocolDevelopment/electronic_applications/ctc.htm. Accessed October 28, 2014.
 
Dhalla S, Kopec JA. The CAGE questionnaire for alcohol misuse: a review of reliability and validity studies. Clin Invest Med. 2007;30(1):33-41. [PubMed]
 
Gust DA, Mosimaneotsile B, Mathebula U, et al. Risk factors for non-adherence and loss to follow-up in a three-year clinical trial in Botswana. PLoS ONE. 2011;6(4):e18435. [CrossRef] [PubMed]
 
Swaminathan S, Menon PA, Gopalan N, et al. Efficacy of a six-month versus a 36-month regimen for prevention of tuberculosis in HIV-infected people in India: a randomized clinical trial. PLoS ONE. 2012;7(12):e47400. [CrossRef] [PubMed]
 
Martinson NA, Barnes GL, Moulton LH, et al. New regimens to prevent tuberculosis in adults with HIV infection. N Engl J Med. 2011;365(1):11-20. [CrossRef] [PubMed]
 
Grant AD, Mngadi KT, van Halsema CL, Luttig MM, Fielding KL, Churchyard GJ. Adverse events with isoniazid preventive therapy: experience from a large trial. AIDS. 2010;24(suppl 5):S29-S36. [CrossRef] [PubMed]
 
Khongphatthanayothin M, Avihingsanon A, Teeratakulpisarn N, et al. Feasibility and efficacy of isoniazid prophylaxis for latent tuberculosis in HIV-infected clients patients in Thailand. AIDS Res Hum Retroviruses. 2012;28(3):270-275. [CrossRef] [PubMed]
 
Durovni B, Saraceni V, Moulton LH, et al. Effect of improved tuberculosis screening and isoniazid preventive therapy on incidence of tuberculosis and death in patients with HIV in clinics in Rio de Janeiro, Brazil: a stepped wedge, cluster-randomised trial. Lancet Infect Dis. 2013;13(10):852-858. [CrossRef] [PubMed]
 
Crum-Cianflone N, Collins G, Medina S, et al. Prevalence and factors associated with liver test abnormalities among human immunodeficiency virus-infected people. Clin Gastroenterol Hepatol. 2010;8(2):183-191. [CrossRef] [PubMed]
 
Lee WM. Drug-induced hepatotoxicity. N Engl J Med. 1995;333(17):1118-1127. [CrossRef] [PubMed]
 
Guidotti LG, Chisari FV. Immunobiology and pathogenesis of viral hepatitis. Annu Rev Pathol. 2006;1:23-61. [CrossRef] [PubMed]
 
Liu ZX, Kaplowitz N. Immune-mediated drug-induced liver disease. Clin Liver Dis. 2002;6(3):755-774. [CrossRef] [PubMed]
 
Boasso A, Shearer GM, Chougnet C. Immune dysregulation in human immunodeficiency virus infection: know it, fix it, prevent it? J Intern Med. 2009;265(1):78-96. [CrossRef] [PubMed]
 
Cohen Stuart JWT, Velema M, Schuurman R, Boucher CA, Hoepelman AI. Occult hepatitis B in people infected with HIV is associated with low CD4 counts and resolves during antiretroviral therapy. J Med Virol. 2009;81(3):441-445. [CrossRef] [PubMed]
 
Hamers RL, Zaaijer HL, Wallis CL, et al; PharmAccess African Studies to Evaluate Resistance (PASER). HIV-HBV coinfection in Southern Africa and the effect of lamivudine-versus tenofovir-containing cART on HBV outcomes. J Acquir Immune Defic Syndr. 2013;64(2):174-182. [CrossRef] [PubMed]
 
Gray D, Nuttall J, Lombard C, et al. Low rates of hepatotoxicity in HIV-infected children on anti-retroviral therapy with and without isoniazid prophylaxis. J Trop Pediatr. 2010;56(3):159-165. [CrossRef] [PubMed]
 
Pekovic V, Mayanja H, Vjecha M, et al; Uganda-Case Western Reserve University Research Collaboration. Comparison of three composite compliance indices in a trial of self-administered preventive therapy for tuberculosis in HIV-infected Ugandan adults. J Clin Epidemiol. 1998;51(7):597-607. [CrossRef] [PubMed]
 
Centers for Disease Control and Prevention (CDC). Treatment of tuberculosis. MMWR Morb Mortal Wkly Rep. 2003;52(RR11):1-77.
 
Saukkonen JJ, Cohn DL, Jasmer RM, et al; ATS (American Thoracic Society) Hepatotoxicity of Antituberculosis Therapy Subcommittee. An official ATS statement: hepatotoxicity of antituberculosis therapy. Am J Respir Crit Care Med. 2006;174(8):935-952. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
Short-Course Rifamycin and Pyrazinamide Treatment for Latent Tuberculosis Infection in Patients With HIV Infection*: The 2-Year Experience of a Comprehensive Community-Based Program in Broward County, Florida
PubMed Articles
Guidelines
Personality disorders in patients with HIV/AIDS.
New York State Department of Health | 7/20/2006
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543