Skip directly to search Skip directly to A to Z list Skip directly to site content
CDC Home

Severe Isoniazid-Associated Liver Injuries Among Persons Being Treated for Latent Tuberculosis Infection --- United States, 2004--2008

Since the 1960s, 6 to 9 months of isoniazid (INH*) has been the mainstay of treatment for latent tuberculosis infection (LTBI), but its application has been limited by concerns about the toxicity of INH and the long duration of treatment. To quantify the frequency of severe adverse events (SAEs) associated with LTBI treatment and to characterize the clinical features of affected patients, in January 2004 CDC began a national project to monitor SAEs associated with treatment for LTBI. State health departments were encouraged to report SAEs associated with any LTBI treatment regimen to a passive surveillance system. This report summarizes the results for 2004--2008, when 17 SAEs in 15 adults and two children (aged 11 and 14 years) were reported. All patients had received INH therapy and had experienced severe liver injury. Five patients, including one child, underwent liver transplantation. Five adults died, including one liver transplant recipient. These findings underscore the risk for an idiosyncratic drug-induced reaction in patients of any age treated with INH, including those with or without a putative predictor for INH-associated liver injury. Patients receiving INH for LTBI therapy should be monitored according to American Thoracic Society (ATS)/CDC recommendations because of the risk for drug-induced hepatoxicity (1,2). Providers should counsel patients to terminate INH therapy promptly and seek medical attention if they experience signs and symptoms of illness.

An SAE was defined as any drug-associated reaction resulting in a patient's hospitalization or death after at least 1 treatment dose for LTBI. Public and private health-care providers notified local health departments of SAEs. Local health departments then submitted standardized reports to CDC through their state health departments. Standardized reports included demographic information, LTBI treatment regimen, dates of treatment initiation and cessation, dates of hospitalization, results of testing for antibodies to viral hepatitis, clinical outcome, and dates of liver transplantation or death. Although the surveillance system was passive, CDC was available upon invitation to conduct extended onsite investigations. Investigations included medical record reviews and interviews of patients or their proxies and medical providers.

During 2004--2008, CDC received 21 reports of LTBI treatment--associated adverse events; however, four did not meet the SAE surveillance definition and were excluded from this analysis. All 17 patients with events meeting the SAE definition had received INH therapy and experienced liver injury. Of the 17 patients, two were children aged <15 years (Table 1). For the 15 affected adults, the median age was 39 years (range: 19--63 years). The SAEs were diagnosed between the second and ninth month of therapy, with the exception of one adult whose regimen spanned 17 months because of repeated treatment interruptions and who was diagnosed with an SAE in the seventeenth month. Sixteen patients tested negative for antibodies to hepatitis A (IgM anti-HAV), hepatitis B (antibody to hepatitis B core antigen) and hepatitis C (anti-HCV); one adult had pretreatment coinfection with hepatitis C virus (HCV) and human immunodeficiency virus (HIV) (Table 1). Of 17 patients, five underwent liver transplantation, including one child. Five of 15 adults died, including a liver transplant recipient.

Onsite clinical investigations

For 10 SAEs, state health departments invited CDC personnel to conduct onsite investigations (Table 2, Table 3). All 10 affected patients had indications for LTBI treatment, were prescribed INH within the recommended dosage range, and took the medication as prescribed. Prescribers followed ATS/CDC guidelines for monthly clinical monitoring of all 10 patients (1,2). Pretreatment serum aminotransferase concentrations were normal for five adults who underwent baseline testing (Table 2). Monthly aminotransferase monitoring was scheduled for two adults: one with HCV/HIV coinfection and another patient aged >35 years.

SAE symptoms began in the 10 patients 1--7 months after INH initiation (Table 3); for all patients, SAE diagnosis was based on symptoms rather than laboratory abnormalities. Seven patients initially experienced excess fatigue, nausea, or abdominal pain, but waited until the onset of jaundice before seeking medical attention. All patients had developed jaundice and markedly abnormal aminotransferase concentrations by the time of clinical evaluation. One patient had markedly abnormal aminotransferase concentrations 2 months before symptom onset, but the laboratory abnormalities were discovered incidentally during routine care by a provider who was unaware of LTBI treatment, and treatment continued until symptom onset. For seven of 10 patients, a provider other than the one who had prescribed the INH detected the SAE (Table 2).

For two patients, treatment was discontinued within 3 days of symptom onset (Table 2). Of the remaining eight patients, all discontinued INH at least 1 week after symptom onset. No patient discontinued INH until specifically instructed by a medical provider. All 10 patients underwent testing to exclude viral infections and other potential causes of liver injury. Liver biopsy or explanted liver histopathologic examination was performed for five patients; results from each revealed the presence of nonspecific changes consistent with drug-induced liver injury (3).

Seven of 10 patients had a putative predictor§ for INH-associated liver injury (Table 3). Of the three patients without a putative risk factor, two had ingested acetaminophen-containing medications during INH therapy; however, the two had taken standard doses for less than 1 week.

Reported by

State health departments; T Harrington, MD, L Manangan, MPH, J Jereb, MD, T Navin, MD, Div of Tuberculosis Elimination, National Center for HIV/AIDS, Viral Hepatitis, STDs, and Tuberculosis Prevention; K Powell, MD, EIS Officer, CDC.

Editorial Note

Approximately 4% of the U.S. population has latent tuberculosis infection (LTBI) (4). Because LTBI can progress to active disease, CDC recommends testing and treatment of LTBI for persons in certain groups (1). The findings in this report underscore the importance of following ATS/CDC recommendations (Box) regarding selection of candidates for LTBI treatment and for following recommendations for sustained clinical monitoring throughout LTBI treatment to detect rare, but severe, adverse events among patients of any age.

The finding that seven of 10 SAEs were diagnosed by medical providers other than the ones that prescribed INH indicates the importance of provider-to-provider and provider-to-patient communication for the safe administration of INH therapy. In this series, a diagnostic delay occurred for at least one patient who sought care from a provider other than the INH prescriber. Also, eight patients continued taking the medication while developing symptoms, a practice that has been noted in other published reports (5). Medical providers should emphasize to patients that INH treatment should be stopped immediately upon the earliest onset of symptoms (e.g., excess fatigue, nausea, vomiting, abdominal pain, or jaundice), even before a clinical evaluation has been conducted, and that initial symptoms can be subtle and might not include jaundice.

Two of the 17 patients in this series were children. Although the condition is thought to be rarer in children than in adults, INH-associated liver injury has been reported previously in children (6), and both clinicians and patients should be aware that SAEs can occur among patients of all ages. Nine of the 17 SAEs occurred beyond the third month of therapy, indicating that INH-associated liver injury is possible anytime during the treatment course. This finding was in contrast to an earlier study that found 10 of 11 episodes of INH-induced hepatotoxicity occurred during the first 3 months of therapy (7).

In this case series, all patients were monitored according to current guidelines (i.e., monthly clinical evaluation, including symptom screening and physical examination) (1,2), and two patients were selected for additional laboratory monitoring. However, despite adherence to current guidelines for monitoring, liver injury occurred, and SAE diagnosis was prompted by symptoms, not laboratory values. Additionally, three patients had no putative predictors of liver injury, indicating that careful monitoring is needed regardless of the patient's risk factor profile. Although all 10 patients in this series were symptomatic, INH-associated liver injury can occur even in the absence of symptoms.

INH-associated liver injury is an idiosyncratic reaction, independent of dosing, and is a diagnosis of exclusion (2). Historically the incidence has been estimated at 1 per 1,000 patients who begin treatment (1,2), but the lack of specific diagnostic criteria and heterogeneous definitions complicate comparisons across studies. The SAE surveillance system is the only national system that collects relevant public health data regarding the appropriateness of testing and treatment for LTBI and monitoring during treatment. However, as with all surveillance systems, underreporting is common in the SAE surveillance system, and LTBI is not reportable in most jurisdictions. In addition, calculation of INH-associated SAE rates is made difficult by the absence of reliable denominators for the number of persons initiating INH treatment, which has been estimated at 291,000 to 433,000 per year (8). Because the demographic characteristics of the patients who begin LTBI treatment with INH remain unknown, the risk factors for INH-associated liver injury cannot be determined conclusively.

LTBI treatment remains a key component of the TB elimination strategy in the United States. One study estimated that LTBI treatment prevented 4,000--11,000 TB cases in 2002 in the United States, substantially reducing the burden of TB (8). In the United States, 9 months of INH therapy is the standard LTBI treatment regimen. Efficacy and safety have not been established for other treatment regimens, such as 4 or 6 months of rifampin (9), 3 months of INH and rifampin (the preferred regimen in the United Kingdom [10]), or 3 months of once-weekly INH and rifapentine, a regimen currently under investigation (CDC, unpublished data, 2010).

Until an equally effective, better-tolerated regimen is developed, 9 months of INH therapy remains the mainstay of LTBI treatment. CDC encourages optimal use of INH by targeting LTBI testing to those patients most likely to benefit from treatment of LTBI (1). No more than a 1-month supply of INH at a time should be prescribed, and treatment should be combined with careful clinical monitoring (1,2). Alcohol consumption, underlying liver disease, and the concurrent use of medications that are metabolized in the liver can increase the occurrence or severity of liver injuries among INH recipients.

Local providers should report possible INH-associated SAEs to their respective health departments and to the Food and Drug Administration's MedWatch (https://www.accessdata.fda.gov/scripts/medwatch). State health departments should report these events to CDC's Division of Tuberculosis Elimination (e-mail: LTBIdrugevents@cdc.gov).

References

  1. CDC. Targeted tuberculin skin testing and treatment of latent tuberculosis infection. MMWR 2000;49(No. RR-06).
  2. American Thoracic Society. An official ATS statement: hepatotoxicity of antituberculosis therapy. Am J Respir Crit Care Med 2006;174:935--52.
  3. Czaja AJ, Carpenter HA. Optimizing diagnosis from the medical liver biopsy. Clin Gastroenterol Hepatol 2007;5:898--907.
  4. Bennett DE, Courval JM, Onorato I, et al. Prevalence of tuberculosis infection in the United States population: the national health and nutrition examination survey, 1999--2000. Am J Respir Crit Care 2008;177:348--55.
  5. CDC. Severe isoniazid-associated hepatitis---New York, 1991--1993. MMWR 1993;42:545--7.
  6. Pediatrics Tuberculosis Collaborative Group. Targeted tuberculin skin testing and treatment of latent tuberculosis infection in children and adolescents. Pediatrics 2004;114:1175--201.
  7. Nolan CM, Goldberg SV, Buskin SE. Hepatotoxicity associated with isoniazid preventive therapy: a 7-year survey from a public health tuberculosis clinic. JAMA 1999;281:1014--8.
  8. Sterling TR, Bethel J, Goldberg S, et al. The scope and impact of treatment of latent tuberculosis infection in the United States. Am J Respir Crit Care Med 2006;173:927--31.
  9. Menzies D, Long R, Trajman A, et al. Adverse events with 4 months of rifampin therapy or 9 months of isoniazid therapy for latent tuberculosis infection: a randomized trial. Ann Intern Med 2008;149:689--97.
  10. Joint Tuberculosis Committee of the British Thoracic Society. Control and prevention of tuberculosis in the United Kingdom: code of practice 2000. Thorax 2000;55:887--901.

* Isonicotinylhydrazine.

Monthly clinical monitoring (including a physical examination) for the signs and symptoms of adverse events is recommended by ATS and CDC for all LTBI treatment patients (1,2). Existing guidelines suggest that patients who have HIV infection, patients who have chronic liver disease, pregnant women, women in the immediate postpartum period (≤3 months of delivery), and patients who use alcohol regularly should be considered for baseline laboratory hepatic testing. Routine laboratory testing is indicated for patients whose baseline testing is abnormal and other persons at risk for hepatic disease (1,2).

§ Predictors of INH-associated liver injury include preexisting liver disease, HIV infection, injection-drug use, concurrent alcohol consumption, pregnancy or the immediate postpartum period (≤3 months of delivery), concomitant administration of medications with hepatotoxic potential, and older age (1,2).

What is already known on this topic?

Since the 1960s, 6 to 9 months of isoniazid (INH) has been the mainstay of treatment for latent tuberculosis infection (LTBI), but its application has been limited by concerns about the toxicity of isoniazid and the long duration of treatment.

What is added by this report?

During 2004--2008, a total of 17 serious liver injuries were reported in patients receiving INH therapy; five patients underwent liver transplantation, and five died, including one liver transplant recipient.

What are the public health implications for public health practice?

Patients receiving INH therapy for LTBI should be told categorically by medical providers to stop taking their medication immediately if they have symptoms such as nausea, vomiting, abdominal discomfort, or unexplained fatigue and to contact their providers for further evaluation.


TABLE 1. Reported severe adverse events (N = 17) associated with isoniazid (INH*) treatment for latent tuberculosis infection (LBTI), by patient characteristics --- United States, 2004--2008

Characteristic

No.

Age group (yrs)

≤15

16--35

>35

2

5

10

Sex

Male

Female

6

11

Race/Ethnicity

Hispanic

Black, non-Hispanic

White, non-Hispanic

8

1

8

Country of birth

United States

Foreign-born

10

7

Duration of INH treatment (days)

Median

Range

104

28--499

Period from initiation of INH treatment to severe adverse event symptoms (days)

Median

Range

109

56--502

Results of testing for viral hepatitis§

Negative

Abnormal

16

1

Outcome

Recovered

Had liver transplant

Died

8

5

5

* Isonicotinylhydrazine.

Includes one patient who received intermittent (>9 months) INH treatment for LBTI.

§ Includes testing to detect antibodies to hepatitis A (IgM anti-HAV), hepatitis B (antibody to hepatitis B core antigen) and hepatitis C (anti-HCV). One adult patient had pretreatment coinfection with hepatitis C virus and human immunodeficiency virus; testing for hepatitis A and B antibodies showed the presence of antibodies consistent with the patient's history of previous vaccination.

Includes one patient who died immediately after receiving a liver transplant.


TABLE 2. Results of onsite case investigations (n = 10) of severe adverse events (SAEs) associated with isoniazid (INH*) treatment for latent tuberculosis infection (LTBI), by case characteristics --- United States, 2004--2008

Characteristics

No.

Treated outside of a public health clinic

2

Had clinical monitoring monthly

10

Had laboratory monitoring of serum aminotransferase levels monthly

2

Results of baseline testing of serum aminotransferase

Within normal limits

Abnormal

Never tested

5

0

5

Period from SAE symptom onset to discontinuation of INH (days)

≤2

3--6

7--10

11--14

15--20

>20

1

1

4

0

2

2

SAE diagnosis by different clinician than the one who prescribed INH

7

Serum aspartate aminotransferase (AST) measurement at SAE diagnosis (international units/liter [IU/L])§

Median

Range

2,200

387--3,000

Serum alanine aminotransferase (ALT) measurement at SAE diagnosis (IU/L)§

Median

Range

2,192

272--3,000

Putative risk factors for INH-induced liver injury

None

Preexisting liver disease

Human immunodeficiency virus (HIV) infection

Concurrent injection-drug use

Concurrent alcohol consumption

Pregnancy or ≤3 months after delivery

Older age

Concurrent use of non-acetaminophen-containing medications with hepatotoxic potential††

3

1

1

0

3**

1

5

4

* Isonicotinylhydrazine.

Includes one patient with HIV infection and four of five patients aged >35 years.

§ The American Thoracic Society and CDC recommend that, in the absence of symptoms, INH should be discontinued if aminotransferase values are five times the upper limit of normal. In the presence of symptoms, INH should be discontinued if aminotransferase values are three times the upper limit of normal. All patients were symptomatic upon presentation when aminotransferase values were examined. All values exceeded the recommended threshold.

Predictors of INH-associated liver injury include preexisting liver disease, HIV infection, injection-drug use, concurrent alcohol consumption, pregnancy or the immediate postpartum period (≤3 months after delivery), older age, and concomitant administration of medications with hepatotoxic potential. Categories were not mutually exclusive.

** Upon prescription of INH, one patient without other predictors for liver injury had reported rare alcohol consumption (i.e., one drink per month). After SAE diagnosis, another patient reported weekly binge drinking with the intent to become intoxicated, and a third patient reported daily alcohol use during LTBI treatment. Neither of those patients reported alcohol use upon prescription of INH.

†† Medications with hepatotoxic potential included antiretroviral medications, a synthetic opioid medication, an antidepressant medication, a lipid-lowering agent, and an antihyperglycemic medication.


TABLE 3. Clinical characteristics of cases (n = 10) in onsite investigations of severe adverse events (SAEs) associated with isoniazid (INH*) treatment for latent tuberculosis infection (LTBI) --- United States, 2004--2008

Age (yrs)

Preexisting medical conditions

Putative predictors for liver injury

Concurrent medications with hepatotoxic potential

Symptoms leading to SAE diagnosis

Period to SAE symptom onset after INH initiation (mos)

Period from INH initiation to SAE diagnosis (days)

Duration of therapy after symptom onset (days)

Outcome

11

None

None

Acetaminophen for 3 days to treat fever 1 mo before symptom onset

Fatigue, mild icterus, depression for 1--2 days, then jaundice, vomiting for 1 day

7

209

3

Liver transplant

19

Morbid obesity,

migraine headaches

Concurrent excess alcohol consumption (about once weekly), reported after SAE diagnosis

Concurrent use of unidentified over-the-counter weight loss product; infrequent use of combination antiemetic and antidiarrheal medication after symptom onset

Diarrhea, nausea and vomiting, abdominal pain for 2--3 days, then fatigue and weakness

3

104

7

Recovery

24

None

None

Use of acetaminophen after onset of SAE-related symptoms (approximately 1 week before SAE diagnosis)

Nausea, abdominal pain, bloating for 17 months (waxing and waning), then fever, headache, myalgias, nausea for 4 days

2

499

438

Recovery

27

Hypothyroidism

None

Fatigue for 2 months, then icterus, dark urine for several days

1

146

107

Liver transplant

29

Eczema

Rare concurrent alcohol consumption

Fatigue, nausea for 2 weeks then icterus, dark urine and jaundice for several days

4

137

16

Death

35

HIV infection,

chronic hepatitis C virus infection, eczema

HIV infection,

chronic hepatitis C virus infection§

Concurrent administration of antiretroviral therapy, antibiotic therapy, and synthetic opioid medication

Pruritic rash and fever, fatigue, decreased appetite, nausea, vomiting, gradual darkening of urine for 1 week, then jaundice

3

87

7

Recovery

39

Morbid obesity, type 2 diabetes mellitus

Older age, ≤3 mos postpartum

Abdominal pain for 3 days, then nausea, diarrhea, dark urine, jaundice

4

121

2

Liver transplant, death

44

Depression, anxiety, obesity

Older age, possible concurrent daily alcohol use (reported after SAE diagnosis)

Concurrent use of selective serotonin reuptake inhibitor

Fatigue, nausea, vomiting, abdominal pain for 7 days, then jaundice for 2 days

3

97

9

Liver transplant

49

Hyperlipidemia, hypothyroidism, asthma

Older age

Concurrent use of lipid-lowering medication (statin)

Abdominal pain, fatigue for 7 days, then jaundice

3

91

9

Liver transplant

62

Type 2 diabetes mellitus

Older age

Concurrent use of sulfonylurea

Severe fatigue, left-sided flank pain for 2 weeks, then icterus, jaundice, dark urine for 5 days

1

56

20

Recovery

* Isonicotinylhydrazine.

Predictors of INH-associated liver injury include preexisting liver disease, HIV infection, injection-drug use, concurrent alcohol consumption, pregnancy or the immediate postpartum period (≤3 months after delivery), older age, and concomitant administration of medications with hepatotoxic potential.

§ Aminotransferase values were within normal limits at initiation of INH.


BOX. American Thoracic Society/CDC recommendations for targeted testing and isoniazid treatment for latent tuberculosis infection (LTBI) and monitoring during treatment

• Existing recommendations emphasize the careful selection of candidates for LTBI testing and treatment based on risk for infection. Persons who are not at risk for TB infection should not undergo testing for LTBI.

• Monthly clinical monitoring, including a brief physical examination, for the signs and symptoms of LTBI treatment--associated adverse events is recommended for all patients.

• Patients who have human immunodeficiency virus (HIV) infection, patients who have chronic liver disease, pregnant women, women in the immediate postpartum period (≤3 months after delivery), and patients who use alcohol regularly should be considered for baseline laboratory hepatic testing.

• Although baseline laboratory testing is not routinely indicated in older persons, it may be considered on an individual basis, especially for patients who are taking medications for chronic medical conditions.

• Routine laboratory testing is indicated for patients whose baseline testing is abnormal and other persons at risk for hepatic disease.

• An evaluation including laboratory testing should be obtained upon the first sign or symptom of a possible adverse event. Providers should educate patients to discontinue treatment immediately, even before an evaluation is conducted.

• In the absence of symptoms, isoniazid should be discontinued if aminotransferase values are five times the upper limit of normal.

• In the presence of symptoms, isoniazid should be discontinued if aminotransferase values are three times the upper limit of normal.

SOURCES: CDC. Targeted tuberculin skin testing and treatment of latent tuberculosis infection. MMWR 2000;49(No. RR-6).

American Thoracic Society. An official ATS statement: hepatotoxicity of antituberculosis therapy. Am J Respir Crit Care Med 2006;174:935--52.


Use of trade names and commercial sources is for identification only and does not imply endorsement by the U.S. Department of Health and Human Services.

References to non-CDC sites on the Internet are provided as a service to MMWR readers and do not constitute or imply endorsement of these organizations or their programs by CDC or the U.S. Department of Health and Human Services. CDC is not responsible for the content of pages found at these sites. URL addresses listed in MMWR were current as of the date of publication.

All MMWR HTML versions of articles are electronic conversions from typeset documents. This conversion might result in character translation or format errors in the HTML version. Users are referred to the electronic PDF version (http://www.cdc.gov/mmwr) and/or the original MMWR paper copy for printable versions of official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

**Questions or messages regarding errors in formatting should be addressed to mmwrq@cdc.gov.

 
USA.gov: The U.S. Government's Official Web PortalDepartment of Health and Human Services
Centers for Disease Control and Prevention   1600 Clifton Rd. Atlanta, GA 30333, USA
800-CDC-INFO (800-232-4636) TTY: (888) 232-6348 - Contact CDC–INFO
A-Z Index
  1. A
  2. B
  3. C
  4. D
  5. E
  6. F
  7. G
  8. H
  9. I
  10. J
  11. K
  12. L
  13. M
  14. N
  15. O
  16. P
  17. Q
  18. R
  19. S
  20. T
  21. U
  22. V
  23. W
  24. X
  25. Y
  26. Z
  27. #