Emergency Medicine

Battling Tuberculosis: A Strategic Update

Who is at risk for active tuberculosis? What are the parameters of a positive tuberculin skin test, and what distinguishes active from latent infection among patients with positive results? What special considerations apply to different therapeutic regimens for latent infection? The authors review the latest guidelines.

By Suzanne Atkin, MD, FACEP, Michael Jaker, MD, and Lee B. Reichman, MD, MPH

Dr. Atkin and Dr. Jaker are associate professors of medicine and Dr. Reichman is a professor of medicine and a professor of preventive medicine and community health at the New Jersey Medical School in Newark. Dr. Reichman is also executive director of the New Jersey Medical School's National Tuberculosis Center.

It was once thought that tuberculosis (TB), like smallpox, would be eliminated from the list of infectious causes of human suffering. Unfortunately, as the AIDS epidemic has progressed, it is now clear that TB infection remains a worldwide health concern. In addition, the development of multidrug-resistant TB in the prisons of the former Soviet Union and other sites has led to a resurgence of interest in this ancient disease.

New guidelines for tuberculin skin testing and for treatment of latent TB infection (LTBI) were developed in 1999 and published in 2000 by the American Thoracic Society and the Centers for Disease Control and Prevention. These guidelines define what is considered to be a positive tuberculin skin test for different groups of people, stratifying individuals both by the size of the tuberculin skin test area of induration and by risk factors for the development of active TB.

NOMENCLATURE CHANGES AND TARGETED TB TESTING

Individuals with positive tuberculin skin tests, using purified protein derivative (PPD), who do not have active TB are considered to have LTBI. In the past, these individuals were treated with isoniazid (INH), which was designed to prevent active TB, and the treatment was referred to as chemoprophylaxis or preventive therapy. Under the new guidelines, those terms should no longer be used to describe TB treatment. The terms have been replaced by "treatment of latent tuberculosis infection," which recognizes the fact that individuals with positive tuberculin skin tests have actually been infected with the tubercle bacillus and not merely exposed to it.

The term "targeted tuberculin testing" in the new guidelines refers to reserving tuberculin skin testing specifically for those groups considered to be at greater relative risk for developing active TB than the general population. In the past, the standard of care had been to test large populations for TB, including those at low risk for developing active disease. In contrast, the current guidelines discourage testing of individuals at low risk and emphasize the need to limit testing to groups at high risk for developing active TB.

Individuals at high risk for active TB are those who have been infected recently and those who have clinical conditions that predispose them to progression to active TB from LTBI. Persons at risk for recent infection include: close contacts of individuals with active pulmonary TB, those with recent skin test conversions from negative to positive, recent immigrants from areas with high incidence rates of TB, children with positive skin tests, those infected with the human immunodeficiency virus (HIV), homeless persons, injection drug users, and individuals who work or reside in institutional settings such as hospitals, homeless shelters, prisons, and nursing homes.

Clinical conditions predictive of progression from LTBI to active TB include: HIV infection, injection drug use (especially injection drug use by individuals who are infected with HIV), long-term corticosteroid or other immunosuppressive drug use; chest x-rays suggestive of prior untreated TB (with pulmonary fibrotic lesions), and others listed in the table.

Target Groups for TB Skin Testing
	Persons at increased risk of recent infection with TB
	Close contacts of persons with active TB Persons whose PPD skin test converts from negative to positive within a two-year period Recent immigrants from areas with high incidence rates of TB Children, especially those younger than five years old, who have a positive PPD skin test; also, adolescents and young adults Homeless persons Persons who are HIV positive Injection drug users People who reside or work in institutional settings such as hospitals, homeless shelters, prisons, nursing homes
	Clinical conditions associated with progression from LTBI to active TB
	HIV infection Injection drug users, with even greater risk among those who are co-infected with HIV and TB Persons with chest x-rays suggestive of prior untreated TB (fibrotic lesions) Underweight persons (loss of >10% of ideal body weight) Silicosis Persons with diabetes, with even greater risk among those who are insulin-dependent or poorly controlled Gastrectomy with weight loss and malabsorption Jejunoileal bypass Kidney or cardiac transplant Cancer of the head, neck, or lung; also, lymphoma and leukemia Persons receiving corticosteroids or other immunosuppressants

DIAGNOSIS OF LTBI

The three diagnostic tests for LTBI involve skin testing, chest x-rays, and sputum analysis.

Skin testing. Although the Mantoux intradermal tuberculin skin test (commonly referred to as the PPD) has been used for more than a century, it remains the best test for diagnosing LTBI. Multiple puncture tests such as the tine test are not considered accurate and are not recommended. The tubercle bacillus usually will not produce a positive skin reaction until 2 to 12 weeks after infection. In patients who are infected, this delayed hypersensitivity skin reaction to the antigenic proteins begins within five to six hours of intradermal injection and peaks at 48 to 72 hours. A positive skin test is not specific to mycobacterium TB; nontuberculous mycobacteria and vaccination with bacille Calmette-Guerin (BCG) may also cause positive tests.

In order for a skin test to be interpreted as positive, both the transverse diameter of skin induration and the relative risk category of the patient must be known (see the table below). The three cut-off levels recommended are: 5 mm or more (for individuals who are immunosuppressed or are at high risk for developing active TB if infected); 10 mm or more (for individuals with an increased likelihood of recent infection); and 15 mm or more for those individuals at lowest risk.

PPD Skin Induration Considered Positive
	>5 mm	• HIV positive • Recent contact with active TB case • Organ transplant recipients • Immunosuppressed patients* • Chest x-ray suggestive of old TB
	>10 mm	• Recent immigrants (within 5 years) from areas with endemic TB • IV drug users • Residents and employees of health care facilities, shelters, and prisons • Patients with comorbid conditions likely to progress to active TB (see table, above) • Children less than four years old • Recent increase in size of PPD induration by >10 mm or conversion from negative to 10 mm.
	>15 mm	Individuals at low risk for TB
	*Individuals receiving the equivalent of 15 mg or more per day of prednisone for two to four weeks or longer

While vaccination with BCG may cause skin induration with tuberculin testing, the reactivity decreases with time and is not likely to be the cause of skin induration of 10 mm or more in an adult. Because there is no way to determine whether skin induration in a BCG-vaccinated patient is due to LTBI or BCG, and because the vaccination does not reliably protect against TB infection, it is recommended that the reaction be considered positive if the patient falls into any of the groups considered to be at high risk for recent infection or for development of active TB.

Chest x-rays. It is recommended that each patient with a newly positive tuberculin skin test have a chest x-ray to exclude the likelihood of active TB or changes suggestive of old TB. There is a greater risk of progression to active TB when the chest x-ray reveals noncalcified nodules and fibrotic scars than if there are calcified nodules or apical or basilar pleural thickening. Pregnant women who have LTBI or who have had recent contact with a person with active TB and who have negative skin tests are advised to have a chest x-ray with uterine shielding (even during the first trimester) because of the increased risk of congenital TB and progression to active TB.

There is no indication for annual or periodic screening chest x-rays in any individuals with LTBI unless they are symptomatic.

Sputum analysis. It is not recommended that patients with LTBI have sputum cultures or smears for TB unless their chest x-rays show evidence consistent with prior TB. In such individuals, three sputum samples should be collected on different days for smear and culture. Patients infected with HIV who have respiratory symptoms and LTBI should have sputum analysis for TB even if they have negative chest x-rays.

TREATMENT OF LTBI

It is recommended that single-drug treatment of LTBI not be initiated until the possibility of active TB has been excluded. The recommended treatment for LTBI in all patients (including HIV-infected, pregnant, and pediatric patients) is currently INH daily for nine months. Several other regimens are acceptable alternative treatments (see table below). There are two alternative regimens that employ twice-weekly dosing, which should always be given as directly observed therapy. It should be recognized that compliance with therapy is often problematic not only because patients with LTBI are asymptomatic but also because the required treatment is lengthy.

Recommended Drug Regimens for LTBI
Drug	Regimen	Comments
isoniazid (INH)	Daily for nine months	Preferred regimen Requires clinical monitoring monthly Toxicities: rash, hepatitis/elevated liver enzymes, peripheral neuropathy Elevates phenytoin level

	Twice weekly for nine months	Must have directly observed therapy (DOT) Requires clinical monitoring monthly

	Daily for six months	Requires clinical monitoring monthly Not indicated for pediatric patients, HIV-positive patients, or those with old untreated fibrotic lesions on chest x-ray

	Twice weekly for six months	Must have DOT Requires clinical monitoring monthly Not recommended for pediatric patients, HIV-positive patients, or those with old untreated fibrotic lesions on chest x-ray
rifampin plus pyrazinamide	Daily for two months	Requires clinical monitoring at two, four, six, and eight weeks; liver enzymes and bilirubin levels at baseline and at two, four, and six weeks. Rifampin toxicities: nausea, vomiting, diarrhea, rash, hepatitis, fever, thrombocytopenia, flu-like syndrome, orange tears/sweat/urine. Contraindicated in HIV-positive patients taking protease-inhibitors or non-nucleotide reverse transcriptase inhibitors (rifabutin can be used instead) Many drug interactions (decreases levels of oral contraceptives, oral hypoglycemics, coumadin, digitalis, anticonvulsants, and ketoconazole, among others) Pyrazinamide toxicities: gastrointestinal upset, hepatitis, rash, arthralgias, hyperuricemia (common), acute gout (rare) Associated with reports of liver injury resulting in clinical and laboratory findings of hepatitis and in hospitalization or death

	Twice weekly for two to three months	Must have DOT
rifampin	Daily for four months	Requires clinical monitoring monthly Toxicities: See above
pyrazinamide plus ethambutol or pyrazinamide plus quinolone	Daily for six months	Recommended for persons likely to be infected with multidrug-resistant TB If patient is immunocompromised, treatment is recommended for 12 months. Pyrazinamide toxicities: see above. Ethambutol toxicities: nausea, vomiting, abdominal discomfort, elevated LFTs, hyperuricemia, rash; may cause peripheral neuropathy, optic neuritis, or neurologic symptoms. Quinolone toxicities: gastrointestinal symptoms, elevated LFTs Note: Quinolone may chelate with calcium, magnesium and aluminum containing antacids, which will decrease its absorption.

In the past, it was recommended that INH treatment not be given to patients with LTBI at low risk for active TB who were over 35 years of age because of the risk of drug-induced hepatotoxicity. Although it is true that the risk of hepatotoxicity increases with age, it is now recommended that INH be given to those individuals with LTBI regardless of age.

It is also important to note that the current recommendation for pregnant women who are at high risk for progression of LTBI to active TB or who are HIV positive is not to delay treatment solely because of pregnancy, even during the first trimester.

For those patients with LTBI who were contacts of individuals infected with multidrug-resistant TB (INH- and rifampin-resistant) and who are at high risk for progression to active TB, the current recommendation is treatment with pyrazinamide and ethambutol or with pyrazinamide and a quinolone for six months for immunocompetent patients and 12 months for immunocompromised patients. For those patients with LTBI who were contacts of individuals with only INH-resistant TB, the current recommendation is treatment with rifampin and pyrazinamide daily for two months.

Isoniazid interferes with the metabolism of vitamin B6 (pyridoxine) and may cause peripheral neuropathy. This is rarely problematic with the standard INH dose of 5 mg/kg, except for those patients with LTBI taking INH who are pregnant, have seizure disorders, or are at significant risk for the development of peripheral neuropathy due to comorbid conditions such as diabetes, uremia, alcoholism, malnutrition, and HIV infection.

PATIENT MONITORING DURING TREATMENT

It is recommended that clinical monitoring be performed at the start of treatment and then monthly in patients being treated with single-drug therapy (INH or rifampin) and at two, four, six, and eight weeks for those receiving both rifampin and pyrazinamide for two months. Clinical monitoring consists of a detailed review for possible signs and symptoms of hepatitis or drug-induced side effects, including rash, fever, jaundice, nausea, vomiting, fatigue, right upper quadrant abdominal pain, and paresthesias.

Routine hepatic function testing is currently not recommended merely on the basis of age greater than 35. Evaluation of hepatic function is indicated at the start of treatment for patients with LTBI in the following cases: HIV infection, pregnancy or within three months postpartum, regular alcohol use, a history suggestive of liver disease, and current use of potentially hepatotoxic medications. Required tests are aspartate aminotransferase (AST), alanine aminotransferase (ALT), and bilirubin levels.

It is recommended that repeat liver function tests be monitored in those patients with abnormal baseline tests and in those with signs or symptoms of or at risk for liver disease. Isoniazid should be discontinued in symptomatic patients whose serum aminotransferases are more than three times normal and in asymptomatic patients whose aminotransferases are more than five times normal.

Due to recently reported cases of hepatotoxicity in patients taking the two-month regimen of pyrazinamide and rifampin, the current recommendation for monitoring of this treatment differs from the recommendation for INH. Liver function tests are recommended at the start of treatment and at two, four, and six weeks. It is important to note that asymptomatic elevations of AST and ALT are expected with pyrazinamide and rifampin treatment and do not require cessation of therapy. This regimen should be discontinued if the AST or ALT is more than five times the upper limit of normal in asymptomatic individuals or greater than the upper limit of normal in individuals with signs or symptoms of hepatitis, or if the bilirubin level is higher than normal.

Pyrazinamide and rifampin are not recommended for patients with INH-associated liver toxicity or underlying liver disease and should be used with caution in patients who are taking other potentially hepatotoxic drugs and those with a history of alcoholism. Although pyrazinamide is commonly associated with elevated serum uric acid levels, acute gouty arthritis is rare. In those patients who develop acute arthritis, uric acid levels should be assessed and treatment of hyperuricemia initiated, if indicated.

Complete blood count and platelet levels are indicated at the start of therapy and during repeat monitoring for those patients taking rifampin or rifabutin.

Examples of Test Interpretation

Case 1. A 27-year-old nurse who has been working for six months in a nursing home is found to have a tuberculin skin test result of 10-mm induration. No previous testing is documented. However, she recalls a negative skin test two years ago at another job. She is asymptomatic and has no other medical history.
Discussion. This nurse is at high risk for recent infection with TB because she is both a recent converter and a health care worker. Therefore, her skin test result of 10-mm induration is considered positive. Her workup should include a chest x-ray but no sputum analysis because she has no respiratory symptoms. If her chest x-ray is negative, she should be started on INH daily for nine months with clinical monitoring each month. At the end of the nine months, she can be discharged from care and treatment should be documented. There is no need for follow-up chest x-rays.

Case 2: A 50-year-old, HIV-positive man is found to have a tuberculin skin test result of 6-mm induration on routine testing at an AIDS center. No previous skin test has been done, and the patient has a dry, chronic cough.
Discussion. This patient is considered to be at highest risk for developing active TB because of his HIV infection. Therefore, his test result of 6-mm induration is considered positive. His workup should include a chest x-ray; sputum analysis for TB should also be done, whether or not the chest x-ray is abnormal, because he has respiratory symptoms. In addition, baseline liver enzyme levels should be assessed. If his chest x-ray does not suggest active TB, his sputum smear and culture are negative, his liver function tests are normal, and any identified infected contact has been determined not to have resistant TB (either by actual assessment or local epidemiologic information), he should be treated with INH and pyridoxine daily for nine months. He should be followed with monthly clinical monitoring during treatment.

Case 3. A 36-year-old man, a Haitian immigrant who came to this country three years ago, is found to have a tuberculin skin test result of 12-mm induration on job-entry testing at a minimum security prison. The man claims to have received BCG vaccine as a child.
Discussion. This man is considered to be at increased risk of progression to active TB because of his recent immigration status and country of origin, and his history of BCG vaccination does not change the therapeutic approach. If his chest x-ray is normal, he should have treatment with INH daily for nine months. Also, the six-month treatment regimen of pyrazinamide and ethambutol should be considered if the contact for this patient was known or strongly suspected to have multidrug-resistant TB (highly prevalent in Haiti). He should be followed with monthly clinical monitoring during treatment and discharged with documentation of treatment and instructions to return if symptoms occur. There is no need for follow-up chest x-rays.

Case 4: A 58-year-old airline pilot is found to have a tuberculin skin test result of 10-mm induration during a routine life insurance physical examination. His last skin test result was 4-mm induration two years ago.
Discussion. This man is at low risk for TB, is not considered to have a positive test, and does not need chest x-rays or therapy.

GLOBAL THREAT

With the progression of the AIDS epidemic, TB infection continues to be a global health threat. The new guidelines issued by the American Thoracic Society and the Centers for Disease Control and Prevention on tuberculin skin testing and treatment of LTBI will help physicians diagnose this disease more accurately and manage patients more effectively.