She is not as alert as usual and has a stiff neck. Her present chest film shows hilar adenopathy with multiple small lesions throughout all lung fields. A head CAT scan demonstrates enlarged lateral ventricles, enhancement of the meninges at the base of the brain and small vessel brain stem infarcts with possible pons involvement. There are no suggestions of significantly increased intra-cranial pressure. A lumbar puncture is performed. The opening pressure is 21 cm, white cell count 525, 85% lymphs, protein 86, glucose 55 (blood glucose 92). Cultures are sent which later grow Mycobacterium tuberculosis.

Known as "consumption", the white plague, and the white death, tuberculosis is perhaps the earliest, documented bacterial disease of humanity. Because of its characteristic pulmonary scarring and bony changes, it has been documented in human remains as ancient and diverse as the Egyptian and Incan mummies to the sand preserved natural remains found in the Chinese deserts of Asia Minor.

In the late part of the 19th and early part of the 20th century, almost all adults showed evidence of exposure and immune response to Mycobacterium tuberculosis. The recognition of TB as a major health issue is even becoming known in the popular press. In a special on War on Disease, National Geographic reported that one third of humanity are now carriers of tuberculosis (1). TB exposure was determined by a positive OT (old tuberculin) skin test. Skin testing by PPD (purified protein derivative) is now the standard (available since the 1930s) (1,2).

In the United States in general, and Hawaii in specific, there has been a rapid decrease in tuberculosis because of dramatic improvement in the living conditions of city dwellers (sociologic), the recognition of the means of spread the disease (epidemiological), isolation of contagious individuals in sanatoriums, as well as the development of anti-tuberculosis chemotherapy (bacteriologic and anti-microbial). There has been a 9-fold decrease in reported TB cases in Hawaii from 1060 cases in 1930 to 136 cases in 2000. This is in spite of a dramatic increase in population over that same time span. Indeed, tuberculosis was considered a vanishing disease and the medical specialists in this field and the public health resources spent on it were vanishing as well. It was thought of as a residual third world problem.

Hawaii has always had rates of infection much higher than the rest of the country. The case rate of TB in Hawaii is now about 2 times the national average 11.2 vs. 5.8 cases per 100,000 population (4). It is also clear that most newly diagnosed cases are imported and occur in foreign-born individuals (4). The spike in cases in 1992 was due to World War II Philippine veterans being required to come to the US to register for veteran benefits and having their TB recognized while here. Foreign born TB cases in Hawaii account for 83% of all cases as compared to 46% of cases on the mainland (1). In Hawaii, these are patients are from Southeast Asia and the West Pacific Islands. Patients from the Philippines account for 63% of TB cases while native born cases account for only 16%. Pacific Island immigrants from COFA (Compact of Free Association) (former trust Territories, Republic of Marshall Islands, Federated States of Micronesia, Palau) made up 44% of the cases from 1995-2000. In the rest of the United States, TB is also found among foreign born, but these individuals are from Mexico and South America in addition to Asia (5). The length of residence in Hawaii at the time of a TB diagnosis was <1 year whereas on the mainland it occurred somewhat later at 1-5 years of residency (6). From these data, the highest risk individuals are foreign born and are diagnosed soon after coming to Hawaii.

The diagnosis of TB depends upon clinical suspicion, isolation of the organism or its specific DNA, the tuberculosis skin test (PPD) or the response to therapy. It can involve any organ, can range from having no symptoms to overwhelming symptoms, progress indolently or become rapidly fatal, can be local or systemic. These facts make it clear that the slogan "Think TB" is correct, especially here in Hawaii.

The TB skin test is the only available standardized method for identifying latent TB infection. It consists of injecting 0.1ml of intermediate PPD (5 TU) intracutaneously into the volar surface of the forearm. It is to be read at 48-72 hours although a positive result can be read up to 7 days after testing. It must be read by a trained health professional. Self-reading is not allowed. Induration, not erythema, is measured in millimeters. The cutaneous reaction depends on an intact cellular immune response to the TB antigens. Tine or multiple puncture tests are no longer allowable. False positive tests do occur with other non-TB mycobacterium antigens. False negatives occur both with improper placement of the antigen as well as when immunity decreases or is interfered with by other disease. The Center for Disease control has suggested different cut off levels for different populations. Positivity is based on a statistical probability that a reaction reflects true Mycobacterium tuberculosis infection. A 5 mm test is considered positive in cases of close contact with a confirmed case, in patients with an abnormal chest film consistent with old or active TB, on immigration visa adjustment examination and in patients with HIV (a 5mm test might still be falsely negative in HIV patients with a CD4 count of less than 100). A 10 mm reaction is considered positive in areas of high TB incidence such as Hawaii. It includes most TB positive patients with a relatively small number of false positives. In all other groups, a 15 mm test is considered positive. This higher level is especially useful in areas that have a high incidence of exposure to atypical Mycobacterium. A booster phenomena is known where the initial test is negative but reverts to its correct positive status with repeat testing in 1-4 weeks. Individuals who are truly negative will not boost. Repeated testing does not sensitize uninfected individuals. Anergy panel skin testing (testing for anergy) is no longer recommended. It was unreliable due to non-standardized antigens and did not seem to have clinical relevance.

BCG (Bacille Calmette-Guerin) vaccination is still used in some other countries. It is effective in preventing disseminated tuberculosis, especially tubercular meningitis in infants. Its efficacy in preventing pulmonary tuberculosis is debated. It is still used in countries with high rates of tuberculosis. It is also recommended as part of a treatment program for newborns of mothers with active pulmonary tuberculosis. It is a live vaccine derived from Mycobacterium bovis strains. The current recommendation is that individuals who have received BCG should have this fact ignored and their PPD tests interpreted the same as other individuals. The only exception is if the immunization was given less than 1 year previously.

Replacement tests for the PPD skin test have been sought to help with the difficulty of correct placement, having the patient return for a reading of the test, correct reading of the PPD and correct evaluation of these results. A newer test, Quantiferon, is a blood test that is in the final stages of FDA testing and will, in all likelihood, replace PPD testing. It recognizes Mycobacterium TB specifically and eliminates the cross-reactions in patient who have been exposed to atypical organisms as well as those who have received BCG immunizations.

Because of all of the problems noted above and because of the emergence of multiple drug resistant TB, it is very important to isolate the organism in suspected cases of tuberculosis. Children rarely produce sputum and this makes the isolation of organisms especially problematic.

Asymptomatic primary tuberculosis (7) is the most frequent clinical presentation in pediatrics. It is recognized during contact work-up or routine tuberculin testing. The child is asymptomatic. The chest film is usually normal but might have some hilar adenopathy. A 25 year study of TB in Baltimore documented the morbidity and mortality of non-treated, asymptomatic TB (8). For this reason, adults with asymptomatic TB are treated with INH for 6 months. The duration of therapy in children is longer than in adults. Converters are treated with INH for 9 months. This minimizes the occurrence of more progressive TB infections. In our case example, the positive PPD led to treatment as is appropriate but the therapy was discontinued. This results in the risk of drug resistance as well as subsequent disease as noted above.

Because treatment with INH for 9 months is associated with completion rates as low as 5-25% (9,10) shorter duration regimes have been advocated for latent TB including a 2 month schedule of rifampin and pyrazinamide. The reports of death from hepatitis associated with this regime have brought this practice into question (11). In addition, because of the compliance issues noted above, direct observed therapy has become more common.

Our patient might well have pulmonary tuberculosis. Pulmonary tuberculosis can be seen in children as well as adults. In adults, sputum smears are positive in 40-60% (i.e., organisms are identified on an acid fast stain of the sputum) of confirmed cases of pulmonary TB. Most cases of TB in adults have positive cultures or smears (82%). 96% of cases with cavitary pulmonary lesions on CXR will have a positive culture, as will 70% of cases with focal infiltrates. A positive smear requires 10,000 to 1,000,000 organisms/ml of sputum. In cavitary disease, the smears are usually positive and a negative smear strongly suggests another etiology. The culture, of course, is the gold standard for tuberculosis. Cultures are usually positive in 4-8 weeks although BACTEC liquid cultures may be positive in 2-4 weeks (12). In these studies, antibiotics to suppress all other growth are added and radiolabeled carbon-14 substrates are used. Radiolabeled carbon dioxide is measured as an indication of the presence of Mycobacterium tuberculosis.

Other technologies have the possibility of more rapid diagnosis for tuberculosis. Nucleic Acid Amplification Tests (NAATs) are available in 2 probes, Gen-Probe MTD and Roche Amplicor. Guidelines for the use of these tests as well as an algorithm are available (13). They are licensed for use on respiratory samples only and are limited to patients who have not been on anti-TB treatment for more than 7 days or who have been treated within the past 12 months. They do not replace cultures. In smear positive cases, they have a sensitivity of 95% and specificity of 98%. In smear negative cases, they have a sensitivity of 50% and a specificity of 95%. If cultures and NAATs are discordant, the cultures remain the gold standard. Additional means of obtaining specimens include fine needle aspiration, bronchoalveolar lavage or transbronchial biopsies. These tests are also useful to uncover a different diagnosis.

Since children do not produce sputum, isolation of the organism is much more difficult. Gastric washings may be used as a specimen source. Even after this, the organism may not be recovered. When all of these tests fail in a pediatric patient, a search should be undertaken for the organism from the person infecting the child. This organism, if found, can serve as a surrogate specimen for drug sensitivity testing.

Pleurisy with effusions, although rare in young children, may be seen in older children and is more common in boys than in girls. Its onset is rapid with signs of pneumonia. Fever is usually present and can last for weeks. Thoracentesis yields fluid with a low glucose, high protein and predominantly lymphocytic cells. Smears of the effusion are frequently negative for TB, as are cultures. Pleural biopsy, perhaps by a thorascope these days, increases the yield of both of these studies (7).

TB by diagnostic site shows that pulmonary makes up of 87% of cases, pleural 3%, lymphatic 6%, bone/joint <1%, genitourinary <1% and meningeal 2%. In pediatric patients, all of these infections are possible but the most likely are pulmonary TB and meningitis.

Our patient has confirmed tubercular meningitis. Note that the cell count is largely mononuclear/lymphocytic, similar to viral meningitis in this regard. The CAT scan noted above is virtually diagnostic. Isolation of the organism is crucial as in all cases of TB. Drug therapy is aimed at the isolated organism but until that occurs 4 or 5 drugs are used.

Multiple drug resistant TB is the major concern. Primary drug resistance to anti-TB medications occurs in nature. For INH, the rate of resistance is 1 in 1 million, for rifampin 1 in 100 million and for both INH and rifampin 1 in 100 trillion. Since cavitary lung lesions contain approximately 1 billion organisms, naturally occurring, primary resistance has not been a real problem. However, partially treated infections either due to inadequate length of therapy or lack of compliance, results in selecting multiple drug resistant TB organisms which is a major issue in treating TB patients and public health prevention programs. The standard treatment now consists of 4 drugs, INH and rifampin for 6 months and pyrazinamide and ethambutol for 2 months. Streptomycin can be substituted for ethambutol. Further treatment is based on cultures and sensitivity. In pediatric patients, treatment is more likely to be empiric. Some providers avoid ethambutol because of the difficult of monitoring for ethambutol associated optic neuritis. There have been no documented cases of optic neuritis in children. Because of this, many still use ethambutol.

Those with meningeal, bone or otherwise disseminated disease are treated for 12 or more months. All patients should receive directly observed therapy. In patients with confirmed drug resistance, changes in their regimes are required such that second and third line drugs are substituted (e.g., ethionamide, cycloserine, PAS, quinolones, clofazimine, amikacin, kanamycin, capreomycin). The length of treatment is also prolonged. These recommendations have changed frequently and it is best to get recommendations from the Department of Health or infectious disease experts.

The long-term consequences of TB meningitis are real and hydrocephalus and decreased functioning are almost universally seen.

TB requires vigilance both for the individual patient and the community. The rule is THINK TB.

Questions

1. True/False: Tuberculosis is a disease of the past and no longer a major health care issue.

2. True/False: Testing with PPD is a useful screening test for patients suspected of having tuberculosis.

3. True/False: A history of BCG vaccination makes PPD testing contraindicated and the results unreliable.

4. True/False: Children with a positive PPD skin test and a positive chest film should be treated with INH alone for 9 months.

5. True/False: The risk of multiple-drug resistant TB is much higher in patients that did not complete initial TB therapy.

6. True/False: Immigrants are at greater risk of having TB than native-born Americans.

7. True/False: Patients with HIV/AIDS have a higher rate of acquiring pulmonary TB than the general population.

8. True/False: Health care workers are at a greater risk of acquiring TB than the general population.

9. True/False: Ethambutol cannot be used in pediatric patients since vision testing is often impossible.

10. True/False: Hawaii has a lower rate of TB than the rest of the US.

References

1. Weiss R. War on Disease. National Geographic 2002;201(2):17.

2. Seibert FB. The isolation and properties of the purified protein derivative of tuberculin. Amer Rev Tuberc 1934;30:713.

3. Seibert FB, Munday B. The chemical composition of the active principle of tuberculosis. XV. A precipitated purified tuberculin protein suitable for the preparation of a standard tuberculin. Amer Rev Tuberc 1932;25:724.

4. www.state.hi.us/doh

5. Zuber PLF, McKenna MT, Binkin NJ, et al. Long term risk of tuberculosis among foreign-born persons in the United States. JAMA 1997;278(4):304-307.

6. www.nationaltbcenter.edu

7. Smith M. Tuberculosis in children and adolescents. Clin Chest Med 1989;10:381.

8. Brailey ME. Tuberculosis in white and negro children. II. The epidemiologic aspects of the Harriet Lane Study. 1996, Cambridge: Harvard University Press.

9. Bock NN, Metzger BS, Tapia JR, Blumberg HM. A tuberculin screening and isoniazid preventive program in an inner-city population. Am J Respir Crit Care Med 1999;159:295.

10. Tulsky JP, Pilote L, Hahn JA, et al. Adherence to isoniazid prophylaxis in the homeless: a randomized controlled trial. Arch Intern Ed 2000;160:697.

11. Centers for Disease Control and Prevention. Update: fatal and severe liver injury associated with rifampin and pyrazinamide for latent tuberculosis, and revisions in the American Thoracic Society/CDC recommendations-United States. Am J Respire Crit Care Med 2001;164:1319.

12. Good RC, Mastro TD. The Modern Mycobacterium Laboratory: How it can help the Clinician. Mycobacterial Diseases. Clin Chest Med 1989;10:315.

13. Update: Nucleic Acid Amplification Test for Tuberculosis. MMWR 2000;48:593.

Answers to questions

1.F. 2.T, 3.F, 4.F, 5.T, 6.T, 7.T, 8.T, 9.F, 10.F