Emerging Infections of International Public Health Importance
Tuberculosis: Global Impact and Drug Resistance
Carrie Horwitch MD, MPH
Dr. Horwitch is a clinical assistant professor of Health Services in the School of Public Health and Community Medicine at the University of Washington. She is also co-investigator for the APEC Emerging Infections Network project and has been co-coordinator and instructor for the Emerging Infectious Diseases of International Public Health Importance course since its inception. She is an internal medicine specialist with experience in HIV/AIDS care and tropical medicine training. She is an Associate Program Director for internal medicine residency and is the Director for the ambulatory and HIV clinic at the Virginia Mason Medical Center.
Tuberculosis (TB) is an infectious disease that has been around for thousands of years. You might wonder why we are discussing TB as part of a course on emerging infectious disease. The focus of this lecture will be to illustrate how an old disease can “re-emerge” and have new factors that increase its burden in the world and its interaction with a true emerging infection.
TB is the fourth most common infectious diseases worldwide and the second most common disease in adults. Approximately 30 million people are infected with TB worldwide. Of those infected, about 10 million will develop active disease and there will be close to 3 million deaths annually. If there are no changes in current levels of infection, it is estimated that 1 billion people will be infected by 2020, responsible for 200 million active cases and 55 million deaths.
Here is a world map of the global distribution of tuberculosis.
Sub-Saharan Africa and Southeast Asia are the countries with the highest incidence. Two-thirds of the world’s cases of TB are in Asia, although some African countries carry the highest incidence (165 cases per 100,000). The former Soviet Union now has an incidence of 78/100,000 (increase from 34/100,000 in 1991) and the United States an incidence of 5.8/100,000.
Between 1985-1992 in the United States, TB cases increased by 20% mostly due to the rise of HIV infection and the breakdown of public health and TB funding. In 2000, there were 16,377 cases of TB; of those the majority was foreign born (25.8/100,000, compared to US born 3.5/100,000). Since 1999, there has been a 7% decrease in incidence.
In Seattle-King County, reported cases of TB rose from 104 cases (1999) to 127 cases (2000). Seventy-nine percent of the cases were in foreign born persons. The highest burden of the disease is in King County (7.3/100,000) compared to the rest of Washington State (4.4/100,000).
TB is a disease that has strong interplay between the environment, the host and the agent.
This schematic shows how these three interact with each other. Not all persons who become exposed to TB get infected, and all those who are infected do not get active disease. Likewise, there are certain strains of TB that appear to be prominent in outbreak situations especially in health care facilities. Like many diseases, TB is more commonly seen in more crowded and impoverished areas.
Pathobiology of TB
The incubation period is 2-10 weeks. Transmission and infection depend on the following:
Certain groups are at higher risk of activating latent TB infection. These are:
Tuberculosis is transmitted by droplets. When an infected person coughs, the droplets are passed into the airway of the exposed person. This is why close proximity to someone who is ill is a factor for acquiring the infection. The bacillus then enters the lung and may cause active TB disease or may be “walled off” by the body’s immune system and form a granuloma. Days, weeks, or years later, the TB bacillus can re-emerge and cause active disease. TB infection (also known as latent infection) is seen in the majority of the cases. These persons are NOT infectious to others, as they have no active disease. Persons with latent TB are the source of most future TB disease. We may be able to identify those persons who were exposed by use of the TB skin test, known as the PPD or TST.
Clinical manifestations of active TB disease include cough, weight loss, fever and night sweats for pulmonary TB. There can also be extrapulmonary TB manifestations such as TB meningitis, military TB, gastric TB, lymphadenopathy and bony abnormalities.
Diagnosis is difficult and relies on signs and symptoms mentioned above, positive sputum smear for acid-fast bacilli, chest x-ray finding compatible with TB (if pulmonary) and a positive polymerase chain reaction (PCR) or culture for TB.
Tuberculosis and HIV co-infection
Co-infection with tuberculosis and HIV virus has changed the dynamic of this disease dramatically. There is an interplay in both directions with these two disease. The presence of HIV infection and immune compromise increases the risk of activating latent TB. The stage of HIV has an impact on the presentation and outcome for TB disease. Treatment of HIV has also been shown to improve TB outcomes.
Here is a map of persons living with HIV/AIDS globally.
Here is a map of estimated TB and HIV co-infection.
You can see from these two slide and the one previously that where we have high incidence of TB, we also have high incidence and prevalence of HIV infection. This is a deadly combination where the largest burden of these diseases again is sub-Saharan Africa and Southeast Asia.
Why is this important?
This slide shows what normally happens when a person gets exposed to TB and what happens with HIV disease.
As you can see, with normal immunity, the majority (98%) of persons will have latent infection. In HIV+ populations, only 60-70% go to latent infection. Up to 30-40% can activate their disease in the early phase. Of those who do not activate in the first several months, those with normal immunity have <10% chance of activating their TB infection whereas those with HIV have a 50% chance of activation. Thus, those without HIV have a 5-10% lifetime risk of activating their latent TB, and those with HIV have a 5-10% annual risk of activation. This is almost a 100x higher risk of lifetime activation for the HIV infected person.
Co-infection with HIV also affects the presentation of TB. In generally healthy persons, the predominant presentation is pulmonary TB. This slide is a study from Zambia that shows the difference between HIV positive and HIV negative persons with TB and their manifestations.
As you can see, those with HIV infection were more likely to have extrapulmonary manifestations. This can lead to delayed diagnosis and treatment.
Diagnosis of TB relies heavily on having a positive sputum smear for acid fast bacillus. This slide shows how HIV co-infection can decrease the positivity of the sputum smear.
Again, making it much more difficult to make a rapid and accurate diagnosis. Similar studies have been done for chest x-ray findings also. These studies have shown that the further immunesuppressed an HIV infected person is, the more likely of having atypical chest xray presentations such as pleural effusion, atypical infiltrate, lymphadenopathy, and miliary pattern.
Treatment of TB consists of using multiple drugs over an extended period of time, usually six months with 2 - 4 drug regimen. The primary four drugs used for initial treatment of TB are isoniazid (INH), ethambutol (EMB), rifampin (RIF), and pyrazinamide (PZA). All four drugs are used for the first two months of treatment on a daily basis and if the patient has responded and if the sputum smear is negative, then this regimen is decreased to only two drugs for the remaining four months of therapy. The usual two-drug regimen is either INH and RIF or INH and EMB depending on availability of RIF and other medications the patient is taking. If the patient has drug resistant TB, then at least two drugs to which the TB is sensitive to are to be continued for the remaining months. In multidrug resistant TB (MDR-TB), defined as resistant to both INH and RIF, treatment needs to consist of 2-3 drugs the bacteria is sensitive to and treated for at least 12 months.
Resistance to TB drugs can be primary where the person initially gets infected from drug resistant bacteria. Or, it can be acquired where the resistance arises to one or more drugs during the course of therapy. One method to reduce risk of drug resistance is to administer the medications by directly observed therapy (DOT). This slide shows the outcomes with using a DOT strategy vs self-administration of TB drugs.
The mortality outcomes for this study are significantly better in the DOT group.
Drug resistance also has an impact on morbidity and mortality outcomes. This slide shows a table of a large cohort study looking at outcomes for drug susceptible compared TB and drug resistant TB.
Success rates decrease when one or more of the potent anti-TB medications are not effective.
Drug resistance also places a serious cost to the public health system. In addition, cure rates for MDR-TB are extremely poor and in many cases are not successful. This slide present data from Seattle-King County comparing the cost and outcome of drug susceptible TB to MDR-TB.
The dollar amount is the cost per single patient for total treatment ($2200 compared to $150,000).
Factors of emergence for TB
We have already seen how microbial adaptation and change has impacted TB dramatically. There has been a worldwide increase in single drug and multi-drug resistance. In the US 8.4 % of isolates are INH resistant and 2.2% of isolates are MDR-TB. Globally, 17-54% of isolates are INH resistant and 4-30% are MDR-TB. Resistance testing is not universal so the data are less concrete. However, it does reveal that we have a potential situation where we may not have effective, affordable treatment for one of the world’s most common and deadly infectious diseases.
In 1991, there was an outbreak of MDR-TB in a NY hospital. Twenty-one of 23 patients were HIV positive, and of those, 19 (83%) died from TB. Twelve of 79 health care workers had new seroconversion from this outbreak. In an MDR-TB outbreak in a Russian prison, 114 isolates were culture positive. Of these, 87 patients had the W-Beijing strain. There were 34% new MDR cases compared to 55% who had been treated previously. In an outbreak in an Argentina hospital, out of 1253 patients, 272 were HIV positive. Of these, 124 of 272 had MDR-TB. 101 of the 124 were resistant to five or more drugs. 77 of 101 had an identical genome and the median survival time was 33 days.
Human behavior and demographics plays a role with increasing crowding and poverty. Increased urbanization can lead to both of these factors. Also, war and drought areas can increase refugee populations moving into new areas. Poor adherence to medicine treatment, as we have seen, increases risk of drug resistance. Many patients also wait to seek care for evaluation and diagnosis, which can delay treatment.
International travel has been identified in some transmission of active TB. A study by Kenyon showed that a single active infected patient was able to transmit TB to other passengers and crew on four different flights that were taken over the course of a few weeks. Other studies have shown that an infected crew member transmitted TB to other crew and to passengers. Despite cabin air flow exchange and HEPA filter use--which decreases the contamination to <100 colonies/160 Liter of air--there was still transmission. It should be noted however, that the concentrations on airplanes is significantly lower than what we normally find on city buses, shopping malls or airport terminals. Persons immigrating into the US are screened for TB; however, tourists, students and business travelers are not screened.
Technology and industry has been identified also as a source of transmission of TB. There have been cases of TB transmission through contaminated instruments, medical waste, and cadaver to embalmer. Technology has, however, given us improved tools of diagnosis, such as the more rapid diagnosis with use of DNA probes and RFLP (restriction fragment length polymerase).
Worldwide, the breakdown of public health has been a consistent and chronic reason for increase of TB. In the US, TB funding was decrease in the mid-1980’s due to a decrease in the number of cases. However, the mid-1980’s was the time when HIV was being identified and there was an increase of 52,000 cases of TB over the expected number during these years. A huge increase in the TB funding needed to be supplied to get the problem back under control. There has also been a large increase of TB cases in the former Soviet Union due to lack and disruption of the public health infrastructure. TB medications and DOT strategy is not available in all areas of the world. This leads to misuse of medication and increased risk of drug resistance. There is also a lack of rapid diagnostic technology and testing which impairs the ability to diagnose and treat active cases. Tracing of exposed persons (contact tracing) and follow-up is rarely done in other countries outside of the developed world due to lack of resources and personnel.
Surveillance, prevention and control of TB
Prevention and control of TB is multifactorial and incorporates the following:
Surveillance for TB is done primarily in a passive fashion. When a person is diagnosed with TB, they are then reported to the TB control. Active finding of cases occurs in some outbreak investigations but is much less frequent and requires additional resources. The following are some of the surveillance systems that exist:
The WHO and International Union Against TB and Lung Disease (IUATLD) have designed a model for improving TB diagnosis and management on a global level. This is called DOTS. DOTS includes the following five criteria:
WHO/IUATLD has also established some global surveillance objectives:
To assist with the standardized testing requires better and expanded use of rapid TB diagnostics. The current TB diagnostics are slow and less accurate especially in the setting of HIV infection. RFLP (restriction fragment length polymerase) has been useful to establish links for outbreak situations to a specific strain but is not as useful as a rapid diagnostic. PCR of specific sites such as sputum and CSF is probably the most widely used rapid test but is expensive and not available in many countries. Nucleic acid amplification has a sensitivity of 83% and specificity of 97%. The positive predictive value for PCR is higher in low risk persons compared to a sputum smear (59% vs. 36%). Negative predictive value for PCR is 91% compared to 37% for sputum smear.
Newer antibody testing from lymphocyte secretions are being studied and may prove even faster and more effective. BCG Ag-IgG-ALS and ESAT-6 are two of those under study at this time.
The appropriate treatment of TB requires not only DOTS but also an adequate and continuous supply of appropriate TB medications. There is a need to limit TB medications in non-controlled areas, such as local pharmacies or tribal doctors. There needs to be a system in place to ensure adherence to long term treatment (such as DOT).
The ability to measure and diagnose early treatment failure and modify therapy to reduce incidence of multidrug resistance is necessary. Also, there has been only one new drug developed for TB(rifapentime) in the past 25 years.
Treatment of latent infection may be one of the most cost-effective methods of TB control. It requires being able to accurately identify those who have latent TB infection and those who might be at higher risk of activating their disease. Currently we use a skin test called the purified protein derivative to identify those who have had TB exposure and infection. The test, however, is cumbersome to administer. It is an intradermal injection, which requires the patient to return in 48-72 hours to read the result. The result is noted by the amount of induration present, not redness, and requires training and skill to do this properly. A positive test result varies depending on the other risk factors. A person with HIV disease is positive at 5 mm induration, whereas a foreign born non-HIV person is positive at 10 mm. The current TB skin test is only positive in 70-90 % of those with previous infection and is less reliable in those with HIV and the immunecompromised.
We do know, however, that if we can identify this group and make sure they do not have active disease, then a course of INH for 9 months can markedly reduce the risk of TB activation in both persons with and without HIV infection.
Vaccine development is probably most desirable as is true for many other diseases. The current TB vaccine, BCG-Bacille Calmet Guerrin, has poor efficacy (0-70%) and is a live attenuated vaccine (concern for vaccine-related infection). The length of the immunity, if it exists, is not long lasting either.
An ideal vaccine would have the following characteristics:
This slide shows the current vaccine products that are in various stages of development.
Some are still in laboratory investigation stage, while others have started phase 1 and 2 trials. We are still a long way from another vaccine at this time. Concentration on the other treatment, prevention, and control strategies is essential.
In summary, TB is the second most common infectious disease in adults and a major public health threat. Diagnosis, treatment and prevention require ongoing resources, training and supplies. Co-infection with HIV has led to an increase in active cases, mortality and acquisition and spread of MDR-TB. Improved global surveillance, newer and more rapid diagnostics and treatments are necessary at this time. TB is targeted as one of the diseases for the Global Fund. Vaccine development is important but is years away.