Tuberculosis Infection, Disease, and Death
Human TB is caused by infection with mycobacteria, principally Mycobacterium tuberculosis. Individuals with pulmonary or laryngeal TB produce airborne droplets while coughing, sneezing, or simply talking. Inhaled infectious droplets lodge in the alveoli, and bacilli are taken up there by macrophages, beginning a series of events that results in either the containment of infection or the progression to active disease (Frieden and others 2003). Following uptake by macrophages, M. tuberculosis replicates slowly but continuously and spreads through the lymphatic system to hilar lymph nodes. In most infected people, cell-mediated immunity, associated with a positive tuberculin test, develops two to eight weeks after infection. Activated T lymphocytes and macrophages form granulomas, which limit the further replication and spread of bacilli. Unless a later defect occurs in cell-mediated immunity, the infection remains contained within the granulomas.
The immune mechanisms are, in their details, far more complex. For example, following antigenic challenge, a suite of different T cells is responsible for the induction and suppression of protective immunity, delayed hypersensitivity, cytolysis, and the production of antibodies and memory cells. Helper T cells mature into two functionally different populations: in M. tuberculosis infection, the TH1 response is associated with granuloma formation and protection, whereas the TH2 response results in tissue-necrotizing hypersensitivity and the progression of disease. The processes that determine the balance of the two responses affect, for example, the interaction between M. tuberculosis and other infectious agents (Grange 2003).
When the immune response cannot suppress replication, primary infection leads to active TB (progressive primary TB). The most common clinical manifestation is pulmonary disease, typically in the parenchyma of the middle and lower lung. In the most infectious patients, bacilli can be seen microscopically on stained sputum smears (60 to 70 percent of pulmonary cases; Marais and others 2004; Styblo 1991). Smear-negative patients may also be infectious but, per patient, contribute relatively little to transmission (Behr and others 1999; Hernandez-Garduno and others 2004). Extrapulmonary tuberculosis accounts for 10 to 30 percent of the disease but is more common among women and children (particularly lymphatic TB) and in people infected with HIV (Aaron and others 2004; Rieder 1999; Rieder, Snider, and Cauthen 1990; Shafer and Edlin 1996).
In the absence of other predisposing conditions, only about 5 percent of infected people develop progressive primary disease within five years of infection (Comstock, Livesay, and Woolpert 1974; Sutherland 1968; Sutherland 1976). After five years, the annual risk of developing TB by the reactivation of latent infection is much lower(10-4 per capita per year). The risk of progressing to active disease is relatively high in infancy and lower in older children; it increases quickly during adolescence (earlier in girls) and then more slowly throughout adulthood (Comstock, Livesay, and Woolpert 1974; Nelson and Wells 2004; Sutherland, Svandova, and Radhakrishna 1982; Vynnycky and Fine 1997). Whether latent bacilli remain viable for the full life span of all infected people is unknown, but the risk of reactivation certainly persists into old age. The lifetime risk of developing TB following infection clearly depends on the prevailing transmission rate; the rule of thumb is 10 percent, but it has been calculated at 12 percent for all forms of pulmonary disease in England and Wales during the second half of the 20th century (Vynnycky and Fine 2000).
Besides the strong innate resistance to developing disease, infection is associated with an acquired immune response. This response is only partially protective (Dye and others 1998; Sutherland, Svandova, and Radhakrishna 1982; Vynnycky and Fine 1997), which helps explain why developing an effective vaccine has been difficult (few manufactured vaccines are more protective than natural immunity; Andersen 2001; Fordham von Reyn and Vuola 2002; Young and Stewart 2002). Consequently, individuals who carry a latent infection and who continue to be exposed are at risk of TB following reinfection. The importance of reinfection remains controversial, but mathematical modeling shows that the decline of TB in Europe cannot easily be explained without reinfection (Dye and others 1998; Vynnycky and Fine 1997). In addition, molecular fingerprinting has produced direct evidence that TB commonly arises from infection and reinfection in endemic areas (de Viedma and others 2002; Richardson and others 2002; van Rie and others 1999; Verver and others 2004), especially where subjects are infected with HIV (Glynn and others 2004).
The low incidence of infection and the low probability of breakdown to disease explain why TB is relatively rare. Its importance among infectious diseases is attributable not so much to the number of cases as to the high case-fatality rate among untreated or improperly treated patients. About two-thirds of untreated smear-positive patients will die within five to eight years, the majority within the first 18 months (Styblo 1991). Most of those who are still alive after eight years will have quiescent TB (self-cures, susceptible to relapse), and a few will become chronic excretors of bacilli. The case-fatality rate for untreated smear-negative cases is lower, but still of the order of 10 to 15 percent (Krebs 1930; Rieder 1999). Even among smear-positive patients receiving antituberculosis drugs, the case-fatality rate can exceed 10 percent if adherence to treatment is low or if rates of HIV infection and drug resistance are high (WHO 2004c).
Online annex 1 contains more information about factors that affect the risk to individuals of contracting infection and developing disease and the distribution of TB in populations.
