Richard Long, MD, FRCPC, FCCP
Kevin Schwartzman, MD, MPH
The pathogenesis and transmission of TB are inter-related. M. tuberculosis is almost exclusively a human pathogen. How it interacts with the human host determines its survival. From the perspective of the bacterium a successful host-pathogen interaction is one that results in pathogen transmission. Initial infection is usually self-limited and followed by a variable period of latency, which ultimately, in a proportion of those infected, results in infectious pulmonary TB. Transmission from a case of infectious pulmonary TB is by the airborne route in minute droplets of moisture that become increasingly reduced by evaporation, creating "droplet nuclei"Footnote 1.
At the time of initial infection, the distribution of inhaled droplet nuclei in the lung is determined by the pattern of regional ventilation. It thus tends to follow the most direct path to the periphery and to favour the middle and lower lung zones, which receive most of the ventilationFootnote 2. In immunocompetent hosts, it is theorized that alveolar macrophages ingest the M. tuberculosis organisms and may or may not destroy them, depending on the degree to which phagocytosing cells are nonspecifically activated, on host genetic factors and on resistance mechanisms in the bacteriaFootnote 3. If bacteria are successfully cleared, then test results will remain negative on the tuberculin skin test (TST) or interferon-gamma release assay (IGRA).
When innate macrophage microbicidal activity is inadequate to destroy the initial few bacteria of the droplet nucleus they replicate logarithmically, doubling every 24 hours until the macrophage bursts to release the bacterial progenyFootnote 3. New macrophages attracted to the site engulf these bacilli, and the cycle continues. The bacilli may spread from the initial lesion via the lymphatic and/or circulatory systems to other parts of the body. After a period lasting from 3 to 8 weeks the host develops specific immunity (cell-mediated immunity [CMI] and delayed-type hypersensitivity [DTH]) to the bacilli, and individuals typically show positive results on the TST or IGRA. The resulting M. tuberculosis-specific lymphocytes migrate to the site of infection, surrounding and activating the macrophages there. As the cellular infiltration continues, the centre of the cell mass, or granuloma, becomes caseous and necrotic. Radiographically demonstrable fibrocalcific residua of the initial infection include a Ghon focus (a calcified granuloma in the lung) alone or in combination with a calcified granulomatous focus in a draining lymph node (Ghon complex)Footnote 4 Footnote 5. Infection and immune conversion are usually asymptomatic; any symptoms that do occur are self-limited. In a small proportion of those infected, erythema nodosum (a cutaneous immunologic response to an extracutaneous TB infection) or phlyctenular conjunctivitis (a hypersensitivity reaction) may develop.
A proportion of those who are recently infected are unable to contain the infection despite the stimulation of CMI and DTH, and there is progression to disease in a matter of months. Such early disease progression is a function of age and immunologic response, disease being especially likely to occur in young children and the immunocompromised. A progressive Ghon focus, disseminated (miliary) disease and central nervous system disease may occur as early as 2 to 6 months after infection in infants and the severely immunocompromised Footnote 6 Footnote 7. Uncomplicated and asymptomatic lymph node disease (hilar or mediastinal lymphadenopathy without airway involvement) may also occur in the first 2-6 months of infection, although there is debate about whether this should be called active disease (refer to Chapter 9, Pediatric Tuberculosis) Footnote 6 Footnote 8.
At 4-12 months after infection, early disease manifestations include complicated lymph node disease (airway compression, expansile caseating pneumonia, infiltration of adjacent anatomic structures), pleural disease (most commonly a lymphocyte-predominant exudative effusion) and peripheral lymphadenitis (usually in the neck)Footnote 6. In immunocompetent children and adolescents early disease is more likely to manifest as intrathoracic adenopathy and in adults as a unilateral pleural effusion. In severely immunocompromised people of any age (e.g. those with advanced HIV or AIDS), early disease may manifest as intrathoracic adenopathyFootnote 9 Footnote 10. Rarely, in newly infected people who are 10 years of age or older (pubertal) adult-type pulmonary disease (refer to Figure 1) or other types of extrapulmonary TB (for example bone and joint TB) may develop within the first 24 months of infectionFootnote 11.
While early disease progression may or may not result from lympho-hematogenous spread, late disease progression (refer to Figure 1) is almost always the result of the lympho-hematogenous spread of bacilli. Recent infection with early disease progression probably accounts for many cases of TB in recently arrived immigrantsFootnote 12. For purposes of disease reporting, everyone with a diagnosis of TB made within 18-24 months of infection is considered to have "primary" disease (on balance about 5% of those infected). Those newly infected people in whom TB does not develop within this period of time will either be left with LTBI and will never experience disease (on balance about 90% of those infected) or, after a variable period of latency, they will develop late disease progression (on balance about 5% of those infected, refer to Figure 1).
In the classical concept of LTBI, M. tuberculosis bacteria are believed to survive for years in Ghon foci and complexes and in the small granulomas or solid caseous material of lympho-hematogenously seeded foci. Presumably, local conditions, an intact CMI or the presence of inhibitors result in conditions unfavourable to replication. Recent mapping of the complete genome sequence of the bacterium demonstrates that the organism has the potential to synthesize enzymes involved in anaerobic metabolismFootnote 13. Although rapid death and autolysis occur after abrupt depletion of oxygen, the organism can shift into a state of dormancy if allowed to settle through gradual reductions in oxygen tensionFootnote 14 Footnote 15. Therefore, although M. tuberculosis thrives in an aerobic environment, it possesses the genetic and biochemical capability of anaerobic survival and can persist experimentally in oxygen-depleted media. Tubercle formation, with its oxygen-depleted environment, is a defining characteristic of TB. LTBI is usually identified by a positive TST or IGRA in the absence of active disease (refer to Chapter 4, Diagnosis of Latent Tuberculosis Infection).
More recently LTBI and active TB have been considered as two ends of a spectrum of states ranging from asymptomatic infection to overt diseaseFootnote 16 Footnote 17. In this more nuanced concept, patients whose LTBI progresses to overt disease may pass through a continuum of asymptomatic intermediate states with detectable manifestations indicative of diseaseFootnote 16. Such asymptomatic disease states frequently remain undiagnosed, and their manifestations and duration are mostly dependent on host immune response. Defining these intermediate states in concrete terms is considered to be important for pragmatic reasons, as they might have an impact upon the performance of TB biomarkers or other diagnostic measures and could also present targets for therapeutic interventionsFootnote 16 Footnote 18.
The elegant studies of Ferguson strongly suggest that it takes up to 18 months after the initial infection for CMI to matureFootnote 19. During this period of time a reinfection carries the same risk of disease as the initial infection, perhaps explaining why disease is much more common in newly infected close contacts of smear-positive cases than it is in newly infected close contacts of smear-negative cases - the former having a greater likelihood than the latter of repeated exposure and reinfectionFootnote 20-22. Reinfection of immunocompetent hosts that occurs 18 months or more after the initial infection carries a much lower risk of progression to TB disease, estimated to be 21% of the risk of an initial infection progressing to diseaseFootnote 22. It is not known whether this is because prior infection without development of overt disease is simply a marker for people who are less susceptible to disease development or better able to overcome it once it has developed. Nevertheless, in highly endemic areas the majority of TB cases occurring in those with prior LTBI may be due to reinfection rather than reactivation; in Canada, where repeat exposure is much less common, most active TB reflects reactivation and not reinfectionFootnote 23-25. In the severely immunocompromised host, reinfection and initial infection carry a similarly high risk of disease regardless of when the reinfection occurred (refer to Chapter 6, Treatment of Latent Tuberculosis Infection).
In Canada, most TB is understood to be "reactivation" TB, i.e. occurring 18-24 months or more after the initial infection. It usually presents as adult-type pulmonary disease (upper lung zone fibrocavitary disease - previously referred to as postprimary TB - beginning in small foci that are the result of remote lympho-hematogenous spread), although it may also present as extrapulmonary TB. As mentioned earlier, adult-type pulmonary TB may on occasion be a manifestation of primary TB or a reinfection. In any population group, reactivation of LTBI, leading to reactivation TB, is much more likely to occur in people who are immunocompromised.
There are a number of theories, most of them speculative, as to why adult-type pulmonary TB tends to localize in the upper lung zones. These are described elsewhereFootnote 2 Footnote 5. People with a history of untreated or inadequately treated pulmonary TB or a "high-risk" lung scar (upper lung zone fibronodular abnormality) on chest radiograph are understood to have a higher bacillary burden than those without such a history/radiograph, and to be at increased risk of reactivation TBFootnote 26 Footnote 27.
From the standpoint of public health and the organism's survival as a species, adult-type pulmonary TB is the most important phenotypic expression of the disease. Patients with adult-type pulmonary TB are much more likely to show lung cavitation, created when caseous material liquefies (possibly related to hydrolytic enzymes released from inflammatory cells during their destruction and DTH to tuberculin-like proteins) and erodes into the bronchiFootnote 28. Within the unique extracellular environment of cavities, host defences are ineffectual, and bacteria multiply in large numbers. Because cavities are open to, and discharge their contents into, nearby bronchi these same bacteria are directly communicable to the outside air when the patient coughs. Transmission from patients with adult-type pulmonary TB is facilitated by the concurrent involvement of both the airways and their contiguous pulmonary blood supply at sites of disease in the lung. This minimizes the respiratory limitation experienced by the patient, extending the life of the host within the community and creating further opportunities for transmission before the patient either seeks medical attention or succumbsFootnote 29.
Outside of the extrapulmonary sites of disease alluded to in the section Early Disease Progression and cases of bone and joint TB, whose timeline from infection to disease in children may be as short as a year, most extrapulmonary TB is reactivation disease. Extrapulmonary TB or combined pulmonary and extrapulmonary TB is more common in those who are severely immunocompromised; in those coinfected with HIV the occurrence of extrapulmonary TB increases as the CD4 count decreases (refer to Chapter 7, Non-respiratory Tuberculosis)Footnot9 Footnote 10.
The risk of transition from LTBI to active TB, primary or reactivation, is largely dependent on the immune competency of the host. Age and sex appear to directly affect the immunologic response and the risk of disease: morbidity is greater among young children (<5 years of age), especially infants, among young adults, especially females, and among older adults, especially males. In high-burden countries, the population attributable fraction of undernutrition for TB is 27% according to the WHOFootnote 30. The seasonality of TB (with the highest incidence in spring and early summer) has been attributed to reduced sunlight and vitamin D deficiency during the winter months in some studies but not in othersFootnote 31-33. Ethnic differences have been offered as factors determining host immune response, with some supportFootnote 34, but differences among ethnic groups in all clinical forms of TB are probably best explained as phase differences in an epidemic waveFootnote 35. All races initially exposed in an epidemic as a group are equally susceptible, but eventually death and survival outcome select out people who are relatively more resistant. A growing body of evidence suggests that host genetic factors are important in determining susceptibility to TBFootnote 36-38. Most important from a clinical perspective are the many medical conditions that are well known to affect host immunologic response and increase the risk of progression from LTBI to active TB disease. These are reviewed in detail in Chapter 6, Treatment of Latent Tuberculosis Infection. To identify entry points for interventions aimed at addressing TB risk factors as well as social determinants, Lönnroth and colleagues developed a framework for proximate risk factors and upstream determinants of TB (refer to Figure 2)Footnote 39.
Figure 2. Framework for proximate risk factors and upstream determinants of TBFootnote 39
M. tuberculosis is communicable from one human to another mainly by the aerosol route and rarely through ingestion or percutaneous inoculation (e.g. through laboratory or hospital accident). Bovine TB, which in the past was caused by ingestion of milk heavily infected by M. bovis that then penetrated the mucosa of the oropharynx or the gastrointestinal tract, has been largely eradicated as a result of the pasteurization of milk and the tuberculin testing of cattle, followed by the slaughter of animals found to be infected.
The reservoir for M. tuberculosis is humans. Other animals, in particular primates, may be infected but are rarely a source of infectionFootnote 40-43. Droplet nuclei, sometimes referred to as "the quanta of contagion", are created by forceful expiratory efforts, such as coughing, sneezing, singing, playing wind instruments and even speaking. Before droplets reach the airspace of a room and have had an opportunity to evaporate down to a "droplet nucleus" their numbers can be reduced by wearing a simple gauze (surgical) mask or covering the mouth and nose during coughing. Certain procedures, for example, bronchoscopy, sputum induction, processing of specimens, autopsy and even irrigation or other manipulation of tuberculous abscesses, may also produce infectious aerosols. The droplets have an extremely slow settling rate (0.5 mm per second or less), which permits their transport by air currents, duct systems or elevator shafts for significant distances from the source case. Large particles settle quickly and are either not inhaled by contacts or, if inhaled, are trapped in the mucus of the upper airway. If the organism reaches the trachea and bronchi it is usually swept back to the larynx by ciliary action and cough, and then swallowed. For practical purposes, only the droplet nuclei in the size range 1 to 5 microns reach the terminal air spaces or alveoli; each is understood to contain only a few bacteria. In most instances only one such droplet nucleus is believed to be responsible for establishing infection in the host. Bacteria that are lodged on fomites (linen, furniture, books, floors) do not constitute a significant source of infection: most die quickly through the action of drying, heat or sunlightFootnote 5 Footnote 40-43.
The rate of transmission can be measured by the percentage of close contacts (household and non-household) whose TST or IGRA responses are converted from negative to positive or in whom active TB disease develops. The percentage will depend on the number of infectious droplet nuclei per volume of air (infectious particle density) and the length of time that the uninfected individual spends breathing that air. In the past, drug susceptibility patterns and phage typing of M. tuberculosis isolates have helped to confirm the transmission between source case and contact. More recently, DNA fingerprinting of M. tuberculosis isolates has greatly refined the identification of this relationFootnote 44.
Because of the highly variable latency period of M. tuberculosis infection it is difficult to precisely document transmission using currently available tools. People found to have positive TSTs and/or IGRAs during contact investigation may have been infected in the past (remotely) rather than by the recent source case of concern, though for contact management and public health purposes these contacts are treated as if recently infected if there is no way to determine the duration of infection. DNA fingerprinting techniques will only detect transmission to the small group of people in whom active disease develops following transmission. If most TB disease in a community reflects recent/ongoing transmission, the first priority for public health authorities should be to prevent further transmission. On the other hand, if most TB reflects reactivation of remotely acquired infection, the priority should shift to identification and treatment of people with LTBI, notably those with risk factors for reactivation.
Several patient, pathogen and environmental factors determine whether transmission occurs, largely by affecting the number of infectious droplet nuclei per volume of air (refer to Table 1). Although the probability of being infected after contact with an infectious source decreases with decreasing duration and decreasing closeness of contact, the absolute number of casual contacts infected may exceed the number of infected close contacts, since the former may far outnumber the latterFootnote 45. DNA fingerprint data have highlighted the limits of contact tracing in settings where there is exposure of a large number of people unknown to source cases and in settings where social connections are tenuous at bestFootnote 46 Footnote 47. At this point very little is known about what, if any, host determinants influence the acquisition of initial infection after inhalation of a droplet nucleus. Some individuals are able to achieve complete or "sterile" elimination of M. tuberculosis bacteria rather than developing latent infectionFootnote 48. Observational studies suggest that BCG vaccination in infancy offers some protection against infection with M. tuberculosis as detected by an IGRAFootnote 49-52.
|Disease type||Strain variability||Indoor/outdoor|
|Extrapulmonary||Proximity to the source case|
|Symptomatology||Duration of exposure|
With rare exception (e.g. transmission related to an inadequately sterilized bronchoscope or a needle stick injury), transmission requires that a TB patient be able to produce airborne infectious dropletsFootnote 41-43. This most often limits the potential for transmission to adolescent or adult patients with adult-type pulmonary TB. Younger children can on occasion be infectiousFootnote 53, but as a general rule they have few bacilli in their lesions, often do not produce sputum and rarely have communicable diseaseFootnote 54. Of patients with TB involving the respiratory tract not all are equally efficient at transmission.
1. Sputum smear status
Patients with smear-positive/culture-positive pulmonary TB are more infectious than patients with smear-negative/culture-positive pulmonary TB, and the latter are more infectious than patients with smear-negative/culture-negative pulmonary TB (refer to Table 2 for a summary of the epidemiologic studies on the risk of infection in household [close] contacts grouped according to the bacteriologic status of the source cases)Footnote55-62. Sputum that is smear-positive contains 5,000 or more organisms per millilitre of sputumFootnote 57 Footnote 58 Footnote 63 Footnote 64 Patients with smear-positive bronchoalveolar lavage fluid are considered just as infectious as those with smear-positive sputumFootnote 65. Smear-positive induced sputum is for practical purposes considered to indicate the same degree of contagiousness as smear-positive spontaneously expectorated sputum, though there are currently no data that prove this assertionFootnote 55. With the use of molecular epidemiologic tools the relative transmission rate of smear-negative compared with smear-positive patients has been determined to be 0.17-0.22 or roughly one-fifth the likelihood of transmissionFootnote 66 Footnote 67. In addition to the greater infectivity of smear-positive cases, as mentioned in the section Pathogenesis, the risk of disease after infection from a smear-positive case is greater, by virtue of the higher probability of repeated infection, than it is after infection from a smear-negative case.
|Ref no.||Year of survey||Location||Contacts||Number and % infected contacts by bacteriologic status of index case||General population % positive PPDBTable 2 - Footnote B|
|58||1963-64||Holland||all ages||858Footnote C||391||20%||467||1%||-||-||<< 1%|
2. Disease type on plain chest radiograph
Pulmonary TB patients with cavitation on chest radiograph are more infectious than pulmonary TB patients without cavitation after bacteriologic findings have been taken into account Footnote 68-70. Pulmonary TB patients with "typical" chest radiographic findings (upper lung zone disease, with or without cavitation, and no discernable intrathoracic adenopathy) are more infectious than pulmonary TB patients with "atypical" chest radiographic findings (all others)Footnote 71.
3. Laryngeal disease
Patients with laryngeal TB are more infectious than those with pulmonary TBFootnote 72. Most patients with laryngeal TB (hoarseness associated with inflammation and ulceration of the vocal cords) have far advanced pulmonary disease upstream from the larynxFootnote 73.
In general, normal breathing produces few infectious particles, a bout of coughing or five minutes of speaking in a normal tone produce many more, and a sneeze produces the mostFootnote 74 Footnote 75. The likelihood that household contacts will be infected increases with the frequency of cough in the source caseFootnote 60. When the aerial infectivity of the droplets from smear-positive patients was evaluated by artificially atomizing sputum and exposing guinea pigs to a standard dose, there was marked variability in the infectivity of aerosolized sputum, perhaps explaining the extraordinary heterogeneity of infectiousness among patients with smear-positive pulmonary TBFootnote 76-78. Thus, although patients may appear to have an equal number of bacteria in their sputum, the physical and chemical properties of their sputum, as well as their effectiveness as an aerosolizer, may determine whether they produce a large or small number of droplet nuclei. The role of smoking, allergy or coincidental viral upper respiratory tract infection in aerosol formation is unknownFootnote 79.
5. Delayed diagnosis
The number of contacts and the duration of exposure of each contact may increase as time to diagnosis increases. The longer the duration of symptoms in the source case the greater the risk of transmissionFootnote 65.
Effective treatment (refer to Chapter 5, Treatment of Tuberculosis Disease) appropriate to the drug susceptibility test results rapidly reduces cough frequency and sputum bacillary countsFootnot60 Footnote 80. Even faster than the rate of decrease of the latter is the rate of decrease of bacillary counts in cough-generated aerosol culturesFootnote 81. With treatment those bacteria that continue to be expectorated may be expected to be less metabolically active and/or are inhibited by the drugs, two effects that may decrease the chances of the organism establishing an infection in the hostFootnote 76 Footnote 82. However, in theory, any residual viable bacteria in respiratory secretions can be transmitted, although the chances of this occurring decrease rapidly with effective treatmentFootnote 83. Given the frequency of drug resistance, the determination that treatment is effective in reducing the infectiousness of a given patient should reflect objective clinical, radiographic and/or microbiologic improvement, and not simply time elapsed since treatment initiation.
Data are emerging to suggest that one or more virulence properties of M. tuberculosis may affect its ability to be transmittedFootnote 84. For example, one strain may be better suited than another to overcoming the innate resistance of the host. Although drug-resistant strains have shown reduced virulence in animal modelsFootnote 85, clinical evidence of their transmissibility is compellingFootnote 86-89, and for practical purposes they should be considered just as transmissible as drug-susceptible strains. Beijing/W strains have been reported to be hypervirulent, but indices of transmission have been found to be no greater in patients with these strains than in those without themFootnote 90.
Outdoor exposures are very unlikely to result in transmission unless the source and the susceptible person are in talking distance. Bacillary dispersion is immediate, and sunlight rapidly kills any viable bacilliFootnote 91 Footnote 92. For practical purposes outdoor exposures are not investigated during a contact tracing exercise.
1. Air circulation and ventilation
Given a defined number of bacteria expelled into the air, the volume of air into which the bacteria are expelled determines the probability that a susceptible individual breathing that air will become infected. A high concentration of viable bacteria in the inhaled air of the contact is favoured by indoor exposure, poor ventilation or recirculation of air, and little sunlight (ultraviolet rays). Ventilation dramatically dilutes the concentration of infectious droplet nuclei (refer to Chapter 15, The Prevention and Control of Tuberculosis Transmission in Health Care and Other Settings, for further information on clearance times).
2. Proximity to the source case
Proximity to the source case is also a determinant of transmission. Related to this is overcrowding: if, as a result of there being many people in a room, an individual is forced into close proximity with an infectious case his or her risk of infection is likely to increase.
3. Duration of exposure
Because of the dilution of infected air and the low concentration of infectious droplet nuclei, the duration of exposure required to ensure that transmission occurs is commonly prolonged (days, months or even years), and yet reports have confirmed that exposures as short as a few minutes may be sufficient to infect a close contact. The latter would appear to be supported by the high proportion of active cases that deny any history of exposure.
The highest priority should be given to early diagnosis and prompt, effective treatment of the source case together with isolation of the patient when necessary. The insidious development of symptoms in most cases of TB commonly results in a delay of weeks or months before the patient presents for diagnosis. At that point, when the patient is often at his or her most infectious, any further delay caused by the physician, nurse or system allows unnecessary transmission to others. Maintaining an appropriate awareness of TB among health care providers is thus critical to reducing transmission and initiating early prevention and treatment. Administrative and engineering controls that aim to reduce exposure in health care and other congregate settings complement--but cannot replace--prompt diagnosis and appropriate therapy. Methods once thought to be important in preventing the transmission of TB - disposing of such personal items as cloths or bedding, sterilizing fomites, using caps and gowns, gauze or paper masks, boiling dishes and washing walls - are unnecessary, because they have no bearing on airborne transmission.