Tuberculosis in the elderly: Keep a high index of suspicion
Tuberculosis in the elderly: Keep a high index of suspicion
Several epidemiologic features differentiate tuberculosis (TB) in the elderly from TB in younger adults. First, case rates for active TB are higher in the elderly, despite the fact that reactivity to tuberculin decreases with increasing age.1-4 For example, in a study from Arkansas, 53% of persons with TB were older than 65 years, despite the fact that this age group represented 14% of the population.5
The high rate of active TB in the elderly is explained in large part by the greater prevalence of latent infection with Mycobacterium tuberculosis in this population.6,7 Case rates of active TB are higher in the elderly because the reactivation rate is increased as a result of waning cell-mediated immunity from aging, cancer, and immunosuppressive drugs.
Second, TB case rates are 4 times higher for residents of nursing homes than for elderly persons living at home (234 versus 60 cases per 100,000, respectively).2,8 The increased risk in nursing home residents is likely caused by associated illnesses that predispose them to reactivation TB and by increased transmission in the clustered environment. In fact, it has been estimated that for every case of active TB in a nursing home, 6 additional residents will be newly infected.9 However, since only a small minority of the geriatric population lives in nursing homes, most elderly patients with TB are not nursing home residents.10
Finally, mortality rates from TB are highest in the elderly. Between 1979 and 1989, patients aged 65 years or older accounted for 60% of deaths from TB, a rate 10 times higher than that of young to middle-aged adults.8,11
In this article, we review the presentation and diagnosis of TB in the elderly, and we delineate treatment options.
Age-related changes in immunity
Elderly persons are at increased risk for infectious diseases as a result of senescent changes in the immune system. Although a decline in innate (neutrophil-mediated) immunity and specific (lymphocyte-mediated) immunity has been observed in the elderly, it is frequently not clear whether the defect is primary or secondary to an underlying systemic disorder or to the use of immunosuppressive drugs. Diet and exercise may also influence age-related changes in immune response.12
In terms of innate immunity, macrophage function (including chemotaxis, adherence, and phagocytosis) appears to be largely unaffected by aging. Tests for natural killer (NK) cell function in humans show little if any age effects, although data from mouse studies are conflicting.12 In contrast, neutrophil chemotaxis and respiratory burst appear to be impaired with aging.13
The adaptive immune system, especially cell-mediated immunity, appears to be most vulnerable to the effects of aging,12,14 as evinced by:
Reduced thrombopoietin and involution of the thymus.
Diminished delayed-type hypersensitivity response caused by decreased T-cell proliferation.
Loss of memory T-cell function.
Decreased number of helper T cells and increased number of T suppressor cells.
Reduced interleukin (IL)-2 production and IL-2 receptor numbers.
Increased reactivation of TB and herpes zoster in elderly persons.
Success in controlling M tuberculosis infection depends on the ability of CD4+ T cells, and to a lesser extent CD8+ T cells, to initiate an immune response by producing interferon-g (IFN-g), a cytokine that activates alveolar macrophages. Activated macrophages secrete additional cytokines IL-12 and tumor necrosis factor a, which slow the growth of the pathogen and help contain the bacilli within granulomas.
It is likely that this decline in CD4+ T cell-mediated response is the principal reason elderly patients are more susceptible to TB. Other factors that may predispose elderly persons to reactivation and primary TB include altered mucociliary clearance; malnutrition; increased risk of exposure for residents of group homes; and the greater prevalence of associated diseases, such as malignancies, that negatively affect immune surveillance.
Experimental studies in mice have corroborated that the aging immune system is less able to defend against M tuberculosis.15 Old mice are less able than young mice to mount an antigen-specific CD4+ T cell-mediated response against M tuberculosis. Orme16 found that old mice were unable to control the growth of M tuberculosis administered intravenously and were more susceptible to disseminated disease, especially in the lung, spleen, and liver. This may be partly the result of slower kinetics in the recruitment of CD4+ T cells to the site of infection.17,18
Poor T-cell migration may be the result of the failure of CD4+ T cells to alter the expression of adhesion molecules (particularly the integrin a chain) on their cell surfaces in response to M tuberculosis. Antigen-specific CD4+ T cell responses are also significantly reduced in old mice.15 In addition, old mice synthesize less IL-2 in response to infection, which results in diminished CD4+ T cell proliferation in the lungs during M tuberculosis infection.17,19
In contrast, old mice challenged with low-dose M tuberculosis via the respiratory route were able to control mycobacterial growth in the first 21 days after infection. In fact, the early effective response by old mice surpassed that of young mice.17,20
This early response is dependent on resident CD8+ T cells. In old mice, these cells are found in greater numbers both before and during infection with M tuberculosis, and they play a different role in the lungs. Because the response of CD4+ T cells is delayed, old mice rely on CD8+ T cells to control bacterial loads in the early stages of infection.21-23
CD8+ T cells secrete IFN-g much earlier in infection in old mice than in young mice. Old mice that had impaired ability to produce IFN-g were unable to express early resistance to M tuberculosis in the first weeks of infection.24 Thus, CD8+ T cells allow old mice to rapidly respond to infection; however, this response is not antigen-specific.
CD8+ T cells in the lungs of old mice had increased expression of several NK-associated molecules, allowing them to mount what is probably an antigen-independent response against M tuberculosis- infected cells; over time, however, the bacterial loads increased, surpassing those seen in young mice.20 The old mice have early resistance to respiratory infection with M tuberculosis but cannot sustain this resistance, resulting in increased susceptibility over time. This increased susceptibility is the result of the differences in the inherent properties of the cells in the lungs of old and young mice.
Purified protein derivative testing
In men, reactivity to purified protein derivative (PPD) drops from 50% at age 65 to 74 years to 10% at 95 years or older; the corresponding decline in women is from 40% to about 5%.4 PPD reactivity rates are higher in nursing home residents. This appears to be the result of more than simply the reactivation of disease.25 Potential explanations include unrecognized nosocomial spread of TB in a clustered environment, earlier demise of anergic patients, improvement in nutrition and general health after admission, and a delayed or booster effect.2
The rate of PPD reactivity for patients at the time of admission to a nursing home (including a booster, if necessary) is only 10% to 15%. However, general screens of residents find higher rates, ranging from 20% to 51%,25-27 and when serial testing is performed, annual PPD conversion rates are high. For example, PPD conversion occurred in nearly 5% of 642 nursing home residents over 1 year in the absence of an outbreak of disease.26 In a study of 9937 PPD-negative patients, annual conversion rates were 5% in residents of nursing homes with recognized infectious cases and 3.5% in residents of nursing homes without recognized cases.2
The unrecognized nosocomial spread of disease is thought to be the major reason for skin test conversion. However, investigators in Belgium demonstrated a progressive booster response with 4-stage testing, suggesting that delayed-type hypersensitivity may be recalled.28 Because of the high rate of conversion in the absence of symptoms, a 2-step approach is recommended on admission to a nursing home; if the initial result is negative or equivocal, the tuberculin test should be repeated 1 to 2 weeks later.29,30
A "positive booster effect" is defined as an increase of 6 mm or more for an induration of less than 10 mm on the first test to 10 mm or more on the second test. This information at baseline helps distinguish newly acquired disease from prior infection. Yearly tuberculin skin testing is recommended for elderly persons living in nursing homes who are PPD-negative, based on the high rates of transmission and disease.2