Dean C. Norman
Infections in the elderly often present in an atypical, nonclassical fashion. Furthermore, the differential diagnosis of infectious diseases in the elderly differs from the young because it is dependent on both the clinical setting and the patient's underlying functional status. For example, "free living," independent, healthy elderly are prone to respiratory infections, such as bacterial pneumonia, genitourinary infections and intraabdominal infections including cholecystitis, diverticulitis, appendicitis, and intraabdominal abscesses. Institution-bound elderly are more likely to develop aspiration pneumonia, urinary tract infection, (especially if a chronic indwelling bladder catheter is present), and skin and soft-tissue infections. Infections in the elderly differ from the young also because infections in this age group are often caused by a more diverse group of pathogens compared with the young. This is best exemplified by urinary tract infection, which in the young occurs almost exclusively in females and is usually caused by Escherichia coli. In the aged, a variety of Gram-positive cocci and Gram-negative bacilli, sometimes in combination, are potential pathogens for this infection, and at least a third of urinary tract infections occur in males. This is the reason that obtaining a urine culture prior to empirical therapy for symptomatic urinary tract infection in the elderly is recommended.
Morbidity and mortality rates for infections are usually higher in the elderly compared with the young. These observed higher mortality and morbidity rates are due in part to factors such as (1) lower physiologic reserve capacity due to biologic changes with age and comorbidities, (2) age and disease-related decrements in host defenses, (3) chronic illness, and (4) higher risk for adverse drug reactions due to multiple medications and age-related physiologic changes that alter the pharmacokinetics and pharmacodynamics of many medications. Additional contributing factors include a greater risk for hospitalization, and therefore exposure to nosocomial pathogens, a greater risk of undergoing invasive procedures, and increased likelihood of suffering a procedure-associated complication. Finally, delay in diagnosis and the initiation of appropriate empirical antimicrobial therapy is an important contributing factor. Diagnostic delays may commonly occur in this population, who can least tolerate this. However, infec-
From: Infectious Disease in the Aging Edited by: Thomas T. Yoshikawa and Dean C. Norman © Humana Press Inc., Totowa, NJ
Altered presentations of infection in the elderly
Potential findings with any infection Potential findings with specific infections
Confusion May be afebrile
Lethargy Dyspnea, confusion, falls,
"Failure to thrive" Pneumonia
May be afebrile
Cough and sputum production may be absent Intraabdominal infection
Stiff neck may be absent Confusion, altered consciousness Tuberculosis Weight loss, lethargy Failure to thrive tions in the elderly may present in an atypical or nonclassical manner, which in turn may make early diagnosis difficult. Thus, infections in this age group provide a unique challenge to clinicians.
Atypical or nonspecific responses to infection are commonly observed in older adults and has been recently reviewed extensively (1). Table 1 lists some of these signs and nonclassical presenting features. Delirium, agitation, confusion, lethargy, anorexia, falls, abnormal movements, focal neurologic signs, and urinary incontinence may all be the sole symptom observed at initial presentation. Furthermore, bacteremia may be afebrile and present with dyspnea, confusion, and/or hypotension. The classical finding of a stiff neck may be absent in geriatric patients with bacterial meningitis, and the older patient with pneumonia may not have cough, sputum production, or fever. Similarly, peritoneal findings may be absent in elderly patients with intraabdominal infection (2). In summary, virtually any change in functional status in the elderly may be an indication of the presence of an acute illness, which often is an infection. Furthermore, there may be a dissociation of clinical findings with severity of illness.
Classical studies of bacteremia (3,4) and pneumonia (5,6), as well as a recent review of tuberculosis (7), demonstrate both that a higher percentage of the elderly compared with young demonstrated a blunted fever response and that up to one third of elderly patients with serious bacterial or viral infections do not mount a robust febrile response
(8,9). This is confirmed by a more recent study of acute intraabdominal infection in octogenarians in whom a large percentage with acute cholecystitis, perforation and appendicitis presented with temperatures less than 37.5°C (10). Additionally, a recent study of nosocomial febrile illness in a geriatric medicine unit showed that the mean "febrile" rectal temperature of elderly subjects was merely 38.1°C (100.6°F). In this last study only 8% of febrile patients had rectal temperatures greater than 38.5°C (101.3°F) (11). The significance of these findings is that although fever is the cardinal sign of infection its absence in elderly patients is not uncommon and may delay diagnosis and the initiation of appropriate antimicrobial therapy.
Not only is the presence of fever the single most important diagnostic feature of infection, but a febrile response or its lack thereof has other important implications. Weinstein and colleagues (12) demonstrated in their review of several hundred cases of bacteremia and fungemia that the more robust the fever response to these serious infections, the more likely was the survival. This and other studies have firmly established for many infectious diseases that the absence of fever in response to a serious acute infection is a poor prognostic sign for all age groups. Although the diagnostic and prognostic implications of fever are clear, it is less well established that in humans fever is an important host defense mechanism.
The best evidence that fever is an important host defense mechanism comes from animal models. Experiments that showed poikilothermic animals move to warmer environments in order to raise body temperatures in response to infection confirmed that this behavior has an impact on survival. For example, in one classic experiment, a species of lizards was placed in terrariums, which were kept at different temperatures. The body temperatures of these poikilothermic animals equilibrated with the environmental temperature of the terrarium in which they were housed. Subsequently, the animals were infected: it turned out that those kept in the higher temperature terrariums (thus higher body temperatures) had a much better chance of survival than those animals kept at lower temperatures (13). Similar findings were demonstrated in goldfish (14). Based on these and other animal data, including additional data generated from mammalian experiments, it can be inferred that fever may be an important host defense mechanism in humans. The mechanism(s) by which fever may enhance host defenses is not due to a direct effect of physiologically achievable temperature elevations for most pathogens. The exceptions are Treponema pallidum, the gonococcus, and certain strains of Streptococcus pneumoniae. In these cases normal physiologically achievable body temperatures in humans can inhibit bacterial growth directly. These exceptions aside, experimental data suggest that normal physiologically achievable elevations in body temperature enhance the production of monokines, cytokines, and other factors. These factors facilitate the adherence of granulocytes and other effector cells of the immune response to endothelial cells. Also, they promote immune effector cell migration into interstitial tissue spaces, which contain pockets of infection.
The reason(s) for the blunted fever response to infection observed in a substantial number of elderly patients have not been completely elucidated. One explanation is that the baseline body temperature is lower for this population (see discussion following) and another is the inherent inaccuracies in oral temperature measurement in a subset of elderly patients with dementia, mouth breathing, and variations in respiratory patterns. Further inaccuracies in temperature measurement may be caused by the ingestion of hot and cold foods during the time of measurement. Moreover, routine rectal temperatures are often impractical in subpopulations of debilitated, poorly cooperative patients. The availability of tympanic membrane thermometers may reduce such error in temperature measurement. Reasons for the blunted fever response other than lower baseline temperatures and difficulties in accurately measuring temperatures have not been completely established.
An understanding of potential mechanisms for the blunted fever response in the elderly may be gained by reviewing the current knowledge of the pathogenesis of fever, which has been recently reviewed (15). First, pathogens activate macrophages, which produce endogenous pyrogens including tumor necrosis factor, interleukin-1 (IL-1), IL-6 and interferon-a. These pyrogens then act on the endothelium of the circumventricular organs of the anterior hypothalamus or hypothalamic cells directly and initiate a complex biochemical cascade including the production of prostaglandin E2. The resulting effect is an elevation of the "hypothalamic thermostat," which in turn results in shivering, vasoconstriction, and certain behavioral responses, all of which elevate core body temperature. When the infection subsides, the "thermostat" is reset back to normal and sweating and temperature lowering behavior ensues, restoring body temperature to baseline. Any of these pathways may be affected by aging. Animal data suggest an impaired response to endogenous pyrogens with normal aging (16-18) as well as diminished production of these pyrogens with age (19). Evidence from one model suggests that the aging brain may respond normally to directly injected endogenous pyrogens, implying that there may be a defect in endogenous pyrogens crossing the blood-brain barrier (20).
4. BASELINE TEMPERATURE, SIGNIFICANCE OF FEVER, AND FUO
It has been known for some time that baseline temperature declines significantly in the old. The clinician can remember this fact by remembering the statement "the older, the colder." Lower baseline temperatures, at least for debilitated elderly, has been reaffirmed by Castle and co-workers (21,22), who demonstrated that baseline temperature was decreased among nursing facility residents. Castle and colleagues' studies further demonstrated that infections often led to "robust" or normal increases in body temperature from baseline. However, because the baseline temperature was lower, the rise in temperature, which accompanied infection, often did not reach an oral temperature of 101°F (38.3°C). Generally, 101°F (38.3°C) is the temperature level that many clinicians consider to be the definition of fever. These studies suggest that new definitions for a fever need to be established for the geriatric patient. Based on these studies, an oral temperature of 99°F (37.2°C) or greater on repeated measurements in an elderly nursing facility resident should be considered to be indicative of a fever. Similarly, a persistent rectal temperature of 99.5°F (37.5° C) would constitute a fever as would an elevation of baseline body temperature of 2°F (1.1°C). Finally, the clinician should always remember that any unexplained acute or subacute change in functional status, regardless of whether or not a fever or change in body temperature is present, may indicate the presence of an infection.
The presence of a "robust" or normal fever response to infection in an elderly person has special significance. An extensive study of over 1200 ambulatory patients showed that in contrast to younger patients in whom"benign" viral infections were common, the older febrile patient is more likely to harbor a serious bacterial infection (19). This finding was confirmed by another study, which also confirmed that leukocyte elevations in response to an infection were less in the aged compared with the young (24). Based on these studies, it is recommended that any elderly patient with an oral temperature of 101 °F (38.3°C) or greater be evaluated for a serious bacterial infection.
Fever of unknown origin (FUO) in the elderly differs from the young because a diagnosis can be made in a higher percentage of cases. Furthermore, infections are more likely to cause FUO in the elderly. It is well worthwhile investigating FUO in the elderly because treatable conditions are often found (25,26).
1. Leinicke, T., Navitsky, R., Cameron, S., et. al. (1999). Fever in the elderly: how to surmount the unique diagnostic and therapeutic challenges. Emerg. Med. Pract. 1(5), 1-24.
2. Norman, D. C. and Yoshikawa, T. T. (1984) Intraabdominal infection: diagnosis and treatment in the elderly patient. Gerontology 30, 327-338.
3. Gleckman, R. and Hibert, D. (1982) Afebrile bacteremia. a phenomenon in geriatric patients. JAMA 248, 1478-1481.
4. Finkelstein, M., Petkun ,W. M., Freedman, M. L., et al. (1983) Pneumococcal bacteremia in adults: age-dependent differences in presentation and outcome. J. Am. Geriatr. Soc. 31,19-27.
5. Bentley, D. W. (1984) Bacterial pneumonia in the elderly: clinical features, diagnosis, etiology and treatment. Gerontology 30, 297-307.
6. Marrie, T. S., Haldane, E. V., Faulkner, R. S., et al. (1985) Community-acquired pneumonia requiring hospitalization: is it different in the elderly? J. Am. Geriatr. Soc. 33, 671-680.
7. Perez-Guzman, C., Vargas, M.H., Torres-Cruz, A., et. al. (1999) Does aging modify pulmonary tuberculosis? A meta-analytical review. Chest 116(4), 961-967.
8. Yoshikawa, T. T. and Norman, D. C. (1998) Fever in the elderly. Infect. Med. 15(10), 704-706.
9. Norman, D. C. (1998) Fever and aging. Infect. Dis. Clin. Pract. 7(8), 387-390.
10. Potts, F. E., IV and Vukov, L. F. (1999) Utility of fever and leukocytosis in acute surgical abdomens in octogenarians and beyond. J. Gerontol. (Med. Sci.) 54A(2), M55-M58.
11. Trivalle, C., Chassagne, P., Bouaniche, M., et al. (1998) Nosocomial febrile illness in the elderly: frequency, causes, and risk factors. Arch. Intern. Med. 158(14), 1560-1565.
12. Weinstein, M. P., Murphy, J. R., Reller, R. B., et al. (1983) The clinical significance of positive blood cultures: a comprehensive analysis of 500 episodes of bacteremia and fungemia II: clinical observations with special reference to factors influencing prognosis. Rev. Infect. Dis. 5, 54-70.
13. Kluger, M. J., Ringler, D. M., and Anver, M. R. (1975) Fever and survival. Science 188,166168.
14. Covert, J. B. and Reynolds, W. M. (1977) Survival value of fever in fish. Nature 267, 43-45.
15. Dinarello, C. A. (1999) Cytokines as endogenous pyrogens. J. Infect. Dis. 179(Suppl 2), S294-S304.
16. Norman, D. C., Yamamura, R. H. and Yoshikawa, T. T. (1988) Fever response in old and young mice after injection of interleukin. J. Gerontol. 43, M80-M85.
17. Miller, D., Yoshikawa, T. T., Castle, S. C., et al. (1991) Effect of age in fever response to recombinant tumor necrosis factor alpha in a murine model. J. Gerontol. 46, M176- M179.
18. Miller, D. J., Yoshikawa, T. T., and Norman, D. C. (1995) Effect of age on fever response to recombinant interleukin-6 in a murine model. J. Gerontol. 50A, M276-M279.
19. Bradley, S. F., Vibhagool, A., Kunkel S. L, et al. (1989) Monokine secretion in aging and protein malnutrition. J. Leukocyte Biol. 45, 510-514.
20. Satinoff, E., Peloso, E., and Plata-Salamn, C. R. (1999) Prostaglandin E2-induced fever in young and old Long-Evans rats. Physiol. Behav. 67(1), 149-152.
21. Castle, S. C., Norman, D. C., Yeh, M., et al. (1981) Fever response in elderly nursing home residents: are the older truly colder? J. Am. Geriatr. Soc. 39, 853-857.
22. Castle, S. C., Yeh, M., Toledo, S., et al. (1993) Lowering the temperature criterion improves detection of infections in nursing home residents. Aging Immunol. Infect. Dis. 4, 67-76.
23. Keating, J. H., III, Klimek, J. J., Levine, D. S., et. al. (1984) Effect of aging on the clinical significance of fever in ambulatory adult patients. J. Am. Geriatr. Soc. 32, 282-287.
24. Wasserman, M., Levenstein, M., Keller, E., et. al. (1989) Utility of fever, white blood cells, and differential count in predicting bacterial infections in the elderly. J. Am. Geriatr. Soc. 37, 537-543.
25. Espositio, A. L. and Gleckman, R. A. (1978) Fever of unknown origin in the elderly. J. Am. Geriatr. Soc. 26, 498-505.
26. Knockaert, D. C., Vanneste, L. J., and Bobbaers, J. H. (1993) Fever of unknown origin in elderly patients. J. Am. Geriatr. Soc. 41, 1187-1192.
Was this article helpful?