TUBERCULOSIS AND SYPHILIS IN THE AIDS ERA
Clinical Presentation and Diagnosis
Since 1986, syphilis and tuberculosis have reemerged as significant health care problems, in part related to the AIDS epidemic. Between 1986 and 1990, what was considered an epidemic of syphilis was noted in the United States. During this time, rates were approximately 12 to 21 cases per 100,000 population per year. Since that time, rates of primary plus secondary syphilis have declined by about 84%, and nationally are at 3.2 cases per 100,000 population. This is attributed to better management and control strategies for both syphilis and HIV/AIDS. Similar recent declines have been noted for tuberculosis. In 1995, approximately 22,800 cases of tuberculosis were reported to the Centers for Disease Control, representing three consecutive years of decline. By 1997, reported cases had further declined to 19,855 (7.4 cases per 100,000 population). This represents a 26% decline since 1992. For cases reported with susceptibilities, 7.6% of isolates were resistant to isoniazid (INH), and 1.3% were multiply drug-resistant. Unfortunately, HIV status was reported only in about 50% of patients with tuberculosis, and the range of coinfection by state was nil to 48%. The simultaneous presence of HIV with either tuberculosis or syphilis is of major concern because of the differences in clinical presentation and the potential for altered responses to therapy.
Although pulmonary and extrapulmonary tuberculosis were described early in the AIDS epidemic, the full impact of this mycobacterial infection has been recognized only within the past several years. Between 1984 and 1991, there were approximately 39,000 more cases of tuberculosis than had been anticipated. By 1991, the number of new cases rose to approximately 26,280, an increase of 2.3% over the previous year, with most cases geographically linked to regions associated with HIV infection. Subsequently, numbers have declined to fewer than 20,000 cases per year. A recent prospective investigation of more than 1,100 HIV-infected persons without AIDS-defining diagnoses demonstrated that tuberculosis developed in 31 of them (16 with pulmonary tuberculosis only, seven with only extrapulmonary disease, and eight simultaneously infected with pulmonary and extrapulmonary tuberculosis) during a median follow-up period of 53 months. These numbers, representing 0.7 cases per 100 patient-years, were related to geographic area and the magnitude of initial results of purified protein derivative testing. The risk for development of tuberculosis was eightfold higher in persons with tuberculin test reactions larger than 5 mm than for those with smaller measurements. Negative results on skin testing for mumps were also statistically associated with the development of tuberculosis. Of eligible patients, only 55% received prophylactic INH, and only 59% of these completed 6 months of prophylaxis.
Studies of the prevalence of HIV among patients with proven or suspected tuberculosis demonstrate rates of up to 46%, with many urban centers reporting rates in excess of 10%. Alternatively, an average of 3.8% of patients with AIDS also appear on tuberculosis registries (range, 0 to 10%). This correlation is apparent mostly in IV drug users and selected minorities.
The pathogenesis of tuberculosis in patients with HIV infection involves either reactivation of previously latent disease or new acquisition of Mycobacterium tuberculosis, most commonly by droplet inhalation. Both modes are probably exacerbated by concomitant HIV infection because of its impact on cell-mediated immunity, especially as related to the decline in numbers of CD4 cells and macrophage and monocyte dysfunction.
Clinical Presentation and Diagnosis
Tuberculosis may present at any time in the course of HIV/AIDS and is therefore one of the few AIDS-defining opportunistic infections associated with CD4-cell counts greater than 200/µL. Thus, many patients may present to tuberculosis clinics and physicians’ offices before being given a diagnosis of HIV infection. Pulmonary involvement is seen in approximately 75% of patients and may present as it does in persons who are not infected with HIV. This is especially true in patients with higher CD4-cell counts. Alternatively, patients may present with atypical disease, which includes disease affecting primarily the lower lobes, dissemination, hilar adenopathy, and extrapulmonary involvement. Several studies now demonstrate a marked increase in extrapulmonary tuberculosis in patients coinfected with HIV; extrapulmonary may coexist with pulmonary involvement. Common extrapulmonary manifestations include lymphadenitis, dissemination (miliary), meningitis, and focal disease at other anatomic locations. A recent investigation of HIV and tuberculosis demonstrated that HIV-positive patients were more likely to have extrapulmonary tuberculosis, but smears for acid-fast bacilli in patients with pulmonary tuberculosis were positive more frequently among the cohort without HIV seropositivity. Other opportunistic infections, such as Pneumocystis carinii pneumonia, cytomegalovirus pneumonia, and fungal infections, need to be considered and may not be distinguishable on clinical or radiographic grounds. Tuberculous lymphadenitis may involve nodes about the neck and affect those of the hilum and mediastinum. The clinical presentation is usually that of fever and tender lymphadenopathy. Lymph node aspiration demonstrates acid-fast bacilli in more than 67% of cases. When disease involves the lung, the diagnosis is based on a high index of suspicion plus positive results on acid-fast smear and culture of sputum. Alternatively, screening of asymptomatic HIV-positive patients by chest roentgenography is not warranted.
Skin testing can provide important clues but does not prove the presence of active infection. Results of skin testing follow the degree of immunosuppression resulting from HIV infection. Tuberculin reactivity is thought to be maintained throughout early HIV infection. Overall, about 40% of patients with HIV infection and tuberculosis react positively to intermediate-strength purified protein derivative (Mantoux test). Current recommendations are to consider a 5-mm induration as a positive result in patients with HIV, and (unless specifically contraindicated) all tuberculin-positive patients should receive INH prophylaxis. Debate continues regarding the value of controls and the determination of anergy. The most clinical experience has been acquired with mumps and Candida antigens as controls; however, it has not been routinely demonstrated that anergy to these is associated with increased risks for the development of tuberculosis. Most recent guidelines from the Centers for Disease Control no longer recommend the routine use of controls in tuberculin testing.
The diagnosis of active tuberculosis requires either identification of the organism or a clinical or radiographic response to specific therapy. The best data demonstrate that smears will be positive in 30% to 80% (expectorated sputa) and 30% to 75% (bronchoscopy) of cases with pulmonary involvement. The likelihood of recovery of organisms is higher in patients with higher CD4-cell counts. For disease outside the lungs, a high index of suspicion is necessary so that appropriate smears and cultures can be obtained. When lysis-centrifugation techniques are used, blood cultures for M. tuberculosis are positive in up to 40% of cases and should be obtained when dissemination is considered.
Chest x-ray findings are generally abnormal in patients with pulmonary tuberculosis but may be normal in those with strictly endobronchial disease. “Usual” changes, such as involvement of the upper lobes with or without cavitation and effusion, are often noted, but atypical presentations, such as infiltrates in lower lobes, miliary patterns, and mediastinal adenopathy, are also regularly seen. No features should be considered pathognomonic of tuberculosis, and virtually all have important differential diagnoses.
The therapy of M. tuberculosis infection has been complicated by the recent emergence in some geographic areas of primary drug resistance to front-line medications such as INH and rifampin. At least 12 such outbreaks have been reported since 1990, and they often involve the development of new infection rather than reactivation. Although to date most cases have occurred in inner-city or prison populations, there is little reason to think that highly resistant strains will not emerge in other venues. This problem has been complicated by recent difficulties in obtaining streptomycin. In the absence of this drug, medications employed in the treatment of tuberculosis in patients with HIV are similar to those in other populations. Most data suggest that patients respond well, and treatment failures or relapses are noted in fewer than 5% of cases. However, most investigators treat patients for more prolonged periods. Table 85-1 summarizes treatment recommendations.
Table 85-1. Therapy of tuberculosis in the HIV-positive patient
Standard therapy for sensitive organisms should include INH, rifampin, and pyra-zinamide for 2 months, followed by INH plus rifampin for either an additional 7 months or 6 months after culture negativity (whichever is longer). Patients with proven or presumed multiply resistant organisms (generally at least to both INH and rifampin) need to be treated with multiple-drug regimens, often containing third-line or investigational agents (e.g., ciprofloxacin, ofloxacin, and amikacin). Mortality rates in these outbreaks have been in excess of 75%, often within 4 months of diagnosis. Of the quinolones, of-loxacin and ciprofloxacin have been best studied. They should not be employed as first-line agents, but rather in conjunction with other agents when the use of INH and rifampin is impossible. Susceptibility testing to these agents should be performed. Compliant patients who tolerate complex regimens can be expected to do better. In one published study, the response rate was better than 50% when patients took medicines for at least 2 consecutive weeks within a month of diagnosis. The optimal length of treatment has not been determined but is at least 12 months following culture negativity. Therapy for at least 18 months is recommended if neither INH nor rifampin can be used because of intolerance.
Tuberculin skin testing should be employed in all HIV-positive patients, and chemoprophylaxis should be offered to all HIV-positive patients with positive (>5 mm) tuberculin test results who do not demonstrate active tuberculosis. Selected anergic patients felt to be at high risk for infection may also be offered prophylaxis. Several options now exist. The standard is INH for 12 months. Alternatives are summarized in Table 85-2. Thus, 2-month regimens with rifampin and pyrazinamide offer the alternative of more pills for shorter periods, but without significantly enhanced toxicity. Issues of drug interactions must be addressed. Although some authorities recommend prophylaxis indefinitely, there are no data to support duration in excess of 12 months. All HIV-positive persons exposed to tuberculosis should receive prophylaxis, regardless of skin test reactivity. Prophylactic formulations for persons exposed to multidrug-resistant tuberculosis are experimental but could involve rifampin alone (resistance to INH only) or combinations of ciprofloxacin or ofloxacin plus pyrazinamide (resistance to multiple drugs).
Table 85-2. Regimens for tuberculosis prevention
Recommendations for the prevention of transmission of tuberculosis in health care settings have been published. These stress the need to employ multifactorial methods: early identification and treatment of patients, prevention of spread of droplet nuclei into the air by use of source-control methodology, reduction of air contamination, decontamination of equipment, and surveillance.
The incidence of syphilis has risen during the past several years; this is in large part related to HIV infection. In 1990, more than 50,000 cases of syphilis were reported, an increase of approximately 75% from 1985. Most cases were transmitted heterosexually. Since this time, rates have fallen considerably, in part because of better methods of detection and management of both HIV infection and syphilis. Interactions between HIV infection and syphilis are well described. These include an enhanced likelihood of HIV transmission in patients with primary syphilis, altered serologic responses, more rapid progression of primary syphilis to neurosyphilis, and potential for delayed response to therapy.
That primary syphilis (and other ulcerative genital lesions) is associated with increased transmission of HIV has been well established. Several investigations in Africa and the United States document increased rates of HIV positivity in persons treated for syphilis or having a history of genital ulcers. The reasons for this association include the disruption of epithelial barriers in ulcers (thus facilitating HIV transmission) and the availability of large numbers of activated T lymphocytes at the base of genital lesions, which may expedite entry of HIV into the immune system.
A modification of serologic response to syphilis in persons infected with HIV has been documented but is the exception rather than the rule. Selected patients may demonstrate marked fluctuations in titers. However, in most instances, both reaginic tests [Venereal Disease Research Laboratory (VDRL), rapid plasma reagin (RPR)] and specific treponemal tests [fluorescent treponemal antibody absorption (FTA-ABS), microhemagglutination specific for Treponema pallidum (MHA-Tp)] should be used in the same manner as in non–HIV-infected populations. The diagnosis of neurosyphilis requires a combination of techniques that includes interpretation of cerebrospinal fluid (CSF) VDRL test results and chemical and cellular results of lumbar puncture. The serologic response to therapy may also be altered in patients who are HIV-positive. Response may be delayed, and a higher percentage of persons may become nonreactive on specific treponemal testing.
Clinical syphilis is clearly modified in patients with HIV. Although no data exist that document more severe primary disease, accelerated progression to neurosyphilis has been noted. Up to 1.5% of patients with AIDS have neurosyphilis. Other studies have noted that 44% of persons with neurosyphilis had AIDS, despite appropriate therapy for early disease.
Neurosyphilis in the AIDS era has changed. Most cases now appear to represent early neurosyphilis, with an average age of onset below 40. Primary manifestations of meningitis, meningovascular disease, and uveitis are noted. Late neurosyphilis manifested as tabes dorsalis or paresis occurs less commonly. The most common manifestations of early neurosyphilis are meningitis, cranial nerve palsies, and meningovascular disease. The cranial nerves most affected include the ophthalmic and auditory nerves. At least 95% of persons have abnormal CSF findings that are nondiagnostic but include elevated protein, pleocytosis, and hypoglycorrhachia. Cell counts range from 40 to 400/mm3, often with a lymphocyte predominance. Virtually all patients with neuro-syphilis demonstrate CSF cell counts of more than 5/mm3. CSF protein levels are generally below 500 mg/dL but may occasionally be higher. Interpretation is complicated by other illnesses, including HIV infection, that can cause similar central nervous system abnormalities.
The treatment of primary and secondary syphilis in patients with HIV infection is similar to that in non–HIV-infected patients. IM injection of 2.4 g of benzathine penicillin is recommended. Some experts would treat with three doses, each a week apart. VDRL studies should be repeated at 6 and 12 months, and a fourfold reduction in titer is expected at 6 months. For patients who fail to respond, CSF examination and (if CSF unremarkable) repeated treatment for 3 weeks is recommended. The need for routine lumbar puncture in HIV-infected patients with syphilis is uncertain. It should be performed diagnostically in all persons with neurologic findings or complaints, both to define the length of parenteral therapy and to monitor response. Patients with advanced syphilis (late latent syphilis or neurosyphilis) of the central nervous system are preferably treated parenterally for 10 to 14 days with 18 to 24 million units of IV aqueous penicillin G, given daily in four to six divided doses. The treatment of patients with early neurosyphilis is associated with high cure rates. Table 85-3 summarizes the recommendations for treatment of syphilis in HIV-infected persons. Few data exist to recommend agents other than penicillin for neurosyphilis or latent syphilis. Recent investigations have documented a disturbingly high failure rate (23%) with IV ceftriaxone administered in doses of up to 2 g daily. Under all circumstances, vigilant follow-up is indicated to define failure or relapse. (R.B.B.)
Table 85-3. Therapy of syphilis in HIV-infected patientst
Ad Hoc Committee of the Scientific Assembly on Microbiology, Tuberculosis, and Pulmonary Infections. Treatment of tuberculosis and tuberculosis infection in adults and children. Clin Infect Dis 1995;21:9–27.
An excellent summary of available data and recommendations for the management of tuberculosis in the United States. Medications employed in management are discussed, along with proposed regimens. The value of directly observed therapy is re-inforced.
Alpert PL, et al. A prospective study of tuberculosis and human immunodeficiency virus infection: clinical manifestations and factors associated with survival. Clin Infect Dis 1997;24:661–668.
More than 100 HIV-positive patients with tuberculosis were identified, prospectively followed, and compared with an HIV-negative cadre. Those who were HIV-positive were more likely to have extrapulmonary tuberculosis and were less likely to have sputum smears positive for acid-fast bacilli. Improved survivorship was correlated with rapid initiation of treatment and less severe HIV disease.
Barnes PF, et al. Tuberculosis in patients with human immunodeficiency virus infection. N Engl J Med 1991;324:1644–1650.
Excellent overview of epidemiology, clinical features, diagnosis, and management of tuberculosis in patients with HIV infection. Provides therapeutic strategies based on the likelihood of drug resistance and ways to distinguish presumptively M. tuberculosis from Mycobacterium avium complex.
Busillo CP, et al. Multidrug-resistant Mycobacterium tuberculosis in patients with human immunodeficiency virus infection. Chest 1992;102:797–801.
This investigation demonstrates that since 1990, approximately one third of patients with tuberculosis have harbored multidrug-resistant strains. Resistance is defined as resistance to INH plus at least one other first-line agent. Often, resistance was noted to at least three agents. Approximately 75% of patients died despite therapy with at least four drugs. Optimal therapy is unknown.
Centers for Disease Control and Prevention. Tuberculosis morbidity—United States, 1997. MMWR Morb Mortal Wkly Rep 1998;47:253–257.
Excellent recent data are provided that demonstrate the continued decline in rates of active tuberculosis in the United States. Most of this decline is attributable to cases in persons born in the United States. Rates among immigrants have actually increased slightly. Of isolates for which susceptibility data are available, about 8% demonstrate resistance to INH, and 1.3% are multidrug-resistant. Furthermore, cases of multidrug-resistant tuberculosis appear to be decreasing.
Centers for Disease Control and Prevention. 1998 Guidelines for treatment of sexually transmitted diseases. MMWR Morb Mortal Wkly Rep 1998;47(RR1):28–49.
An extremely contemporary review of guidelines for the management of syphilis, with especial reference to HIV/AIDS. Best single source of recommendations for the management of syphilis. Provides information for treatment that includes tables for penicillin desensitization.
Centers for Disease Control and Prevention. Primary and secondary syphilis—United States, 1997. MMWR Morb Mortal Wkly Rep 1998;47:493–497.
This article provides recent data on the epidemiology of primary and secondary syphilis. During the past several years, the incidence has declined by up to 84%. Rates remain higher for blacks and are highest in the South.
Chaisson RE, et al. Tuberculosis in patients with the acquired immunodeficiency syndrome: clinical features, response to therapy, and survival. Am Rev Respir Dis 1987;136:570–574.
This vintage investigation documents the interaction between tuberculosis and HIV infection, demonstrating the frequency of extrapulmonary disease and the fact tuberculosis may antedate the diagnosis of HIV. A good clinical response to therapy was generally noted.
Flood JM, et al. Neurosyphilis during the AIDS epidemic, San Francisco, 1985–1992. J Infect Dis 1998;177:931–940.
The authors review their experience with 117 patients who had neurosyphilis, of whom approximately 65% were HIV-infected. Most were young and presented with early neurosyphilis. High VDRL titers were often noted. Treatment of this cadre was generally curative.
Hook EW III, Marra CM. Acquired syphilis in adults. N Engl J Med 1992;326:1060–1069.
Provides excellent demographic data on the impact of HIV on the prevalence of syphilis and insights into the natural history of syphilis and the impact of HIV. Points out difficulties in defining “cure” and the need for further information regarding optimal treatment.
Hopewell PC. Impact of human immunodeficiency virus infection on the epidemiology, clinical features, and control of tuberculosis. Clin Infect Dis 1992;15:540–547.
Provides timely data on epidemiology, clinical features, and pathogenesis of tuberculosis in the setting of HIV. Defines reasonable management strategies and describes important drug interactions. Redefines the need for relevant procedures of infection control to limit the spread of tuberculosis, including disease caused by multidrug-resistant strains.
Musher DM, Hamill RJ, Gaughn RE. Effect of human immunodeficiency virus (HIV) infection on the course of syphilis and on the response to treatment. Ann Intern Med 1990;113:872–881.
This article represents an exhaustive review of the English literature dealing with syphilis before and after the introduction of penicillin, and provides excellent insights into the natural history of the disease. Also extensively reviews data on the impact of HIV on syphilis and provides strategies for management.