1999 USPHS/IDSA Guidelines for the Prevention of Opportunistic Infections in Persons Infected with Human Immunodeficiency Virus FREE

In August 1999, agencies of the U.S. Public Health Service (USPHS), in collaboration with the Infectious Diseases Society of American (IDSA), published 1999 Guidelines for the Prevention of Opportunistic Infections in Persons Infected with Human Immunodeficiency Virus (HIV) (1). These guidelines are an update of those prepared in 1995 and 1997 and published in several venues, including Annals of Internal Medicine (2, 3). They are intended primarily for health care providers who care for HIV-infected persons and are reproduced here in an effort to reach and as a service to all internists who care for HIV-infected patients.

Single copies of the 1999 USPHS/IDSA guidelines can be obtained from the AIDS Treatment Information Service (ATIS) by calling 800-448-0440, 301-217-0023 (international), or 800-243-7012 (TYY) or by downloading the document from the ATIS Web site at http://www.hivatis.org.

The working group was chaired by Henry Masur, MD, National Institutes of Health, Bethesda, MD; Jonathan E. Kaplan, MD, Centers for Disease Control and Prevention, Atlanta; and King K. Holmes, MD, PhD, University of Washington, Seattle.

Members of the group included Beverly L. Alston, MD (National Institutes of Health, Bethesda, MD); Neil Ampel, MD (University of Arizona, Tucson); Jean R. Anderson, MD (Johns Hopkins University, Baltimore); A. Cornelius Baker (National Association of People with AIDS, Washington, DC); David Barr (Forum for Collaborative HIV Research, Washington, DC); John G. Bartlett, MD (Johns Hopkins University, Baltimore); John E. Bennett, MD (National Institutes of Health, Bethesda, MD); Constance A. Benson, MD (University of Colorado, Denver); Samual A. Bozzette, MD (University of California, San Diego); Richard E. Chaisson, MD (Johns Hopkins University, Baltimore); Clyde S. Crumpacker, MD (Harvard Medical Center, Boston); Judith S. Currier, MD, MSc (University of California-Los Angeles Medical Center, Los Angeles); Lawrence Deyton, MD, MSPH (U.S. Department of Veterans Affairs, Washington, DC); W. Lawrence Drew, MD, PhD (Mt. Zion Medical Center, University of California-San Francisco, San Francisco); William R. Duncan, PhD (National Institutes of Health, Bethesda, MD); Robert W. Eisinger, PhD (National Institutes of Health, Bethesda, MD); Wafaa El-Sadr, MD, MPH, MPA (Harlem Hospital, New York); Judith Feinberg, MD (Holmes Hospital, Cincinnati); Kenneth A. Freedberg, MD, MSc (Boston University School of Medicine, Boston); Hansjakob Furrer, MD (University Hospital, Berne, Switzerland); John W. Gnann Jr., MD (University of Alabama, Birmingham); Mark J. Goldberger, MD, MPH (U.S. Food and Drug Administration, Rockville, MD); Sue Goldie, MD, PhD (Harvard School of Public Health, Boston); Eric P. Goosby, MD (U.S. Department of Health and Human Services, Washington, DC); Peter A. Gross, MD (Hackensack Medical Center, Hackensack, NJ); Richard Hafner, MD (National Institutes of Health, Bethesda, MD); Diane Havlir, MD (University of California, San Diego); Thomas M. Hooton, MD (Harborview Medical Center, Seattle); Douglas A. Jabs, MD (Johns Hopkins University, Baltimore); Mark A. Jacobson, MD (University of California, San Francisco); Edward Janoff, MD (Veterans Administration Medical Center, Minneapolis); Mari Kitahata, MD, PhD (University of Washington, Seattle); Joseph V. Kovacs, MD (National Institutes of Health, Bethesda, MD); Catherine Leport, MD (Hospital Bichat-Claude Bernard, Paris); Myron J. Levin, MD (University of Colorado Health Science Center, Denver); Juan C. Lopez, MD (Hospital Universatario Gregorio Maranon, Madrid); Michael Marco (Treatment Action Group, New York); Douglas L. Mayers, MD (Henry Ford Hospital, Detroit); David A. Melnick, MD (Kaiser Permanente, Springfield, VA); Lynne M. Mofenson, MD (National Institutes of Health, Bethesda, MD); Julio S.G. Montaner, MD (St. Paul's Hospital, Vancouver); Richard Moore, MD (Johns Hopkins University, Baltimore); James Neaton, PhD (University of Minnesota, Minneapolis); Charles Nelson (National Association of People with AIDS, Washington, DC); Joseph F. O'Neill, MD, MS, MPH (Health Resources and Services Administration, Rockville, MD); Joel Palefsky, MD (University of California, San Francisco); Alice Pau, PharmD (National Institutes of Health, Bethesda, MD); John P. Phair, MD (Northwestern University, Chicago); Stephen Piscitelli, PharmD (National Institutes of Health, Bethesda, MD); Michael A. Polis, MD, MPH (National Institutes of Health, Bethesda, MD); Thomas C. Quinn, MD (Johns Hopkins Hospital, Baltimore); Peter Reiss, MD, PhD (University of Amsterdam, Amsterdam); David Rimland, MD (Veterans Administration Medical Center, Atlanta); Cynthia L. Sears, MD (Johns Hopkins University, Baltimore); Leonard Seeff, MD (National Institutes of Health, Bethesda, MD); Kent A. Sepkowitz, MD (Memorial Sloan-Kettering Cancer Center, New York); Thomas G. Slama, MD (National Foundation for Infectious Diseases, Indianapolis); Elaine M. Sloand, MD (National Institutes of Health, Bethesda, MD); Stephen A. Spector, MD (University of California, La Jolla); David L. Thomas, MD, MPH (Johns Hopkins University, Baltimore); Russell B. Van Dyke, MD (Tulane School of Medicine, New Orleans, LA); D. Heather Watts, MD (National Institutes of Health, Bethesda, MD); L. Joseph Wheat, MD (Indiana University School of Medicine, Indianapolis); Scott M. Whitcup, MD (National Institutes of Health, Bethesda, MD); Paige Williams, PhD (Harvard School of Public Health, Boston); Thomas C. Wright Jr., MD (Columbia University College of Physicians and Surgeons, New York).

Participants from the Centers for Disease Control and Prevention, Atlanta, included Kenneth G. Castro, MD, Kevin M. DeCock, MD, DTM&H, Scott F. Dowell, MD, MPH, Mark S. Dworkin, MD, MPHTM, Clare Dykewicz, MD, MPH, Tedd Ellerbrock, MD, Rana Hajjeh, MD, Scott Holmberg, MD, MPH, David R. Holtgrave, PhD, Harold W. Jaffe, MD, Jeffrey L. Jones, MD, Dennis D. Juranek, DVM, MSc, Eric Mast, MD, MPH, Thomas Navin, MD, Phil E. Pellett, PhD, William C. Reeves, MD, MPH, John A. Stewart, MD, and M. Elsa Villarino, MD, MPH.

This report was prepared by Jonathan E. Kaplan, MD (Division of AIDS, STD, and TB Laboratory Research, National Center for Infectious Diseases and Division of HIV/AIDS Prevention—Surveillance and Epidemiology, National Center for HIV, STD, and TB Prevention) in collaboration with Henry Masur, MD (National Institutes of Health) and King K. Holmes, MD, PhD (University of Washington).

In 1995, the U.S. Public Health Service (USPHS) and the Infectious Diseases Society of America (IDSA) developed guidelines for preventing opportunistic infections (OIs) in persons infected with human immunodeficiency virus (HIV) (1 - 3). These guidelines, written for health-care providers and patients, were revised in 1997 and published in MMWR (4), Clinical Infectious Diseases (5), Annals of Internal Medicine (6), American Family Physician (7), and Pediatrics (8); an accompanying editorial appeared in JAMA (9). Response to these guidelines (e.g., the many requests for reprints and observations from health-care providers) suggests they have served as a valuable reference for HIV care providers. Because the 1995 and 1997 guidelines included ratings indicating the strength of each recommendation and the quality of supporting evidence, readers were able to assess the relative importance of each recommendation.

Since AIDS was first recognized nearly 20 years ago, remarkable progress has been made in improving the quality and duration of life of HIV-infected persons. During the first decade of the epidemic, this improvement occurred because of better recognition of opportunistic disease processes, better therapy for acute and chronic complications, and the introduction of chemoprophylaxis against Pneumocystis carinii pneumonia (PCP), toxoplasmosis, Mycobacterium avium complex (MAC) disease, and bacterial infections. Trimethoprim-sulfamethoxazole (TMP-SMZ) was shown to reduce the incidence not only of PCP but also of toxoplasmosis and bacterial infections.

The second decade of the epidemic has witnessed extraordinary progress in developing highly active antiretroviral therapies (HAARTs) as well as continuing progress in preventing and treating individual OIs. HAART has reduced the incidence of OIs and extended life substantially (10). HAART is the most effective approach to preventing OIs and should be considered for all HIV-infected persons who qualify for such therapy. However, some patients are not ready or able to take HAART, and others have tried HAART regimens, but therapy has failed. Such patients will benefit from prophylaxis against OIs. In addition, prophylaxis against specific OIs continues to provide survival benefits even among persons who are receiving HAART (11).

Because important new data concerning the prevention of opportunistic diseases have emerged since 1997, the USPHS and the IDSA reconvened the Prevention of Opportunistic Infections Working Group on March 4 and 5, 1999, to determine which recommendations warranted revision. Participants included representatives from federal agencies, universities, and professional societies, as well as community health-care providers and patient advocates. Much attention was focused on recent data related to the advisability of discontinuing OI prophylaxis (primary prophylaxis and prophylaxis against recurrence) among persons whose CD4⁺ T-lymphocyte counts have increased to above prophylaxis thresholds because of HAART. The OI Working Group also addressed two pathogens not previously considered—human herpesvirus type 8 (HHV-8) and hepatitis C virus (HCV). In addition, working group members reviewed data concerning the prevention of all common HIV-associated OIs. In revising these current guidelines, as in earlier editions of the guidelines, the group considered factors such as incidence of disease; severity of disease in terms of morbidity and mortality; level of immunosuppression at which disease is most likely to occur; feasibility, efficacy, and cost of preventive measures; impact of intervention on quality of life; and drug toxicities, drug interactions, and the potential for drug resistance to develop.

During the development of these revised guidelines, working group members reviewed published manuscripts as well as abstracts and material presented at professional meetings if complete manuscripts providing data were available for review. A review of the data that served as the basis for the revisions and additional information discussed at the meeting but not deemed sufficient to justify a revision of the recommendations will be published separately in Clinical Infectious Diseases.

Primary changes in the disease-specific recommendations that follow include

These guidelines developed by the OI Working Group were made available for public comment through announcements in the Federal Register and MMWR. The final document is endorsed by the USPHS and IDSA as well as by the Infectious Diseases Society of Obstetrics and Gynecology and the National Foundation for Infectious Diseases.

For each of the 19 diseases covered in this report, specific recommendations are provided that address a] prevention of exposure to the opportunistic pathogen, b) prevention of the first episode of disease, and c) prevention of disease recurrence. Recommendations are rated by a revised version of the IDSA rating system (Table 1) (12). In this system, the letters A through E signify the strength of the recommendation for or against a preventive modality, and Roman numerals I through III indicate the quality of evidence supporting the recommendation.

Because of their length and complexity, the rest of the tables and figure in this report are grouped together, following the references. The tables and figure appear in the following order: dosages for prophylaxis to prevent first episode of opportunistic disease in HIV-infected adults and adolescents (Table 2); dosages for prophylaxis to prevent recurrence of opportunistic disease in HIV-infected adults and adolescents (Table 3); effects of food on drugs used to treat OIs (Table 4); effects of medications on drugs used to treat OIs (Table 5); effects of OI medications on drugs commonly administered to HIV-infected persons (Table 6); adverse effects of drugs used to manage HIV infection (Table 7); dosages of drugs for prevention of OIs for persons with renal insufficiency (Table 8); costs of agents recommended for the prevention of OIs in adults with HIV infection (Table 9); immunologic categories for HIV-infected children (Table 10); immunization schedule for HIV-infected children (Figure); dosages for prophylaxis to prevent first episode of opportunistic disease in HIV-infected infants and children (Table 11); dosages for prophylaxis to prevent recurrence of opportunistic disease in HIV-infected infants and children (Table 12); and criteria for discontinuing and restarting OI prophylaxis for adult patients with HIV infection (Table 13). Recommendations advising patients how to prevent exposure to opportunistic pathogens appear in the Appendix at the end of the text.

This report is oriented toward the prevention of specific OIs in HIV-infected persons in the United States and other industrialized countries. Recommendations for use of antiretroviral therapy, which is designed to prevent immunologic deterioration and delay the need for many of the chemoprophylactic strategies described in this report, are published elsewhere (10), as are integrated approaches to the care of HIV-infected persons (13).

Single copies of this report can be obtained from the AIDS Treatment Information Service (ATIS) by calling 800-448-0440, 301-217-0023 (international), or 800-243-7012 (TTY), and the report can be downloaded from the ATIS website at http://www.hivatis.org. In addition, pamphlets for patients are available from ATIS and also can be accessed on the Centers for Disease Control and Prevention's Division of HIV/AIDS Prevention homepage at http://www.cdc.gov/hiv.

New data on prevention of OIs in HIV-infected persons are emerging, and randomized controlled trials addressing some unresolved issues in OI prophylaxis are ongoing. The OI Working Group has therefore developed a mechanism for routinely and periodically reviewing emerging data and for updating these guidelines on a regular basis. The most recent information will be available from the ATIS website at http://www.hivatis.org.

(1) Although some authorities recommend that persons with HIV infection who are at risk for PCP not share a hospital room with a patient who has PCP, data are insufficient to support this recommendation as standard practice (CIII).

(2) Adults and adolescents who have HIV infection (including pregnant women and those on HAART) should receive chemoprophylaxis against PCP if they have a CD4⁺ T-lymphocyte count of <200/µL (AI) or a history of oropharyngeal candidiasis (AII) (14). Persons who have a CD4⁺ T-lymphocyte percentage of <14% or history of an AIDS-defining illness but do not otherwise qualify should be considered for prophylaxis (BII) (15 - 16). When monitoring the CD4⁺ T-lymphocyte count at least every 3 months is not possible, initiation of chemoprophylaxis at a CD4⁺ T-lymphocyte count of >200 but <250 cells/µL also should be considered (BII) (15).

(3) TMP-SMZ is the recommended prophylactic agent (AI) (16 - 18). One double-strength tablet per day is the preferred regimen (AI) (17). However, one single-strength tablet per day (19) is also effective and might be better tolerated (AI). One double-strength tablet three times per week is also effective (BI) (20). TMP-SMZ at a dose of one double-strength tablet per day confers cross-protection against toxoplasmosis (21) and some common respiratory bacterial infections (17,22). Lower doses of TMP-SMZ also might confer such protection. For patients who have an adverse reaction that is not life-threatening, treatment with TMP-SMZ should be continued if clinically feasible; for those who have discontinued such therapy because of an adverse reaction, reinstitution of TMP-SMZ should be strongly considered after the adverse event has resolved (AII). Patients who have experienced adverse events, especially fever and rash, might better tolerate reintroduction of the drug with a gradual increase in dose (desensitization) as per published regimens (BI) (23 - 24) or reintroduction of TMP-SMZ at a reduced dose or frequency (CIII); up to 70% of patients can tolerate such reinstitution of therapy (22).

(4) If TMP-SMZ cannot be tolerated, prophylactic regimens that can be recommended as alternatives include dapsone (BI) (17), dapsone plus pyrimethamine plus leucovorin (BI) (25 - 26), aerosolized pentamidine administered by the Respirgard IITM nebulizer (Marquest, Englewood, Colorado) (BI) (18), and atovaquone (BI) (27 - 28). Atovaquone appears to be as effective as aerosolized pentamidine (28) or dapsone (BI) (27) but is substantially more expensive than the other regimens. For patients seropositive for Toxoplasma gondii who cannot tolerate TMP-SMZ, recommended alternatives to TMP-SMZ for prophylaxis against both PCP and toxoplasmosis include dapsone plus pyrimethamine (BI) (25 - 26) or atovaquone with or without pyrimethamine (CIII). The following regimens generally cannot be recommended as alternatives because data regarding their efficacy for PCP prophylaxis are insufficient for a firm recommendation: aerosolized pentamidine administered by other nebulization devices, intermittently administered parenteral pentamidine, oral pyrimethamine plus sulfadoxine, oral clindamycin plus primaquine, and intravenous trimetrexate. However, clinicians may consider using these agents in unusual situations in which the recommended agents cannot be administered (CIII).

(5) Initial reports from three prospective observational studies (29 - 31), one retrospective review (32), and one randomized trial (33) suggest that PCP prophylaxis can be safely discontinued in patients responding to HAART with a sustained increase in CD4⁺ T-lymphocyte counts from <200 cells/µL to >200 cells/µL. Such reports have mostly included patients receiving primary prophylaxis (no prior episode of PCP) and protease inhibitor-containing regimens. In these studies, median follow-up ranged from 6 to 12 months and the median CD4⁺ T-lymphocyte count at the time prophylaxis was discontinued was >300 cells/µL. At the time PCP prophylaxis was discontinued, many patients had sustained suppression of HIV plasma RNA levels below detection limits of the available assays. Although optimal criteria for discontinuing PCP prophylaxis are still being assessed, providers may wish to discontinue prophylaxis when patients have sustained a CD4⁺ T-lymphocyte count of >200 cells/µL for at least 3-6 months (CII). Additional criteria might include sustained reduction in viral load for at least 3-6 months (CIII).

(6) No data are available to guide recommendations for reinstituting primary prophylaxis. Pending the availability of such data, a reasonable approach would be to use the criteria for initiating prophylaxis described on page 876 (CIII).

(7) Adults and adolescents who have a history of PCP should be administered chemoprophylaxis (i.e., secondary prophylaxis or chronic maintenance therapy) with the regimens described on page 876 in order to prevent recurrence (AI) (16).

(8) Although patients receiving secondary prophylaxis (prior episode of PCP) might also be at low risk for PCP when their CD4⁺ T-lymphocyte counts increase to >200 cells/µL, inadequate numbers of patients have been evaluated to warrant a recommendation to discontinue prophylaxis in such patients.

(9) Children born to HIV-infected mothers should be administered prophylaxis with TMP-SMZ beginning at 4-6 weeks of age (34) (AII). Prophylaxis should be discontinued for children who are subsequently found not to be infected with HIV. HIV-infected children and children whose infection status remains unknown should continue to receive prophylaxis for the first year of life. The need for subsequent prophylaxis should be determined on the basis of age-specific CD4⁺ T-lymphocyte count thresholds (Table 11) (AII). The safety of discontinuing prophylaxis in HIV-infected children receiving HAART has not been studied.

(10) Children who have a history of PCP should be administered lifelong chemoprophylaxis to prevent recurrence (AI) (34).

(11) Chemoprophylaxis for PCP should be administered to pregnant women as is done for other adults and adolescents (AIII). TMP-SMZ is the recommended prophylactic agent; dapsone is an alternative. Because of theoretical concerns regarding possible teratogenicity associated with drug exposures during the first trimester, providers may choose to withhold prophylaxis during the first trimester. In such cases, aerosolized pentamidine may be considered because of its lack of systemic absorption and the resultant lack of exposure of the developing embryo to the drug (CIII).

(1) HIV-infected persons should be tested for immunoglobulin G (IgG) antibody to Toxoplasma soon after the diagnosis of HIV infection to detect latent infection with T. gondii (BIII).

(2) All HIV-infected persons, but particularly those who lack IgG antibody to Toxoplasma, should be counseled about the various sources of toxoplasmic infection. They should be advised not to eat raw or undercooked meat, particularly undercooked pork, lamb, or venison (BIII). Specifically, meat should be cooked to an internal temperature of 150 °F (65.5 °C); meat cooked until it is no longer pink inside generally has an internal temperature of 165 °F (73.8 °C) and therefore satisfies this requirement. HIV-infected persons should wash their hands after contact with raw meat and after gardening or other contact with soil; in addition, they should wash fruits and vegetables well before eating them raw (BIII). If the patient owns a cat, the litter box should be changed daily, preferably by an HIV-negative, nonpregnant person; alternatively, the patient should wash the hands thoroughly after changing the litter box (BIII). Patients should be encouraged to keep their cats inside and not to adopt or handle stray cats (BIII). Cats should be fed only canned or dried commercial food or well-cooked table food, not raw or undercooked meats (BIII). Patients need not be advised to part with their cats or to have their cats tested for toxoplasmosis (EII).

(3) Toxoplasma-seropositive patients who have a CD4⁺ T-lymphocyte count of <100/µL should be administered prophylaxis against toxoplasmic encephalitis (TE) (AII) (21). The double-strength tablet daily dose of TMP-SMZ recommended as the preferred regimen for PCP prophylaxis appears to be effective against TE as well and is therefore recommended (AII) (21). If patients cannot tolerate TMP-SMZ, the recommended alternative is dapsone-pyrimethamine, which is also effective against PCP (BI) (25 - 26). Atovaquone with or without pyrimethamine also may be considered (CIII). Prophylactic monotherapy with dapsone, pyrimethamine, azithromycin, or clarithromycin cannot be recommended on the basis of current data (DII). Aerosolized pentamidine does not protect against TE and is not recommended (EI) (17,21).

(4) Toxoplasma-seronegative persons who are not taking a PCP prophylactic regimen known to be active against TE should be retested for IgG antibody to Toxoplasma when their CD4⁺ T-lymphocyte count declines below 100/µL to determine whether they have seroconverted and are therefore at risk for TE (CIII). Patients who have seroconverted should be administered prophylaxis for TE as described above (AII).

(5) Limited data suggest that discontinuing prophylaxis for patients whose CD4⁺ T-lymphocyte counts increase to >100 cells/µL in response to HAART is associated with a low risk for TE. However, the numbers of patients who have been evaluated are insufficient to recommend routine discontinuation of prophylaxis in such patients. Persons whose CD4⁺ T-lymphocyte count remains <200 cells/µL or who have a history of PCP or oropharyngeal candidiasis still require prophylaxis against PCP, as noted previously.

(6) Patients who have had TE should be administered lifelong suppressive therapy (secondary prophylaxis or chronic maintenance therapy) with drugs active against Toxoplasma to prevent relapse (AI) (35 - 36). The combination of pyrimethamine plus sulfadiazine and leucovorin is highly effective for this purpose (AI) (35 - 36). A commonly used regimen for patients who cannot tolerate sulfa drugs is pyrimethamine plus clindamycin (BI); however, only the combination of pyrimethamine plus sulfadiazine appears to provide protection against PCP as well (AII).

(7) The numbers of patients who have stopped maintenance therapy after responding to HAART are insufficient to warrant recommending discontinuation of maintenance therapy.

(8) TMP-SMZ, when administered for PCP prophylaxis, also provides prophylaxis against toxoplasmosis. Atovaquone might also provide protection (CIII). Children aged >12 months who qualify for PCP prophylaxis and who are receiving an agent other than TMP-SMZ or atovaquone should have serologic testing for Toxoplasma antibody (BIII), because alternative drugs for PCP prophylaxis might not be effective against Toxoplasma. Severely immunosuppressed children who are not receiving TMP-SMZ or atovaquone who are found to be seropositive for Toxoplasma should be administered prophylaxis for both PCP and toxoplasmosis (i.e., dapsone plus pyrimethamine) (BIII).

(9) TMP-SMZ can be administered for prophylaxis against TE as described for PCP (AIII). However, because of the low incidence of TE during pregnancy and the possible risk associated with pyrimethamine treatment, chemoprophylaxis with pyrimethamine-containing regimens can reasonably be deferred until after pregnancy (CIII). For prophylaxis against recurrent TE, the health-care provider and clinician should be well informed about the benefit of lifelong therapy and the concerns about teratogenicity of pyrimethamine. Most clinicians favor lifelong therapy for the mother, given the high likelihood that disease will recur promptly if therapy is stopped (AIII).

(10) In rare cases, HIV-infected pregnant women who have serologic evidence of remote toxoplasmic infection have transmitted Toxoplasma to the fetus in utero. Pregnant HIV-infected women who have evidence of primary toxoplasmic infection or active toxoplasmosis (including TE) should be evaluated and managed during pregnancy in consultation with appropriate specialists (BIII). Infants born to women who have serologic evidence of infections with HIV and Toxoplasma should be evaluated for congenital toxoplasmosis (BIII).

(1) HIV-infected persons should be educated and counseled about the many ways that Cryptosporidium can be transmitted (BIII). Modes of transmission include having direct contact with infected adults, diaper-aged children, and infected animals; drinking contaminated water; coming into contact with contaminated water during recreational activities; and eating contaminated food.

(2) HIV-infected persons should avoid contact with human and animal feces. They should be advised to wash their hands after contact with human feces (e.g., diaper changing), after handling pets, and after gardening or other contact with soil. HIV-infected persons should avoid sexual practices that might result in oral exposure to feces (e.g., oral-anal contact) (BIII).

(3) HIV-infected persons should be advised that newborn and very young pets might pose a small risk for transmitting cryptosporidial infection, but they should not be advised to destroy or give away healthy pets. Persons contemplating the acquisition of a new pet should avoid bringing any animal that has diarrhea into their households, should avoid purchasing a dog or cat aged <6 months, and should not adopt stray pets. HIV-infected persons who wish to assume the small risk for acquiring a puppy or kitten aged <6 months should request that their veterinarian examine the animal's stool for Cryptosporidium before they have contact with the animal (BIII).

(4) HIV-infected persons should avoid exposure to calves and lambs and to premises where these animals are raised (BII).

(5) HIV-infected persons should not drink water directly from lakes or rivers (AIII).

(6) Waterborne infection also might result from swallowing water during recreational activities. HIV-infected persons should be aware that many lakes, rivers, and salt-water beaches and some swimming pools, recreational water parks, and ornamental water fountains might be contaminated with human or animal waste that contains Cryptosporidium. They should avoid swimming in water that is likely to be contaminated and should avoid swallowing water while swimming or playing in recreational waters (BIII).

(7) Several outbreaks of cryptosporidiosis have been linked to municipal water supplies. During outbreaks or in other situations in which a community “boil-water” advisory is issued, boiling water for 1 minute will eliminate the risk for cryptosporidiosis (AI). Use of submicron personal-use water filters(1) (home/office types) and/or bottled water(2) also might reduce the risk (CIII). The magnitude of the risk for acquiring cryptosporidiosis from drinking water in a nonoutbreak setting is uncertain, and current data are inadequate to recommend that all HIV-infected persons boil water or avoid drinking tap water in nonoutbreak settings. However, HIV-infected persons who wish to take independent action to reduce the risk for waterborne cryptosporidiosis may choose to take precautions similar to those recommended during outbreaks. Such decisions should be made in conjunction with health-care providers. Persons who opt for a personal-use filter or bottled water should be aware of the complexities involved in selecting appropriate products, the lack of enforceable standards for the destruction or removal of oocysts, the cost of the products, and the logistic difficulty of using these products consistently.

(8) Patients who take precautions to avoid acquiring cryptosporidiosis from drinking water should be advised that ice made from contaminated tap water also can be a source of infection (BII). Such persons also should be aware that fountain beverages served in restaurants, bars, theaters, and other places also might pose a risk because these beverages, as well as the ice they contain, are made from tap water. Nationally distributed brands of bottled or canned carbonated soft drinks are safe to drink. Commercially packaged noncarbonated soft drinks and fruit juices that do not require refrigeration until after they are opened (i.e., those that can be stored unrefrigerated on grocery shelves) also are safe. Nationally distributed brands of frozen fruit juice concentrate are safe if they are reconstituted by the user with water from a safe source. Fruit juices that must be kept refrigerated from the time they are processed to the time of consumption might be either fresh (unpasteurized) or heat-treated (pasteurized); only those juices labeled as pasteurized should be considered free of risk from Cryptosporidium. Other pasteurized beverages and beers also are considered safe to drink (BII). No data are available concerning survival of Cryptosporidium oocysts in wine.

(9) HIV-infected persons should avoid eating raw oysters because cryptosporidial oocysts can survive in oysters for more than 2 months and have been found in oysters taken from some commercial oyster beds (BIII). Cryptosporidium-infected patients should not work as food handlers, especially if the food to be handled is intended to be eaten without cooking (BII). Because most foodborne outbreaks of cryptosporidiosis are believed to have been caused by infected food handlers, more specific recommendations to avoid exposure to contaminated food cannot be made.

(10) In a hospital, standard precautions (i.e., use of gloves and hand washing after removal of gloves) should be sufficient to prevent transmission of cryptosporidiosis from an infected patient to a susceptible HIV-infected person (BII). However, because of the potential for fomite transmission, some experts recommend that HIV-infected persons, especially those who are severely immunocompromised, should not share a room with a patient with cryptosporidiosis (CIII).

(11) No agents have been proven to be effective as chemoprophylaxis against cryptosporidiosis. Rifabutin or clarithromycin, when taken for MAC prophylaxis, were associated with a reduced risk for cryptosporidiosis in one study (37), but data are insufficient to warrant a recommendation for using these drugs.

(12) No drug regimens are known to be effective in preventing the recurrence of cryptosporidiosis.

(13) At present, no data indicate that formula-preparation practices for infants should be altered in an effort to prevent cryptosporidiosis (CIII). However, in the event of a “boil-water” advisory, similar precautions for the preparation of infant formula should be taken as for drinking water for adults (AII).

(1) Other than general attention to hand washing and other personal hygiene measures, no precautions to reduce exposure can be recommended at this time.

(2) No chemoprophylactic regimens are known to be effective in preventing microsporidiosis.

(3) No chemotherapeutic regimens are known to be effective in preventing the recurrence of microsporidiosis.

(1) HIV-infected persons should be advised that certain activities and occupations might increase the likelihood of exposure to tuberculosis (BIII). These include volunteer work or employment in health-care facilities, correctional institutions, and shelters for the homeless, as well as in other settings identified as high risk by local health authorities. Decisions about whether to continue with activities in these settings should be made in conjunction with the health-care provider and should be based on factors such as the patient's specific duties in the workplace, the prevalence of tuberculosis in the community, and the degree to which precautions are taken to prevent the transmission of tuberculosis in the workplace (BIII). Whether the patient continues with such activities might affect the frequency with which screening for tuberculosis needs to be conducted.

(2) When HIV infection is first recognized, the patient should receive a TST by administration of intermediate-strength (5-TU) purified protein derivative (PPD) by the Mantoux method (AI). Routine evaluation for anergy is not recommended. However, there are selected situations in which anergy evaluation might assist in guiding individual decisions about preventive therapy (38).