Resource Logo

Opportunistic Infections: Is it Safe to Stop Prophylaxis?


In the early years of the epidemic, AIDS was characterized by several opportunistic infections (OIs) that were responsible for much of the morbidity and mortality associated with the disease. Today, less is heard about these conditions, but OIs remain an important consideration for people with HIV, especially those whose immune systems have sustained considerable damage.

Traditionally, the term "opportunistic infection" has referred to conditions that rarely affect people with healthy immune systems, but can be devastating to those whose immune systems have been weakened, for example by HIV, cancer chemotherapy, or drugs used to facilitate organ transplants. Classic examples of such diseases include cryptococcal meningitis, cytomegalovirus (CMV) disease, Mycobacterium avium complex (MAC), Pneumocystis carinii pneumonia (PCP), and toxoplasmosis. Malignancies such as Kaposi's sarcoma (KS) and non-Hodgkin's lymphoma are also often included.

In some cases, the term is also used to describe diseases such as bacterial pneumonia, anal and cervical cancer, herpes simplex infection, and tuberculosis that occur in both HIV negative and HIV positive people, but that tend to be more frequent, more severe, and more difficult to treat in people with HIV. A description of some common opportunistic conditions in people with HIV/AIDS can be found at the end of this article.

Prophylaxis is the use of drugs to prevent the occurrence (primary prophylaxis) or recurrence (secondary prophylaxis or maintenance therapy) of a disease. OI prophylaxis was in part responsible for decreases in HIV disease progression and AIDS-related deaths in the late 1980s and the first half of the 1990s.

In the past couple of years, the widespread use of highly active antiretroviral therapy (HAART) has dramatically reduced OI incidence and progression, and has led many people with HIV and their physicians to reconsider traditional thinking about OI prophylaxis. It remains to be seen whether long-term suppression of HIV will allow the immune systems of people with HIV to recover enough to keep OIs in check indefinitely.

HAART and the Decline in OIs

For many people with HIV, HAART -- combination antiretroviral regimens of three or more drugs, generally including a protease inhibitor -- has led to decreases in viral load and increases in CD4 T-cells. These T-cells are the mainstay of the immune system's defense against pathogens, and higher CD4 cell counts are associated with a decreased risk of OIs. With the widespread use of HAART beginning in 1996, the overall incidence of OIs has declined. In addition to contributing to the decline in OI incidence, HAART has also led to improvements in people who had previously failed to respond to OI treatment. Many experts believe that HAART -- rather than specific OI prophylaxis regimens -- is now the most important intervention to prevent these illnesses.

The first reports of decreases in OI rates due to HAART – along with declines in hospitalization and AIDS-related deaths – were presented at the 4th Conference on Retroviruses and Opportunistic Infections (CROI) in January 1997. In a 1997 Treatment Issues survey of AIDS clinicians, all 37 responding physicians reported that they had seen declines in OIs, hospitalizations, and AIDS-related deaths among their patients.

Since early 1997, newer reports have continued to confirm and extend the earlier findings. As reported by Judith Currier, MD, of the University of Southern California at the 5th CROI in February 1998, the results of AIDS Clinical Trials Group (ACTG) study 320, a clinical trial that compared AZT/3TC/indinavir to AZT/3TC alone, showed that people with advanced HIV disease receiving the three-drug regimen had half as many AIDS-related illnesses as those taking the two-drug regimen. The largest decreases were seen in those with the lowest CD4 cell counts (under 50 cells/mm3). According to more recent follow-up data from this cohort, no participants whose CD4 cell counts had risen above 200 cells/mm3 on HAART experienced any cases of CMV, MAC, or PCP. Other researchers reported OI decreases among HIV positive people in New Orleans and Baltimore. According to R. Moore of Johns Hopkins University, HAART was associated with a 60% reduction in the risk of contracting an OI, and no participants in the Baltimore cohort whose CD4 counts rose above 200 cells/mm3 on HAART developed any such illnesses.

At the 12th World AIDS Conference in Geneva in June 1998, Matthias Egger, MD, with the Swiss HIV Cohort Study reported on two large observational studies looking at OI incidence in participants receiving HAART. His group found that the occurrence of new OIs fell from 22 per 100 person-years in 1992 to five per 100 person-years in 1997-98, a decrease of 86%. The researchers saw declines in incidence of 80-90% for all OIs except non-Hodgkin's lymphoma, which decreased by 50%.

Declines in Specific OIs

At the 1998 CROI, researchers from the University of California at San Francisco (UCSF) reported that at San Francisco General Hospital (SFGH) between 1994 and 1997, incidence of cryptococcal meningitis fell by 63%, PCP incidence decreased by 71.4%, MAC incidence fell by 83.5%, and CMV incidence declined by 93.8% (see Table 1, below). By 1997, nearly three-quarters of those being treated for HIV disease at SFGH were taking combination therapy that included a protease inhibitor.

Table 1: OI Cases at San Francisco General Hospital, 1994-1997

Opportunistic infection 1994 1995 1996 1997
Cryptococcal meningitis 27 33 23 10
CMV 48 46 30 3
MAC 165 142 76 27
PCP 224 163 119 64

Some of the most dramatic results have been seen in the case of CMV disease. Several researchers at various conferences have reported that the incidence of CMV disease has decreased in people taking HAART, and that HAART has halted disease progression in people with existing CMV retinitis. In a study reported in the April 1998 issue of AIDS, John Walsh, MD, and colleagues at Chelsea and Westminster Hospital in London reported that HAART led to significant increases in survival time in people with CMV retinitis. Among 147 patients, the median survival time increased from eight and a half months pre-HAART to nearly two years with HAART.

Rates of PCP have also decreased dramatically. At the 1998 CROI, L. Huang and colleagues from UCSF reported on a decrease in PCP cases at SFGH from a peak of 290 cases in 1992 to 63 cases in 1997. Among the five people who developed PCP with a CD4 count above 200 cells/mm3, none had received antiretroviral therapy. The decline was attributed to a combination of PCP prophylaxis and HAART.

Various researchers have also reported declines in incidence and resolution of symptoms of cryptosporidiosis, KS, microsporidiosis, molluscum contagiosum, and toxoplasmosis.

Researchers have reported conflicting results regarding the effect of HAART on progressive multifocal leukoencephalopathy (PML). At the 38th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC) in September 1998, A. De Luca of Catholic University in Rome reported that in a cohort of 35 people with PML, those on HAART did no better than those not taking potent antiretroviral regimens. In contrast, B. Elliott, M.D. Mileno, and K. Henry all have reported that HAART induced prolonged remission of PML, with shrinking brain lesions and improved cognitive function. Studies have also been less clear about whether HAART reduces rates of candidiasis and non-Hodgkin's lymphoma, with different studies giving contradictory data.

Immune System Recovery

The basis for the decrease in OIs in people taking HAART is the restoration of immune system function, including an increase in CD4 T-cells. CD4 helper T-cells direct the body's defenses against invaders. These cells produce a variety of cytokines (chemical messengers) that orchestrate the activity of various other types of immune cells. These cells include neutrophils, monocytes, and macrophages, which engulf and destroy pathogens, and B-cells, which give rise to antibodies. As HIV disease progresses and the number of CD4 cells falls, the body is less able to fight invading organisms such as bacteria, fungi, parasites, and viruses.

Effective HAART reduces – or in the best case, halts – HIV replication. Once the virus is suppressed, the immune system can begin to recover. Studies have shown that HAART leads to the restoration of lymph node architecture and the reappearance of CD4 T-cells and B-cells. C. Mastroianni and colleagues of La Sapienza University in Rome found that HAART leads to increased neutrophil and macrophage function in people with advanced HIV, and to significant improvements in anti-fungal activity, chemotaxis (response to cytokines), and oxidative metabolism. This improved immune function in turn reduces the risk of OI incidence, progression, and recurrence.

While it makes sense that higher CD4 cell counts would be associated with a lower risk of OIs, the connection between viral load levels and OI risk is not as well understood. Studies have shown that decreases in viral load and increases in CD4 cell counts are independently associated with reduced OI incidence. Reduced viral load predicts decreased OI risk even if CD4 cell counts remain low -- especially in people with fewer than 200 CD4 cells/mm3 -- and increases in CD4 cell count predict reduced OI risk even if viral load remains detectable. Several studies have shown that some people who experience an increase in viral load while on HAART maintain a high CD4 count that protects them against OIs; thus, HAART "failure" does not necessarily imply lack of clinical benefit. According to Currier, studies suggest that "the degree of viral suppression needed to prevent clinical events over the short term may be less than what is needed for durable antiretroviral response."

Naive and Memory Cells

There are two types of CD4 T-cells that play a role in immune defenses: naive cells and memory cells. Naive CD4 cells are immune system cells that are not yet committed to fighting a specific pathogen. All T-cells start out naive when they are produced in the bone marrow. From there, they migrate to the thymus (an organ in the upper chest), where they learn to recognize specific invaders. Each T-cell becomes specialized to respond to a single pathogen. When a T-cell recognizes its specific pathogen, it becomes active and proliferates, creating many more cells that are active against the same invader. Once the immune response has run its course, some of these specialized T-cells remain as memory cells. These "primed" CD4 cells can trigger an immediate immune response if the same pathogen is encountered again.

When CD4 cell counts increase in response to HAART, whether the new cells are naive or memory cells is an important consideration. As untreated HIV disease progresses, the CD4 cell count decreases. Since each memory CD4 cell responds to only one specific pathogen, it is possible that all the cells that recognize a given invader – say CMV – may be lost. When this happens, the body cannot mount an immune response against that pathogen. If the increase in CD4 cells brought about by HAART is simply a multiplication of existing types of CD4 cells (clonal replication), the immune system may still be lacking defenses against certain invaders. The immune system needs new naive CD4 cells in order to defend against pathogens for which all the specific memory CD4 cells have been lost, or against new pathogens that the person has never before encountered.

When effective HAART is initiated, immune restoration occurs gradually over time. Researchers have shown that there are two phases of CD4 cell recovery. The first phase – characterized by a rapid increase in blood CD4 cell counts -- results from the mobilization and redistribution of existing memory CD4 cells from the lymph nodes to the bloodstream; this occurs during the first six months or so of immune recovery. The second phase occurs later and more slowly, and involves the production of new naive CD4 cells.

Thus, a person who has some anti-MAC memory CD4 cells but no remaining anti-CMV memory CD4 cells in his or her blood when starting HAART may end up with many more anti-MAC cells, but still no anti-CMV cells during the first phase of immune recovery. However, some of the new naive CD4 cells produced during the second phase of recovery may become specialized to recognize CMV.

In order to know whether a person's immune system can defend against a given pathogen, it is necessary to determine whether the person has CD4 cells that recognize and respond to that invader. Fred Valentine, MD, of New York University Medical Center, has described a lymphocyte proliferation assay (LPA) that can be used to determine immune response to a specific organism. Lymphocytes (T-cells and B-cells) are placed in a tissue culture with antigens from a pathogen. If the cells recognize the organism, they multiply. If no cells recognize the pathogen, no proliferation occurs. Several researchers have shown that CD4 cell proliferation responses increase when a person starts HAART. For example, K. Komanduri and colleagues noted a restoration of CMV-specific CD4 cell response after potent anti-HIV therapy was initiated.

LPAs are currently available to detect responses to cytomegalovirus, Mycobacterium avium, Pneumocystis carinii, and Toxoplasma gondi, but not for several other pathogens. The tests used today are research tests, and are not yet widely available commercially.

Can People on HAART Stop Prophylaxis?

Not surprisingly, news of decreasing OI incidence in people on HAART and reports of immune system recovery have led many people with HIV to wonder whether they can stop taking OI prophylaxis.

There are certainly many benefits to stopping prophylaxis. Preventive drugs add considerably to a person's "pill burden," which may make adherence more difficult. In addition, many OI prophylaxis drugs have adverse side effects and drug interactions, especially with the protease inhibitors. Also, the widespread, long-term use of prophylaxis has led to the development of drug-resistant organisms (for example, fluconazole-resistant Candida and acyclovir-resistant herpes simplex virus). Yet these potential benefits cannot justify stopping prophylaxis if the risk of getting an OI remains high.

Historically, decisions about when to start OI prophylaxis have been based on CD4 counts. For example, PCP prophylaxis has been recommended for people with fewer than 200 CD4 cells/mm3 and MAC prophylaxis has been recommended for those with fewer than 50-75 cells/mm3. Is it therefore safe to stop prophylaxis when CD4 counts rise above these thresholds due to HAART?

Research to date has not yet provided a definitive answer. Although it is clear that HAART has reduced OI incidence in certain populations, whether it will do so for a specific individual is not known. Richard Chaisson, MD, of Johns Hopkins University School of Medicine notes that in patients responding to HAART with increased CD4 cell counts, the risk of OIs "becomes very small." However, in order to know whether it is safe for an individual to stop prophylaxis for a given OI, it is necessary to determine if that person has CD4 cells that can respond to the specific pathogen in question.

Several studies have shown that some people who are responding well to HAART and have stopped taking preventive or suppressive drugs remain free of OIs. However, the 1997 guidelines from the U.S. Public Health Service/Infectious Disease Society of America (USPHS/IDSA) continue to recommend that decisions about prophylaxis be based on a person's lowest, or nadir, CD4 count (see Table 2, below).

Table 2: USPHS/IDSA Guidelines for the Prevention of Selected OIs in HIV Positive Adults

OI Indication* Primary** Maintenance**
Candidiasis routine primary px not generally recommended, nadir <50 cells/mm3 fluconazole, itraconazole fluconazole, ketoconazole, itraconazole
routine primary px not generally recommended, nadir <50 cells/mm3 fluconazole, itraconazole fluconazole, amphotericin B, itraconazole
Cryptococcosis nadir <50 cells/mm3, many do not recommend routine primary px fluconazole, itraconazole fluconazole, amphotericin B, itraconazole
CMV nadir <50 cells/mm3, CMV antibody positive oral ganciclovir optional oral or IV ganciclovir, ganciclovir eye implant, IV foscarnet, IV cidofovir
Histoplasmosis nadir <100 cells/mm3, endemic area itraconazole itraconazole, amphotericin B, fluconazole
MAC nadir <50 cells/mm3 clarithromycin or azithromycin, rifabutin clarithromycin or azithromycin + ethambutol and/or rifabutin
PCP nadir <200 cells/mm3, unexplained fever, history of thrush TMP-SMX, dapsone, dapsone +pyramethamine, aerosolized pentamidine via Respirgard nebulizer same regimens
Toxoplasmosis nadir <100 cells/mm3, Toxoplasma antibody positive TMP-SMX, dapsone + pyramethamine pyramethamine + sulfadiazine or clindamycin
Tuberculosis positive TB skin test, contact with person with active TB 12*** months isoniazid, pyrazinamide + rifampin, rifabutin not necessary

* The nadir is a person's lowest ever CD4 count; recommendations apply if a person has ever had a CD4 cell count below the threshold

** Preferred regimens are listed first

*** The most recent guidelines for people co-infected with HIV and TB recommend nine months of isoniazid preventive therapy

Px = prophylaxis

Adapted from 1997 USPHS/IDSA guidelines for the prevention of opportunistic infections in persons infected with human immunodeficiency virus. MMWR June 27, 1997 / 46(RR12);1-46.

Community physicians who treat HIV/AIDS have varying philosophies about whether and when to discontinue OI prophylaxis. The 1997 Treatment Issues survey of clinicians found that a majority (20 of 37 respondents) believe that data is not yet sufficient to justify stopping OI prophylaxis. However, several physicians noted that they had discontinued prophylaxis for CMV, MAC, and PCP on an individualized basis. According to Chaisson, there is growing evidence that stopping preventive therapy is a "reasonable maneuver" if a person's CD4 cell count after starting HAART is at least 50% above the traditional thresholds for prophylaxis, even if the person does not have an undetectable HIV viral load.

Some physicians prefer to wait until a person's CD4 cell count has remained elevated and HIV viral load has remained suppressed for three to six months before they consider stopping prophylaxis. For example, Currier states that, "emerging data suggests that the lowest-risk patients are those with no prior history of an opportunistic infection who have been responding to antiretroviral therapy for more than four months with stable to increasing CD4 counts and sustained HIV viral load levels below 500 copies/mL." This delay is supported by a study by Egger, who reported at the Geneva AIDS conference that for most OIs, incidence decreased immediately after starting HAART, but for MAC and CMV, three months elapsed before a decline in incidence was seen. Clinicians may also test blood or tissue levels of pathogens (for example, CMV viral load) to help determine which individuals may be at most at risk of developing OIs.

It is important to distinguish between primary prophylaxis, used to prevent a first case of an OI, and secondary prophylaxis (suppressive or maintenance therapy), used to prevent relapse, recurrence, or disease progression in a person who has already experienced an OI. Physicians tend to be less willing to consider discontinuing maintenance therapy. People with HIV who do discontinue either primary or secondary prophylaxis should be monitored regularly, and anti-OI treatment should be resumed at the first sign of symptoms

Studies of Prophylaxis Discontinuation

With each major AIDS conference, more evidence accumulates to indicate that it may be safe for some people taking HAART to stop primary or secondary prophylaxis. The most data exist for CMV, MAC, and PCP.

Various research teams have studied participants who stopped taking CMV maintenance therapy after responding well to HAART. Cristina Tural and colleagues of Badalona, Spain reported in the Journal of Infectious Diseases that after three months on HAART, CMV incidence decreased in people whose CD4 counts had once been below 50 cells/mm3, but had risen due to antiretroviral therapy. The group concluded that for selected patients, discontinuation of CMV maintenance therapy may not result in retinitis progression. At the 1998 CROI, W. Freeman of the University of California at San Diego (UCSD) reported on eight HIV positive persons on HAART who stopped CMV maintenance therapy. After a median of five months, only one had a relapse of CMV retinitis; this person also experienced an increase in viral load and a decrease in CD4 count, suggesting HAART failure.

At the 1998 ICAAC, J. Deayton and colleagues of the Royal Free Hospital and School of Medicine in London reported that 16 patients not taking CMV maintenance therapy had no evidence of CMV in their blood after a median of 13.5 weeks on HAART. All but two were still CMV negative after a median eight months of follow-up; the two that did have detectable CMV in their blood also experienced an increase in HIV viral load. The researchers suggested that CMV relapse may be an early sign of HAART failure. Based on the results of a study reported at the same conference, T. Chiller and colleagues from the University of Texas concluded that persons taking HAART who have more than 120 CD4 cells/mm3 and an undetectable viral load "may be candidates for discontinuation of CMV prophylaxis."

Some participants in the various studies did experience relapse, progression, or new cases of CMV retinitis even while taking HAART. However, these cases tended to occur within two to three months after starting combination antiretroviral therapy, suggesting that it may take some time before HAART improves immune system function enough to keep CMV in check.

In the pre-HAART era, lifelong maintenance therapy was required to prevent relapse of symptomatic MAC disease, and there were no reports of MAC eradication (elimination of MAC bacteria from the body). At the 1998 CROI, Judith Aberg and colleagues from UCSF reported on four patients who had previously had symptomatic MAC disease. All four stopped MAC prophylaxis after responding to HAART and achieving CD4 cell counts over 100 cells/mm3. After five to ten months of follow-up, none had MAC symptoms, and all had MAC-free blood and bone marrow cultures.

Studies have also shown promising results of stopping PCP prophylaxis in people responding to HAART. As reported at the Geneva AIDS conference by H. Furrer and colleagues with the Swiss HIV Cohort Study, 53 participants on HAART discontinued PCP prophylaxis after achieving CD4 cell counts over 200 cells/mm3 on at least two tests. After an average five months of follow-up, none developed PCP.

Similarly, at the 1998 ICAAC, M. Schneider and colleagues of the University Hospital in Utrecht presented the results of a study of 73 HAART responders whose CD4 counts had risen above 200 cells/mm3 on at least two consecutive tests; 60 stopped primary prophylaxis and 13 discontinued maintenance therapy. Again, no cases of PCP occurred after a median follow-up period of 12 months. Finally, B. Yangco and colleagues of the Infectious Disease Research Institute in Tampa, FL, reported on a retrospective study of participants in the APACHE cohort. Twenty-one people stopped PCP prophylaxis after responding to HAART and achieving CD4 counts above 200 cells/mm3; the median CD4 nadir was 105 cells/mm3. After a median of 13 months of follow-up, no cases of PCP occurred.

Despite the promising results seen so far, large prospective clinical trials are needed to definitively demonstrate that discontinuation of OI prophylaxis is safe. Several such studies are currently underway.

Changing Clinical Presentation of OIs

Along with reports of decreased OI incidence in people taking HAART, several physicians and researchers have also observed that some people using potent antiretroviral therapy have experienced unusual OI symptoms. In some cases, it appears that effective anti-HIV treatment actually worsens certain OI manifestations. Some physicians have seen unusual eye inflammation in people with CMV, while others have noted lymph node swelling associated with MAC. Martin French reported at the Geneva AIDS conference that 33 of 132 patients he studied had atypical inflammatory symptoms associated with CMV, MAC, herpes simplex virus, molluscum contagiosum, or shingles.

According to Francesca Torriani of UCSD, nine of 13 patients on HAART whose CMV maintenance therapy was stopped experienced vitritis, an inflammation of the eye which was in some cases associated with reduced visual acuity. This occurred most often in people with higher CD4 counts, suggesting a stronger anti-CMV immune response. Similar findings of vitritis and macular edema (swelling) were reported by Deayton at the 1998 ICAAC. Writing in the February 1998 issue of Archives of Ophthalmology, Marietta Karavellas, MD, also with the UCSD group, termed this phenomenon "immune recovery vitritis," and noted that no cases of similar symptoms were seen in the pre-HAART era.

E. Race of Harvard Medical School reported in the January 24, 1998 issue of The Lancet on five cases of MAC-related lymphadenitis (lymph node inflammation) and fever within one to three weeks after starting HAART. He suggested that such reactions may be due to "significant numbers of functionally competent immune cells becoming available to respond to heavy mycobacterial burden." Some experts have termed these symptoms "MAC reversal syndrome." At the 1998 ICAAC, P. Phillips of St. Paul's Hospital in Vancouver reported similar symptoms, and noted that they may represent a local immune response to MAC before the infection becomes disseminated through the bloodstream. Mark Kaplan, of North Shore University Hospital in Manhasset, NY, compared these symptoms to the reversal reactions sometimes seen when treatment is started for leprosy, a disease caused by a related mycobacterium.

Although these symptoms may seem alarming, in reality they appear to be the result of improved immune function. As HAART strengthens the immune system, it may enable the body to mount an inflammatory response that was not previously possible. Such reversal reactions do not indicate antiretroviral failure, and should not lead to the discontinuation of HAART. According to Chaisson, such reactions are "a reassuring indication of HAART-mediated immune recovery." The reactions typically occur shortly after HAART is initiated (usually within one to two months) and resolve as antiretroviral therapy continues. In some cases, anti-inflammatory drugs can help relieve symptoms.

A Shifting Consensus

Although there have been many reported cases in which people who have responded well to HAART have remained free of OIs when they discontinue preventive or suppressive therapy, it is still too soon to make a blanket recommendation that people stop OI prophylaxis or maintenance therapy when their CD4 cell counts rise and their viral load levels fall.

People with HIV are still getting OIs, and some people are still dying from them. People in much of world do not have access to HAART, and many people in the U.S. have poor access to health care and are not receiving state-of-the-art anti-HIV treatment and OI prophylaxis. In addition, HAART does not work for everyone; many people experience antiretroviral failure, intolerable side effects, or the development of drug-resistant HIV.

Today's climate is characterized by cautious optimism. While many community physicians are experimenting with stopping OI prophylaxis, the general consensus seems to be that many patients should continue their prophylactic regimens. This is reflected in the current USPHS/IDSA guidelines, which recommend initiating or continuing OI prophylaxis on the basis of a person's lowest, or nadir, CD4 cell count. However, it is clear that the consensus is shifting. According to Chaisson, "while it is not yet standard practice to discontinue prophylaxis when CD4 T-cells levels rise, a growing body of evidence suggests that this would be appropriate."

Although HAART is clearly successful in reducing OI incidence in the population as a whole, its effect on a given individual is unknown. To underscore the potential danger in prematurely stopping OI prophylaxis, Valentine has reported on a person taking HAART who had a CD4 count that had risen to 440 cells/mm3. LPA tests showed that he had CD4 cells that could respond to cytomegalovirus, Mycobacterium avium, and other pathogens, but not to Pneumocystis carinii, and the person did in fact develop a mild case of PCP.

Immune restoration takes place slowly, and it is not yet known whether immune system recovery sufficient to prevent OIs will occur in all people. It is likely that in the near future, tests that reveal CD4 cell function and gauge immune system recovery will become more widely available to help guide clinical decision-making. But as Currier notes, "for now, patient history and changes in both viral load and CD4 levels are the best available markers" of OI risk. As yet, there are no agreed-upon new CD4 thresholds for stopping primary prophylaxis, nor is there a well-defined idea of when maintenance therapy can be discontinued and an OI can be considered "cured."

For now, the best practice seems to be careful individualization of prophylactic regimens. It is important that HIV positive people and their physicians be aware of and discuss the potential risks and benefits of stopping prophylactic therapy. For example, a case of CMV retinitis could lead to permanent vision loss, and it has been shown that some OIs tend to increase HIV viral load. Some prophylaxis regimens (for example, TMP-SMX to prevent PCP and toxoplasmosis) are inexpensive and have a minimal effect on quality of life; the small cost may be worth the peace of mind of knowing that one is protected. On the other hand, having to take a few less pills may have a considerable positive effect on a person's ability to adhere to a complex combination antiretroviral regimen.

People with HIV are likely to feel a combination of both fear of a potential OI, and relief at the a prospect of reducing the number of medications they must take. HIV positive people and their healthcare providers must balance the benefits -- fewer pills, fewer side effects, and fewer drug interactions -- with the very real risks of OI incidence and progression.

Appendix: Opportunistic Conditions in People with HIV/AIDS

Anal and cervical cancer --Anal and cervical cancer affect both HIV positive and HIV negative people, but tend to progress more rapidly and recur more often in people with HIV. Anal and cervical cancer are associated with certain strains (e.g., 16, 18, 31) of the human papillomavirus (HPV). The earliest stage of anal or cervical cancer is dysplasia, or abnormal cell proliferation and differentiation. Later stages include intraepithelial neoplasia and squamous intraepithelial lesions. Early stages are often asymptomatic; later stages may be characterized by pain, bleeding, and discharge. People with HIV should receive regular (every 6-12 months) Pap smears, a test that can detect precancerous cells at an early stage. Follow up tests include colposcopy and tissue biopsy. Early stages may be treated by freezing, burning, or surgically removing the abnormal tissue. Once the cancer has spread, treatment may involve chemotherapy and radiation.

Aspergillosis -- Aspergillus is a fungus found on decaying vegetable matter. It has been found in marijuana, so people who use medicinal cannabis should exercise caution. Aspergillosis occurs in people with late-stage HIV disease, typically those with low CD4 counts and neutropenia. It often causes a respiratory disease characterized by cough, shortness of breath, and fever. It may also affect the central nervous system (brain and spinal cord), leading to seizures and partial paralysis. Historically, the prognosis of people with aspergillosis has been poor. See "Treatment for HIV-Related Fungal Infections," BETA, June 1995 (not on SFAF website).

Bacterial respiratory diseases -- With the widespread use of PCP prophylaxis, an increasing proportion of pneumonia in people with HIV is due to bacteria including Pneumococcus and Streptococcus species, Haemophilus influenzae, Staphylococcus aureus, and Pseudomonas aeruginosa. Drug-resistant strains of Streptococcus pneumoniae are increasingly common. Symptoms may include cough (typically productive), shortness of breath, chest pain, and fever. Some bacterial infections may become disseminated, or spread throughout the body in the bloodstream.

Bartonella infection -- People with HIV may be susceptible to infection with Bartonella species of bacteria. Such an infection is sometimes called cat scratch fever, and is associated cat flea bites. HIV positive people should be cautious around cats and keep fleas under control. The disease is characterized by fever, a rash of pus-filled blisters, and swollen lymph nodes.

Candidiasis -- Candidiasis is caused by the fungus Candida albicans or a related species, common pathogens that a healthy immune system usually can keep under control. Candidiasis is often the first sign of immune system decline. The condition may recur more often and be harder to treat as CD4 cell count decreases. Candidiasis affects the skin, nails, and mucous membranes, most commonly in the mouth (thrush), throat (esophageal candidiasis), and vagina ("yeast infection"). Rarely candidiasis may become disseminated and spread throughout the body. Thrush is characterized by red or white patches in the mouth. Vaginal candidiasis is characterized by thick white discharge accompanied by itching and burning. Symptoms of esophageal candidiasis include heartburn, pain, and difficulty swallowing. Mild candidiasis of the mouth and vagina may be treated with topical medications applied to the affected area; topical treatment is preferred over systemic (whole body, e.g., oral) treatment when possible to reduce the risk of developing drug-resistant Candida. Esophageal candidiasis and candidiasis that does not respond to topical medication may require systemic treatment. See "Vaginal Candidiasis and Other Types of Vaginitis," BETA, September 1995 (not on SFAF website).

Coccidioidomycosis -- Also known as valley fever, coccidioidomycosis is caused by the fungal pathogen Coccidioides immitis. In the U.S., it is most prevalent in the southwest. The organism usually only causes disease in people with suppressed immune systems, especially those with fewer than 50 CD4 cells/mm3. The pathogen is typically inhaled, leading to a lung infection. Symptoms include fever, fatigue, weight loss, and cough. Coccidioidomycosis may also affect the membranes surrounding the brain (meningitis), and may become disseminated and spread throughout the body.

Cryptococcosis -- Cryptococcosis is caused by a yeast-like fungus, Cryptococcus neoformans, found in soil and bird droppings. It is transmitted by inhalation of contaminated dust. Cryptococcosis usually occurs in people with fewer than 50 cells/mm3. The most common manifestation is cryptococcal meningitis, an inflammation of the membranes surrounding the brain and spinal cord. Symptoms include fever, fatigue, nausea, malaise, headache, stiff neck, mental confusion, memory loss, vision and movement difficulties, and personality changes; if untreated, coma and death may result. Cryptococcosis may also affect other parts of the body including the skin, lungs (cryptococcal pneumonia) and brain (encephalitis); the disease may become disseminated. Cryptococcosis is diagnosed by blood and cerebrospinal fluid (CSF) cultures, CSF pressure measurement, and CT or MRI scans. Traditionally, lifelong maintenance therapy has been recommended to prevent relapse.

Cryptosporidiosis -- Cryptosporidium parvum is a protozoan parasite found in water or soil contaminated by feces. It is not killed by water treatments such as chlorine. In 1993 in Milwaukee, WI, 100 people died following an outbreak of cryptosporidiosis in the city's water supply. Use of water filters or boiled water can help prevent infection. The disease is also transmitted through oral/anal sex. The disease tends to last longer as immunosuppression worsens. Cryptosporidiosis primarily affects the gastrointestinal tract. Symptoms include watery diarrhea, nausea, vomiting, abdominal cramps, bloating, fatigue, and low-grade fever; prolonged diarrhea may lead to weight loss. Diagnosis may be difficult because many other conditions have similar symptoms. Cryptosporidiosis can be difficult to treat; in addition to drugs that fight the parasite, other medications may be used to relieve symptoms. See "Avoiding Cryptosporidium: Safe Food, Safe Water, Safe Sex," BETA, March 1996 (not on SFAF website); "Cryptosporidium and Other Environmental Pathogens," BETA, December 1994 (not on SFAF website).

Cytomegalovirus (CMV) disease -- CMV is a herpesvirus that infects most people early in life. The virus remains latent in the body and usually only causes disease in people with weakened immune systems. CMV disease occurs most often in people with fewer than 50 cells/mm3. CMV may cause retinitis, an inflammation of the retina, leading to blurred vision, floaters and, ultimately, blindness. People with low CD4 cell counts should have regular eye exams and vision screenings. CMV may also affect the gastrointestinal tract (CMV colitis, CMV esophagitis), lungs (CMV pneumonia), brain (CMV encephalitis), and spinal cord (polyradiculopathy). CMV disease is difficult to treat. The older systemic drugs cause severe side effects and require a permanent catheter for intravenous infusion. More recently, oral drugs and eye implants have become available. Traditionally, lifelong maintenance therapy has been recommended for people who have had an episode of CMV disease. See "Advances in Treatment for Cytomegalovirus Retinitis," BETA, September 1995 (not on SFAF website).

Gastrointestinal bacterial infections -- Many pathogens that affect both HIV negative and HIV positive people -- such as Salmonella, Shigella, Campylobacter, and Listeria species -- may cause more severe disease and be harder to treat in people with HIV. Bacterial pathogens may cause nausea, diarrhea, and abdominal cramps, and may be difficult to diagnose because symptoms are similar to those of various other diseases. The best way to avoid infection is to take precautions when handling and preparing food.

Herpes Simplex Virus (HSV) -- HSV is common in HIV negative and HIV positive people, but may be more severe, more frequent, and harder to treat in people with HIV. In addition, HSV may facilitate HIV transmission. HSV-1 typically causes lesions around the mouth ("cold sores"), while HSV-2 typically causes genital or anal lesions; both may be sexually transmitted. HSV may also affect the esophagus, brain (HSV encephalitis or meningitis), and lungs (HSV pneumonia). HSV causes fluid-filled blisters which may be accompanied by itching, burning, redness, and discharge. The lesions scab over and heal, but the virus remains latent in the nerves and may reactivate due to various factors such as stress, fever, or hormonal fluctuations. Reactivation may be signaled by a prodrome (preceding symptoms) of tingling, burning, or itching. HSV may be treated with topical medications or systemic drugs. See "Herpesviruses," BETA, December 1995 (not on SFAF website).

Histoplasmosis -- Histoplasmosis is infection with the fungus Histoplasma capsulatum, which lives in soil and bird droppings. In the U.S., it is most common in the midwest. The disease occurs in both HIV positive people (usually those with fewer than 100 CD4 cells/mm3) and HIV negative people (especially children and the elderly). People become infected by breathing airborne spores, and the disease most often affects the lungs. It may also affect the skin, gastrointestinal tract, and central nervous system. Symptoms include fever, weight loss, fatigue, dry cough, shortness of breath, skin rash, and enlarged lymph nodes. Histoplasmosis is diagnosed by lung fluid culture, blood or urine tests, or tissue biopsy. It is treated with antifungal medications.

Isosporiasis -- Isosporiasis is caused by the parasite Isospora belli, which is present in contaminated food and water. It is spread by the fecal-oral route, and may be transmitted by oral/anal sex. It primarily infects the gastrointestinal tract and gall bladder, with symptoms including diarrhea, abdominal cramps, bloating, fever, and weight loss. Symptoms resemble those of many other diseases. Diagnosis is made by stool sample evaluation or endoscopy (use of a lighted tube to view the gastrointestinal tract). Severe cases may require surgery to remove the gall bladder.

Kaposi's sarcoma (KS) -- In the early 1980s, KS was one of first OIs that signaled AIDS. AIDS-related KS is more aggressive than the older form seen previously in elderly Mediterranean and Eastern European men. KS incidence is higher in gay men than in other populations with HIV, suggesting that it is sexually transmitted. KS was recently discovered to be associated with -- and likely caused by -- a herpesvirus called KSHV or HHV-8. Cytokines (chemical messengers) and growth factors are believed to play a role in the development of the disease. KS involves abnormal growth of the linings of the blood and lymph vessels. It is characterized by purplish or brownish lesions on the skin or mucous membranes, and may also affect internal organs such as the lungs, liver, gastrointestinal tract, and lymph nodes. Many different types of treatment have been tried for KS, and there is no definitive cure. Therapies include freezing and radiation of the lesions, as well as a wide range of chemotherapy options and immune modulators. See "Kaposi's Sarcoma," BETA, March 1996 (not on SFAF website).

Lymphoma -- Lymphoma is a cancer of the lymphatic system. Non-Hodgkin's or B-cell lymphomas are most often associated with HIV disease, and are characterized by uncontrolled multiplication of certain immune system white blood cells. The disease usually affects people with fewer than 75 CD4 cells/mm3. The cause of the disease is not know, but it appears to be associated with the Epstein-Barr virus. Lymphoma may affect the lymph nodes, gastrointestinal tract, liver, and bone marrow. Symptoms may include swollen lymph nodes, fever, fatigue, weight loss, pain, and enlarged spleen and liver. Primary central nervous system lymphoma affects the brain and spinal cord, with symptoms that may include confusion, headache, seizures, lethargy, weakness, memory loss, speech and vision difficulties, and personality changes. Diagnostic methods include CT and MRI scans and brain biopsy. Treatment typically involves chemotherapy and radiation; however, there is no cure, and lymphoma is often fatal.

Microsporidiosis -- Microsporidiosis is caused by a group of parasites found in contaminated water. It may also be transmitted by oral/anal sex. Both HIV negative and HIV positive people may be infected, but the disease tends to be more severe and harder to treat in people with HIV. Microsporidiosis typically affects the gastrointestinal tract, with symptoms including diarrhea, nausea, bloating, abdominal pain, rapid weight loss, and dehydration. The disease may also affect the eye, liver, kidneys, and brain. Because these symptoms resemble those of other diseases, microsporidiosis may be hard to diagnose. Diagnostic methods include stool sample examination, endoscopy, and biopsy.

Molluscum contagiosum -- Molluscum is a skin condition caused by a pox virus. It becomes more common and recurs more frequently as immune function decreases. The disease is characterized by raised, flesh-colored lesions with a central depression. Lesions occur most often on the face and neck, but also on the arms, legs, and anogenital region. Treatment involves freezing or surgery.

Mycobacterium avium complex (MAC) disease -- MAC disease is caused by Mycobacterium avium or Mycobacterium intracellulare, two species of acid-fast bacteria found in contaminated water, food, and soil. HIV positive people with fewer than 50 CD4 cells/mm3 are most at risk. MAC bacteria enter the gastrointestinal tract or lungs. From there, the bacteria may enter the bloodstream and the disease may become disseminated and affect almost any organ in the body. Symptoms include fever, fatigue, weight loss, night sweats, appetite loss, diarrhea, abdominal cramps, and enlarged lymph nodes. Diagnosis involves a blood culture or biopsy. Like the related Mycobacterium tuberculosis, MAC bacteria rapidly become resistant to a single drug, so treatment requires a combination of at least two drugs. Traditionally, lifelong maintenance therapy has been recommended to prevent relapse. Other species of atypical mycobacteria, including Mycobacterium kansasii and Mycobacterium xenopi may also cause disease in people with HIV.

Progressive multifocal leukoencephalopathy (PML) -- PML is caused by the JC papovavirus, which is common in healthy adults but usually only causes disease in people with suppressed immune systems. PML usually occurs in people with advanced AIDS. The disease is the result of the breakdown of the myelin that insulates nerve fibers in the central nervous system. PML causes lesions in the brain and spinal cord that worsen over time. Symptoms include weakness, disturbances of speech, motion, balance, and vision, memory loss, and changes in personality and mental status. Because symptoms are similar to those of other conditions, diagnosis may be difficult, and may involve CSF PCR assay, CT or MRI scans, and brain biopsy. There is no standard treatment for PML, although many therapies have been tried with varying degrees of success.

Pneumocystis carinii pneumonia (PCP) -- Several reports of the previously rare PCP in gay men in 1981 were the first harbingers of the disease that would later come to be known as AIDS. Early in the epidemic, PCP was the leading cause of AIDS-related death. The disease is caused by a pathogen called Pneumocystis carinii, which has characteristics of both a protozoan and a fungus. The organism infects most people early in life, but only causes disease if the immune system is weakened. PCP usually occurs in people with fewer than 200 CD4 cells/mm3. The organism usually infects the lungs, causing pneumonia. Symptoms include fever, dry cough, shortness of breath, fatigue, night sweats, and weight loss. The organism may also infect other organs including the lymph nodes, gastrointestinal tract, and liver. PCP is diagnosed by means of sputum culture, chest X-ray, or bronchoscopy with biopsy. PCP was one of the earliest prevention success stories, and there are currently several prophylaxis and treatment options.

Toxoplasmosis -- Toxoplasma gondii is a parasite found in cat feces, soil, and contaminated meat. People with HIV are advised to use caution when handling raw meat and cleaning cat litter boxes. Toxoplasmosis is most likely to occur in people with fewer than 100 CD4 cells/mm3. The disease may affect many organs, including the lungs, eye, heart, and gastrointestinal tract. However, the most common site is the brain (toxoplasmic encephalitis). Symptoms may include headache, fever, seizures, dizziness, nausea, disturbances in vision, speech, and motion, changes in mental status, and coma. Toxoplasmic encephalitis may be hard to diagnose due to its similarity to other brain diseases; diagnostic methods include blood tests, CT or MRI scans, and brain biopsy. Some of the drugs used to prevent PCP and MAC can also help prevent toxoplasmosis.

Tuberculosis (TB) -- Mycobacterium tuberculosis is an acid-fast bacteria that is spread through the air. TB infection is not the same as active TB disease. It is estimated that HIV positive people who are infected with TB bacteria are 100 times more likely to develop active TB disease than HIV negative people. TB is the most common cause of death of people with HIV/AIDS worldwide. People with HIV are more likely to have disseminated TB and to have extrapulmonary (outside the lungs) TB. The disease may affect the lymph nodes, bones, kidneys, and brain. Symptoms of TB include a productive cough, bloody sputum, fever, fatigue, and weight loss. TB infection can be detected by a skin test, and active TB disease is diagnosed by sputum cultures and chest X-rays. If a person has a positive skin test, active TB disease can be prevented with drugs. Active TB disease must be treated with a combination of drugs taken for at least six months. Drug resistance develops rapidly if treatment is not consistent and complete. Multidrug-resistant TB requires prolonged treatment with a larger number of drugs.

Varicella Zoster Virus Infection (Herpes Zoster) -- Herpes zoster is caused by the varicella-zoster virus (VZV). VZV causes chickenpox, a disease characterized by a rash of red, itchy blisters, usually in children. After an episode of chickenpox, the virus remains latent in the nerves and may reactivate later in life if the immune system becomes weakened. Reactivation leads to herpes zoster or shingles, characterized by a rash of painful fluid-filled blisters on the skin along a nerve pathway. Shingles is a common first sign of HIV disease progression. An outbreak may be preceded by a prodrome characterized by pain, itching, malaise, and fever. VZV may become disseminated and may affect the lungs, liver, eyes, and brain. An episode of shingles may be followed by post-herpetic neuralgia, or nerve pain. Herpes zoster may be treated with antiherpes drugs; it is harder to treat and may become chronic in people with HIV.

Liz Highleyman is Acting Editor of BETA.

Aberg, J.A. and others. Eradication of disseminated Mycobacterium avium complex (DMAC) in four patients after twelve months on mycobacterial therapy and response to highly active antriretroviral therapy (HAART). 5th Conference of Retroviruses and Opportunistic Infections. Chicago, February 1-5, 1998. Abstract 729.

Centers for Disease Control and Prevention. 1997 USPHS/IDSA guidelines for the prevention of opportunistic infections in persons infected with human immunodeficiency virus. MMWR June 27, 1997 / 46(RR12);1-46.

Chaisson, R.E. The changing natural history of HIV/AIDS in the HAART era: clinical implications. Report from the 8th Clinical Care Options for HIV Symposium. Clinical Care Options for HIV Online Journal 4(2) ( June 1998.

Currier, J.S. OI prophylaxis in antiretroviral responders: stay the course or bite the bullet? AIDS Clinical Care 10 (10): 73. October 1998.

Currier, J.S. Incidence rates and risk factors for opportunistic infections in a phase III trial comparing indinavir+ZDV+3TC to ZDV+3TC. 5th Conference of Retroviruses and Opportunistic Infections. Chicago, February 1-5, 1998. Abstract 257.

Deayton, J. Highly active antiretroviral therapy (HAART) including protease inhibitors can completely suppress asymptomatic CMV viremia in the absence of specific anti-CMV therapy. 38th Interscience Conference on Antiretroviral Agents and Chemotherapy. San Diego, September 24-27, 1998. Abstract 268.

Egger, M. Opportunistic infections in the era of HAART. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Abstract 76.

Falkenberg, J. Treatment Issues' second survey of physicians' treatment practices. Treatment Issues 12(1): 3-17. Winter 1997-98.

Freeman, W.R., and others. Ophthalmological manifestations of immune recovery in AIDS patients on HAART therapy. 5th Conference of Retroviruses and Opportunistic Infections. Chicago, February 1-5, 1998. Abstract 757.

Hengel, R.L. and others. Reconstitution of "naive" (CD45RA+CD62L+) CD4+ T-cells after potent anti-HIV-1 therapy. 5th Conference of Retroviruses and Opportunistic Infections. Chicago, February 1-5, 1998. Abstract 157.

Holtzer, C.D. and others. Decline in specific opportunistic infections at San Francisco General Hospital. 5th Conference of Retroviruses and Opportunistic Infections. Chicago, February 1-5, 1998. Abstract 183.

Moore, R.D. and others. Decline in CMV and other opportunistic disease with combination antiretroviral therapy. 5th Conference on Retroviruses and Opportunistic Infections. Chicago, February 1-5, 1998. Abstract 184.

Phillips, P. Mycobacterial lymphadenitis associated with the initiation of effective antiretroviral therapy. 38th Interscience Conference on Antiretroviral Agents and Chemotherapy. San Diego, September 24-27, 1998. Abstract I-264.

Schneider, M.M.E. and others. No Pneumocystis carinii pneumonia after discontinuation of prophylaxis in HIV-infected patients with a CD4+ count higher than 200 cells/mL induced by HAART. 38th Interscience Conference on Antiretroviral Agents and Chemotherapy. San Diego, September 24-27, 1998. Abstract I-269.

Tural C. and others. Long-lasting remission of cytomegalovirus retinitis without maintenance therapy in human immunodeficiency virus-infected patients. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Abstract 22244. Also Journal of Infectious Diseases 177(4):1080-1083.

Vullo, V. and others. Restoration of the innate immunity in HIV-infected patients after protease inhibitor therapy. 38th Interscience Conference on Antiretroviral Agents and Chemotherapy. San Diego, September 24-27, 1998. Abstract I-182.


Copyright © 1999 -BETA, Publisher. All rights reserved to the San Francisco AIDS Foundation. Reproduced by permission. Reproduction of this article (other than one copy for personal reference) must be cleared through BETA: PO Box 426182, San Francisco, CA 94142-6182. Tel: 415 487 8060 Fax: 415 487 8069 San Francisco AIDS Foundation, Mail SFAF..

Information in this article was accurate in January 10, 1999. The state of the art may have changed since the publication date. This material is designed to support, not replace, the relationship that exists between you and your doctor. Always discuss treatment options with a doctor who specializes in treating HIV.