Clinical aspects of HIV infection

Juan Sierra

Introduction.

In the last year there has been a drastic change in the knowledge of the clinical concepts related to HIV infection. It can be said that for the first time in the epidemic of HIV / AIDS the concept of a chronic, controllabe disease similar to other chronic diseases is being handled. For the first time also, the possibility of complete virologic remission has been discussed, not to mention the optimists that have even sketched the possibility of a virological cure. Although these concepts are, in my opinion, not immediately attainable as yet, the fact alone that they are mentioned in scientific forums means a turn in the concepts and previsions that existed only a few years ago. Indeed, the vision of the optimists must be taken with caution in light of the important limitations that have been recognized when the new therapeutic modalities are intended to be applied in real life outside the controlled situations of controlled clinical trials. The aspect of accessibility to the new technology and drugs has been the object of extensive discussions, since the achievement of a greater therapeutic efficiency invariably brings along an exponential increase in the cost of the therapeutic measure. As a consequence of this, the gap between patient expectations in developed countries (better each time ) and patient expectations in developing countries (practically the same as several years ago or slightly better) will increase substantially. The economic capacity will determine more than ever the clinical course which a patient will follow. The medical decisions will be based more frequently on economic aspects and not on scientific aspects. Though this polarization is not new in the AIDS epidemic, neither is it in general medicine, and it is important to mention it because now more than ever the economic capacity of the individual or the health system as a whole will condition big differences in the control and treatment of the disease. That is why the pharmacoeconomic aspect has received special attention in the evaluation of most of the new technologies and therapeutic interventions. In the discussion that follows, this aspect will be treated in relation to each one of the diagnostic interventions or therapies that are discussed. Most of the changes that have emerged in the last year are derived from new knowledge in four main clinical areas: a) HIV primary infection b) better development of methods to predict the course of infection and to evaluate the efficacy of antiviral drugs, c) improvement in the treatment and prevention of opportunistic infections, and d) development of new antiviral drugs. Indeed, the knowledge derived from the pathogenic mechanisms of HIV infection has been crucial to influence this change in the strategy of managing HIV infected patients. In this document, we will revise the clinical aspects, while pathogenic ones will be mentioned only in their relationship to the therapeutic aspects since they will be dealt with in another review. Also, the clinical aspects of HIV infection in women will be intentionally treated separately to emphasize the importance that this topic has attained, as a result of the little emphasis that this area usually receives.

Primary HIV infection

Review of knowledge previous to the International AIDS Conference in Vancouver (July of 1996).

In 53 to 93% of patients with recently-acquired HIV infection an acute disease occurs during the stage of seroconversion. This syndrome is clinically similar to influenza and has been named the acute retroviral syndrome. The appearance of this syndrome is independent from the route of transmission of HIV since it has been found in patients that acquired the virus by any route. The variable frequency of the syndrome seems to depend exclusively on the accessibility to health services, and on the experience that primary attention levels have when recognizing this syndrome. It is not unexpected, therefore, that the primary syndrome is recognized in a very high percentage (more than 90%) in developed countries and almost not recognized in Latin American countries. The time that elapses between exposure to HIV-1 and development of the acute retroviral syndrome is usually between 2 and 4 weeks.

The clinical manifestations of the syndrome are wide. The most frequent clinical picture is an influenza-like illness. Tindall and Cooper emphasize that the clinical picture must not necessarilly resemble mononucleosis, and, in fact, frequently it is only a nonspecific febrile illness with or without headache, rash or other manifestations(1). The signs and reported symptoms in various series are fever, lymphadenopathy, faringytis, rash , myalgias and arthralgias. During the primary event an acute meningoencephalitis has been described. Also, oral candidiasis, mucocutaneous ulcerations and loss of weight can occur. These clinical manifestations generally last 1 or 2 weeks, and spontaneously heal. Following this, the individual enters a clinical latency phase in which no symptoms exist although there is active viral replication mainly in the lymph nodes(2) . In rare occasions, severe depletion in the CD4 T cell count has been described as well as the occurrence of serious opportunistic infections such as Pneumocystis carinii pneumonia(3). The occurrence of the acute viral syndrome associated with HIV-1 seroconversion has been found to be an adverse factor for the subsequent course of the disease. Sinicco et al found that 68% of 23 patients who presented with the acute viral syndrome developed AIDS at 56 months, compared to 20% at 66 months in a group of patients who seroconverted with no symptoms(4). Similarly, some virologic factors such as the presence of syncicium inducing isolates (SI) and the presence of P24 antigen in serum during the primary infection have correlated with the accelerated development of AIDS(5). Although there has been a significant advance in the knowledge of the pathogenesis, clinical manifestations and the effect on the subsequent clinical course of the infection, little is known about the effect of therapeutic interventions in this stage of the infection. Will it be possible through intensive antiviral treatment based on combinations of drugs to induce a total viral suppression and to affect the subsequent course? Since the virologic factors, including the viral load, probably contribute to the subsequent accelerated progression to AIDS, it is likely that the suppression of viral replication is an adequate strategy for treatment. In fact, a recently published study(6) found a significative lower progression to AIDS in those individuals treated during the primary infection. After this study, however, there are no published studies yet in which combination of antiviral drugs are used for the treatment of the syndrome. A logical question that follows is, if the virologic supression during this phase thus affects the outcome and subsequent course of these patients, will it be necessary as a health-care strategy to carry out clinical and epidemiological surveillance to detect cases in this phase that will be affected by treatment and in that way perhaps impact the pathologic process that would appear afterwards?

Review of information presented in the conference

Several authors have confirmed a direct relationship between the presence of symptomatology, the type of symptomatology, and the duration of symptomatology during the primary infection with the subsequent course of the disease. Pedersen et al in Denmark found that duration of symptomatology of more than 14 days predicted progression to AIDS and rapid decline of CD4 T cells(7). Wallace et al in San Diego(8) described that the presence of the acute viral syndrome increased the risk of neurological deterioration and confirmed an association with rapid progression. Vanhems et al(9) in a multinational study in Europe, Australia and Canada showed that the presence of encephalitis in the acute viral syndrome was a strong predictor of subsequent non AIDS defining opportunistic complications. The presence of supraclavicular adenopathy and faringitis predicted the rapid development of AIDS or death. Several studies presented at the conference demonstrated a significant suppression of the viral load with the use of a triple combination of antiretroviral drugs during the acute viral syndrome(10-12).

In these studies the question of what is the long-term effect of such therapy in patients who achieved total suppression of viral load during the acute phase has not been answered. However, based on findings in previous studies, the recent recommendations of the International AIDS Society(13) suggest the use of double or triple combinations to achieve a total virologic suppression if possible during this phase.

Markers to predict disease progression and monitoring of efficacy of antiviral therapy

Since the early years of the epidemic, the variable course of HIV infection among different individuals has been recognized. While some patients develop severe immunosupression and infectious complications in an accelerated form, others remain stable immunologically and free of infections for long periods of time. This variable course has made it necessary to rely on tests that serve as better markers than the ones availabe to predict the course of the disease. Not only this, but with the development of new and more potent antiviral drugs there is a need to assess the efficacy of these drugs without having to wait several years for the occurrence of clinical events that indicate therapeutic failure. The markers used until recently were mainly of inmunologic type (CD4 lymphocyte count, beta 2-microglobulin and neopterin in serum) and of virologic type (quantitation of virus or viral products by different methods such as detection of viral antigens, quantitative culture assay, and quantitative detection of viral nucleic acids). The most frequently used marker in clinical practice and in therapeutic studies has been, until recently, the CD4 cell count. In fact, in the last Center for Disease Control (CDC) clinical classification of HIV infection, CD4 cells were used as the only laboratory marker in which the system was based (CDC 1993). A count lower than 200 per mm3, according to this classification, defines the presence of AIDS and indicates a high risk for opportunistic infections. Counts greater than 500, on the other hand, suggest a relatively early stage of the disease. In view of the current knowledge on pathogenesis, however, it has become clear that the central event that occurs previous to the decrease in CD4 cells is increased viral replication in plasma and lymph nodes. It is a common observation that there are patients with large numbers of CD4 cells who rapidly progress to AIDS, and patients with much lower initial counts who stay stable for longer periods of time. The difference, as it is now known, is the quantity of virus, or "viral load" that exists in plasma. The higher viral replication that occurs, the higher viral load there is in plasma and a more accelerated course of the disease. Of the available methods to measure viral load , the quantitative coculture assay has good sensitivity but it is laborious, costly and results may delay many weeks. The p24 antigen determination in plasma has a low sensitivity , although it is very reproducible and simple to perform. The immune-complex acid-dissociation method increases the sensitivity significantly although it does not seem to be as sensitive as the nucleic acid amplification techniques. There are currently three amplification techniques to measure viral load available: RT-PCR ( Polymerase chain reaction to amplify RNA in quantitative form), B-DNA (Branched chain DNA), and NASBA. All these techniques have demonstrated to be reproducible and relatively simple to perform in laboratories trained appropriately. Clinical studies have demonstrated that the values obtained are modified with antiviral treatment .

Main research questions in this topic

Although the use of virologic markers of progression has become general practice, it was not clear, until recently, what is the exact role of this marker as predictor of future clinical events, and whether it can be used as a means to base a new classification of the disease. If the viral load is useful to predict future clinical events in a given patient, is a change in viral load, produced by the effect of antiviral drugs, also associated with a corresponding change in the outcome of that given patient? What is the real use of these markers as primary endpoints for efficacy in clinical studies? And, what is their role in daily clinical practice? Finally, an important question is, can other less expensive markers, such as ICD p24 antigen, be used with similar efficacy, and in what situations can they be applied?

Review of information presented in the conference

One of the most relevant points regarding clinical aspects during all the conference was precisely the presentation of data confirming the usefulness of the viral load tests in plasma to predict long-term clinical events and to evaluate the response to antiviral treatment. The importance of using these markers in daily clinical practice to determine when there is failure in antiviral treatment¾ and consequently make the necessary adjustments to the treatment¾ was stressed. In a multicited study during the conference, J Mellors et al. of the University of Pittsburgh(14) presented their data about the analysis of viral load by bDNA in plasma from subjects infected with HIV enrolled in a cohort study since 1984 and followed for a maximum of 11 years (Multicenter AIDS cohort study). Subjects were stratified according to basal measurements of viral load and CD4 T+ cell counts and these measurements were correlated with global survival and survival without AIDS using Kaplan Meier curves. The viral load in plasma was a better predictor of long-term events than the CD4 T cell count. The 5 year survival was 93%, 89%, 68% and 44% in subjects with viral load of < 4500, of 4501 to 13000, of 13001 to 36300 and more than 36300 copies/ml, respectively. The adjusted risk for death by basal CD4 cell count (decline of 100) and by HIV-RNA (two fold increase) were 1.04 (CI 95% 1.01-1.13) and 1.26 (1.16-1.38) respectively. When the analysis was made averaging two measurements of RNA with difference of 6 months, the discriminatory value to predict subsequent events was still higher. The Kaplan Meier curves show obvious differences among the four groups according to the number of HIV RNA copies measured at baseline. The author makes the resemblance with the systems for estadification of other chronic diseases such as Hodgkin lymphoma or colon cancer in which an elaborate and generally costly evaluation protocol allows the clinician to establish survival differences similar to those which are obtained using the HIV RNA test in plasma in patients with HIV infection. Sabine et al, from Switzerland, presented a similar study whereby using the Roche test (RT-PCR) they obtained similar data in terms of prediction of death and opportunistic infections regardless of the CD 4 T cell count(15). Graham and colleagues from John Hopkins University evaluated another test to quantify viral load: quantitative microculture assay in peripheral mononuclear cells, in order to determine if this test also had predictive value on the subsequent clinical events. The presence of > 100 infectious units was strongly associated with the progression to AIDS at 2 years (38%) while the presence of less than 16 units was associated with an incidence of only 8%. These values were independent of the CD4 T cell count in those subjects with CD 4 < 500 but not in those with more than 500. This method can be used as an alternative to establish the prognosis when HIV RNA tests are not available(16). The data obtained in these studies on viral load do not indicate a correlation between the effect of antiviral treatment on the viral load and the long-term outcome. This was one of the important questions related to the clinical use of these tests. Two studies mainly have defined a correlation among the changes in viral load in response to treatment, and the long-term clinical outcome. These two studies are the analysis of the virologic data in studies ACTG 175 and Delta, both of which were presented during the conference. In the first study(17), whose clinical results have been reported previously, the relative importance of CD4 cell counts, HIV RNA and the isolation of a virus with a syncitium inducing phenotype (SI) on the occurrence of AIDS or death was evaluated in subjects treated with AZT alone, DDI alone, and combinations. Using proportional Hazzard models, the change in CD4 counts and viral RNA in the first 8 weeks of treatment was evaluated. Changes in one log10 of viral RNAviral had a Hazzard risk of 0.27 (p < 0.001) for a future 50% decline in CD4 cell counts; of 0.38 (p=0.001) for AIDS, or death and of 0.4 (p=0.013) for death, while the risk of a decline of 100 CD4 cells /mm3 was 0.62 (p=0.002), 0.72 (p=0.09) and 0.71 (p=0.15) for the same variables, respectively. The presence of SI phenotype vs NSI was a better predictor of the three events than the changes in the CD4 cell count. In the Delta study(18) Brun-Vézinet and colleagues studied 230 patient in three groups of treatment: AZT alone, AZT plus DDI, and AZT plus DDC. As in the ACTG 175 study, they found a correlation between the more potent antiviral effect of the combinations with the subsequent development of clinical events. A third multicenter study, presented by Phillips(19), examined the antiviral effect of the combination of 3TC and Zidovudine in the NUCA studies 3001 and 3002, finding a similar correlation between the antiviral effect in plasma and clinical benefit. All these data clearly indicate that it is possible to use the viral load in plasma as a measure of efficacy in studies that evaluate antiviral drugs. Also, they suggest that it is possible to use these measurements to determine in daily practice whether a patient is responding to treatment in order to change the antiviral drugs. In fact, in the recommendations issued by the International AIDS Society(6) the routine use of HIV-RNA measurements in plasma is recommended in order to make therapeutic decisions. The specific management according to these recommendations will be discussed further.

Opportunist Infections and malignancies associated with AIDS

Review of knowledge previous to the conference

The preventive management, diagnosis and therapy of the different opportunist infections has been crucial for the appropriate management of patients with HIV infection. Several authors have described the relationship between the appearance of the different opportunistic infections with the level of the CD4 cells. Above 200 CD4 / mm3 cells, pulmonary tuberculosis, Kaposi Sarcoma and some bacterial infections can occur. Below 200 CD4 cells the incidence of PCP and infections by Candida increase, and under 100 and especially 50 CD4 cells, infections by Citomegalovirus (CMV), Mycobacterium avium bacteremia and cerebral toxoplasmosis, and neoplasias as primary cerebral lymphoma. Knowledge of the risk that occurs in each individual patient depending on the CD4 cell level has allowed the design of appropriate plans for prevention in those subjects with greater risk. The use of prevention strategies for opportunist infections, especially PCP, and the longer survival of patients has changed the spectrum of clinical presentation in the last years, increasing the number of cases with infections of later stages such as Citomegalovirus and Mycobacterium avium bacteremia.

Prevention and management of the principal opportunistic infections. Main research questions on the topic, for decision makers and perspectives of advance

Pneumocystis carinii. One of the interventions with the greatest cost-benefit ratio in patients with HIV infection has been the prevention of Pneumocystis carinii pneumonia with thrimetroprim-sulfametoxazole. Before the era of prevention, approximately 60% of the individuals with AIDS had PCP as the first manifestation of their disease and 80% had PCP during the course of their disease. At the present time, even with widely used prophylaxis for PCP, this infection continues to occur. The individuals with the greatest risk to develop PCP are those who already presented a previous episode, those who have other opportunist infections as oral candidiasis, chronic diarrhea or fever, and those who have CD4 cell count lower than 200 / mm3. (20). The most effective agent for prophylaxis of PCP is thrimetroprim / sulfametoxasol (TMP/SMX), which has the advantage of being useful to prevent cerebral toxoplasmosis also(21,22). Although a large percentage of subjects infected with HIV develop hypersensitivity to this drug, it has recently been demonstrated that most of them can be desensitized to the drug(23,24). Other useful options for prevention are aerosolized pentamidine and regimes containing dapsone which at 100 mg a day seems to be as effective as TMP /SMX with the drawback of not being effective for prevention of toxoplasmosis. For treatment of PCP, as for prophylaxis, TMP /SMX is the agent of choice. The recommended dose is 20mg/kg of trimetroprim, for three weeks, although some authors have suggested that 15 mg may be equally effective(25). Other drugs that may be good alternatives for TMP/SMX, when there is toxicity, are Pentamidine at 4mg/kg IV once a day, Clindamicin+Primaquine, Dapsone 100mg once a day combined with trimetroprim 4 mg/kg tid(26) and, more recently, especially for mild-to-moderate cases, the use of atavaquone 750mg every 6 hours(27). An important improvement in the management of PCP has been the demonstration of the effectiveness of adjuvant glucocorticoids in severe cases to treat the respiratory failure that develops generally during the first days of treatment with antimicrobial agents. Several studies have demonstrated efficacy when glucocorticoids are begun early in the admission of the patient and when there is moderate-to-severe respiratory failure(28,29).

Tuberculosis. The immunologic deterioration that occurs in patients with HIV infection results in a high frequency of reactivation tuberculosis in infected individuals. Tuberculosis in HIV infected patients is commonly extrapulmonary and with atypical presentations. The new methods for laboratory diagnosis, where they are available, have allowed better detection and more rapid diagnosis in these patients. The sensitivity of the tuberculin test to detect infected cases is low in patients with CD4 counts < 200/mm3 , which limits its benefit in this group of patients. The current recommendations for treatment of tuberculosis in patients with HIV infection are similar to those in patients without HIV infection. The American Thoracic Society recommended in 1994 the use of a four drug regime (INH, rifampin, pirazinamide, and ethambutol) for 2 months followed by INH and rifampin for 4 or 5 months. The treatment must be extended if the clinical or microbiologic response is slow. There is no evidence that suppressive treatment with INH needs to be extended for life.

Emphasis has been made on the need to implement treatment programs of directly observed therapy to ensure compliance. It is important to consider the prevalence of resistance to one or several drugs in a given area in order to issue therapeutic guidelines adapted locally to each region. Reports of multirresistant M. tuberculosis in HIV infected patients have been received from different parts of the world. In terms of prevention, the benefit of using INH for one year in HIV infected patients with positive tuberculin skin reactions has been clearly demonstrated. The use of INH in anergic patients residing in areas with high prevalence of positive PPD may be justified.

Toxoplasmosis. Persons with HIV infection and advanced immunodeficiency are at high risk of developing serious infection by Toxoplasma gondii. Currently Toxoplasma gondii is the most frequent cause of space-occupying lesions in the central nervous system in patients with AIDS. The most frequent clinical manifestations are encephalitis, retinitis and, less commonly, myocarditis and pneumonitis. Most cases of toxoplasmosis occur in patients with a CD4 count lower than 100/mm3. Toxoplasma encephalitis generally appears as multiple abscesses in image studies of the central nervous system. The diagnosis is based currently on the histologic identification of parasites on cerebral biopsies. However, due to the potential morbidity associated with cerebral biopsies, current practice is to give treatment with specific antitoxoplasma drugs to patients with suspicion of toxoplasmosis, based on image studies, and with positive serology. Cerebral biopsy is recommended when the instituted therapy fails. New methods for diagnosis, which are more sensitive and specific, need to be evaluated; for example, detection of nucleic acids by amplification methods in CSF. The role of serology for diagnosis of toxoplasmosis is limited. The proportion of patients with cerebral toxoplasmosis who have a positive specific IgG in serum is 97%. However, the presence of high antibody titers or the isolated finding of antibodies in serum are not enough to confirm the diagnosis; therefore, they are not useful for diagnosis in the patient with suspicion of toxoplasmosis. Effective treatment for cerebral toxoplasmosis includes the use of pirimethamine plus sulfadiazine and pirimethamine plus clyndamicin. Both combinations of drugs have shown similar efficacy in comparative studies(30,31). Other drugs such as atavaquone and the new macrolides clarithromicin and azithromicin do not yet have a defined role as initial therapy. Use of drugs for primary prophylaxis is indicated in patients with serology positive for toxoplasma and a CD4 cell count lower than 100. For this purpose, TMP /SMX has been evaluated in retrospective studies which showed efficacy, and has the advantage of conferring protection also for PCP. The studies that have evaluated pirimethamine and clyndamicin or pirimethamine alone for prevention have shown little efficacy either for disease occurrence or for toxicity.

Disseminated Mycobacterium avium (MAC) infection. Disseminated MAC infection occurs in advanced stages of HIV infection, generally when CD4 cell counts are lower than 50. It is currently one of the most common infections and an important cause of morbidity. New antimicrobial agents and combinations have been developed for the treatment and prevention of this infection. The clinical picture is characterized by constitutional symptoms with fever, diaphoresis and weight loss. Diarrhea and liver function abnormalities can occur. The diagnosis is made with blood cultures in media for Mycobacterium growth. The use of radiometric methods, such as Bactec, has improved the rate of isolation of Mycobacteria in blood in patients with AIDS. The isolation of Mycobacteria from respiratory specimens rarely corresponds to MAC and most likely suggests M. tuberculosis. Effective treatment against disseminated MAC is needed to improve the quality of life, even though it probably does not affect the survival time. Although there are several drugs with in vitro activity against MAC, the ideal plan or the duration of the treatment has not been defined yet. In the most recent recommendations, the proposal was an initial treatment combination consisting of chlaritromicin or azithromicin as primary drugs, ethambutol as a second drug. and a third drug that may be ciprofloxacin or ofloxacin(32). The effect of the addition of rifabutin to this plan is currently being studied. The duration of treatment must be at least 12 weeks and most likely a more simplified regimen will have to be maintained subsequently in order to prevent relapse. For prevention of disseminated MAC infection, several drugs have been evaluated. Rifabutin at 300 mg each day was compared to placebo in patients with less than 200 CD4 cells and showed a decreased incidence of bacteremia from 18 to 9%, although no benefit on survival was observed(33). More recent studies, whose results have not been published, have evaluated clarithromicin and azithromicin as prophylaxis.

Citomegalovirus (CMV) infection in HIV infected patients. CMV infection is another opportunist infection that affects patients in advanced stages of inmunosuppresion. Up to 30% of patients with AIDS and CD4 cell counts lower than 100 develop CMV retinitis or involvement in other organs such as gastroenteritis, esophagitis and pneumonitis. CMV disease occurs as a consequence of reactivation of a latent infection when there is severe inmunosuppression. CMV retinitis appears gradually as loss of vision in one or both eyes, appearance of blind spots, and is usually painless. Diagnosis is made by the ophtalmologist based on the characteristic appearance of retinal lesions, which are described as whitish granular exudates that coalesce to form larger lesions; followed later by the appearance og hemorrhagic lesions. If no treatment is given, the lesion progresses until there is complete loss of vision. There are currently two antiviral drugs approved with proved efficacy against CMV: ganciclovir and foscarnet. Several new drugs are being developed and one of them (HPMPC) will probably be approved shortly for use in humans. Induction therapy with ganciclovir 5mg / kg twice a day or foscarnet 60 mg / kg three times a day results in a rate of response of 80 to 100%. The induction phase lasts between 14 and 21 days and should be followed by maintenance treatment with the purpose of avoiding relapse of the disease, which most likely will occur if discontinued. The initial treatment choice has not yet been defined, since in a comparative study between both drugs, the efficacy in inducing a remission in the retinal lesions was similar; however, survival in patients who received foscarnet was 4 months longer(34). In spite of these results, because of the higher cost and greater toxicity of foscarnet, there is not yet a consensus about the initial drug for treatment. In fact, since the survival benefit of foscarnet in this study was probably due to its antiretroviral effect, it is likely that with the current availability of more potent antiretroviral drugs this difference will no longer be relevant. Recently, the use of oral ganciclovir was approved ( 1 g tid) for maintenance treatment in patient with CMV retinitis. The initial studies apparently did not show differences in the mean time for relapse among patients who received oral ganciclovir, against those that were receiving intravenous maintenance therapy. In spite of these results, there is still doubt on the efficacy of oral ganciclovir to avoid progression of retinal disease. There are several methods for viral diagnosis of CMV infection, which include blood and urine viral cultures, detection of viral antigens in peripheral blood and PCR in plasma and in peripheral blood mononuclear cells. The use of these methods to diagnose and predict disease is not generalized, but there is great interest in evaluating them for this purpose. There are currently 3 possible strategies for prevention of CMV retinitis.

  1. Periodic ophtalmologic monitoring for early detection of peripheral lesions that may indicate the start of therapy before severe retinal damage occurs.
  2. Use of virologic methods to monitor reactivation of infection before disease appears.
  3. Antiviral prophylactic therapy to any patient with CD4 cell count lower than 100 / mm3 using oral ganciclovir. The relative benefit of these three strategies is currently being evaluated.

Research questions on this topic and perspectives of advance in opportunist Infections

Clearly the new trends in prevention require the use of drugs or combinations of drugs with proven efficacy but which are also practical regarding form of administration, tolerability and, very especially, cost. The ideal prophylaxis regimen should be that which may be useful to prevent several infections simultaneously; for example, TMP/SMX for toxoplasmosis, PCP and bacterial infections. It is more and more important to consider in the evaluation of drugs for prophylaxis the cost-benefit of their administration. In the area of PCP the main questions currently relate to better diagnostic non-invasive methods such as PCR. An important question is which drugs used for prophylaxis of other infections may be beneficial for the prevention of PCP. The debate about the appropriate use of intensive care for patients with PCP continues. Better information about the optimum use of resources in intensive care units for these patients, without neglecting care quality, is needed. In relation to tuberculosis, the question whether occurrence of tuberculosis affects the course of HIV infection is outstanding. Can the immune activation that occurs during tuberculosis affect the viral load , and, as a consequence, the subsequent course of HIV infection? If so, how can antituberculous treatment modify this effect? Is there a role for immunomodulators as adjuncts to antituberculosis therapy in HIV-infected patients? Better prophylactic regimens for prevention, of shorter duration, are required to improve compliance. The ideal drugs for prevention of tuberculosis where infection was likely caused by multirresistant tuberculosis need to be defined. In relation to toxoplasmosis, better diagnostic methods are required, such as those based on nucleic acid amplification to make non-invasive diagnosis a possibility. The role of the new macrolides in the prevention and treatment of toxoplasmosis needs to be defined. In relation to disseminated MAC infection, there is a need to have simpler and cheaper prevention regimes. The role of the new macrolides in this area must be defined. Concerning CMV, the best treatment regimen for retinitis needs to be defined. Is the combination of foscarnet and ganciclovir a better option initially? What is the effect of this combination on development of resistance? The role of HPMPC (a new antiviral for CMV) in treatment needs to be defined. At present, the best prevention strategy for CMV retinitis is not defined. Probably, monitoring with virologic markers may be used to direct preventive therapy in patients with high risk to develop retinitis. Regarding diarrhea, the prevalence of different pathogens in different regions has to be determined to design empirical treatment regimens. For instance, the frequency of Cyclospora and Microsporidium as cause of diarrhea in Latin American countries needs to be known. A little studied area in Latin American countries is the best cost-effective approach for diagnosis of the most common pathogens causing diahrrea. Non invasive, simple, laboratory tests need to be designed to avoid the use of more expensive, invasive and commonly non accesible procedures such as endoscopies and histopathological studies for diagnosis of disease caused by different microorganisms. For instance, it would be feasible to design a panel for immunologic or molecular diagnosis of most pathogens involved in chronic diarrhea in these patients. There is still little reliable, effective treatments for Microsporidiasis and Criptosporidiosis, the latter being responsible for a large proportion of cases of chronic diarrhea in developing countries. Concerning wasting syndrome, it is necessary to characterize the epidemiology of the syndrome and its most frequent causes in Latin American countries. New potential treatment options for wasting and their application should be investigated. The use of different types of nutritional support such as oral supplements, total parenteral nutrition, multivitamins, and other micronutrients, or enteral feedings must be investigated in regard to efficacy, as well as the cost-benefit ratio of these interventions.

Review of information presented in the conference

One of the issues raised during most of the presentations on prophylaxis for opportunistic infections was that of the cost-benefit ratio of interventions. Such an evaluation was conducted by Freedberg and colleagues from Harvard School of Public Health who used an analytical decision model, to analyze the cost and efficacy of the most common interventions to prevent opportunistic infections(35). The administration of TMP/SMX for prevention of PCP and Toxoplasmosis in fact saves money. The cost of prevention of disseminated MAC infection with rifabutin is 210,600 dollars per year of life saved; the cost of prevention of fungal infections with fluconazol is of 2,484,000 dollars per year of life saved; and the cost of prevention of CMV disease is of 3,790,200 dollars. The cost of prevention of disseminated MAC infection with azithromicin (a recently evaluated regimen) is aproximately of 60000 dollars. According to the model, the most significant cost-effective maneuver is not the reduction in the cost of the prophylactic drugs, but rather the reduction in the risk of CD4 cell counts decline. This has important implications, since the promotion of antiretroviral use in a given population to mantain adequate levels of CD4 cells may have a much more beneficial economic impact than the prevention of opportunistic infections. In a study using azithromicin for prevention of disseminated MAC, Dunne and colleagues from the University of San Diego(36) found that in patients with < of 100 CD4 cells, 1200 mg of azithromicin given once a week reduced the risk of PCP in half , in patients who were already taking prophylaxis with TMP /SMX . This finding supports the use of this macrolide for prevention of disseminated MAC, which will have a significant additional protection for PCP. In presentations on PCP, the risk factors for PCP were analyzed in the current prevention era. The conclusions of two studies(37,38) are that most of the PCP cases that still occur are associated with inadequate prophylaxis, either because of late diagnosis of HIV infection, or because of lack of compliance to treatment. Survival of patients with PCP in intensive care units was analyzed by Curtis of the University of Washington in Seattle(39). He found that the actors that predicted a poor survival were a prolonged stay (> of 14 days) in the intensive care unit and the history of other AIDS defining illness at the time of admission to the unit. In spite of the poor prognosis of those patients with prolonged stays, survival in this group was of 24%, which is higher than previously reported. This study concludes that patients with PCP may benefit from intense care and they should not be rejected solely on the basis of this diagnosis. Atzori and colleagues from Milan(40) evaluated the use of PCR in different body fluids for diagnosis of PCP. The sensitivity of the test , performed in serum, was 90%, 70% in peripheral blood mononuclear cells, and 75% in oral secretions. This study demonstrates the feasability of creating a non-invasive diagnostic test for PCP.

Several studies presented analyzed different strategies for prevention of CMV disease. In a study to evaluate the efficacy of oral ganciclovir compared with placebo, Brosgart and colleagues of the CPCRA stuy(41) enrolled 994 patients with CD4 cell counts of < than 100/mm3 without evidence of CMV disease and with positive serology or culture for CMV. In this study, the authors did not find any difference in development of CMV disease between both groups. These data contrast with the results of a previously published study by Spector et al in which there was a definite benefit. This author and collaborators presented a study(42) in which detection of DNA of CMV in plasma was evaluated to identify those subjects with greater risk of developing the disease. They used plasma of subjects enrolled in the prophylaxis study with oral ganciclovir. The relative risk for patients who received placebo and had a positive PCR was of 3.84. In patients with negative PCR on baseline, only 1% of the group treated with ganciclovir developed retinitis, as compared with 26% of the group not treated, while in patients with positive PCR on baseline, 58% of those receiving placebo and 31% of those receiving ganciclovir developed CMV disease. These results suggest the need to conduct studies using preventive therapy in subjects with positive PCR for CMV and with low CD4 counts. Another study carried out in Canada found similar results using the pp65 antigenemia test for CMV as well as PCR(43). Two studies evaluated the combination therapy with ethambutol, rifabutin and clarithromicin for treatment of disseminated MAC infection and found a greater efficacy of this combination than regimens based on clarithromicin and clofazimide(44) and rifampin, ethambutol, clarithromicin and ciprofloxacin(45). In both studies, this regimen showed a faster clearance of MAC from blood, more rapid improvement of symptoms, and better quality of life. Also, the relapse rate and development of resistance to clarithromicin was lower in the ethambutol, rifabutin and clarithromicin group. In a prospective study for prevention of disseminated MAC infection , clarithromicin was more effective than rifabutin in preventing disseminated disease(46). In the same study, the combination of clarithromicin and rifabutin did not increase the efficacy or prevented the development of resistance. In patients receiving rifabutin alone, there were no cases of resistance to rifabutin, while 29% of the strains isolated from patients who received clarithromicin and developed disseminated disease were resistant to clarithromicin. In relation to tuberculosis, two outbreaks of infection with multirresistant M. tuberculosis were reported in hospitals from Argentina(47) and from London(48). The risk factors in these outbreaks were delay in diagnosis and lack of isolation of patients. In the study from Argentina, the need of isolation facilities for developing countries is emphasized in order to prevent future outbreaks. In a longitudinal study carried out in Uganda, Nelson et al analized the dynamics of viral load in HIV-infected patients with tuberculosis compared with controls without tuberculosis and similar stage of HIV infection. In this study, the viral load (HIV RNA) was approximately 5 times higher in patients with tuberculosis than in controls, but 70 times higher in patients after 6 months of adequate antituberculous therapy. Theses results suggest that M. tuberculosis infection activates replication of HIV, and that this effect remains in spite of adequate antituberculous treatment. In relation to diarrhea, there were two studies presented on intestinal infection with Microsporidium(49, 50). In the first study, by Briner and collaborators from Switzerland, 55 patients with intestinal infection with E. bieneusi were analized; 83% of these patients had chronic persistent or intermitent diarrhea and in 17% the diarrhea was controlled in spite of persisting excretion of the microorganism in feces. These patients had CD4 cell counts higher than the group with persistent diarrhea. One year survival was of 52.8%, and at 2 years, 27%. Treatment with albendazole did not affect excretion of the parasite or the outcome. In the second study, Albrecht et al from Hamburg, presented information from a prospective study on microsporidium infection. They found that the best recovery of the parasite was achieved from the stools using the calcofluor method. In this study, as in the previous one, there was a small group of patients without diarrhea that were excreting the parasite in feces.

Treatment with antiretrovirals

Known information before the Vancouver AIDS conference

The concepts guiding the use of antiviral agents in HIV infection have changed rapidly in the last year. One year ago, there were only 4 antiviral compounds available approved for use in HIV infection (Zidovudine, DDI, DDC, and Stavudine or D4T). In the last year there have been 4 more antiviral agents added (3 TC, and 3 protease inhibitors: saquinavir, ritonavir and indinavir). Other compounds will most likely be approved shortly. This explosive increase in the number of antiviral drugs has occurred along with a better knowledge of their utilization based on the pathogenesis of the disease and on new techniques to determine the efficacy of the antiviral treatment as well as the appearance of resistance to the drugs. In the first years of the era of antivirals, the efficacy of monotherapy with Zidovudine on survival of patients with AIDS and on development of AIDS and CD4 cell decline in asymptomatic patients was documented, although in these patients the results of the studies suggested only a modest, transient benefit, and, according to some studies such as the Concorde, no effect on survival was seen(51-53). Afterwards, a greater efficacy¾ in terms of survival, CD4 cell counts and incidence of opportunistic infections¾ was seen in patients who had received at least 16 weeks of treatment with AZT, when they were changed to DDI(54). The beneficial effect of the change in treatment seemed to be associated with development of resistance to AZT and absence of cross-resistance between both drugs. Finally, the use of monotherapy was definitely abandoned in favor of combined therapy at the end of 1995 when the results of the ACTG 175 study and the Delta study were known. In both these studies, there was a better survival and lower rates of progression to AIDS in subjects treated with combination of DDI/AZT or DDC/AZT compared with those treated with DDI alone or AZT alone. Most importantly, in these studies there was a clear correlation between the effect of therapy on viral load and a clinical benefit.

Research questions on the topic and perspectives for advance

In spite of the great advance in the knowledge on antiretroviral therapy derived from ACTG 175 and Delta studies, which resulted in the definitive abandonment of monotherapy as a standard practice, new questions emerge now, such as, what is the ideal time to start antiretrovirals in a given patient? Is it necessary to begin combined antiviral treatment in all patients in whom HIV infection is detected, independently of the inmunologic status or the viral load? What is the best antiviral regimen to start treatment with? Knowing that virologic markers should be used as endpoint to measure efficacy of treatment, what is the desired level of viral suppresion after which the risk of development of resistance is minimal or non-existent? Is it always necessary to attempt total viral suppression using the most potent combinations from the begining? Can a total viral load suppresion be achieved with the most potent combinations? And, if this is true, how long can this be sustained? Is there a correlation between the inhibition of the viral load in plasma with inhibition of viral replication in lymph nodes? Are there sites of virological latency capable of originating new cycles of viral production after the infected cells that sustain active replication of the virus have been eliminated by treatment? Is there a possibility of immunologic restoration in patients with severe immunosuppression in whom a total suppression of viral load has been achieved? A relevant question for the management of opportunistic infections is: in patients with severe immunosuppresion (< 50 CD4 cells) if the combined treatment resulted in a significant increase of CD4 cells at levels higher than 200, can the secondary or primary prophylaxis regimens against opportunistic infections such as MAC, PCP, CMV or others be modified? Finally, in the patient that started appropriate antiviral treatment, when should it be changed? And, what is the ideal regimen to continue the patient with?

Review of new information presented at the conference

Multiple original papers and symposiums presented by international experts touched on the new aspects of antiviral therapy in light of the current knowledge on pathogenesis, use of viral load markers, and the availability of greater number of antiviral drugs. A common theme in most of the presentations was that current management strategies with antivirals are based on the evidence that massive viral replication is a central part in the pathogenesis of the immunodeficiency observed in patients with HIV infection. This replication occurs in lymph nodes, in peripheral blood mononuclear cells. and in some organs. This concept suggests that antiviral therapy must have as a primary objective the substantial, and, if possible, total virologic suppression. It has already been shown in clinical and in vitro studies that it is not possible to achieve that objective with the use of one drug alone; therefore, combined therapy is the only current alternative. Therapeutic failure of antiviral drugs is due largely to the emergence of viral variants with specific mutations that confer these viruses the capacity to grow in the presence of increased concentrations of the drug to which they have been exposed. The emergence of these variants can be retarded, although possibly not avoided, by using different combinations of antivirals and achieving a maximum suppression in viral replication. Although the above mentioned principles are accepted by most experts, controversy still exists over what is the best regimen to begin with and over what is the best regimen to continue treatment. In spite of the impressive results on viral load and CD4 cells of the combination regimens containing protease inhibitors, it is not yet clear whether the initial regimen should include a protease inhibitor, or if protease inhibitors should be reserved for continued treatment of failing initial schemes. When it is decided to use a protease inhibitor, it is not yet clear which of the three available should be used

Among the evaluated different combination regimes the following is a selection of the most relevant presented at the conference:

  1. Double blind, comparative study of AZT / DDI against AZT / DDI and nevirapine(55). In this study, 152 patients without previous antiviral treatment and with CD4 cell counts between 200 and 600 /mm3 were evaluated. The triple combination showed a reduction in viral RNA of up to 2 logs, and this reduction was more pronounced than with the double therapy.
  2. Antiviral effect of the combination of D4T with DDI(56). In this study, there was a reduction of 2.8 logs of HIV RNA in some subjects, all of them were naïve to previous antiviral treatment. Tolerance was acceptable and no greater neurological toxicity was observed than that reported with individual drugs.
  3. Antiviral effect of combination of hydroxiurea and DDI(57, 58). Some patients showed a very potent antiviral effect using combination of DDI and Hydroxiurea. In the first study, 11 of 20 patients had levels of virus not detectable by PCR at 180 days and 6 of 10 at 360 days. The rest of the patients showed a reduction in the viremia of 1.5 and 1.7 logs, respectively.
  4. Evaluation of the antiviral effect of the combination indinavir /AZT / 3TC after one year(59). In this study, patients with more than 50 CD4 cells and more than 6 months of previous antiviral treatment with AZT were studied. One group of patients received the triple combination, another group indinavir alone, and the third group AZT/3TC. After 44 weeks, 83% of the patients in the three drug groups had levels of HIV RNA below the detection limit, against 22% in the group of indinavir alone and 0% in the group of AZT / 3TC. The corresponding increase in CD4 cells was of 218 at 44 weeks in the group receiving the triple combination, 158 in indinavir alone, and 14 in the AZT / 3TC group.
  5. Triple combination of AZT / 3TC / ritonavir (60). Twelve patients recently infected with HIV were included. All patients showed RNA HIV levels below the limit of detection at variable times during follow-up and negative cultures up to one year into therapy.
  6. Combination of ritonavir with saquinavir(61). In this study, the combination of ritonavir 600mg two times a day and saquinavir 600 mg two times a day in patients previously treated with nucleoside analogs showed important reductions in the viral load. In 70 to 80% of the patients levels of HIV RNA were reduced to non-detectable.

David Ho and colleagues presented an estimate of the time needed to treat patients who have been aviremic for extended periods of time after effective suppressive therapy(62). Based on a mathematical model, which takes into account the life span of infected macrophages and latently infected CD4 lymphocytes, they conclude that a totally inhibitory treatment must be given for at least 1.5-to-3 years before considering discontinuation.

In an official meeting of the International AIDS Society, the current recommendations for the use of antiretrovirals were presented, based on existing evidence. These recommendations have been published recently. Briefly, they include the initial use of antivirals, the continuation treatment after the initial one needs to be changed, and special situations such as primary infection, post exposure prophylaxis and vertical transmission. According to these recommendations, therapy must be started in patients with less than 500 CD4 cells, and those with symptomatic disease. In patients with more than 500 CD4 cells the recommendation is to use the viral load as the decision factor; in patients with more than 30 000 copies /ml or with rapid decline in CD4 cells , treatment must be started. In patients with more than 10 000 copies, it must be considered; and in patients with less than 10 000 copies treatment is not recommended at that time. The initial regimens recommended by the IAS are combination with two nucleoside analogs such as AZT / DDI, AZT / DDC, or AZT / 3TC. The decision of using a protease inhibitor as part of the initial plan is considered reasonable in those patients with rapidly progressive disease. The type of protease inhibitor must be selected based in factors such as tolerability, cost, cross-resistance with other inhibitors, and the possibility of having open options for the future. At the moment there is no specific recommendation for one single protease inhibitor. A change in the initial antiretroviral scheme is recommended when there is treatment failure, toxicity, intolerance or lack of compliance, and current use of a suboptimal regimen. Treatment failure is considered when the viral load returns to levels within 0.3 to 0.5 logs from the pretreatment level, or when there is a significant decrease in the CD4 cell count or clinical progression. Once the decision to change therapy is made, several aspects need to be considered, such as toxicity, why the change is made, and the drugs the patient has been taking previously. If the change is made because of treatment failure, the drugs chosen should be the most potent; preferably they should have different mechanisms of action and without cross resistance with the previous drugs. In patients who have received two nucleoside analogs, the treatment change must be made to two different nucleosides and a protease inhibitor. If in the initial regimen there was a protease inhibitor, the subsequent regimens should include at least two new different drugs. Finally, consideration to stop antivirals should be given to patients with advanced disease in whom quality of life becomes priority over a less likely benefit derived from the drugs. Concerning the different protease inhibitors, it is not yet clear which one of the three available drugs should be used initially when a decision is made to use a protease inhibitor. Saquinavir is the most potent inhibitor in vitro; however, its bioavailability is of only 4% and it has a high percentage binding to proteins, which makes it less potent than indinavir and ritonavir in vivo. Newer preparations of this drug which have greater bioavailability or the use of the drug concomitantly with other drugs that increase its levels in blood by inhibiting hepatic metabolism will likely improve the profile of saquinavir. On the other hand, the mutations that confer resistance to indinavir are associated with resistance to all the other protease inhibitors currently available and in development; therefore, its use as a first line agent must be seen with caution. The major drawback of ritonavir are the frequent interactions with other drugs commonly used in patients with AIDS, which makes its use much more complex.

Conclusions and recommendations

New concepts in the pathogenesis of HIV infection and its consequences have emerged, causing a radical change in the management of patients. The current paradigm is centered in the virus and its replication as the main cause of the inmunologic deterioration that occurs during HIV disease. Therefore, therapy must be directed towards control of this replication. The efficacy of therapy must be evaluated by determination of viral replication . In that same context, we now know that monotherapy with antivirals is no longer recommended in managing HIV infection in any stage of the disease. The availability of new drugs with extraordinary antiviral potency has made possible a new series of combinations with strong antiviral activity, which delay development of resistance and which are translated into palpable long-term clinical benefits. It is likely that the outcome of HIV infected patients managed with these new treatment modalities will improve substantially, and that the disease will be a manageable chronic process similar, from the medical standpoint, to what occurs with other chronic diseases. The lack of access to the new drugs and to sofisticated laboratory tests in developing countries is an issue of great concern since now, more than ever, the differences in economic resources will result in differences in the way in which people can live with VIH. This aspect has been dealt with repeatedly in international forums; facing it represents a challenge for the health systems in various regions of the world. Regional studies should be done to determine the gains that may result from investing considerable sums in providing financial support for comprehensive treatments in patients with HIV infection. For example, the cost of providing antiviral treatment during prolonged time should be compared to the cost of using expensive hospital resources for treatment of opportunistic infections, and to the cost of productive time lost from work, and to the cost of shorter survival time in young individuals, in order to advocate, based solely on economic reasons, the implementation of this type of program. One must also recognize that making this type of resources available to entire populations will likely bring as a consequence an increase in the number of detected cases with HIV infection since a greater number of people that consider themselves exposed will want to have a detection test done. There is also the danger that society may have unrealistic expectations from the new treatments which may allow for complacency with regard to preventive measures. With respect to healthcare, systems should be organized to offer medical care for the unique problem that AIDS represents, which is different to most of the health problems that currently occur. The health system must have the flexibility to change quickly to face the challenge that a significant increase in the number of sick people with AIDS demanding attention represents. The systems should be organized for efficient ambulatory-based care and move away from predominantly hospital based care. The medical care of these patients has become extraordinarily complex; therefore, care should be provided by highly competent and motivated personel. Improvisation should not be permitted or encouraged , since the resources that are invested will most likely be wasted in settings where providers are inexperienced. Care of HIV infected individuals should be concentrated in places with proven expertise in patient care, and attempts to generate resources with research and development should be made. The existence of multiple small sites for the care of patients should be discouraged , since this will cause dispersion of resources , and possibly dilution in the quality of care. Finally, the preventive and normative functions of the government agencies should be separated from those related to medical care.

References

1.Tindall B, Cooper D.A. Primary HIV-1 infection. In Management of the HIV-1 infected patient. Crow S., Hoy J, Mills J.Cambridge University Press, Melbourne Australia, 1996.

2. Pantaleo G, Graziosi C , Demarest JF, et al. HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease. Nature 1993; 362:355-358.

3. Vento S, Di Perri G, Garofano T, concia E, Bassetti D, Pneumocystis carinii pneumonia during primary HIV infection. Lancet; 1993:342:24-25.

4. Sinicco A, Fora R, Sciandra M, Lucchini A, Caramello P, Gioannini P. Risk of developing AIDS after primary actue HIV-1 infection. J AIDS 1993; 6:575-581.

5. Roos MTL, Lange JMA, Goede REY, et al. Viral phenotype and immune response in primary Human Immunodeficiency Virus type 1 infection. J Infect Dis 1992; 165:427-432.

6. Kinloch-de-Loes S, Hirschel BJ, Hoen B. et al. A controlled trial of zidovudine in primary human immondeficiency virus infecton. N Engl J Med 1995, 333: 408-413

7. Pedersen, Court, Katzenstein T, Nielsen C, Lundgren JD, Gersoft J. Events in early HIV infection determine the long term outcome of the infection. XI International Conference on AIDS. Vancouver, 1996. (Abstract MOB1341).

8. Wallace Mrk R, Nelson JA, McCuitchan JA, Heaton RK, Miller LK, Grant Y, the HNRC group. Symptomatic seroconverting illness is associated with more rapid neurologic impairment. XI International Conference on AIDS. Vancouver, 1996. (Abstract MOB 1336.

9. Vanhems Phillippe, Allard R, Lambert J, Cooper DA, Hirschel B, Carr A, Perrin L, Vizzard J, Kinloch-de-Loes S, Hoen B. Occurrence of non-AIDS opportunistic complications, AIDS and death by clinical presentation of primary HIV infection. XI International Conference on AIDS. Vancouver, 1996. (Abstract MOB1330).

10. Lafeuillade Alain, Pellegrino P, Poggi C, Costes O, Profizi N, Tamalet C. Triple drug ombination in primary HIV-1 infection. XI International Conference on AIDS. Vancouver, 1996. (Abstract MOB 13319).

11. Tamalet Catherine, Lafeuillade A, Tourres C, Cuclos N, Tivoli N, Raoult D. Triple nucleoside analog combination therapy in primary HIV-1 infection suppresses lymph node infectious viral burden., XI International Conference on AIDS. Vancouver, 1996. (Abstract MOB 1337).

12. Saimot AG, Simon F, Landman R, Girard PM, Fauveau V, Mariot P, Leibowitch J, Mathez D. A triple nucleoside analogue combination in four patients with primary HIV-1 infections: towards complete virological remissions?. XI International Conference on AIDS. Vancouver, 1996. (Abstract MOB 1332).

13. Carpenter C, Fischl M, Hammer S, Hirsch M, et al. Antiretroviral therapy for HIV infection in 1996. Recommendations of an International Panel. JAMA 1996; 276:146-154.

14. Mellors JW, Kingsley L, Gupta P, Rinaldo C, Phair J, Saah A, Detels R, Schrager L, Margolick J., Giorgi J, Muñoz A. Prognostic value of plasma HIV-1 RNA quantification in seropositive adult men. XI International Conference on AIDS. Vancouver, 1996. (Abstract We B 410)

15. Sabine Yerly, Perneger T, Hirschel B, Mater L, Malinverni R, Perrin L, the Swiss HIV cohort study. HIV viremia influences survival in HIV infected patients. XI International Conference on AIDS. Vancouver, 1996. (Abstract We B 413).

16. Graham NMH, Asxtemborski J, Saah AJ, Vlahov D, Margolick JB, Farzadegan H. Infectious HIV viral load predicts clinical progression and survival among HIV infected adults. XI International Conference on AIDS. Vancouver, 1996. (Abstract Wed B 411).

17. Katzenstein, David A, Hammer S, Hughes M, Gundacker H, Jackson B, Fiscus S, Lathery J, Rasheed S, Reichman R, Elbeik T, Japour A, D’Aquila R, Scott W, Griffiths B, Winters M, Merigan T, Hirsch . Suppression of plasma HIV RNA by reverse transcriptase inhibitors prevents AIDS and death in ACTG 175; combination and monotherapy with ZDV, DDI and DDC. XI International Conference on AIDS. Vancouver, 1996. (Abstract MoB 293).

18. Brun-Vezinet F. The Delta Virology Group. HIV viral load changes in delta patients. XI International Conference on AIDS. Vancouver, 1996. (Abstract MoB 292).

19. Phillips AN, Eron J, Bartlett J, Hill AM, for the North American Lamivudine HIV working group. Effect of nucleoside analogue RT inhibitors on plasma HIV RNA and CD4 count as an indicator of clinical effect. XI International Conference on AIDS. Vancouver, 1996. (Abstract MoB 290).

20. Phair J, Muñoz A, Deterls R, et al. The risk of Pneumocysits carinii pneumonia among men infected with human immunodeficiency virus type 1. N Engl J Med 1990; 322:161-165

21. Bozzete, SA, Finkelstein DM, Spector SA, et al. A randomized trial of three antipneumocystis agents in patients with advanced human immunodeficiency virus infection. N Engl J Med 199;532:693-699.

22. Masur H. Prevention and treatment of Pneumocystis penuemonia. N Engl J Med 1992;327: 1853-1860.

23. Maclean Smith R, Iwamoto GK, Richerson HB, Flaherty JP. Trimethroprim sulfamethoxazole desentization in the acquired immune deficency syndrome. Ann Intern Med 1987;106:335

24. White MV, Haddad ZH, Prunner E, Sainz C. Desensitizatino to trimethroprim-sulfamethoxazole in patients with acquired immunodeficiency syndrome and Pneumocystis pneumonia. Ann Allergy 1989; 62:177-179.

25. Sattler Fr, Cowan R, Nielsen DM, et al. Trimethoprim-sulfamethoxazole compared with pentamidine for treatment of Pneumocystis carinii pneumonia in the acquired immondeficiency syndrome: a prospective, non crossover study. Ann Intern Med 1988;109:280-287.

26. Medina I, Mills J, Leoung G, et al.. Oral therapy for Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome: a controlled trial of trimetroprim-sulfamethoxazole versus trimethorpim-dapson. N Engl J Med 1990; 323:776-782.

27. Hughes W, Leoung G, Kramer F, et al. Comparison of atovaquone (566C89) and trimethroprim-sulfamethoxazole for the treatment fo Pneumocystis carinii pneumonia in patients with the acquired immonodeficiency syndrome (AIDS) N Engl J Med 1993;328:1521-1527.

28. Bozzette SA, Sattler FR, Chiu J, et al. A controlled trial of early adjunctive treatment with cortixosteroids for PCP in the acquired immondeficiencys syndrome. N Engl J Med 1990;323:1451-1457.

29. Montaner JSG, Lqwson LM, Levitt N, et al. Corticosteroids prevent early deterioration in patients with moderately severe Pneumocystisis carinii pneumonia and the acquired immunodeficiencys syndrome. Ann Intern Med 1990;113:14-20.

30. Porter S, Sande M.A. Toxoplasmosis of the central nervous system in the acquired immunodeficiency syndrome. N engl J Med 1992;326:1226.

31. Katlama C, De Wit S, Guichard A, et al. A ramdomized European trial comparing pyrimethamine-clindamycin to pyrimethamine-sulfadiazine. In: Proceedings of the 32nd Interscience Conference on Antimicrobial Agents and Chemotherapy. Anaheim, CA, 1992 Oct 12-14. Abstract 1215.

32. Masur H. and the Public Health Sevice Task Force on Prophylaxis and Therapy for Mycobacterium avium complex.. Recommendations on prophylaxis and therapy for disseminated Mycobacterim avium complex disease in patients infected with the human immunodeficiency virus. N Engl J Med. 329:898-904.

33. Nightingale SD, Cameron DW, Gordin FM, et al. 1993. Two controlled trials of rifabutin prophylaxis against Mycobacterium avium complex infection in AIDS. N Engl J Med 1993; 329:828-833.

34. Studies of Ocular Complications of AIDS Reseach Group, in Collaboration with the AIDS Clinical Trials Group. Mortality in patients with the acquired immunodeficiency syndrome treated with either foscarnet or ganciclovir for cytomegalovirus retinitis. N Engl J Med 1992;326: 213-220.

35. Kenneth A. Freedberg , JL Alpher, GR Seage, E. Losina, MC Weinsteri, DE Craven, AD Paltiel. Cost-effectiveness of preventing AIDS-related opportunistic infections. XI International Conference on AIDS. Vancouver, 1996. (AbstractTuB415).

36. Dunne MW, Havlir D, Dube M, Sattrler F, Forthal D, Kemper C, McCutchan A, the 066-174 study investigators. Prevention of Pneumocystis carinii pneumonia with azithromycin. XI International Conference on AIDS. Vancouver, 1996. (Abstract Tu B 410).

37. Duchin Jeffrey S, Sohlberg B, Buskin S, Hopkins S, Simon P. Risk factors for Pneumocystis carinii pneumonia:Delayed diagnosis of HIV-infection and failure to receive prophylactic therapy. XI International Conference on AIDS. Vancouver, 1996. (Abstract TuB 114).

38. Flepp Markus, Ledergerber B, Schenker C, Egger M, Gebhardt M, Luthy R, the Swiss HIV cohort study (SHCS). Pneumocystis carinii pneumoniaas first AIDS indicator disease in the era of primary prophylaxis. Why does it still occur? XI International Conference on AIDS. Vancouver, 1996. (Abstract TuB 113).

39. Curtis J. Randall, Horner RD, Bennett CL. Survival from intensive care for patients with HIV-related Pneumocystis carinii pneumonia (PCP). XI International Conference on AIDS. Vancouver, 1996. (AbstractTuB111).

40. Atzori C, Agostoni F, Zambelli A, Cargnel Antonietta. P. Carinii DNA detected by ITSs nested PCR in orofpharingeal and blood samples of HIV pts with PCP. XI International Conference on AIDS. Vancouver, 1996. (Abstract TuB115).

41. Brosgart Carol, Craig C, Hillman D, Louis TA, Alston B, Fisher E., El-Sadr W, the Terry Beirn Community Programs for Clinical Research on AIDS (CPCRA). Final results from a randomized, placebo-controlled trial of the safety and efficacy of oral ganciclovir for prophylaxis of CMV retinal and gastrointestinal mucosal disease. XI International Conference on AIDS. Vancouver, 1996. (Abstract THB301).

42. Spector Stephen A, Pilcher M, Lamy P, Hsia K, Wong R, Stempien MJ, the Roche Cooperative Oral Ganciclovir study group. XI International Conference on AIDS. Vancouver, 1996. (Abstract Th B302).

43. Mazzulli Tony, Wood S, Chua R, Phillips A, Fong IW, Rachlis A, Krajden M, Mortimer C, Clark J, Walmslley S, Prospective evaluation of cytomegalovirus (CMV) antigenemia and DNA hybridization assays in HIV infected persons at risk for CMV disease. XI International Conference on AIDS. Vancouver, 1996. (AbstractTh A392).

44. May Th, Vincent V, Brel F, Beuscart CI, Perronne C, Doco-Lecompte T, Dautzenberg B, Grosset J. CURAVIUM group. The French randomized clinical trial of combination therapy with clarithromycin for MAC bacteremia in AIDS patients: Final results. XI International Conference on AIDS. Vancouver, 1996. (Abstract WeB422).

45. Shafran SD, Singer Joeal, Zarowny DP, Phillips P, Salit Y, Walmsley S, Fong Y, Gill J. Rifabutin, ethambutol and claritrhromycin is superior to rifampin, ethambutol, clofazimine and ciprofloxacin fro MAC bacteremia (CTN 10). XI International Conference on AIDS. Vancouver, 1996. (Abstract WeB423).

46. Cohn DL, Benson CA, Williams P, Nevin T, Korvick J, Hafner R, Bourland D, Kopek E, et al. ACTG 196/CPCRA 009. Protocol Team. A prospective, randomized, doubleblind, comparative study of the safety and efficacy of clarithromycin vs rifabutin vs the combination for the prevention of mycobacteriumavium complex bacteremia or disseminated MAC disease in HIV-infected patients with CD4 counts<100 cells/mm3. XI International Conference on AIDS. Vancouver, 1996. (Abstract WeB 421).

47. Gonzalez Montaner LJ, Palmero D, Alberti F, Ambroggi M, Gonzalez Montaner PJ, Abbate E. Nosocomial outbreak of multi-drug resistant tuberculosis (MDR-TB) among AIDS patients in Buenos Aires, Argentina. XI International Conference on AIDS. Vancouver, 1996. (Abstract We B304).

48. Easterbrook Phillippa J, Bell A., Hannan M, Hayward A, Troop M, Shave A, Nelson M, Hawkins D, Gazzard BG, Azadian B, Lau YK. Nosocomial outbreak of multidrug resistant tuberculosis in a London HIV unit: outbreak investigation and clinical follow-up. XI International Conference on AIDS. Vancouver, 1996. (Abstract WeB305).

49. Briner David, Meister Th, Luthy R, Wber R and the Swiss HIV cohort study (SHCS). Clinical manifestations, extraintestinal complications and long term course of intestinal Enterocytozoon bieneusi microsporidiosis. XI International Conference on AIDS. Vancouver, 1996. (Abstract MoB114 ).

50. Albrecht Helmut, Sobottka Y, Ziellmann M, Meyer S, Jackle S, Stellbrink HJ, greten H. Prospective evaluation of the clinical significance of intestinal microsporidiosis in 2 cohorts of HIV-infected patients. XI International Conference on AIDS. Vancouver, 1996. (Abstract MoB 115).

51. Volberding, PA, Lagakos SW, Koch Ma, et la. Zidovudine in asymptomatic human immunodeficiency virus infection. N Engl J Med 1990;322:941-949.

52. Hamilton JD, Hartigan PM, Simberkoff MS, et al. A controlled trial of early versus late treatmendt with zidovudine in symptomatic human immunodeciency virus infection. N Engl J Med 1992;326: 437-443.

53. Fischl MA, Richman DD, Grieco MH, et al. The efficacy of azidothymidine (AZT) in the treatment of patients with AIDS and AIDS-related complex.N Engl J Med 1987;317:185-191.

54. Kahn J, Lagakos SE, Richman DD, et al. A controlled trial comparing continued zidovudine with didanosine in human immunodeficiency virus infection. N Engl J Med 1992;327:581-587.

55. Myers Maureen W, Julio G. Montaner . The Incas Study Gropu. A randomized, doubel blinded comparative trial of the effects of zidovudine, didanosine and nevirapine combinations in antiviral naive, AIDS free HIV infectedc patients with CD4 counts 200-600/mm3. XI International Conference on AIDS. Vancouver, 1996. (Abstract MoB 294).

56.Pollard Richard, Peterson D, Hardy D, Pedneault L, Rutkiewcz V, Pottage J, Murphy R, Gathe J, Beall G, Skovronski J, Cross A, Dunkle L. Stavudine (D4T) and Didanosine (DDI) combination therapy in HIV infected subjects: antiviral effect and safety in an on-going pilot randomized double-blinded trial. XI International Conference on AIDS. Vancouver, 1996. ( Abstract Th B293).

57. Biron F, Peyramond D, Lucht F, Fresard A, Nugier F, Vallet T, Grange J, Hamedi-Sangsari, Villa Jorge. Anti HIV activity of the combination of Didanosine and Hydroxiurea in HIV-1 infected individuals. XI International Conference on AIDS. Vancouver, 1996. (Abstract ThB 291)

58. Rae S, Montaner JSG, Raboud JM, Conway B, Szala C, Patenaude P, Shillington A, Predictors of response in a pilot study of Hidroxyurea as adjuvant therapy among patients with advanced HIV disease receiving Didanosine therapy. XI International Conference on AIDS. Vancouver, 1996. (Abstract Th B 292).

59. Gulick Roy, M, Mellors J, Havlir D, Eron J, Gonzalez C, McMahon D, Richman D, Vantine F, Rooney J, Jonas L, Meibohm A, Emini E, Chodakewitz J. Potent and sustained antiretroviral activity of indinavir, zidovudine, and lamivudine. XI International Conference on AIDS. Vancouver, 1996. (Abstract Th B931).

60. Markowitz Martin, Cao Y., Hurley A., O’Donovan R, Heath-Chiozzi M, Leonard J, Smiley L, Keller A, Johnson D, Johnson P, Ho DD. Triple therapy with AZT, 3TC and Ritonavir in 12 subjects newly infected with HIV-1. XI International Conference on AIDS. Vancouver, 1996. (Abstract ThB 933).

61. Cameron William, Sun E., Markowitz M, Farthing C, McMahon D, Poretz D, Cohen C, Follansbee S, Ho D, Mellors J, Hsu A, Grannemman GF, Maki R, Salgo M, Court J, Leonard J. Combination use of ritonavir and saquinavir in HIV-infected patients: preliminary safety and activity data. XI International Conference on AIDS. Vancouver, 1996. (Abstract ThB 934).

62. Perelson AS, Essunger P, Markowitz M, Ho David D. How long should treatment be given if we had an antiretroviral regimen that completely blocks HIV replication?. XI International Conference on AIDS. Vancouver, 1996. (Abstract ThB 930).