In HIV-1-infected patients who start highly active antiretroviral therapy (HAART), CD4 cell counts and plasma HIV-1 RNA concentration (viral load) at baseline are strongly associated with the risk of subsequent progression to AIDS or death.1-5 Once HAART has been initiated, these parameters are monitored regularly to assess patient response to therapy. The initial response is generally determined at 6 months.6 Accelerated approval of new drugs by licensing authorities is also based on HIV-1 RNA reductions at month 6.7,8 In most patients, HIV-1 RNA concentrations decrease and CD4 cell counts increase during the first months of HAART, but the degree of these initial virologic and immunologic responses varies widely among patients.
Several studies have addressed the prognostic significance of the early response to therapy. These were based on cohorts with limited follow-up, generally less than 3 years, and small numbers of clinical events.3,9-13 Most cohorts were primarily composed of antiretroviral-experienced patients, and, so far, no study has developed prognostic models separately for treatment-naive and pretreated patients. Furthermore, comparison of results between studies is hampered because investigators have used different approaches to modeling the immunologic and virologic responses to HAART.14 For example, some authors have investigated the prognostic significance of treatment-induced changes (differences from baseline)15 and others have used the absolute values at a given point in time,9 whereas we and still others have classified patients according to whether the response was complete (virologic and immunologic), incomplete (discordant virologic or immunologic only), or absent.10,16,17 It thus remains unclear at present which definitions of early immunologic and virologic responses to HAART are most strongly associated with clinical outcome in treatment-naive and pretreated patients.
Here, we identified from the literature definitions that have been used to characterize immunologic and virologic responses to HAART at 6 months. We then examined the relative prognostic value of these definitions using the French Hospital Database on HIV (FHDH), a large national cohort study,4,10 updating the data previously described10 and considering the long-term clinical progression of patients having initiated HAART as naive or antiretroviral-experienced patients.
The FHDH is a large nationwide cohort that is based in 68 hospitals across France.4 Standardized procedures and specific French Ministry of Health software are used by trained research assistants to prospectively collect clinical, laboratory, and treatment data from medical records. All participants provided informed consent for participation in the FHDH. For the present analysis (as well as for a previous analysis10), we selected all protease inhibitor-naive adult patients who started a HAART regimen with 1 protease inhibitor and 2 nucleoside reverse transcriptase (NRTI) inhibitors between July 1996 and March 1997. In addition, patients had to have CD4 cell counts and plasma HIV-1 RNA measurements recorded at the start (baseline) and after 6 months of HAART. Exclusion criteria were enrollment in a trial of antiretroviral therapy, infection with HIV-2, or treatment with nelfinavir or nonnucleoside reverse transcriptase inhibitors (NNRTIs), drugs that were used only in clinical trials at the time of enrollment.
From the literature, we identified 3 categories of definitions for immunologic and virologic responses: (1) defined changes from baseline in the CD4 cell count (eg, ≥50 cells/μL gained) and plasma HIV-1 RNA level (eg, ≥1 log10 reduction)15; (2) CD4 cell count and HIV-1 RNA load levels attained at 6 months, expressed as continuous values or as categories (thresholds) (eg, ≥200 CD4 cells/μL or undetectable HIV-1 RNA9); and (3) classification of patients in groups with complete, incomplete (discordant), or absent virologic and immunologic responses based on changes from baseline10,17 or thresholds at 6 months.3
Based on these categories, we compared the 6 following models. Model 1 considered only the CD4 cell count and plasma HIV-1 RNA level at baseline. Model 2 defined the immunologic response at 6 months as an increase of at least 50 CD4 cells/μL and the virologic response as a decline of plasma HIV-1 RNA level of at least 1 log10 or a plasma HIV-1 RNA level of less than 1000 copies/mL at 6 months. The threshold of 1000 copies/mL was chosen to overcome the heterogeneity in detection limits of assays used to quantify plasma HIV-1 RNA in the different hospitals.18 In models 3 and 4, the CD4 cell count and plasma HIV-1 RNA level attained at 6 months were used as log-transformed continuous variables or as 4 categories (≥350, 200-349, 100-199, and <100 CD4 cells/μL and <3, 3-3.9, 4-4.9, and ≥5 log10 HIV-1 RNA copies/mL).12 In model 5, the response was classified as complete, immunologic response only, virologic response only, and absent based on whether 50 or more CD4 cells/μL had been gained and whether the HIV-1 RNA level declined by at least 1 log10 or became undetectable (<1000 copies/mL). Model 6 defined the same categories based on whether or not the CD4 cell count at 6 months was at least 200 cells/μL and the plasma HIV-1 RNA was less than 3 log10 copies/mL.
Clinical progression was defined as new AIDS-defining events or death that occurred later than 6 months after the initiation of HAART. We included deaths from all causes and used the clinical part of the 1993 Centers for Disease Control and Prevention revision of the AIDS case definition (ie, people without an AIDS-defining disease but with a CD4 cell count less than 200 cells/μL were not classified as having AIDS).19 In patients with a previous AIDS diagnosis, we considered only new AIDS-defining events and did not consider recurrences.
Survival patterns were displayed using Kaplan-Meier plots of cumulative survival probabilities, and hazard ratios (HRs) for the association between prognostic factors and clinical outcomes were estimated using Cox proportional hazards models. Time was measured from 6 months after starting HAART to the date the endpoint occurred; the date of the most recent follow-up visit; or January 1, 2002. Patients seen event-free in the 6 months before the date of the last database update (clinic visit between June 2001 and January 2002) were censored on January 1, 2002. Patients with no follow-up visit in these 6 months were censored at the date of their last visit.
All models were adjusted for characteristics at baseline, including age, clinical stage (with AIDS or free of AIDS), transmission group, baseline CD4 cell count, and plasma HIV-1 RNA level. Baseline CD4 cell counts and plasma HIV-1 RNA levels were modeled as continuous variables, using log2 transformations for CD4 counts and log10 transformations for HIV-1 RNA levels. Models were stratified on starting period, at 3-month intervals. Separate analyses were conducted according to patients' previous antiretroviral therapy experience (ie, antiretroviral therapy naive or pretreated). In all analyses, we used an “intent-to-continue-treatment” approach and thus ignored subsequent changes of treatment, including treatment interruptions and terminations.
We compared the strength of the association between various definitions of immunologic and virologic responses at 6 months and subsequent clinical progression in multivariate Cox regression models separately for treatment-naive and experienced patients. We used an information-theoretic approach based on Akaike's Information Criteria (AIC) to identify the best approximating model.20 The model with the smallest AIC is considered to be closest to the reality that generated the data and should be selected for making inferences. According to an empiric rule suggested by Burnham and Anderson,20 a model for which differences in the AIC are lower than 2 should be most useful, whereas those with differences from 4 to 7 are less supported by the data and those with a difference in the AIC greater than 10 could be omitted from further consideration. In addition, for each definition and model, we compared the range in the risk of clinical progression at 5 years, using Kaplan-Meier life table estimates, from the lowest to the highest category of risk.
Statistical analyses were performed using SAS software, version 8 (SAS Institute, Cary, NC).
The patients included in the present analysis are the same than those who were included in a previous study.10 The description of the cohort at baseline and 6 months is provided in Table 1. The only differences between data presented in the present paper and those previously reported are the longer duration of follow-up and, consequently, the larger number of clinical outcomes (also shown in Table 1). In brief, a total of 2236 protease inhibitor-naive patients were followed for a median time of 58 months (8882 person-years) after initiation of HAART. The rate of dropouts was estimated as 6.4 per 100 person-years. During follow-up, 325 patients experienced a new AIDS-defining event and 161 died, of whom 64 had experienced a new AIDS-defining event. The incidence of new AIDS-defining events or death was 3.37 per 100 person-years among treatment-naive patients and 3.65 per 100 person-years among pretreated patients, for an incidence rate ratio of 0.90 (95% confidence interval [CI]: 0.69-1.18).
The results from the different Cox models with the corresponding AIC values are shown in Table 2. AIDS at baseline, older age, and intravenous drug use were significantly associated with a more rapid clinical progression (not shown). HRs for progression to AIDS or death according to CD4 cell count and viral load at baseline and 6 months are presented. In pretreated and treatment-naive patients, the reduced values for the AIC demonstrated that all models including information on CD4 cell count and plasma HIV-1 RNA level at 6 months are superior to model 1. Model 3, which included 6-month CD4 cell count and plasma HIV-1 RNA level as continuous variables, had the lowest AIC in both patient groups. Model 4, which considered 6-month CD4 cell count and plasma HIV-1 RNA level in 4 categories, also had a low AIC value, although in naive patients, the difference in AIC between models 3 and 4 reached 10. The other 2 models, which considered the completeness of the response, were less adequate. Interestingly, in models 3 and 4, neither the CD4 cell count nor the plasma HIV-1 RNA level at baseline was significantly associated with clinical progression in treatment-naive patients, whereas in pretreated patients, the viral load but not the CD4 cell count at baseline remained strongly associated with clinical progression. The risk of clinical progression increased progressively with lower 6-month CD4 cell counts and higher 6-month plasma HIV-1 RNA levels. No difference was demonstrated between patients who exhibited a plasma HIV-1 RNA level at 6 months <1000 copies/mL and those whose plasma HIV-1 RNA level was less than 10,000 copies/mL, however.
Kaplan-Meier estimates of the probability of clinical progression are shown in Figure 1 for pretreated patients stratified by CD4 cell count and plasma HIV-1 RNA levels at 6 months; they are presented separately for patients who had viral loads at baseline ≥10,000 copies/mL (≥4 log10) and patients who had lower HIV-1 RNA levels at baseline. The number of events was too small to allow a similar analysis for treatment-naive patients. Treatment-experienced patients with 6-month viral loads of 3 to 3.9 log10 and 4 to 4.9 log10 were combined in a single group, because HRs were similar in these patients (see Table 2). It is emphasized in Figure 1 that the prognosis in pretreated patients was different according to baseline plasma HIV-1 RNA.
At 5 years and for the entire study population, the range in the risk of clinical progression from the lowest to the highest category of risk (Table 3) was wider in model 4 than in other models, again indicating that model 4 better discriminates the data than other models. The risk of clinical progression ranged from 7% in patients whose CD4 count at 6 months was ≥350 cells/μL and whose HIV-1 RNA level was <3 log10 to 63% in patients whose CD4 count at 6 months was <100 cells/μL and whose HIV-1 RNA level was ≥5 log10 (model 4), whereas the risk ranged roughly from 10% to 30% in other models.
In the present study, we aimed to determine the importance of the initial immunologic and virologic responses to HAART as a determinant of clinical progression over 5 years in a large cohort of patients with HIV-1 infection who started protease inhibitor-based HAART. Because existing definitions of immunologic and virologic responses vary, we compared different prognostic models that were based on these definitions. We found that the initial response was prognostic of the further clinical outcome at 5 years and that the initial response, defined as CD4 cell count and viral load attained after 6 months of HAART, was most strongly associated with subsequent clinical progression. When considering naive and treatment-experienced patients together according to viral load and CD4 cell count at 6 months, the probability of clinical progression at 5 years ranged from 7% in patients in the lowest risk stratum to 63% in patients in the group at highest risk. Finally, in patients who had previously received antiretroviral therapy but not in treatment-naive patients, the viral load at the time of starting therapy was also of prognostic importance.
We and others in the Antiretroviral Therapy Cohort Collaboration have recently described clinical progression according to initial virologic and immunologic responses in a large sample of treatment-naive patients from different cohort studies.21 We found that baseline CD4 count and viral load were no longer of prognostic relevance once the 6-month values had been taken into account. The present study confirms these results with a longer follow-up in naive patients and demonstrates that prognosis is also affected by plasma HIV-1 RNA at baseline in treatment-experienced patients. This may reflect the impact of mutations conferring resistance to nucleoside reverse transcriptase inhibitors in patients who previously received only an NRTI regimen and not HAART. The prognostic impact of pretreatment viral load in patients pretreated with HAART may be somewhat different, because the number of potentially active antiretroviral drugs left may be lower in patients having received drugs from 2 or 3 of the main antiretroviral classes available. One may assume, however, that the situation of patients pretreated by NRTI dual therapy would be closer to the situation of patients initiating second-line HAART, particularly for patients who had failed a first-line regimen combining 2 NRTIs and 1 NNRTI, who also harbor resistance to NNRTIs. This needs to be verified by further studies.
Our results have important implications for the management of antiretroviral treatment-experienced patients who start HAART. Prognosis depends on the CD4 cell count at 6 months rather than on the number of CD4 cells gained. Baseline and 6-month plasma HIV-1 RNA levels are of prognostic importance. The prognosis of a patient with a viral load of less than 1000 copies/mL at 6 months and a viral load at baseline equal to or greater than 10,000 copies/mL is worse, on average, compared with that of a patient who also reached a viral load of less than 1000 copies/mL but whose baseline viral load was less than 10,000 copies/mL. These results suggest that individuals on monotherapy or dual therapy or patients with a history of such therapy should start HAART before HIV-1 RNA concentrations reach 10,000 copies/mL.
In a previous report,10 when classifying patients as complete, incomplete (discordant), or absent virologic and immunologic responders, we found no significant difference on clinical progression after 24 months of follow-up between immunologic responders only and complete responders, whereas after a median follow-up of 58 months, the difference became significant (HR = 1.77, 95% CI: 1.25-2.50), in accordance with results from others,11 although the value of the HR has not changed much (1.65 vs. 1.77). Similar results were observed when model 3 was run on follow-up data that had not been updated. This result is mainly attributable to the power gained by the extended follow-up and the increased number of clinical events. In fact, with longer follow-up, there seems to be little interaction between initial immunologic and virologic responses. Immunologic response is directly related to the risk of opportunistic events. Indeed, after 4 months of successful HAART, each 50-cell/μL increase in the CD4 count was associated with a 60% reduction in the incidence of opportunistic infections.22 The independent impact of early virologic response on clinical outcome is of particular interest, because it probably emphasizes the essential roles of adherence to treatment, treatment absorption, and the presence or absence of relevant resistance mutations, which are 3 factors known to be among the main determinants of early virologic response.
Interestingly, there was no significant difference in clinical outcome between patients whose viral load at 6 months was less than 1000 copies/mL and patients with a viral load greater than 1000 copies/mL but less than 10,000 copies/mL. Clearly, over 5 years of follow-up, even a partial virologic response is beneficial. Findings from the Antiproteases Cohorte (APROCO) cohort suggest that only when viral replication rebounds to more than 10,000 copies/mL does a decrease in the CD4 cell count follow.23 A partial virologic response to HAART could thus be defined as a viral load at 6 months of between 1000 and 10,000 copies/mL.24 In protease inhibitor-treated patients, who have only a few treatment options left because of drug intolerance or resistance, delaying changes to other regimens until the viral load increases to greater than 10,000 copies/mL may thus be reasonable, although there is clearly a risk of selecting mutant resistant viruses.25-27 Some preliminary results28 have shown that in patients with persistent low viremia (<10,000 copies/mL), protease inhibitor mutations were found in less than one third of the patients. It would be of particular interest to compare the rate of viral mutations according to levels of persistent viral replication (<1000 copies/mL vs. 1000-10,000 copies/mL) and the subsequent virologic response if the treatment regimen is modified.
The fact that the CD4 cell response is strongly associated with clinical outcome in treatment-naive and treatment-experienced patients has important implications for clinical trials. The recommendations of the European and American licensing authorities on efficacy endpoints are exclusively based on virologic outcomes at 6 months.7,8 The results from this and other studies21 indicate that immunologic and virologic responses are important when assessing the efficacy of antiretroviral drugs and that both should be considered as independent efficacy endpoints.
Despite the fact that our analyses were based on the intent-to-continue-treatment principle, which ignores modifications or interruptions of HAART during follow-up, the initial response was strongly associated with clinical outcome up to 5 years after initiation of HAART. Maintenance of a durable response is certainly a primary goal of therapy in patients treated with HAART. The association between the 6-month response and the long-term outcome could possibly be driven by the fact that patients who have stronger initial immunologic and virologic responses also have more durable responses.29 Nevertheless, our results indicate that the time at which therapy is started and the nature of the initial regimen should be carefully considered and that the motivation of patients and caregivers should be maximized to ensure the best adherence to therapy possible,30 because the long-term prognosis seems to be highly related to the early success of the first regimen.
The plasma HIV-1 RNA level and CD4 cell count at 6 months should be taken into account independently when assessing the early response of patients to antiretroviral therapy. The persistent impact of an early response on clinical progression at 5 years emphasizes the major importance of success of first-line HAART.
The authors are grateful to all participants of the FHDH and to the research assistants, without whom this work would not have been possible.
1. Egger M, May M, Chene G, et al. Prognosis
of HIV-1-infected patients starting highly active antiretroviral therapy
: a collaborative analysis of prospective studies. Lancet
2. HIV Surrogate Marker Collaborative Group. Human immunodeficiency virus type 1 RNA level and CD4 count as prognostic markers and surrogate endpoints: a meta-analysis. AIDS Res Hum Retroviruses
3. Ledergerber B, Egger M, Erard V, et al. AIDS-related opportunistic illnesses occurring after initiation of potent antiretroviral therapy
4. Grabar S, Pradier C, Le Corfec E, et al. Factors associated with clinical and virological failure in patients receiving a triple therapy including a protease inhibitor. AIDS
5. Miller V, Sabin CA, Phillips AN, et al. The impact of protease inhibitor-containing highly active antiretroviral therapy
on progression of HIV disease and its relationship to CD4 and viral load
6. National Institutes of Health. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Available at: http://AIDSinfo.nih.gov
. Accessed 2004.
7. Committee for Proprietary Medicinal Products. Note for Guidance on the Clinical Development of Medicinal Products for Treatment of HIV Infection
. London: EMEA; 2003.
8. US Food and Drug Administration. Guidance for Industry. Antiretroviral Drugs Using Plasma HIV RNA Measurements. Clinical Considerations for Accelerated and Traditional Approval
. US Food and Drug Administration; Rockville, MD, 2002.
9. d'Arminio Monforte A, Testori V, Adorni F, et al. CD4 cell counts at the third month of HAART may predict clinical failure. AIDS
10. Grabar S, Moing VL, Goujard C, et al. Clinical outcome of patients with HIV-1 infection according to immunologic and virologic response after 6 months of highly active antiretroviral therapy
. Ann Intern Med
11. Piketty C, Weiss L, Thomas F, et al. Long-term clinical outcome of human immunodeficiency virus-infected patients with discordant immunologic and virologic responses to a protease inhibitor-containing regimen. J Infect Dis
12. Chene G, Sterne JA, May M, et al. Prognostic importance of initial response in HIV-1 infected patients starting potent antiretroviral therapy
: analysis of prospective studies. Lancet
13. Williams PL, Currier JS, Swindells S. Joint effects of HIV-1 RNA levels and CD4 lymphocyte cells on the risk of specific opportunistic infections. AIDS
14. Thiebaut R, Thiessard F, Merchadou LD, et al. Antiretroviral combination therapy and HIV infection
. Long term follow up of patients under triple therapy is necessary. BMJ
15. Mezzaroma I, Carlesimo M, Pinter E, et al. Clinical and immunologic response without decrease in virus load in patients with AIDS after 24 months of highly active antiretroviral therapy
. Clin Infect Dis
16. Kaufmann D, Pantaleo G, Sudre P, et al. CD4-cell count in HIV-1-infected individuals remaining viraemic with highly active antiretroviral therapy
(HAART) [letter]. Lancet
17. Piketty C, Castiel P, Belec L, et al. Discrepant responses to triple combination antiretroviral therapy
in advanced HIV disease. AIDS
18. Le Moing V, Chêne G, Masquelier B, et al. Definition of virological response and the type of assay used for quantification of viral load
may influence the proportion of responders to antiretroviral therapy
[letter]. J Acquir Immune Defic Syndr
19. Centers for Disease Control and Prevention. 1993 revised classification system for HIV infection
and expanded surveillance case definition for AIDS among adolescents and adults. Morb Mortal Wkly Rep
20. Burnham K, Anderson D. Model selection and Inference. A Practical Information-Theoretic Approach
. New York: Springer-Verlag; 1998.
21. The ART Cohort Collaboration. Prognostic importance of initial response in HIV-1 infected patients starting potent antiretroviral therapy
: analysis of prospective studies. Lancet
22. Binquet C, Chene G, Jacqmin-Gadda H, et al. Modeling changes in CD4-positive T-lymphocyte counts after the start of highly active antiretroviral therapy
and the relation with risk of opportunistic infections: the Aquitaine Cohort, 1996-1997. Am J Epidemiol
23. Le Moing V, Thiebaut R, Chene G, et al. Predictors of long-term increase in CD4(+) cell counts in human immunodeficiency virus-infected patients receiving a protease inhibitor-containing antiretroviral regimen. J Infect Dis
24. Abgrall S, Duval X, Joly V, et al. Clinical and immunologic outcome in patients with human immunodeficiency virus infection, according to virologic efficacy in the year after virus undetectability, during antiretroviral therapy
. Clin Infect Dis
25. Deeks SG. Treatment of antiretroviral-drug-resistant HIV-1 infection. Lancet
26. Havlir DV, Hellmann NS, Petropoulos CJ, et al. Drug susceptibility in HIV infection
after viral rebound in patients receiving indinavir-containing regimens. JAMA
27. Belec L, Piketty C, Si Mohamed A, et al. High levels of drug-resistant HIV variants in patients exhibiting increasing CD4 T cell counts despite virological failure of protease inhibitor-containing antiretroviral combination therapy. J Infect Dis
28. Duval X, Darmon S, Longuet P, et al. Evolution of protease and reverse transcriptase inhibitor-associated mutations in HIV-1 protease inhibitor-treated patients with persistent low viremia. Presented at: 10th Conference on Retroviruses and Opportunistic infections; 2003; Boston.
29. Palella FJ, Jr, Chmiel JS, Moorman AC, et al. Durability and predictors of success of highly active antiretroviral therapy
for ambulatory HIV-infected patients. AIDS
30. Carrieri MP, Raffi F, Lewden C, et al. Impact of early versus late adherence to highly active antiretroviral therapy
on immuno-virological response: a 3-year follow-up study. Antivir Ther
Clinical Epidemiology Group of the French Hospital Database on HIV
- Scientific Committee: E. Billaud, F. Boué, D. Costagliola, X. Duval, C. Duvivier, P. Enel, S. Fournier, J. Gasnault, C. Gaud, J. Gilquin, S. Grabar, M. A. Khuong, J. M. Lang, M. Mary-Krause, S. Matheron, M. C. Meyohas, G. Pialoux, I. Poizot-Martin, C. Pradier, E. Rouveix, D. Salmon-Ceron, A. Sobel, P. Tattevin, H. Tissot-Dupont, and Y. Yasdanpanah
- DMI2 coordinating center: French Ministry of Health (E. Aronica, V. Tirard-Fleury, and I. Tortay)
- Statistical analysis center: INSERM U720 (S. Abgrall, D. Costagliola, S. Grabar, M. Guiguet, E. Lanoy, H. Leneman, L. Lièvre, M. Mary-Krause, V. Potard, and S. Saidi)
- Centres d'Information et de Soiur de l'immunodéficience humaine (CISIH), Paris area: CISIH de Bichat-Claude Bernard (Hôpital Bichat-Claude Bernard: S. Matheron, J. L. Vildé, C. Leport, P. Yeni, E. Bouvet, C. Gaudebout, B. Crickx, C. Picard-Dahan), CISIH de Paris-Center Ouest (Hôpital Européen Georges Pompidou: L. Weiss, D. and Tisne-Dessus; G. H. Tarnier-Cochin: D. Sicard and D. Salmon; Hôpital Saint-Joseph: J. Gilquin and I. Auperin; Hôpital Necker adultes: J. P. Viard and L. Roudière), CISIH de Paris-Sud (Hôpital Antoine Béclère: F. Boué and R. Fior; Hôpital de Bicêtre: J. F. Delfraissy and C. Goujard; Hôpital Henri Mondor: Ph. Lesprit and C. Jung; Hôpital Paul Brousse), CISIH de Paris-Est (Hôpital Saint-Antoine: M. C. Meyohas, J. L. Meynard, O. Picard, and N. Desplanque; Hôpital Tenon: J. Cadranel, C. Mayaud, G. Pialoux, and W. Rozenbaum), CISIH de Pitié-Salpétrière (GH Pitié-Salpétrière: F. Bricaire, C. Katlama, S. Herson, and A. Simon), CISIH de Saint-Louis (Hôpital Saint-Louis: J. M. Decazes, J. M. Molina, J. P. Clauvel, and L. Gerard; GH Lariboisière-Fernand Widal: P. Sellier and M. Diemer), CISIH 92 (Hôpital Ambroise Paré: C. Dupont, H. Berthé, and P. Saïag; Hôpital Louis Mourier: E. Mortier and C. Chandemerle; Hôpital Raymond Poincaré: P. de Truchis), and CISIH 93 (Hôpital Avicenne: M. Bentata and P. Honoré; Hôpital Jean Verdier: S. Tassi and V. Jeantils; Hôpital Delafontaine: D. Mechali and B. Taverne)
- CISIH, area outside Paris: CISIH Auvergne-Loire (CHU de Clermont-Ferrand: H. Laurichesse and F. Gourdon; CHRU de Saint-Etienne: F. Lucht and A. Fresard); CISIH de Bourgogne-Franche Comté (CHRU de Besançon; CHRU de Dijon; CH de Belfort: J. P. Faller and P. Eglinger; CHRU de Reims); CISIH de Caen (CHRU de Caen: C. Bazin and R. Verdon), CISIH de Grenoble (CHU de Grenoble), CISIH de Lyon (Hôpital de la Croix-Rousse: D. Peyramond and A. Boibieux; Hôpital Edouard Herriot: J. L. Touraine and J. M. Livrozet; Hôtel-Dieu: C. Trepo and L. Cotte), CISIH de Marseille (Hôpital de la Conception: I. Ravaux and H. Tissot-Dupont; Hôpital Houphouët-Boigny: J. P. Delmont and J. Moreau; Institut Paoli Calmettes: J. A. Gastaut; Hôpital Sainte-Marguerite: I. Poizot-Martin, J. Soubeyrand and F. Retornaz; CHG d'Aix-En-Provence: P. A. Blanc and T. Allegre; Center pénitentiaire des Baumettes: A. Galinier and J. M. Ruiz; CH d'Arles; CH d'Avignon: G. Lepeu; CH de Digne Les Bains: P. Granet-Brunello; CH de Gap: L. Pelissier and J. P. Esterni; CH de Martigues: M. Nezri and R. Cohen-Valensi; CHI de Toulon: A. Laffeuillade and S. Chadapaud), CISIH de Montpellier (CHU de Montpellier: J. Reynes; CHG de Nîmes), CISIH de Nancy (Hôpital de Brabois: T. May, C. Rabaud), CISIH de Nantes (CHRU de Nantes: F. Raffi and E. Billaud), CISIH de Nice (Hôpital Archet 1: C. Pradier and P. Pugliese; CHG Antibes Juan les Pins), CISIH de Rennes (CHU de Rennes: C. Michelet and C. Arvieux), CISIH de Rouen (CHRU de Rouen: F. Caron and F. Borsa-Lebas), CISIH de Strasbourg (CHRU de Strasbourg: J. M. Lang, D. Rey, and P. Fraisse; CH de Mulhouse), CISIH de Toulouse (CHU Purpan: P. Massip, L. Cuzin, E. Arlet-Suau, and M. F. Thiercelin Legrand; Hôpital la Grave; CHU Rangueil), CISIH de Tourcoing-Lille (CH Gustave Dron; CH de Tourcoing: Y. Yasdanpanah), and CISIH de Tours (CHRU de Tours; CHU Trousseau)
- CISIH overseas: CISIH de Guadeloupe (CHRU de Pointe-à-Pitre), CISIH de Guyane (CHG de Cayenne: M. Sobesky and R. Pradinaud), CISIH de Martinique (CHRU de Fort-de-France), and CISIH de La Réunion (CHD Félix Guyon: C. Gaud and M. Contant)
Keywords:© 2005 Lippincott Williams & Wilkins, Inc.
HIV infection; antiretroviral therapy; treatment outcome; CD4 lymphocyte count; viral load; prognosis