Share this article on:

Continued CD4 cell count increases in HIV-infected adults experiencing 4 years of viral suppression on antiretroviral therapy

Hunt, Peter Wa,b; Deeks, Steven Ga; Rodriguez, Benignoc; Valdez, Hernanc; Shade, Starley Ba,d; Abrams, Donald Ia,d; Kitahata, Mari Me; Krone, Melissaf; Neilands, Torsten Bf; Brand, Richard Jf,g; Lederman, Michael Mc; Martin, Jeffrey Na,f,g

Clinical Science

Objective: To determine the extent to which HIV-infected patients, including those with advanced immunodeficiency, can reverse peripheral CD4 T-cell depletion while maintaining long-term viral suppression on highly active antiretroviral therapy.

Design: Cohort study.

Participants: Four-hundred and twenty-three HIV-infected patients who initiated HAART prior to 1998 and achieved a viral load ≤ 1000 copies/ml by 48 weeks were evaluated for up to 4 years or until plasma HIV RNA levels increased to > 1000 copies/ml.

Main outcome measure: CD4 count changes.

Results: Among patients who maintained plasma HIV RNA levels ≤ 1000 copies/ml, CD4 counts continued to increase through year 4 of HAART. In the last year examined, from year 3 to 4 of HAART, mean CD4 count gains were +89 × 106, +86 × 106, +95 × 106, and +88 × 106/l in patients with pre-therapy CD4 counts of < 50 × 106, 50 × 106–199 × 106, 200 × 106–349 × 106, and ≥ 350 × 106/l, respectively (all gains were significantly greater than zero; P < 0.05). Among those with a pre-therapy CD4 count of < 50 × 106/l, 88% achieved a CD4 cell count of ≥ 200 × 106/l and 59% achieved a count of ≥ 350 × 106/l by year 4. Factors associated with increased CD4 cell count gains from month 3 to year 4 included lower pre-therapy CD4 cell count, younger age, female sex, and infrequent low-level viremia (versus sustained undetectable viremia).

Conclusions: Most patients who achieve and maintain viral suppression on HAART continue to experience CD4 T-cell gains through 4 years of therapy. The immune system's capacity for CD4 T lymphocyte restoration is not limited by low pre-therapy CD4 counts.

From the aPositive Health Program, San Francisco General Hospital, University of California, the bDivision of General Internal Medicine, Veterans Affairs Medical Center, San Francisco, California, the cCenter for AIDS Research, Case Western Reserve University, University Hospitals, Cleveland, Ohio, the dCommunity Consortium, San Francisco, California, the eDepartments of Medicine and Health Services, University of Washington, Seattle, Washington, the fCenter for AIDS Prevention Studies, University of California, and the gDepartment of Epidemiology and Biostatistics, University of California, San Francisco, California, USA.

Requests for reprints to: J. N. Martin, Center for AIDS Prevention Studies, Prevention Sciences Group, 74 New Montgomery Street, Suite 200, San Francisco, CA 94105, USA.

Received: 21 November 2002; revised: 11 March 2003; accepted: 18 March 2003.

Back to Top | Article Outline


Current Department of Health and Human Services guidelines cite that one potential risk of delaying highly active antiretroviral therapy (HAART) in HIV-infected patients is that the damage to the immune system observed in persons with low CD4 cell counts could be irreversible [1]. The evidence for this concern appears to come from several studies that show a plateau in CD4 T-cell gains after the second year of therapy [2–4]. However, these observations were based on patients with a wide variety of virologic responses to HAART. Since the extent of viral suppression is a powerful predictor of CD4 T-cell gains [5–8], these observations may underestimate the immune system's capacity for CD4 T-cell restoration. Two recent studies have reported on long-term CD4 T-cell trends in patients with sustained HAART-induced viral suppression, and both also suggested plateaus in CD4 T-cell gains after 2 years with failure to achieve a CD4 cell count > 500 × 106/l in the majority of patients with pre-therapy CD4 cell counts < 200 × 106/l [9,10]. However, a limited number of patients represented in the final year of observation may have led to an underestimation of the CD4 T-cell gains after year 3.

To evaluate the immune system's capacity for CD4 T-cell restoration on HAART, we observed a large number of patients who achieved and maintained viral suppression on HAART, including those starting with low CD4 counts, pooling data from four clinic-based cohorts. We observed CD4 count changes through 4 years of therapy while the plasma HIV RNA remained ≤ 1000 copies/ml. We also evaluated factors associated with CD4 T-cell gains to identify subsets of patients who may experience poor immunologic responses despite favorable virologic responses to HAART.

Back to Top | Article Outline


Study design

Using data collected from four clinic-based cohorts in San Francisco, Cleveland, and Seattle, we examined HIV-infected adults who initiated HAART and achieved a plasma HIV RNA level ≤ 1000 copies/ml by 48 weeks. Patients were observed for up to 4 years as long as they maintained a plasma HIV RNA level ≤ 1000 copies/ml.

Back to Top | Article Outline


The San Francisco General Hospital cohort is a study of HIV-infected patients who were started on protease inhibitor-based HAART before 1997 [11]. The Community Consortium cohort is a study of HIV-infected patients followed in several private practices and public clinics in the San Francisco Bay Area who had a plasma HIV RNA level > 1500 copies/ml prior to HAART and a plasma HIV RNA < 500 copies/ml after 120 days of therapy. The Cleveland cohort consists of all patients who receive care through the Case Western Reserve University Hospitals of Cleveland HIV clinic (John Carey Special Immunology Unit) [12]. The Seattle cohort consists of all HIV-infected patients followed at the University of Washington Harborview Medical Center HIV Clinic.

From these four cohorts, we selected all patients who initiated HAART prior to 1 January 1998, achieved a plasma HIV RNA level ≤ 1000 copies/ml by 48 weeks (± 8 weeks), and had a CD4 cell count available within the year prior to HAART initiation. HAART was defined as three or more antiretroviral drugs including either a protease inhibitor, a non-nucleoside reverse transcriptase inhibitor, or abacavir. Saquinavir hard gel-based regimens were considered HAART only if they were combined with ritonavir. Patients were excluded if they became pregnant or used hydroxyurea or interleukin-2 during the first 48 weeks of HAART. Patients may have had antiretroviral experience prior to initiating HAART. Each cohort was approved by the local institution's review board on human subjects research.

Back to Top | Article Outline


Demographic characteristics and medication data were collected by chart review at three cohort sites and from the University of Washington HIV Information System that captures clinical information including automated pharmacy and diagnosis data. Although clinical diagnoses were not uniformly available for each cohort, reported deaths were investigated for their cause by review of the medical chart and death certificate. CD4 cell counts, plasma HIV RNA measurements, and serum hepatitis C virus antibody determinations, collected during the course of clinical care, were performed by the clinical laboratories associated with each of the cohort sites. All pre-therapy plasma HIV RNA levels that were > 1000 copies/ml and measured by branched-chain DNA (bDNA) amplification version 1 or version 2 assays (Quantiplex® HIV RNA, Chiron Corporation, Emeryville, California, USA) were converted to bDNA version 3 equivalents by multiplying by a factor of 1.75 [13]. Pre-therapy HIV RNA measurements obtained by PCR or bDNA version 3 assays were left as reported. Plasma HIV RNA measurements after HAART initiation were not converted as the relationship between bDNA versions 1 and 2 to version 3 no longer holds below levels of 1000 copies/ml [14]. Pre-therapy CD4 cell count and plasma HIV RNA level were defined as the last available measurement prior to the initiation of HAART. These measurements occurred at a median of 0.6 months prior to HAART initiation (interquartile range: 0–1.5 months).

Back to Top | Article Outline

Statistical analysis

Eligible patients were followed for CD4 cell count changes over time. Observations were censored after the last plasma HIV RNA known to be ≤ 1000 copies/ml, at the first known date of pregnancy, or at the date of hydroxyurea or interleukin-2 use. Intermittent viremia < 1000 copies/ml was allowed so that the impact of low-level viremia on CD4 cell count changes could be assessed.

Several approaches were used to determine CD4 cell count changes. The first was to determine the median CD4 cell count 3 months after HAART initiation (within a 10-week window) and then at 6-month intervals (within a 3-month window) for groups of patients classified by their pre-therapy CD4 cell count (< 50 × 106, 50–199 × 106, 200–349 × 106, and ≥ 350 × 106/l). The median CD4 count 3 months prior to HAART initiation (within a 10-week window) was also determined for comparative purposes. Within CD4 T-cell count stratum and window, the medians were based on all patients with one or more measurements; for patients with multiple measurements in a window, the median for the patient was first used to get a single value in the window. This method is simple but susceptible to bias due to possible selective dropout of patients with minimal CD4 T-cell gains. The second approach used a normal linear mixed model (using PROC MIXED in SAS) employing square-root transformed CD4 cell counts to conform to model assumptions. A five-piece (0–3 months, 3–12 months, 1–2 years, 2–3 years, 3–4 years) segmented linear model was used in each pre-therapy CD4 cell count stratum to allow for changing slopes over time. Akaike information criterion model selection was used to choose the covariance model [15]. The between-subjects component allows for a random intercept and random slopes in all five time segments, common to all groups. The within-subjects component is a spatial exponential autocorrelation model with group-specific parameters. This method is relatively robust to bias due to loss to follow-up since it only requires the missing at random assumption, which allows the probability of loss to be related to past CD4 cell counts [16,17]. For interpretability, the predicted square root-transformed CD4 cell counts were back-transformed into the native unit scale (×106 cells/l) [18]. The final approach to evaluate CD4 T-cell changes over time was to determine the proportion of patients achieving two successive CD4 counts ≥ 200 × 106/l, ≥ 350 × 106/l, and ≥ 500 × 106/l using Kaplan–Meier analysis. Landmarks were considered achieved at the date of the first successive CD4 cell count at or above the cut-off.

Since the initial gains in peripheral CD4 cell counts are largely due to redistribution from lymphoid tissue [19–21], we analyzed separately factors associated with early and late CD4 T-cell gains. The early change in CD4 cell count (baseline to month 3) was calculated as the CD4 count at month 3 (the median of all values for a particular patient within a 10-week window) minus the pre-therapy CD4 cell count. The late change in CD4 cell count (month 3 to year 4) was calculated as the CD4 cell count at year 4 (the median of all values for a particular patient within a 3-month window) minus the CD4 count at month 3.

Factors associated with early CD4 T-cell changes and late changes were analyzed using linear regression. Age, sex, pre-therapy CD4 cell count, log10 pre-therapy HIV RNA level, hepatitis C serostatus, and cohort site were evaluated as potential predictors of the early CD4 T-cell changes. In addition to these factors, the month 3 CD4 cell count and the frequency of low-level viremia (the percentage of all HIV RNA measurements after week 48 that were detectable between 50 and 1000 copies/ml) were considered as potential factors associated with the CD4 T-cell change from month 3 to year 4.

Back to Top | Article Outline


Pre-therapy characteristics of patients and follow-up

A total of 423 patients achieved a plasma HIV RNA ≤ 1000 copies/ml by week 48 of HAART and were followed for long-term CD4 count changes. Approximately half were from San Francisco General Hospital, most were men between the ages of 30 and 50 years, and one-fifth were co-infected with hepatitis C virus (Table 1). A wide range of pre-therapy CD4 counts and plasma HIV RNA levels were represented. Through 4 years of HAART, 153 patients maintained a plasma HIV RNA ≤ 1000 copies/ml for the entire period, 178 patients were censored for a plasma HIV RNA > 1000 copies/ml (because of either breakthrough viremia on HAART or treatment interruption), two patients were censored for initiating hydroxyurea therapy, 15 patients were censored because of death (their last known plasma HIV RNA level was ≤ 1000 copies/ml; this determination occurred a median of 3 months prior to death; interquartile range, 2–7 months), and 75 patients were censored because of loss to follow-up. The 15 deaths occurring in patients whose last known plasma HIV RNA was ≤ 1000 copies/ml were attributed to lung cancer (four), cardiac causes (three), end-stage liver disease (two), unspecified pneumonia (one), and unknown causes (five). Among the 423 participants included in the analysis, the median number of CD4 cell count and plasma HIV RNA determinations per year prior to censoring was four (interquartile range, 3–5 for each measurement).

Back to Top | Article Outline

CD4 T-cell changes in response to HAART

Following an initial rapid increase in CD4 counts in the first 3 months of therapy, there continued to be a less marked, but nonetheless sustained increase in median CD4 counts through year 4, particularly in patients with pre-therapy CD4 counts < 350 × 106/l (Fig. 1). We next evaluated CD4 T-cell changes within each time interval and pre-therapy CD4 T-cell stratum using a mixed effects repeated measures model (Table 2). There were statistically significant increases in the CD4 cell count in all pre-therapy CD4 T-cell count strata ≤ 350 × 106/l in all time intervals, including from year 3 to year 4. For example, the group of patients with a pre-therapy CD4 cell count < 50 × 106/l had a mean gain of 89 × 106/l (P < 0.001 versus a gain of zero) between year 3 and year 4. Furthermore, there were no differences in CD4 cell count gains comparing the year 1 to year 2 interval to the year 3 to year 4 interval within each pre-therapy CD4 T-cell strata (P > 0.25 for each stratum). Lastly, there were no significant differences in CD4 cell count changes between year 3 and year 4 across the pre-therapy CD4 T-cell stratum (P = 0.99).

Among patients with pre-therapy CD4 cell counts < 350 × 106/l, the majority [76% by Kaplan–Meier estimation; 95% confidence interval (CI), 69–82%] was able to achieve a CD4 cell count > 350 × 106/l on two successive occasions after 4 years of sustained virologic suppression. By year 4 of HAART, 59% (95% CI, 45–74%) of patients with pre-therapy CD4 cell counts < 50 × 106/l, 70% (95% CI, 56–82%) of those with pre-therapy CD4 T-cell counts between 50 × 106 and 199 × 106/l, and 93% (95% CI, 85–98%) of those with pre-therapy CD4 T-cell counts between 200 × 106 and 350 × 106/l achieved a CD4 cell count > 350 × 106/l on two successive occasions (Fig. 2). Fig. 2 also shows that the proportion of patients with pre-therapy CD4 cell counts < 200 × 106/l who achieve a CD4 cell count > 350 × 106/l is continuing to increase as of year 4. Using other CD4 cell count landmarks, 88% (95% CI, 77–95%) of patients with a pre-therapy CD4 count < 50 × 106/l achieved a CD4 cell count of 200 × 106/l and 32% (95% CI, 21–49%) achieved a CD4 cell count of 500 × 106/l by year 4 of HAART.

To evaluate the possibility that the observed increases in CD4 cell counts were the result of selective drop-out of patients experiencing minimal CD4 T-cell gains, we compared the slopes of CD4 T-cell changes from month 3 to month 18 across groups of patients who were censored at different time points (Fig. 3). There was no evidence that patients who remained in the analysis at year 4 were selectively enriched for those experiencing the highest CD4 T-cell gains. Instead, there was a trend toward decreased CD4 T-cell slopes in those who remained at year 4.

Back to Top | Article Outline

Factors associated with CD4 T-cell changes

We next evaluated factors associated with early and late CD4 T-cell changes. The median increase in CD4 T-cell counts from pre-therapy to month 3 was 65 × 106/l (interquartile range, 22–128 × 106/l) and the median increase in CD4 T-cells from month 3 to year 4 was 273 × 106/l (interquartile range, 144–404 × 106/l). The CD4 T-cell changes from pre-therapy to month 3 were associated with the pre-therapy plasma HIV RNA level (+ 20 × 106/l per log10 copies/ml increase; P < 0.001), but not age, pre-therapy CD4 T-cell count, sex, or hepatitis C serostatus in a multivariable model (P > 0.10 for all associations).

Several factors were independently associated with CD4 T-cell count changes from month 3 to year 4. Of the 153 patients who maintained a plasma HIV RNA level ≤ 1000 copies/ml for 4 years, 135 had available CD4 cell count determinations at month 3 and year 4. Pre-therapy CD4 cell counts were < 50 × 106/l in 35, 50 × 106–199 × 106/l in 37, 200 × 106–349 × 106/l in 34, and ≥ 350 × 106/l in 29. Among these patients and after adjustment for other significant factors, every 100 × 106/l increase in the pre-therapy CD4 cell count was associated with 63 × 106 fewer CD4 T-cells/l (P = 0.005) gained from month 3 to year 4 (Table 3). Likewise, for every 10-year increase in age, there were 60 × 106 fewer CD4 T-cells/l (P = 0.02) gained from month 3 to year 4. Women gained a mean of 188 × 106 CD4 T-cells/l (P = 0.03) more than men. Moderate (3.5–4.5 log10 copies/ml) but not high (> 4.5 log10 copies/ml) pre-therapy plasma HIV RNA levels were associated with increased CD4 T-cell gains compared to patients with low pre-therapy plasma HIV RNA levels (< 3.5 log10 copies/ml). However, hepatitis C virus serostatus was not predictive. Interestingly, patients with infrequent low-level viremia between week 48 and year 4 (defined as those having between 1 and 20% of all plasma HIV RNA levels during this time between 50 and 1000 copies/ml) gained a mean of 134 × 106 more cells/l (P = 0.006) than patients who had no episodes of detectable viremia during this time. Patients with infrequent low-level viremia also gained a mean of 173 × 106 more cells/l (P = 0.005) than patients with frequent low-level viremia (defined as those having > 20% of all plasma HIV RNA levels during this time between 50 and 1000 copies/ml). Adjustment for total number of plasma HIV RNA determinations did not alter this association (data not shown). Finally, there was a trend toward an association between CD4 cell count at month 3 of HAART and subsequent CD4 T-cell gains (P = 0.07). Substituting the absolute change in CD4 cell count from baseline to month 3 for the month 3 CD4 cell count in the multivariable model did not change any of the adjusted associations and was also positively associated with subsequent CD4 T-cell gains (P = 0.004, data not shown).

Back to Top | Article Outline


The capacity for CD4 T-cell regeneration during long-term HAART has not been well defined. Specifically, there remains a concern that CD4 cell counts often reach a plateau after the first few years of viral suppression [4,9,10]. This concern has been put forth as one reason to consider administering HAART earlier rather than later in the course of HIV infection. To assess the immune system's capacity for CD4 T-cell restoration, we studied HIV-infected patients experiencing durable viral suppression on HAART and observed that most patients continue to experience significant CD4 cell count increases for at least 4 years. There was no evidence of a plateau in CD4 T-cell gains in the majority of patients with pre-therapy CD4 counts < 350 × 106 cells/l who maintained viral suppression ≤ 1000 copies/ml. Consequently, irreversible depletion of circulating CD4 T-cells appears to be uncommon in patients with advanced immunodeficiency as long as durable treatment-mediated viral suppression can be maintained.

Our results differ from those of Tarwater et al. who recently reported a plateau in CD4 T-cell gains after 2 years of HAART [4]. However, that study included patients with a wide variety of virologic responses to HAART, including those with high-level detectable viremia. Our results also differ from a recent report that found a plateau in CD4 T-cell gains after 3 years of HAART among patients who maintained viral suppression [9]. Since half of the patients in that study were followed for less than 45 months and the last available CD4 cell count was carried forward to month 48, the report of a plateau after year 3 is probably an underestimation of the true CD4 T-cell gains experienced by those patients. Lastly, our results differ from a recent report from the ACTG 375 study team, which also reported a plateau in CD4 T-cell gains after year 2 of suppressive HAART [10]. However, this study was limited by a small sample size and allowed for higher levels of viremia, which may have resulted in decreased observed CD4 T-cell gains.

While we present robust evidence that CD4 cell counts continue to increase through year 4 of suppressive HAART, it is notable that the rate of recovery is slow and that steady state has not yet been reached by 4 years. This slow rate of recovery may reflect the inherent slow repopulation rate of naive CD4 T-cells by the adult thymus as has been suggested by others [22–24]. Indeed, the rate of CD4 T-cell gains we observed after year 1 of HAART is comparable to the slow rate of naive CD4 T-cell increases experienced by HIV-uninfected adults in the first 2 years following intensive chemotherapy [25]. Alternatively, the slow rate of CD4 T-cell recovery may be the result of a countervailing HIV-induced process that either limits T-cell production or enhances destruction, such as ongoing T-cell activation despite viral suppression [26,27]. Since T-cell activation is independently associated with clinical progression in untreated HIV infection [28], a slow decay in levels of T-cell activation with long-term viral suppression might explain the slow rate of CD4 T-cell recovery.

We also determined several factors associated with HAART-mediated CD4 T-cell changes. Lower pre-therapy CD4 cell counts were associated with greater CD4 T-cell gains, an effect most prominent in the first 2 years of suppressive HAART. This suggests homeostatic mechanisms regulating CD4 T-cell synthesis or apoptosis that are sensitive to the degree of CD4 T-cell depletion. Alternatively, the redistributive phase of T-cells may be prolonged in patients with advanced pre-HAART immunodeficiency. Consistent with others’ findings [9,29,30], we also found that younger age predicted greater late CD4 T-cell gains on suppressive HAART, supporting the importance of thymic function (or its surrogate) in immune reconstitution. While few women were in our sample, the increased CD4 T-cell gains we observed in women are consistent with other recent reports [31–33], and may indicate an important sex difference in immune reconstitution that needs further exploration. That patients with infrequent low-level viremia have higher CD4 T-cell gains than patients with sustained undetectable plasma HIV RNA levels has recently been shown by another cohort [34]. The mechanism for this is not known, but conceivably, higher CD4 T-cell gains may provide a larger pool of infectable cells, thereby driving rather than being a consequence of detectable bursts of viral replication.

A potential limitation of this study is the possibility that the continued CD4 cell count gains we observed were due to selective drop-out of patients with the worst immunologic responses, thereby enriching the sample over time with patients with the best immunologic responses. However, there was a trend toward decreased CD4 T-cell changes from month 3 to month 18 in the patients who remained at year 4 compared to patients who were censored earlier. Furthermore, assuming that patients who leave the analysis would exhibit similar CD4 T-cell changes as those who stay in the analysis after controlling for prior changes, our use of random slopes and intercepts in the repeated measures model should prevent inferences being driven solely by those who remain in the analysis [16,17].

Despite the sustained CD4 T-cell increases over 4 years of HAART observed in this study, our results should not be interpreted as evidence supporting the delay of antiretroviral therapy in patients with CD4 cell counts < 200 × 106/l. Indeed, several cohort studies have reported an increased risk of progression to AIDS and death in patients initiating HAART at CD4 cell counts below < 200 × 106/l [35–37]. Patients with low pre-HAART nadirs may also have persistent defects in immune function despite achieving normal CD4 cell counts on suppressive HAART [38]. Moreover, we evaluated a selected group of patients who achieved and maintained long-term viral suppression. Since prior reports from clinic-based cohorts have found that half of HIV-infected patients experience virologic failure within the first year of HAART [11,39,40], our results likely overestimate the CD4 T-cell gains experienced by all patients initiating HAART. However, our results provide definitive evidence that the immune system is capable of slowly repopulating circulating CD4 T lymphocytes, even after advanced HIV-associated immune depletion, at least through 4 years of suppressive HAART.

Back to Top | Article Outline


Sponsorship: Supported in part by the University of California AIDS Research Center (CC99-SF-001), the UCSF-Gladstone Institute of Virology & Immunology Center for AIDS Research (P30 MH59037), the Case Western Reserve University Center for AIDS Research (P30 AI 36219), the NIH (RO1 AI052745), the Center for AIDS Prevention Studies (P30 MH62246) and the General Clinical Research Center at San Francisco General Hospital (5-MO1-RR00083-37).

Back to Top | Article Outline


1. DHHS and Kaiser Family Foundation Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. 2002.
2. Notermans DW, Pakker NG, Hamann D, Foudraine NA, Kauffmann RH, Meenhorst PL, et al. Immune reconstitution after 2 years of successful potent antiretroviral therapy in previously untreated human immunodeficiency virus type 1-infected adults. J Infect Dis 1999, 180:1050–1056.
3. Gulick RM, Mellors JW, Havlir D, Eron JJ, Meibohm A, Condra JH, et al. 3-year suppression of HIV viremia with indinavir, zidovudine, and lamivudine. Ann Intern Med 2000, 133:35–39.
4. Tarwater PM, Margolick JB, Jin J, Phair JP, Detels R, Rinaldo C, et al. Increase and plateau of CD4 T-cell counts in the 3½ years after initiation of potent antiretroviral therapy. J Acquir Immune Defic Syndr 2001, 27:168–175.
5. Staszewski S, Miller V, Sabin C, Schlecht C, Gute P, Stamm S, et al. Determinants of sustainable CD4 lymphocyte count increases in response to antiretroviral therapy. AIDS 1999, 13:951–956.
6. Wood E, Yip B, Hogg RS, Sherlock CH, Jahnke N, Harrigan RP, et al. Full suppression of viral load is needed to achieve an optimal CD4 cell count response among patients on triple drug antiretroviral therapy. AIDS 2000, 14:1955–1960.
7. Le Moing V, Thiebaut R, Chene G, Leport C, Cailleton V, Michelet C, 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 2002, 185:471–480.
8. Deeks SG, Barbour JD, Grant RM, Martin JN. Duration and predictors of CD4 T-cell gains in patients who continue combination therapy despite detectable plasma viremia. AIDS 2002, 16:201–207.
9. Kaufmann GR, Bloch M, Finlayson R, Zaunders J, Smith D, Cooper DA. The extent of HIV-1-related immunodeficiency and age predict the long- term CD4 T lymphocyte response to potent antiretroviral therapy. AIDS 2002, 16:359–367.
10. Valdez H, Connick E, Smith KY, Lederman MM, Bosch RJ, Kim RS, et al. Limited immune restoration after 3 years’ suppression of HIV-1 replication in patients with moderately advanced disease. AIDS 2002, 16:1859–1866.
11. Deeks SG, Hecht FM, Swanson M, Elbeik T, Loftus R, Cohen PT, et al. HIV RNA and CD4 cell count response to protease inhibitor therapy in an urban AIDS clinic: response to both initial and salvage therapy. AIDS 1999,13:F35–F43.
12. Valdez H, Chowdhry TK, Asaad R, Woolley IJ, Davis T, Davidson R, et al. Changing spectrum of mortality due to human immunodeficiency virus: analysis of 260 deaths during 1995–1999. Clin Infect Dis 2001, 32:1487–1493.
13. Deeks SG, Barbour JD, Martin JN, Swanson MS, Grant RM. Sustained CD4+ T-cell response after virologic failure of protease inhibitor-based regimens in patients with human immunodeficiency virus infection. J Infect Dis 2000, 181: 946–953.
14. Anastassopoulou CG, Touloumi G, Katsoulidou A, Halzitheadoru H, Pappa M, Paraskevis D, et al. Comparative evaluation of the QUANTIPLEX HIV-1 RNA 2.0 and 3.0 (bDNA) assays and the AMPLICOR HIV-1 MONITOR v1.5 test for the quantitation of human immunodeficiency virus type 1 RNA in plasma. J Virol Methods 2001, 91:67–74.
15. Burnham KP, Anderson DR. Model Selection and Inference: A Practical Information-Theoretic Approach. New York: Springer-Verlag, Inc.; 1988.
16. Jennrich RI, Schluchter MD. Unbalanced repeated-measures models with structured covariance matrices. Biometrics 1986, 42:805–820.
17. Laird NM. Missing data in longitudinal studies. Stats Med 1988, 7:305–315.
18. Duan N. Smearing estimation: a non parametric retransformation method. J Am Stat Assoc 1983, 78:605–610.
19. Pakker NG, Notermans DW, de Boer RJ, Roos MT, de Wolf F, Hill A, et al. Biphasic kinetics of peripheral blood T-cells after triple combination therapy in HIV-1 infection: a composite of redistribution and proliferation. Nature Med 1998, 4:208–214.
20. Sousa AE, Chaves AF, Doroana M, Antunes F, Victorino RM. Kinetics of the changes of lymphocyte subsets defined by cytokine production at single cell level during highly active antiretroviral therapy for HIV-1 infection. J Immunol 1999, 162:3718–3726.
21. Bucy RP, Hockett RD, Derdeyn CA, Saag MS, Squires KE, Sillers M, et al. Initial increase in blood CD4(+) lymphocytes after HIV antiretroviral therapy reflects redistribution from lymphoid tissues. J Clin Invest 1999, 103:1391–1398.
22. Smith KY, Valdez H, Landay A, Spritzler J, Kessler HA, Connick E, et al. Thymic size and lymphocyte restoration in patients with human immunodeficiency virus infection after 48 weeks of zidovudine, lamivudine, and ritonavir therapy. J Infect Dis 2000, 181:141–147.
23. Teixeira L, Valdez H, McCune JM, Koup RA, Badley AD, Hellerstein MK, et al. Poor CD4 T-cell restoration after suppression of HIV-1 replication may reflect lower thymic function. AIDS 2001, 15:1749–1756.
24. Douek DC, McFarland RD, Keiser PH, Gage EA, Massey JM, Haynes BF, et al. Changes in thymic function with age and during the treatment of HIV infection. Nature 1998, 396: 690–695.
25. Hakim FT, Cepeda R, Kaimei S, Mackall CL, McAtee N. Constraints on CD4 recovery postchemotherapy in adults: thymic insufficiency and apoptotic decline of expanded peripheral CD4 cells. Blood 1997, 90:3789–3798.
26. Anthony K, Sereti I, Orenstein J, Der Simonian R, Herpin B, Metcalf JA, et al. T-cell turnover does not normalize after 24 weeks of HAART and remains elevated in patients with failure of CD4 recovery despite effective viral suppression. XIV International Conference on AIDS. Barcelona, July 2002 [abstract WePeA5734].
27. Hunt PW, Martin JN, Sinclair E, Bredt B, Hagos E, Lampiris H, et al. T cell activation is associated with lower CD4+ T cell gains in HIV-infected patients with sustained viral suppression during antiretroviral therapy. J Infect Dis 2003, 187:1534–1543.
28. Giorgi JV, Hultin LE, McKeating JA, Johnson TD, Owens B, Jacobson LP, et al. Shorter survival in advanced human immunodeficiency virus type 1 infection is more closely associated with T lymphocyte activation than with plasma virus burden or virus chemokine coreceptor usage. J Infect Dis 1999, 179:859–870.
29. Lederman MM, McKinnis R, Kelleher D, Cutrell A, Mellors J, Neisler M, et al. Cellular restoration in HIV infected persons treated with abacavir and a protease inhibitor: age inversely predicts naive CD4 cell count increase. AIDS 2000, 14: 2635–2642.
30. Viard JP, Mocroft A, Chiesi A, Kirk O, Roge B, Panos G, et al. Influence of age on CD4 cell recovery in human immunodeficiency virus-infected patients receiving highly active antiretroviral therapy: evidence from the EuroSIDA study. J Infect Dis 2001, 183:1290–1294.
31. Giordano TP, Visnegarwala F, Wright JA, Adu-Oppong A, Graviss EA. Gender but not ethnicity predicts CD4 cell increase in response to sustained virologic suppression (SVS) from highly active antiretroviral therapy (HAART). XIV International Conference on AIDS. Barcelona, July 2002 [abstract WePeB5965].
32. Keiser P, Nassar NN, Koen G, Moreno S. Comparison of virologic response in a cohort of women and men treated with nelfinavir containing HAART regimens. XIV International Conference on AIDS. Barcelona, July 2002 [abstract WePeB5962].
33. Palella FJ, Jr., Gathe J, Brutus A, Sension M, Allen B, Morrow P. Gender comparisons in long term responses among patients receiving nelfinavir and dual nucleoside reverse transcriptase inhibitors as first ever highly active antiretroviral therapy. XIV International Conference on AIDS. Barcelona, July 2002 [abstract WePeB5967].
34. Sklar PA, Ward DJ, Baker RK, Wood KC, Gafoor Z, Alzola CF, et al. Prevalence and clinical correlates of HIV viremia (`blips') in patients with previous suppression below the limits of quantification. AIDS 2002, 16:2035–2041.
35. Hogg RS, Yip B, Chan KJ, Wood E, Craib KJP, O'Shaughnessy MV, et al. Rates of disease progression by baseline CD4 cell count and viral load after initiating triple-drug therapy. JAMA 2001, 286:2568–2577.
36. Sterling TR, Chaisson RE, Moore RD. HIV-1 RNA, CD4 T-lymphocytes, and clinical response to highly active antiretroviral therapy. AIDS 2001, 15:2251–2257.
37. Egger M, May M, Chene G, Phillips AN, Ledergerber B, Dabis F, et al. Prognosis of HIV-1-infected patients starting highly active antiretroviral therapy: a collaborative analysis of prospective studies. Lancet 2002, 360:119–129.
38. Lange C, Valdez H, Medvik K, Asaad R, Wild M, Kalayiian R, et al. Earlier initiation of HAART better preserves functional immune competence even in persons who achieve ‘normal’ CD4+ T-cell counts. XIV International Conference on AIDS. Barcelona, July 2002 [abstract Lb0r09].
39. Lucas GM, Chaisson RE, Moore RD. Highly active antiretroviral therapy in a large urban clinic: risk factors for virologic failure and adverse drug reactions. Ann Intern Med 1999, 131:81–87.
40. Valdez H, Lederman MM, Woolley I, Walker CJ, Vernon LT, Hise A, et al. Human immunodeficiency virus 1 protease inhibitors in clinical practice: predictors of virological outcome. Arch Intern Med 1999, 159:1771–1776.

HIV-1; CD4 cell count; drug effects; antiretroviral therapy; HAART

© 2003 Lippincott Williams & Wilkins, Inc.