Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Acquir Immune Defic Syndr. Author manuscript; available in PMC 2013 April 5.
Published in final edited form as:
PMCID: PMC3617548

Poor Initial CD4+ Recovery With Antiretroviral Therapy Prolongs Immune Depletion and Increases Risk for AIDS and Non-AIDS Diseases

Jason V. Baker, MD, MS,* Grace Peng, MD,* Joshua Rapkin, MS,* David Krason, MD, Cavan Reilly, PhD,* Winston P. Cavert, MD,* Donald I. Abrams, MD,§ Rodger D. MacArthur, MD,|| Keith Henry, MD,* and James D. Neaton, PhD*, for the Terry Beirn Community Programs for Clinical Research on AIDS (CPCRA)



Low CD4+ increases risk for both AIDS- and non–AIDS-related morbidity and mortality. The magnitude of CD4+ recovery early after initial antiretroviral therapy (ART) is important in the ultimate duration of immune depletion.


We examined CD4+ recovery among 850 participants in the Community Program for Clinical Research on AIDS Flexible Initial Retrovirus Suppressive Therapies study with virologic suppression (ie, achieved an HIV RNA level <400 copies/mL) with 8 months of initial ART and determined subsequent risk for AIDS, non-AIDS diseases (non-AIDS cancers and cardiovascular, end-stage renal, and liver diseases), or death using Cox regression during a median 5-year follow-up.


Mean pretreatment CD4+ was 221 cells/μL; 18% (n = 149) had a poor CD4+ recovery (<50 cells/μL) after 8 months of effective ART, resulting in lower CD4+ over 5 years. Older age (hazard ratio 1.34/10 yrs, P = 0.003) and lower screening HIV RNA (hazard ratio 0.65 per log10 copies/mL higher, P = 0.001), but not screening CD4+, were associated with a poor CD4+ recovery. After 8 months of effective ART, 30 patients experienced the composite outcome of AIDS, non-AIDS, or death among participants with a poor CD4+ recovery (rate = 5.8/100 person-years) and 74 patients among those with an adequate recovery (≥50 cells/μL; rate = 2.7/100 personyears) (adjusted hazard ratio = 2.24, P < 0.001). The risk of this composite outcome associated with a poor CD4+ recovery declined when ART was initiated at higher CD4+ counts (P < 0.01).


Impaired immune recovery, despite effective ART, results in longer time spent at low CD4+, thereby increasing risk for a broad category of HIV-related morbidity and mortality conditions.

Keywords: CD4 recovery, immune reconstitution, AIDS, non-AIDS conditions, HIV disease progression, antiretroviral therapy, cardiovascular disease, liver disease, kidney disease, non–AIDS-defining malignancies


Treatment of HIV infection with combination antiretroviral therapy (ART) suppresses viral replication, leading to recovery of CD4+ cells in most individuals,1,2 accounting for large reductions in AIDS-related morbidity and mortality over the past decade.35 Recent data from the Strategies for Management of Antiretroviral Therapy (SMART) study and observational cohorts indicate that the duration of time spent at low CD4+ levels is an important determinant of both AIDS and non-AIDS morbidity and mortality.68 Thus, rapid CD4+ recovery immediately after initiation of ART is clinically important. Previous studies have reported that 15%–20% of HIV-infected patients initiating combination ART fail to achieve an adequate CD4+ cell rise (25–50 cells/μL) within 6 months of beginning ART, despite virologic suppression.9,10 This poor immunologic response has been associated with increased risk for AIDS or death.10

We examined the risk associated with a poor “initial CD4+ recovery” with effective ART, in terms of both AIDS- and non–AIDS-related conditions—specifically common non- AIDS diseases consisting of atherosclerotic cardiovascular disease, end-stage renal disease, cirrosis, and non–AIDS-defining malignancies. We also examined predictors of a poor initial immune response and assessed the potential for long-term CD4+ recovery in these patients. The Flexible Initial Retrovirus Suppressive Therapies (FIRST) trial, conducted by the Terry Beirn Community Program for Clinical Research on AIDS, was well suited to study this relationship in that both fatal and severe nonfatal AIDS and non-AIDS events, along with CD4+ counts and HIV RNA levels, were collected in a standardized manner after initiation of ART.



We used follow-up data (median of 5 years) for participants in FIRST, which involved 80 research sites in the United States. The study design and primary results of FIRST have been reported.11,12 Briefly, between 1999 and 2002, 1397 ART-naive HIV-infected participants were randomized equally to 1 of 3 ART strategies—nucleoside reverse transcriptase inhibitors plus protease inhibitors (PIs), nucleoside reverse transcriptase inhibitors plus a nonnucleoside reverse transcriptase inhibitor (NNRTI), or the use of a 3-class strategy (PI + NNRTI + nucleoside reverse transcriptase inhibitor). Clinical, immunologic, and virologic outcomes were ascertained. CD4+ counts were assessed at 2 time points before ART initiation; the screening value was first, and the baseline value was second. The median time between measures was 10 days (interquartile range 7–20). After randomization and initiation of ART, participants were seen at month 1, month 4, and every 4 months thereafter for data collection. At these visits, standardized clinical end point collection forms were completed, and CD4+ and HIV RNA levels (Roche Amplicor 1.0) were measured.

HIV RNA and CD4+ responses at month 8 are shown in Figure 1 for participants in FIRST. This study cohort was limited to the subset with evidence of “effective ART treatment,” defined as HIV RNA <400 copies/mL at month 8. An 8-month time frame was chosen based on the visit schedule during FIRST, and it incorporates the period of rapid CD4+ recovery immediately after ART initiation. The initial CD4+ recovery at month 8 was defined as “poor” if the change from baseline (the second CD4+ measurement before initiating ART) was <50 cells/μL and “adequate” if the recovery was ≥50 cells/μL. A threshold of 50 cells/μL was based on current guidelines defining immunologic failure as a CD4+ recovery <25–50 cells/μL during the first year of therapy, with most of this increase occurring within months of ART initiation.13

CD4+ and HIV RNA response in FIRST study participants assessed at month 8.


All new or recurrent AIDS events, deaths from any cause, malignancies, and nonfatal cardiovascular, renal, and liver disease events were collected during follow-up. Qualifying AIDS events included probable or confirmed cases defined by US Centers for Disease Control and Prevention AIDS criteria (1993),14 adapted to include additional conditions related to immunodeficiency (footnote in Fig. 3). Non-AIDS disease events included fatal and nonfatal cardiovascular (myocardial infarction, stroke, coronary artery intervention, or death from chronic atherosclerotic cardiovascular disease); renal (end-stage renal disease or death from chronic kidney disease); liver (cirrhosis or death from liver failure); and non-AIDS cancers (all cancers excluding Kaposi sarcoma, lymphoma, and invasive cervical cancer). Structured case report forms were completed for AIDS and non-AIDS events. A Clinical Events Committee reviewed and adjudicated all AIDS and non-AIDS events.

Incidence of AIDS, non-AIDS disease, or death based on initial CD4+ recovery. Rate per 100 person-years (with 95% CI) of morbidity and mortality (defined as AIDS, non-AIDS diseases, or death) after 8 months of effective ART is plotted. Data are presented ...

Statistical Analysis

Logistic regression was used to examine associations between a poor initial CD4+ recovery and the following covariates: age, sex, race, screening CD4+ count (the first CD4+ measurement and not used to define a poor recovery) and HIV RNA level, prior AIDS event, and coinfection with hepatitis B or C. An odds ratio (OR) >1 implies that the covariate is associated with an increased risk for a poor CD4+ recovery at month 8. Using longitudinal regression analyses15 adjusted for the same baseline covariates listed above, rates of change in CD4+ beginning at month 8 were estimated for those with and without an adequate initial CD4+ recovery.

To assess the relationship of pretreatment CD4+ count and CD4+ response at 8 months, screening CD4+ count was used as the predictor instead of baseline values to avoid a spurious association (ie, regression to the mean) that could result from assessing the association of baseline CD4+ count with change from baseline.16 Cox’s proportional hazards models were used to study the relationship of the initial CD4+ recovery and risk (after month 8) for AIDS, non-AIDS disease, or death adjusted for the same covariates used for logistic models. These outcomes did not vary by treatment group in FIRST; therefore, analyses are pooled over the 3 groups.17 Findings are summarized with hazard ratios (HRs) and 95% confidence intervals (CIs), corresponding to risk of an event among participants with a poor initial CD4+ recovery compared with participants with an adequate recovery. Rates per 100 person-years are also cited. Analyses were performed using SAS (Version 8.2). All reported P values are 2 sided.


Patient Characteristics

The baseline characteristics of participants in FIRST have been reported.12 Our cohort (effective ART at month 8, n = 850; Fig. 1) generally included participants whowere older (39 vs 37 years; P < 0.001), more commonly white race (32% vs 18%; P < 0.001), less likely intravenous drug use users (13% vs 18%; P = 0.01), and with higher mean baseline CD4+ counts (221 vs 196 cells/μL; P = 0.02) and lower mean baseline HIV RNA levels (5.0 vs 5.1 log10 copies/mL; P = 0.05), when compared with remaining FIRST participants (largely those with HIV RNA levels ≥400 copies/mL at 8 months). Within our cohort, 12% had a CD4+ recovery <25 cells/μL, 18% had a CD4+ recovery <50 cells/μL, and 34% had a CD4+ recovery <100 cells/μL at 8 months. Of those assigned to a PI-based regimen, 61.6% were prescribed nelfinavir with 24.9% using a ritonavir-boosted PI. Efavirenz accounted for 60.4% of NNRTI use, and zidovudine was part of the ART regimen in 54.0%. The median follow-up was 60 months.

Predictors of a Poor Initial CD4+ Recovery

Table 1 presents a comparison of baseline characteristics by CD4+ recovery, along with results from multivariate logistic regression examining predictors of a poor initial CD4+ recovery, despite effective ART. Older age and lower HIV RNA levels before initiation of ART were associated with a poor CD4+ recovery at month 8. Screening CD4+ count was associated with a poor CD4+ recovery in univariate analyses (OR 1.10 per 100 cells/μL higher; 95% CI 1.02 to 1.19) but not in the multivariate models. However, baseline CD4+ count was associated with a poor CD4+ recovery in both univariate (OR 1.18; 95% CI 1.09 to 1.28) and multivariate (OR 1.16; 95% CI 1.05 to 1.27) models. Randomized ART strategy was not associated (P = 0.54 for comparison of 3 groups) with the initial CD4+ recovery (OR for PI vs NNRTI = 0.82; 95% CI 0.53 to 1.28; and OR for 3-class strategy vs NNRTI = 0.81; 95% CI 0.53 to 1.23) nor zidovudine prescribed at study entry (OR 1.33; 95% CI 0.93 to 1.90). While taking ART during follow-up, approximately 75% of participants reported 100% adherence to their regimen, with no difference between the poor and adequate recovery groups (P = 0.61). Furthermore, the HIV RNA response after 1 and 4 months of ART did not predict immune recovery. Among participants with a poor initial CD4+ recovery, HIV RNA was<400 copies/mL in 64% at month 1 and 83% at month 4, whereas for participants with an adequate CD4+ recovery the HIV RNA was <400 copies/mL in 45% at month 1 and 90% at month 4. Fifty-four percent of participants with poor CD4+ recovery had an HIV RNA level <400 copies/mL at 1 and 4 months and at 8 months compared with 42% of the patients with an adequate CD4+ recovery (P = 0.007).

Predictors of a Poor Initial CD4 Recovery (<50 Cells/μL) Despite 8 Months of Effective ART (HIV RNA <400 Copies/mL)

CD4+ Recovery With Longer Follow-up

CD4+ recovery over the entire follow-up was examined to characterize the long-term immunologic consequences of a poor initial CD4+ recovery (Fig. 2). Mean CD4+ counts for participants with a poor and adequate recovery, respectively, were: 287 and 207 cells/μL at baseline (P < 0.001), 268 and 410 cells/μL at month 8 (P < 0.001), and 441 and 556 cells/μL after 5 years of follow-up (P < 0.001). The rate of CD4+ rise after month 8 was similar between the 2 groups (slope after a poor recovery = 27 cells per μL per year; slope after an adequate recovery = 27 cells per μL per year; P = 0.99). Over a median follow-up of 52 months after month 8, the percent of time spent at CD4+ counts <350 cells/μL was 50% for those with a poor initial CD4+ recovery and 34% among participants with an adequate initial CD4+ recovery. In addition, 46.2% of participants with a poor CD4+ recovery, and 63.3% with an adequate CD4+ recovery, achieved a CD4+ count ≥500 cells/μL at least twice during follow-up (at or after month 8; P < 0.01).

CD4+ count over time stratified by initial CD4+ recovery with ART. Mean CD4+ count (with 95% CI) is plotted over 5 years of follow-up among FIRST participants with virologic suppression (HIV RNA <400 copies/mL) after 8 months of initial ART and ...

Morbidity and Mortality Subsequent to the Initial CD4+ Recovery

Among the 814 participants who had not experienced an event during the first 8 months, there were 16 participants with AIDS events, 8 with non-AIDS disease events, and 7 deaths from other causes among 143 participants with a poor initial CD4+ recovery and 33 AIDS events, 31 non-AIDS disease events, and 19 deaths from other causes among 671 with an adequate CD4+ recovery. There were 32 nonfatal and 17 fatal AIDS events overall, and the most common AIDS-related illnesses were esophageal or pulmonary candidiasis (n = 11), lymphoma (n = 8), and Pneumocystis jirovecii infection (n = 7). Among the fatal and nonfatal non-AIDS diseases, there were 14 participants with cardiovascular, 6 with renal, 16 with non-AIDS cancers, and 5 with liver-related events.

Rates for the composite endpoint of AIDS, non-AIDS disease, or death were examined by the initial CD4+ recovery at 3 different cutoffs (25, 50, or 100 cells/μL; Fig. 3). Risk after month 8 was greater with a poor initial CD4+ recovery (<50 cells/μL) compared with an adequate recovery (≥50 cells/μL; adjusted HR = 2.24; 95% CI 1.44 to 3.49); this relationship was similar with a CD4+ cutoff of 25 cells/μL (HR 1.76; 95% CI 1.04 to 3.01) or 100 cells/μL (HR 1.59; 95% CI 1.06 to 2.36) at month 8. Approximately half of subsequent clinical events among those with a poor CD4+ recovery occurred during the first year (or 8–20 months after initiating ART). The adjusted risk associated with a poor CD4+ recovery, compared with an adequate recovery, was also estimated for components of the composite outcome: AIDS (HR 2.84; 95% CI 1.52 to 5.29), non-AIDS diseases (HR 1.57; 95% CI 0.73 to 3.39), and death from other causes (HR 1.36; 95% CI 0.58 to 3.19). The risk for AIDS is the strongest, though point estimates for non-AIDS events are in the same direction. Further, the most common non-AIDS events, cardiovascular (HR 1.31; 95% CI 0.35 to 4.87) and non–AIDS-defining malignancies (HR 2.09; 95% CI 0.67 to 6.51), suggest a similar potential for higher risk. The low numbers and more modest risk increase for non-AIDS events, compared with AIDS, account for the wider confidence bounds and nonsignificant risk estimates.

Risk Associated With a Poor Initial CD4+ Recovery Decreases With Earlier ART Initiation

Risk of the composite including AIDS, non-AIDS, or death after a poor, compared with an adequate, initial CD4+ recovery was examined by screening CD4+ levels (<100, 100–199, 200–349, ≥350 cells/μL) in Figure 4. The risk associated with a poor initial CD4+ recovery declines when ART is started at higher CD4+ counts (P < 0.01 for interaction). In addition, rates of the composite outcome were higher when follow-up CD4+ counts were <350 cells/μL (6.4/100 person-years; 95% CI 4.9 to 7.9) as compared with ≥350 cells/μL (1.5/100 person-years; 95% CI 1.0 to 2.0).

Risk of AIDS, non-AIDS, or death stratified by pretreatment CD4+ count. The risk of morbidity and mortality (defined as AIDS, non-AIDS disease, or death) after 8 months of effective ART is presented and stratified by pretreatment screening CD4+ count. ...


ART treatment strategies that minimize time spent at lower CD4+ levels are important in reducing risk for HIV-related morbidity and mortality, including both AIDS and non-AIDS diseases. We examined the association of initial CD4+ recovery after 8 months of effective ART (HIV RNA <400 copies/mL) and risk for a composite of AIDS, common non- AIDS disease events, or death among 850 previously ART-naive HIV-infected participants during a 5-year median follow-up. A poor initial CD4+ recovery (<50 cells/μL), despite effective ART, was associated with increased risk for this composite outcome. Lower pretreatment CD4+ counts did not alter the likelihood of a poor initial CD4+ recovery, but the consequences of a poor CD4+ recovery on subsequent morbidity and mortality were greater when ART was initiated at lower CD4+ counts. This is likely due to the prolonged duration of low CD4+ counts for participants with a poor initial CD4+ recovery—32% more time spent with CD4+ counts <350 cells/μL compared with those with an adequate initial CD4+ recovery (≥50 cells/μL).

The degree of HIV-related immune depletion is now recognized as an important contributor to risk for many non-AIDS diseases and traditional AIDS events. In the D:A:D study, rates of liver-related mortality and all non-AIDS causes of death decrease at higher CD4+ levels.8 Previous analyses using follow-up data from FIRST demonstrated that risk for cardiovascular, renal, liver, and non-AIDS cancer events was independently associated with latest CD4+ levels.18 This study adds to these observations and indicates that the duration of time spent at lower CD4+ levels after the initiation of ART contributes to risk for both AIDS and non-AIDS diseases.

The first phase of immune recovery after initiation of ART is characterized by a redistribution of both naive and memory CD4+ cells from lymphatic tissues, whereas the subsequent gradual CD4+ recovery over time is primarily naive CD4+ cells.19 Lymphatic tissues are the major viral reservoirs during HIV infection, and the majority of infected cells are T cells.20,21 Higher HIV RNA levels may therefore be associated with greater numbers of CD4+ T cells being sequestered within lymphatic tissues, resulting in a greater redistribution of cells after ART-associated viral suppression. Indeed, higher pretreatment HIV RNA levels are consistently associated with greater CD4+ cell recovery after starting ART.2225 In addition, thymus output during sustained HIV replication is important for repopulating naive CD4+ cells in lymphatic tissues, consistent with research showing that age is an important determinant of both early- and late-immune recovery after ART.22,2628 Our results agree with other studies that younger age and higher pretreatment HIV RNA level are associated with a better CD4+ cell recovery with effective ART.22,24,27,28

The clinical importance of achieving specific CD4+ count thresholds after initiation of ART has received increasing attention in recent years. A threshold of 500 cells/μL is often used to define a normal CD4+ count because this represents the lower limit of the normal range in HIV-negative adults,2931 though a goal of 800 cells/μL has also been suggested as this represents the mean CD4+ count among HIV-negative adults.32 Symptomatic HIV-related events (Centers for Disease Control and Prevention category B and C events) may be more frequent among individuals who fail to achieve CD4+ counts >500 cells/μL, though previous risk assessments have been limited by the low number of events overall at these CD4+ levels.30 An observational multicenter study showed that mortality rates among 2435 HIV-infected patients with a CD4+ count >500 cells/μL were the same as for the general population after 6 years of ART use.33 Furthermore, the notion that CD4+ levels >500 cells/μL are normal is consistent with the observation from the EuroSIDA study that the rate of CD4+ recovery with ART slows once levels >500 cells/μL are achieved.34 The ability to achieve normal CD4+ counts is reflected in the degree of early recovery.30 Indeed, in this study, participants with a poor, compared with an adequate, initial CD4+ recovery were less likely to achieve a CD4+ count >500 cells/μL and spent more time at CD4+ counts <350 cells/μL.

CD4+ recovery was not predicted by the pretreatment CD4+ count among participants in this study. This is in contrast to a previous study, where patients with a baseline CD4+ count <200 cells/μL were less likely to have poor recovery, defined as an increase <25 cells/μL at 6 months, than patients with baseline CD4 +≥200 cells/μL.10 Our use of screening CD4+ count as a predictor, versus the baseline measure used to define change, reduces the effects of regression to the mean and this may explain the difference. Because of the within-patient variability of CD4+ count, even when there is no relationship between baseline level and change from baseline, a spurious negative correlation is induced. This can largely be circumvented by using a previous measurement (eg, screening in our case) as the predictor variable instead of the baseline value used for measuring change.16,35,36

In summary, a potential benefit associated with earlier initiation of ART, and the one not previously emphasized, may be avoiding the risk associated with a poor CD4 recovery and the consequent greater time spent at lower CD4+ levels associated with both AIDS and non-AIDS diseases. Further, a poor CD4 recovery can be expected in 15%–20% of patients irrespective of CD4+ level at the time of initiation of ART. Additional research is needed examining reasons for a poor CD4+ recovery and to study immune-modulating interventions that increase CD4+ responses in these patients.


The authors would like to sincerely thank all the participants in the Flexible Initial Retrovirus Suppressive Therapies study and the network of Community Program for Clinical Research on AIDS investigators for their commitment to conducting consistently high-quality HIV clinical trials.

Supported by National Institute of Allergy and Infectious Diseases (NIAID), National Institutes of Health grants U01AI042170, U01AI46362, and T32 AI055433. K.H. receives research support from GSK, BMS, Tibotech, Serono, Thera, Merck, and Pfizer and is on the speakers bureau for GlaxoSmithKline, Bristol-Myers Squibb, Roche, and Gilead.


No other authors have disclosures to report.


1. Hammer SM, Squires KE, Hughes MD, et al. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. AIDS Clinical Trials Group 320 Study Team. N Engl J Med. 1997;337:725–733. [PubMed]
2. Autran B, Carcelain G, Li TS, et al. Positive effects of combined antiretroviral therapy on CD4+ T cell homeostasis and function in advanced HIV disease. Science. 1997;277:112–116. [PubMed]
3. Mocroft A, Vella S, Benfield TL, et al. Changing patterns of mortality across Europe in patients infected with HIV-1. EuroSIDA Study Group. Lancet. 1998;352:1725–1730. [PubMed]
4. Palella FJ, Jr, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med. 1998;338:853–860. [PubMed]
5. Epidemiology of HIV/AIDS—United States, 1981–2005. MMWR Morb Mortal Wkly Rep. 2006;55:589–592. [PubMed]
6. El-Sadr WM, Lundgren JD, Neaton JD, et al. CD4+ count-guided interruption of antiretroviral treatment. N Engl J Med. 2006;355:2283–2296. [PubMed]
7. Smit C, Geskus R, Walker S, et al. Effective therapy has altered the spectrum of cause-specific mortality following HIV seroconversion. AIDS. 2006;20:741–749. [PubMed]
8. Weber R, Sabin CA, Friis-Moller N, et al. Liver-related deaths in persons infected with the human immunodeficiency virus: the D:A:D study. Arch Intern Med. 2006;166:1632–1641. [PubMed]
9. Grabar S, Le Moing V, 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. 2000;133:401–410. [PubMed]
10. Moore DM, Hogg RS, Chan K, et al. Disease progression in patients with virological suppression in response to HAART is associated with the degree of immunological response. AIDS. 2006;20:371–377. [PubMed]
11. MacArthur RD, Chen L, Mayers D, et al. The rationale and design of the CPCRA (Terry Beirn Community Programs for Clinical Research on AIDS) 058 FIRST (Flexible Initial Retrovirus Suppressive Therapies) Trial. Control Clin Trials. 2001;22:176–190. [PubMed]
12. MacArthur RD, Novak RM, Peng G, et al. A comparison of three highly active antiretroviral treatment strategies consisting of non-nucleoside reverse transcriptase inhibitors, protease inhibitors, or both in the presence of nucleoside reverse transcriptase inhibitors as initial therapy (CPCRA 058 FIRST Study): a long-term randomised trial. Lancet. 2006;368:2125–2135. [PubMed]
13. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Department of Health and Human Services (DHHS); 2008. [Accessed June 19, 2008]. pp. 1–128. Available at
14. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Recomm Rep. 1992;41:1–19. [PubMed]
15. Laird NM, Ware JH. Random-effects models for longitudinal data. Biometrics. 1982;38:963–974. [PubMed]
16. Ederer F. Serum cholesterol changes: effects of diet and regression toward the mean. J Chronic Dis. 1972;25:277–289. [PubMed]
17. Kalbfleisch J, Prentice R. The Statistical Analysis of Failure Time Data. 2. New York: John Wiley; 2002.
18. Baker JV, Peng G, Rapkin J, et al. for the Terry Beirn Community Programs for Clinical Research on AIDS (CPCRA) CD4+ cell count and risk of non-AIDS diseases following initial treatment for HIV infection. AIDS. 2008;22:841–848. [PMC free article] [PubMed]
19. Pakker NG, Notermans DW, de Boer RJ, et al. Biphasic kinetics of peripheral blood T cells after triple combination therapy in HIV-1 infection: a composite of redistribution and proliferation. Nat Med. 1998;4:208–214. [PubMed]
20. Haase AT. Population biology of HIV-1 infection: viral and CD4+ T cell demographics and dynamics in lymphatic tissues. Annu Rev Immunol. 1999;17:625–656. [PubMed]
21. Schacker T, Little S, Connick E, et al. Productive infection of T cells in lymphoid tissues during primary and early human immunodeficiency virus infection. J Infect Dis. 2001;183:555–562. [PubMed]
22. Lederman MM, Valdez H. Immune restoration with antiretroviral therapies: implications for clinical management. JAMA. 2000;284:223–228. [PubMed]
23. Teixeira L, Valdez H, McCune JM, et al. Poor CD4 T cell restoration after suppression of HIV-1 replication may reflect lower thymic function. AIDS. 2001;15:1749–1756. [PubMed]
24. Florence E, Lundgren J, Dreezen C, et al. Factors associated with a reduced CD4 lymphocyte count response to HAART despite full viral suppression in the EuroSIDA study. HIV Med. 2003;4:255–262. [PubMed]
25. Smith CJ, Sabin CA, Youle MS, et al. Factors influencing increases in CD4 cell counts of HIV-positive persons receiving long-term highly active antiretroviral therapy. J Infect Dis. 2004;190:1860–1868. [PubMed]
26. Smith KY, Valdez H, Landay A, 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. [PubMed]
27. Viard JP, Mocroft A, Chiesi A, 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. [PubMed]
28. Douek DC, McFarland RD, Keiser PH, et al. Changes in thymic function with age and during the treatment of HIV infection. Nature. 1998;396:690–695. [PubMed]
29. Kaufmann GR, Perrin L, Pantaleo G, et al. CD4 T-lymphocyte recovery in individuals with advanced HIV-1 infection receiving potent antiretroviral therapy for 4 years: the Swiss HIV Cohort Study. Arch Intern Med. 2003;163:2187–2195. [PubMed]
30. Kaufmann GR, Furrer H, Ledergerber B, et al. Characteristics, determinants, and clinical relevance of CD4 T cell recovery to <500 cells/microL in HIV type 1-infected individuals receiving potent antiretroviral therapy. Clin Infect Dis. 2005;41:361–372. [PubMed]
31. Kaufmann GR, Bloch M, Finlayson R, et al. 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. [PubMed]
32. Gras L, Kesselring AM, Griffin JT, et al. CD4 cell counts of 800 cells/ mm3 or greater after 7 years of highly active antiretroviral therapy are feasible in most patients starting with 350 cells/mm3 or greater. J Acquir Immune Defic Syndr. 2007;45:183–192. [PubMed]
33. Lewden C. Groups AaAS. Responders to antiretroviral treatment over 500 CD4/mm3 reach same mortality rates as general population (APROCO and Aquitaine cohorts, France). 10th European AIDS Conference/ EACS; Dublin, Ireland. 2005.
34. Mocroft A, Phillips A, Gatell J, et al. Normalisation of CD4 counts in patients with HIV-1 infection and maximum virological suppression who are taking combination antiretroviral therapy: an observational cohort study. Lancet. 2007;370:407–413. [PubMed]
35. Blomqvist N, Svardsudd K. A new method for investigating the relation between change and initial value in longitudinal blood pressure data. II. Comparison with other methods. Scand J Soc Med. 1978;6:125–129. [PubMed]
36. Hayes RJ. Methods for assessing whether change depends on initial value. Stat Med. 1988;7:915–927. [PubMed]