Sage Journals: Discover world-class research

Abstract

We used the Veterans Health Administration (VA) HIV Clinical Case Registry (CCR) to evaluate the association between annual CD4 averages and all-cause mortality in HIV-infected veterans during their initial episode of suppressive highly active antiretroviral therapy (HAART). We observed 1083 deaths in 14 769 patients. Unadjusted mortality rates in the top and bottom CD4 quintiles differed significantly from the mid CD4 strata. Mortality in the top CD4 quintile (≥720 cells/mm³) was 14.1/1000 patient-years, 95% confidence interval (CI): 10.1-18.2, compared with 20.4 (CI: 15.5-25.3) in the next lower CD4 stratum (530-719 cells/mm³). This difference was significant in Cox proportional hazards model, controlling for demographics, hepatitis co-infections, low-level viremia, HAART adherence, and refill rates of individual antiretrovirals (HR: 1.4, CI: 1.13-1.73). Our results support early HAART initiation as advocated by the current US treatment guidelines for HIV infection.

Keywords

CD4 count mortality HAART veterans stavudine

Introduction

Owing to the recent development of safer and more potent antiretroviral therapy (ART) regimens, long-term stable suppression of viremia below the level of detection has become the therapeutic default for HIV-infected patients. It is unclear whether permanent virologic suppression using highly active antiretroviral therapy (HAART) will suffice to fully restore life expectancy. According to a recent Dutch publication, only 0.4 to 1.5 years of life are lost attributable to HIV infection if HAART is initiated early.¹ Other large European retrospective analyses recently failed to show the differences in age-adjusted mortality rates between the general and HIV-infected population with CD4 counts >500 cells/mm³.^2
–4 In contrast, other cohort analyses recently indicated that up to 10 to 15 years of life may be lost due chronic HIV infection even if viral replication is fully controlled and HAART was started in more recent years.^5
–7

The CD4 values achieved during long-term HAART are better predictors of all-cause mortality than baseline CD4 counts,⁸ even though they are strongly correlated.⁹ It is not known whether there is a mortality impact of different CD4 counts within the normal range (≥500 cells/mm³).

Herein, we evaluated the contribution of peripheral CD4 counts on all-cause mortality in a large population of older, virologically suppressed predominantly male Americans with a high prevalence of comorbidities including hepatitis B (HBV) and C virus (HCV) coinfection.¹⁰ Our patients may be well suited to serve as a model for the aging HIV-infected population in resource-rich settings.

Methods

We used the Clinical Case Registry (CCR) for HIV-infected US Veterans, which contains all nonnarrative clinical data for patients receiving care in the Veterans Health Administration network.¹¹ It includes laboratory values and vital signs as well as clinic utilization codes, pharmacy benefits management codes, codes for procedural terminology, and International Classification of Diseases, Ninth Revision (ICD-9) codes. Our observation period was fron January 1, 1995 to December 31, 2009. Patients were included in the analysis if they met the following criteria: (1) prescription of first-time HAART ≥14 days within the VA system, (2) clinical follow-up time ≥18 months, (3) ≥1 available HIV viral load (VL) result <400 copies/mL within 18 months after HAART initiation, and (4) ≥1 CD4 measurement during virologic control. Highly active ART was defined as simultaneous prescription of either ≥2 nucleoside reverse transcriptase inhibitors (NRTIs) and ≥1 antiretroviral (ARV) drug of a different class, triple-class therapy, zidovudine (ZDV) containing triple-NRTI regimens, or dual-therapy with a boosted protease inhibitor and a nonnucleoside reverse transcriptase inhibitor.

Follow-up time started at first time initiation of HAART within the VA system. Death date and occurrence of clinical AIDS were recorded and updated locally at VA facilities. End of clinical follow-up was defined as death or the last record for CD4 count or VL measurement, sodium or hematocrit measurement, vital sign record, or clinic visit code. Patients were censored at the time of virologic failure defined as the first of 2 consecutive VL measurements ≥400 copies/mL or any measurement ≥1000 copies/mL.

Presence or absence of HBV infection was determined by evidence for positive surface antigen or HBV-DNA polymerase chain reaction (PCR) ever, HCV status was determined by a combination of ICD-9 codes, serologic, and PCR laboratory values (proof of chronicity was required for neither). Baseline VL and CD4 counts were defined as the closest value before or at the day of HAART initiation (if within 90 days). When absolute CD4 values were missing, they were calculated from CD4 percentages and absolute lymphocyte values of differential blood counts. For the statistical analysis, we used time updated annual CD4 averages derived by the area under the curve (AUC) method. This was updated every 6 months after HAART initiation (0-1 years, 0.5-1.5 years, 1-2 years, 1.5-2.5 years, etc), incorporating all available CD4 counts after HAART initiation. The CD4 area under the curve (AUC) averages were categorized into quintiles determined by the value reached 1 year before the end of follow-up.

As patients with isolated VL blips 400 to 1000 copies or consecutive VL values <400 copies/mL were only censored if they met the criteria for virologic failure, we calculated cumulative and annual percentages of detectable low level viremia <1000 copies/mL per year in a similar fashion as that of CD4 AUC averages.

Annual and cumulative use rates of all licensed individual ARV components were tabulated by refill history, accounting for early refills and ARV drugs dispensed during hospitalizations. Highly active ART adherence was defined as the percentage of days in which the patient had sufficient antiretroviral supplies in his or her possession to take a complete HAART regimen.

Statistical Analysis

Data cleaning and statistical analyses were conducted with SPSS version 18 and 20 (IBM SPSS Inc, Chicago, Illinois). We calculated the unadjusted annual mortality rates and tabulated them by 1 year time lagged CD4 quintile reached. We compared our rates with a calculated mortality rate of a race, gender, and age-matched segment of the US population derived from 2007 CDC life tables.¹² To assess the association between CD4 quintile reached and death, we constructed a Cox model using low-level viremia, ARV use rates, and HAART adherence as continuous time-dependent covariates. As immunologic and virologic deterioration and altered ARV use could be a consequence of terminal disease, all time-dependent variables were entered by 1 year time lag. Missing values for time-dependent variables were handled by linear extrapolation wherever framed by existing values; otherwise the last value was carried forward.

As the CCR contains no narrative data, it is inherently difficult to exclude patients that received HAART outside of the VA system before entering the cohort. This subjects all CCR analyses to potential lead-in bias. We addressed this issue with 2 sensitivity analyses (1) by excluding patients with prior non-HAART ARV drug exposure within the VA system and (2) by the additional exclusion of patients whose VL was <10 000 copies/mL before HAART initiation.

Figure 1.

Unadjusted mortality rate for 14 767 patients during virologic suppression after ≥ 18 months of HAART ± CI (dotted lines) and median age of the study population under follow up in years (irregularly broken line, right y-axis). * Only months 19-24 of follow up included in 2nd year on HAART.

Results

Of the 32 195 veterans who had ever received HAART inside the VA system, 14 769 achieved virologic suppression for ≥18 months and were eligible for analysis, contributing 68 606 patient-years. The median duration of virologic suppression was 3.3 years (interquartile range [IQR] 2-5.9 years) during which we observed 1083 deaths ≥18 months after HAART initiation. Demographic characteristics are displayed in Table 1.

Table 1.

Baseline Characteristics and Number of Deaths by Year of HAART Initiation.^a

N	14 769
Events	1083
Median age at HAART initiation	47.7 (41.4-54.0)
Female gender	2.2%
Race, African American	42%
European American	39%
Missing or other	19%
Hepatitis C coinfection	40%
Hepatitis B coinfection	13%
Clinical AIDS at baseline	18%
ARV naive	69%
Baseline CD4 count, cells/mm³	260 (120-418)
Missing	19%
<200	30%
200-349	21%
350-499	14%
≥500	16%
Baseline viral load, log₁₀/mL	4.27 (2.7-5)
Missing	32%
Undetectable	13%
Detectable <10 000	15%
≥10 000	38%
Year of HAART initiation
1995	0.03%
1996	7%
1997	15%
1998	10%
1999	9%
2000	8%
2001	7%
2002	7%
2003	7%
2004	7%
2005	7%
2006	7%
2007	7%
2008	3%
2009

^a Values are either medians (± interquartile range) or percentage.

The overall unadjusted mortality rate after ≥18 months of HAART was 22.5 (confidence interval [CI]: 22.1-22.4) per 1000 patient-years and remained stable for the next 8 years, while the median age of the study population steadily increased from 50.2 to 55.4 years (Figure 1). The relationship between time lagged CD4 quintile reached and annual mortality rate is displayed in Figure 2. The unadjusted overall mortality rate of patients in the lowest quintile (CD4 averages < 250 cells/mm³) was 48.1 (CI 39.3-56.9) per 1000 patient-years. Patients in the highest CD4 quintile (>720 cells/mm³) had an annual mortality rate of 14.1 per 1000 patient-years (CI: 10.1-18.2), which was closest to an age-matched segment of the US population (10.7 per 1000 patient-years). Patients in the three middle quintiles (250-389, 390-529, and 530-719 CD4+ cells/mm³) had overall unadjusted mortality rates of 26.6 (CI: 20.9-32.4), 23.7 (CI: 18.4-29.1), and 20.4 (CI: 15.5-25.3) per 1000 patient-years, respectively.

In univariate analysis, time updated CD4 averages carried higher hazard ratios (HRs) and narrower CIs for death than baseline CD4 values. Older age, baseline viremia, HBV, and HCV coinfection were associated with shorter survival, while higher cumulative and recent HAART adherence and the use of certain antiretroviral components were associated with longer survival.

In multivariate analysis, baseline CD4 counts were not predictive of death. Compared with the top CD4 quintile, patients in the next highest quintile (530-719 CD4 cells/mm³) had an HR of 1.4 (CI: 1.13-1.73) for death. The CIs of the HRs within the 3 middle CD4 quintiles (250-719 CD4 cells/mm³) overlapped, but patients in the lowest CD4 quintile (<250 CD4 cells/mm³) had a significantly higher risk of death (HR: 2.46; CI: 1.99-3.01). Older age, higher cumulative HAART adherence, HBV and HCV coinfection, and calendar year of HAART initiation remained predictive of death in the multivariate model (Table 2). Recent (but not cumulative) stavudine (d4T) use (HR: 1.32, CI: 1.12-1.55) was independently associated with death. In contrast, neither the baseline HIV RNA values nor the extent of low-level viremia while on treatment significantly altered the risk of death (Table 2).

Table 2.

Uni- and Multivariate Analyses of Predictors of Death in ARV-Naive Patients.

	Univariate		Multivariate
Baseline variables	P	HR (95% CI)	P	HR (95% CI)
Age at HAART initiation (per decade)	<.0001	1.8 (1.7-1.91)	<.0001	1.93 (1.81-2.05)
Year of HAART initiation (per calendar year)	<.0001	0.96 (0.94-0.98)	<.0001	0.96 (0.94-0.98)
African American race	ns		ns
Female gender	.052	0.6 (0.36-1.004)	ns
Clinical AIDS at baseline	ns		ns
Baseline HIV VL (per log10)	ns		ns
HCV coinfection	<.0001	1.73 (1.54-1.95)	<.0001	1.63 (1.44-1.85)
HBV coinfection	<.0001	1.37 (1.16-1.6)	.027	1.21 (1.02-1.43)
CD4 baseline (= ref) ≥500 cells/mm³	<.0001	1	ns
350-499 cells/mm³	.039	1.29 (1.01-1.63)	ns
200-349 cells/mm³	.001	1.47 (1.18-1.82)	ns
<200 cells/mm³	<.0001	1.55 (1.27-1.9)	ns
Time-dependent variables			ns
Rate of low-level viremic episodes recent	.005	1.36 (1.1-1.68)	ns
Cumulative	.035	1.34 (1.02-1.77)	ns
HAART adherence (per 10%) past year	<.0001	0.94 (0.92-0.95)	ns
HAART adherence (per 10%) cumulative	<.0001	0.9 (0.88-0.92)	<.0001	0.9 (0.88-0.92)
ARV component use^a (per whole year)
Zidovudine past year	.005	0.83 (0.72-0.94)
Tenofovir past year	<.0001	0.71 (0.59-0.84)	ns
Cumulative	.001	0.9 (0.84-0.96)	ns
Efavirenz past year	.001	0.78 (0.67-0.9)
Cumulative	.012	0.96 (0.93-0.99)
Stavudine past year	<.0001	1.31 (1.13-1.52)	.001	1.32 (1.12-1.55)
Cumulative	ns		ns
Emtricitabine past year	<.0001	0.69 (0.56-0.84)	ns
Cumulative	<.0001	0.84 (0.76-0.92)	ns
Abacavir cumulative	.042	0.95 (0.9-0.998)	ns
Indinavir cumulative	.030	0.96 (0.92-0.996)	ns
Delavirdine cumulative	.053	1.13 (0.998-1.28)	.009	1.19 (1.04-1.35)
CD4 AUC average reached (= ref) ≥720 cells/mm³	<.0001	1	<.0001	1
530-719 cells/mm³	<.0001	1.48 (1.19-1.85)	.003	1.4 (1.13-1.73)
390-529 cells/mm³	<.0001	1.6 (1.29-2)	.002	1.42 (1.14-1.76)
250-389 cells/mm³	<.0001	2 (1.62-2.48)	<.0001	1.58 (1.28-1.94)
<250 cells/mm³	<.0001	3.26 (2.64-4.02)	<.0001	2.46 (1.99-3.04)

Abbreviations: HR, hazard ratio; CI, confidence interval; ref, reference; ns, not significant.

^aOnly antiretroviral drugs with significant hazard ratios are displayed.

In the 2 sensitivity analyses restricted to ARV-naive patients and additionally to patients with baseline VLs ≥10 000 copies mL, the baseline CD4 counts were again not statistically significant in multivariate analysis. The values and distribution of the HRs for the time-updated CD4 quintiles were similar to the overall study population (Supplemental Table 1).

We tabulated the recovery of CD4 AUC averages in the subgroup of ARV-naive patients with baseline viremia ≥10 000 copies/mL. Although few patients starting HAART with baseline CD4 counts <350 cells/mm³ reached CD4 averages ≥720 cells/mm³ and the majority of patients initiating HAART between 350 and 499 CD4 cells/mm³ remained <720 CD4 cells/mm³ for most of the study period, almost 75% of patients with baseline CD4 counts ≥500 reached CD4 AUC averages ≥720 cells/mm³ beyond 5 years of HAART (Figure 3).

Discussion

In contrast to earlier⁹ and more recent¹³ reports of a CD4 count plateau after 4 to 5 years on HAART, several cohort analyses^14,15 including our own data show continuing CD4count increases for 8 to 10 years on uninterrupted HAART, regardless of the baseline CD4 count. Both time-updated CD4 values and time-averaged or cumulative CD4 exposure have previously been shown to be predictive of incident non-AIDS-defining cancers.¹⁶ To address the volatility of single CD4 count measurements that fluctuate with postural state or rest¹⁷ and intermittent episodes of inflammation,¹⁸ we calculated the annual AUC averages that were updated semiannually.

In a study on German immune-discordant patients with CD4 counts <200 cells/mm³, new AIDS-defining events were rare beyond the first year of fully suppressive HAART.¹⁹ Deaths after serious non-AIDS events have been found to outnumber AIDS-related deaths 2:1 in virologically suppressed patients on HAART.²⁰ Most of these deaths have been shown to be related to increased rates of either non-AIDS-defining cancers¹⁶ or cardiovascular disease,²¹ which may be partially driven by chronic immune activation²² that may persist in lymphatic tissues for up to 10 years on continuous HAART.²³

We found that HRs for all-cause mortality at both ends of the CD4 spectrum (<250 and ≥720 CD4 cells/mm³) markedly deviated from the mid CD4 range. While increased mortality for the low end of the CD4 spectrum is well established for stably suppressed HIV infection, the finding of a mortality gradient at an upper CD4 threshold >700 cells/mm³ is not. Most large cohort analyses have not differentiated between different CD4 strata within the “normalized” CD4 range >500 cells/mm³.^3,4 However, what is considered “normal” for HIV-negative populations may not apply to chronic HIV infection. Possible explanations for the mortality difference in patients with normalized CD4 counts include the persistence of subclinical immunodeficiency at the lower end of the normal CD4 spectrum and a negative correlation between CD4+ counts and immune activation at the higher end.²⁴

Figure 2.

Yearly mortality during virologic suppression by 1-year time-lagged CD4 quintile reached. The arrow at 1.07% serves as reference and indicates a calculated mortality rate of a race, gender, and age matched segment of the US population (derived from 2007 CDC life tables). * Only months 19-24 of follow up included in during 2nd year on HAART. The ratio of mortality events and patient years of follow up per time-lagged CD4 quintile are listed below the chart.

Figure 3.

Median annual CD4 average (bold line) ±Inter-Quartile range (fine lines) by year since HAART initiation (x-axis) for different baseline CD4 strata of ARV naïve patients with baseline virermia ≥10 000 copies/mL. The dotted red line indicates an optimal CD4 average ≥720 CD4+ T-cells/μL.

We decided against the inclusion of a comorbidity index as a covariate as the incidence of HIV-associated non-AIDS complications may lie in the causal pathway between CD4 recovery and risk of death. As have other investigators, we also found an increased risk of death in patients with HBV and HCV coinfection.²⁵ Our finding of an independent mortality risk for recent d4T use is novel and concerning, as this drug is still widely prescribed in resource-limited settings. In concordance with other studies, higher baseline HIV RNA values did not significantly alter the risk of death,²⁶ nor did the extent of low-level viremia on treatment. Higher HAART adherence was protective of death, possibly because it is a better surrogate marker of stable interruption of viral replication than infrequently measured plasma viremia. Although adherence was not directly measured, pharmacy refill data have been shown to be a reliable surrogate.²⁷

While the baseline CD4 counts were not our primary focus of interest and the number of ARV-naive patients in our study was not big enough to discern mortality differences between the different baseline CD4 strata, our results indirectly support treatment initiation at high CD4 counts, as patients with baseline CD4 counts 350 to 499 cells/mm³ had a much reduced likelihood of reaching optimal CD4 averages ≥720 CD4 cells/mm³. This may explain their higher mortality risk that was reported in the NA-ACCORD study.²⁸ Our results thus indirectly support early HAART initiation as advocated by the current US treatment guidelines for HIV infection.^29,30

The strengths of our study include uniform data collection on exposures and outcomes across the VA system, a novel approach to address the intraindividual variability in CD4 values, and the inclusion of measured HAART adherence and low-level viremia as continuous covariates. Our study has several limitations. Our data did not allow us to directly identify patients already taking HAART prescribed from the outside of the VA system. Extensive chart reviews have previously shown that their share comprises <25%.¹⁰ We analyzed all patients after the first initiation of HAART within the VA system including those with potential HAART exposure from the outside and prior non-HAART ARV exposure within the VA system. While this subjects our analysis to lead-in bias, sensitivity analyses showed that the mortality risk conferred by suboptimal CD4 AUC averages was reduced neither by excluding patients with prior HAART ARV exposure within the VA system nor by the additional exclusion of patients who could have taken HAART from outside the VA system (VL < 10 000/mL at HAART initiation).

We only analyzed the first episode of virologic suppression after initial HAART. Although this approach did not allow for extended follow-up time and may result in an overrepresentation of events during the early years of HAART, it emulates current conditions as most contemporary ARV regimens have become very durable and virologic failure has become the exception.

Our data did not allow us to control for intravenous drug use (IVDU), which has been shown to be a strong predictor of death.¹ Even though the proportion of HCV coinfected patients in our cohort was very high, a recent analysis of HCV-infected US veterans revealed that only 15% engaged in ongoing drug abuse of any kind.³¹

While absolute CD8 values or CD4/CD8 ratios may provide additional prognostic information for death in patients on HAART,³² indirectly confirming an earlier report for CD4 percentage values,³³ we have recently shown that the inclusion of CD8 and CD4/CD8 subpopulations in the model did not substantially change the mortality risk conferred by suboptimal CD4 recovery.³⁴ Also, as our study population comprised 98% male, our results may not be generalizable to women.

In conclusion, “low-normal” CD4 counts >500 cells/mm³ did not confer the same mortality benefit as “high-normal” CD4 counts, that is, at or above the median value for HIV-uninfected populations. Yet, these high-normal CD4 counts were only reliably reached in patients who started HAART at a CD4 count >500 cells/mm³.

Footnotes

Authors’ Note

HD was involved in the design and conduct of the study; in the collection, management, analysis, and interpretation of the data; and in the preparation of the manuscript. SZ was involved in the design and conduct of the study, in the supervision and interpretation of the statistical analysis, and the review of the manuscript. MH was involved in the acquisition of the data, in the revision of the manuscript for intellectual content, and in the final approval for the submitted manuscript. RB was involved in the design and conduct of the study, in the analysis and interpretation of the data, and in the preparation and review of the manuscript. This study was approved by the Institutional Review Board of the Dallas VA Medical Center and its results were partially presented at the XIX International AIDS conference in Washington, DC, on July 23, 2012. The views expressed in this manuscript are those of the authors and do not necessarily reflect those of the Department of Veterans Affairs or the United States government. RB received research grants from Merck & Co, Tibotec Therapeutics, Bristol Myers Squibb and Abbott Pharmaceuticals. He also served as scientific advisor or consultant for EMD Serono, Merck & Co, Gilead, Tibotec Therapeutics and AIDS Arms.

Acknowledgements

We would like to thank Dr. Naim Maalouf for his helpful comments and review of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Supplemental Material

The online appendix is available at .

References

van Sighem

Gras

Reiss

Brinkman

de Wolf

; ATHENA national observational cohort study. Life expectancy of recently diagnosed asymptomatic HIV-infected patients approaches that of uninfected individuals. AIDS. 2010;24(10):1527–1535.

Lewden

Bouteloup

De Wit

. All-cause mortality in treated HIV-infected adults with CD4 ≥500/mm³ compared with the general population: evidence from a large European observational cohort collaboration. Int J Epidemiol. 2012;41(2):433–445.

Rodger

Lodwick

Schechter

. Mortality in well controlled HIV in the continuous antiretroviral therapy arms of the SMART and ESPRIT trials compared with the general population. AIDS. 2013;27(6):973–979.

Young

Psichogiou

Meyer

, et al;COHERE. CD4 cell count and the risk of AIDS or death in HIV-Infected adults on combination antiretroviral therapy with a suppressed viral load: a longitudinal cohort study from COHERE. PLoS Med. 2012;9(3):e1001194.

Losina

Schackman

Sadownik

. Racial and sex disparities in life expectancy losses among HIV-infected persons in the united states: impact of risk behavior, late initiation, and early discontinuation of antiretroviral therapy. Clin Infect Dis. 2009; 49(10):1570–1578.

May

Gompels

Delpech

. Impact of late diagnosis and treatment on life expectancy in people with HIV-1: UK Collaborative HIV Cohort (UK CHIC) Study. BMJ. 2011;343:d6016.

Antiretroviral Therapy Cohort Collaboration. Life expectancy of individuals on combination antiretroviral therapy in high-income countries: a collaborative analysis of 14 cohort studies. Lancet. 2008;372(9635):293–299.

Maman

Pujades-Rodriguez

Nicholas

. Response to antiretroviral therapy: improved survival associated with CD4 above 500 cells/μl. AIDS. 2012;26(11):1393–1398. doi: 1310.1097/QAD.1390b1013e328352d328054.

Kaufmann

Perrin

Pantaleo

. 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(18):2187–2195.

10.

Justice

McGinnis

Skanderson

, et al. Towards a combined prognostic index for survival in HIV infection: the role of 'non-HIV' biomarkers. HIV Med. 2010;11(2):143–151.

11.

Backus

Mole

Chang

Deyton

. The immunology case registry. J Clin Epidemiol. 2001;54(suppl 1):S12–S15.

12.

Arias

. United States life tables, 2004. Natl Vital Stat Rep. 2007;56(9):1–39.

13.

Wright

Carr

Woolley

. CD4 cell responses to combination antiretroviral therapy in patients starting therapy at high CD4 cell counts. J Acquir Immune Defic Synd. 2011;58(1):72–79.

14.

Guihot

Bourgarit

Carcelain

Autran

. Immune reconstitution after a decade of combined antiretroviral therapies for human immunodeficiency virus. Trends Immunol. 2011;32(3):131–137.

15.

Anonymous. Pneumocystis carinii pneumonia in HIV infections. Committee of the 1st Consensus Conference of anti-infectious therapy organized in May 1990 in Paris by the Societe de Pathologie Infectieuse de Langue Francaise [in French]. Revue de Pneumologie Clinique 1990;46(4):141–153.

16.

Kesselring

Gras

Smit

. Immunodeficiency as a risk factor for non-AIDS-defining malignancies in HIV-1-infected patients receiving combination antiretroviral therapy. Clin Infect Dis. 2011; 52(12):1458–1465.

17.

Campbell

Aurelius

Blowes

Harvey

. Decrease in CD4 lymphocyte counts with rest; implications for the monitoring of HIV infection. Int J STD AIDS. 1997;8(7):423–426.

18.

Skiest

Machala

. Comparison of the effects of acute Influenza infection and Influenza vaccination on HIV viral load and CD4 cell counts. J Clin Virol. 2003;26(3):307–315.

19.

Zoufaly

an der Heiden

Kollan

. Clinical outcome of HIV-infected patients with discordant virological and immunological response to antiretroviral therapy. J Infect Dis. 2011;203(3):364–371.

20.

Neuhaus

Angus

Kowalska

. Risk of all-cause mortality associated with nonfatal AIDS and serious non-AIDS events among adults infected with HIV. AIDS. 2010;24(5):697–706.

21.

Lichtenstein

Armon

Buchacz

. Low CD4+ T cell count is a risk factor for cardiovascular disease events in the HIV outpatient study. Clin Infect Dis. 2010;51(4):435–447.

22.

Subramanian

Tawako

Burdo

. Arterial inflammation in patients with HIV. JAMA. 2012;308(4):379–386.

23.

Sheth

Chege

Shin

. Immune reconstitution in the sigmoid colon after long-term HIV therapy. Mucosal Immunol. 2008;1(5):382–388.

24.

Ciccone

Read

Mannon

. Cycling of gut mucosal CD4+ T cells decreases after prolonged anti-retroviral therapy and is associated with plasma LPS levels. Mucosal Immunol. 2010;3(2):172–181.

25.

Branch

Van Natta

Vachon

Dieterich

Meinert

Jabs

. Mortality in hepatitis C virus–infected patients with a diagnosis of AIDS in the era of combination antiretroviral therapy. Clin Infect Dis. 2012;55(1):137–144.

26.

Wood

Hogg

Yip

Moore

Harrigan

Montaner

. Impact of baseline viral load and adherence on survival of HIV-infected adults with baseline CD4 cell counts >= 200 cells/[mu]l. AIDS. 2006;20(8):1117–1123.

27.

Peterson

Nau

Cramer

Benner

Gwadry-Sridhar

Nichol

. A checklist for medication compliance and persistence studies using retrospective databases. Value Health 2007;10(1):3–12.

28.

Kitahata

Gange

Abraham

. Effect of early versus deferred antiretroviral therapy for HIV on survival. N Engl J Med. 2009;360(18):1815–1826.

29.

Thompson

Aberg

Hoy

. Antiretroviral treatment of adult hiv infection: 2012 recommendations of the international antiviral society–USA panel. JAMA. 2012;308(4):387–402.

30.

HHS Panel on Antiretroviral Guidelines of Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1 infected adults and adolescents. Washington, DC: US Department of Health and Human Services; March 27, 2012. http://www.aidsinfo.hih.gov/Guidelines.

31.

Seal

Currie

Shen

. Hepatitis C treatment candidacy and outcomes among 4318 US veterans with chronic hepatitis C virus infection: does a history of injection drug use matter? J Clin Gastroenterol. 2007;41(2):199–205.

32.

Stebbing

Bower

Mandalia

Nelson

Gazzard

. Highly active anti-retroviral therapy (HAART)-induced maintenance of adaptive but not innate immune parameters is associated with protection from HIV-induced mortality. Clin Exp Immunol. 2006; 145(2):271–276.

33.

Burgess

Price

James

. Interferon-gamma responses to Candida recover slowly or remain low in immunodeficient HIV patients responding to ART. J Clin Immunol. 2006;26(2):160–167.

34.

Nurutdinova

Chrusciel

Zeringue

. Mental health disorders and the risk of AIDS-defining illness and death in HIV-infected veterans. AIDS. 2012;26(2):229–234.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB

CD4 Counts and Mortality in Virologically Suppressed US Veterans