The benefits of HAART in the management of HIV disease are well established. Through the sustained suppression of plasma HIV RNA, HAART has been shown to decrease morbidity and mortality among HIV-infected patients [1–3].
Because of challenges associated with HAART, including issues of adherence and side-effects [4,5], therapeutic guidelines have recently shifted towards a recommendation that HAART be delayed until the CD4 cell count approaches 200 cells/μl [6–8]. This strategy aims to avoid premature exposure to antiretroviral therapy in order to prevent the early emergence of antiretroviral resistance and side-effects [4,5,9], while ensuring the initiation of HAART before the benefits of therapy have been shown to be diminished [3,10–12].
However, several studies have recently suggested that a baseline plasma HIV RNA level ≥ 100 000 copies/ml is independently associated with elevated mortality even after adjustment for CD4 cell count [12–14]. This finding has been a cause for major concern and uncertainty among clinicians, since the impact of plasma HIV RNA among patients with CD4 cell counts ≥ 200 cells/μl has not been well described. It is also not known if the association between plasma HIV RNA ≥ 100 000 copies/ml and mortality is explained by patient non-adherence [15,16]. This finding would not be unexpected, since higher plasma HIV RNA has been shown to be among the strongest determinants of HIV disease progression among untreated patients [17,18]. Therefore, the present study was conducted to examine the impact of plasma HIV RNA ≥ 100 000 copies/ml on survival of adherent and non-adherent antiretroviral-naive HIV-infected adults initiating HAART when their CD4 cell counts were ≥ 200 cells/μl.
The HAART Observational Medical Evaluation and Research (HOMER) study administered through the British Columbia Centre for Excellence in HIV/AIDS Drug Treatment Program has been described in detail elsewhere [3,10]. Briefly, the British Columbia Centre is the only free source of antiretroviral medications in the province of British Columbia, and pharmaceutical sales suggest that < 1% of HIV-infected patients obtain antiretroviral drugs outside the programme . For all programme participants, a complete prospective profile of antiretroviral therapy is maintained .
In the present study, analyses were restricted to HIV-infected men and women who were antiretroviral drug naive, were first prescribed triple drug antiretroviral therapy between 1 August 1996 and 30 September 2003 and were followed until 30 September 2004. Study subjects were initially prescribed HAART with regimens including two nucleoside reverse transcriptase inhibitors and either a protease inhibitor (boosted or unboosted) or a non-nucleoside reverse transcriptase inhibitor, at the discretion of the enrolling physician.
The primary end-point in this analysis was time to death. Deaths occurring during the follow-up period were identified on a continuous basis from physician reports and through record linkages carried out with the British Columbia Division of Vital Statistics . The primary analysis was designed to be conservative and to evaluate all-cause mortality; subanalyses considered the combined end-point of AIDS or death, or censored deaths from accidental causes at the time of death and classified them as non-events .
The Kaplan–Meier analysis restricted evaluation to patients who had a baseline CD4 cell count ≥ 200 cells/μl since the study was intended to examine the impact of plasma HIV RNA in patients with CD4 cell counts > 200 cells/μl, as guidelines universally recommend that patients should be treated when the CD4 cell count declines below this level [6,16]. Patients were stratified into two groups using their baseline plasma HIV RNA categories, low (< 100 000 copies/ml) and high (≥ 100 000 copies/ml), since this has also been shown to be the only clinically significant cut-off among HAART-treated patients [3,6,10,12]. Patients were further stratified into adherent and non-adherent categories using data on prescription refill compliance [3,13,21]. As previously, the definition of adherence was based on the time that medication dispensed would last as a proportion of follow-up time [3,13,21]. This calculation was restricted to each patient's first year on therapy to avoid reverse causation that could occur among patients who ceased antiretroviral therapy after they had become too sick to take medication. It has been previously demonstrated that adherence defined in this way strongly predicts virological response and mortality, and that it can adjust for the potentially confounding effect of treatment interruption [3,13,21]. In the primary analysis, patients were a priori defined as non-adherent if they received antiretroviral medications < 95% of the time during the first year of therapy, in order to derive an estimate of the impact of adherence on survival [3,13].
Cox regression analyses
For the Cox regression analyses, all patients were considered based on previous analyses and the thresholds identified in therapeutic guidelines; patients were stratified into combined low (< 50 cells/μl), medium (50–199 cells/μl), and high (≥ 200 cells/μl) CD4 cell count strata [3,6,10,12]. As above, patients were stratified into low (< 100 000 copies/ml) and high (≥ 100 000 copies/ml) plasma HIV RNA groups. Additional variables examined in these analyses included protease inhibitor use in the initial regimen (yes versus no), a prior diagnosis of AIDS (yes versus no), age, gender, physician experience (six or more patients previously enrolled in the programme) [19,22] and date of therapy initiation (before or after July 1997) .
In addition, in order to derive adjusted relative hazards (RH) for mortality among patients with low (< 100 000 copies/ml) and high (≥ 100 000 copies/ml) baseline plasma HIV RNA in each of the CD4 cell count strata, a fixed model was built with indicator variables for each combined baseline plasma HIV RNA and CD4 cell count strata while adjusting for adherence (≥ 95% versus < 95% adherent) and other relevant covariates. Here, patients with a baseline CD4 cell count ≥ 200 cells/μl and a baseline plasma HIV RNA < 100 000 copies/ml served as the reference category. The assumption of proportional hazards was validated by inspection of plots of –log (survival function estimates) against log time. All multivariate models described were fit using the same protocol of adjusting for all variables that were statistically significant (P < 0.05) in univariate analyses.
Between 1 August 1996 and 30 September 2003, 2405 antiretroviral-naive participants aged 18 years and older began triple combination therapy. Of these, 188 (7.8%) were excluded from this analysis for not having both baseline CD4 cell count and plasma HIV-1 RNA level measures available within 6 months prior to the start of antiretroviral therapy. The study sample was based on the remaining 2217 subjects [1818 (82%) men and 399 (18%) women]. No differences in initial HAART regimen, history of injection drug use or subsequent mortality were observed between the study sample and those excluded. However, persons excluded from this analysis were more likely to be female (P = 0.018) and younger (P = 0.025) than those included in the study group. The overall median follow-up time was 50.4 months [interquartile range (IQR), 25.7–72.6]. At baseline, the median age of participants was 38 years, the median CD4 cell count was 210 cells/μl and the median plasma HIV RNA level was 100 010 copies/ml. Overall, 847 (38.2%) patients initiated therapy with a non-nucleoside reverse transcriptase inhibitor, 1046 (47.2%) initiated therapy with an unboosted protease inhibitor and 324 (14.6%) patients initiated therapy with a boosted protease inhibitor. There were 368 deaths during the study period, among these 63 (17%) were identified as accidents/suicides or illicit drug overdoses and 305 (83%) were from non-accidental causes, in the vast majority AIDS related.
Figure 1 shows the Kaplan–Meier cumulative mortality estimates for the 1166 (53%) patients with a baseline CD4 cell count ≥ 200 cells/μl stratified by plasma HIV RNA above or below 100 000 copies/ml and for the ≥ 95% adherent and < 95% adherent populations. When the 516 (44%) non-adherent patients were considered, the mortality rate was statistically elevated among the 235 (46%) patients who had a baseline plasma HIV RNA ≥ 100 000 copies/ml (log-rank P = 0.032). Conversely, when the 650 (56%) adherent patients were considered, the mortality rate was similar between those with a plasma HIV RNA above or below 100 000 copies/ml (log-rank P = 0.690). In subanalyses (data not shown), there were no differences in cumulative mortality based on plasma HIV RNA among patients with CD4 cell counts 200–350 and ≥ 350 cells/μl when this population was considered as a whole (P = 0.450), or when restricted to ≥ 95% adherent (P = 0.441) or < 95% adherent patients (P = 0.569).
Cox regression analyses
The results of the unadjusted and adjusted Cox regression analyses that considered all patients are shown in Table 1. As shown here, as a predictor of mortality amongst the entire cohort, plasma HIV RNA ≥ 100 000 copies/ml was associated with elevated mortality in unadjusted analyses [RH, 1.54; 95% confidence interval (CI), 1.24–1.91; P = < 0.001] and after adjustment for all variables that had P < 0.05 in univariate analyses (RH, 1.29; (95% CI, 1.03–1.61; P = 0.028). The model was adjusted for age, physician experience, CD4 cell count and non-adherence, and the statistical significance of these covariates is indicated in the table.
Table 2 shows the results of the Cox regression analyses that examined the impact of plasma HIV RNA ≥ 100 000 copies/ml in each of the CD4 cell count strata. With patients with a CD4 cell count ≥ 200 cells/μl and a plasma HIV RNA < 100 000 copies/ml as the reference category, the mortality rate was statistically similar among patients with a baseline CD4 cell count ≥ 200 cells/μl and a baseline plasma HIV RNA ≥ 100 000 copies/ml (RH, 1.21; (95% CI, 0.89–1.65; P = 0.232). Again, the other covariates adjusted for in the Cox model and their statistical significance are shown in the table. Results were consistent when accidental causes of death were censored from the analysis, when patients who were free of a clinical AIDS diagnosis at baseline were considered, and when the combined end-point of AIDS or death was considered (data not shown).
Because of limitations inherent in the adherence measure [3,20], several confirmatory analyses using logistic regression analyses were conducted without taking into account the time progression to death, and in each case the results were identical. Specifically, among the 516 non-adherent individuals with baseline CD4 cell counts ≥ 200 cells/μl, the odds ratio for mortality for patients with baseline plasma HIV RNA ≥ 100 000 c/ml was 1.85 (95% CI, 1.18–2.92; P = 0.008) in univariate analyses. Among the 650 adherent individuals with baseline CD4 cell counts ≥ 200 cells/μl, the odds ratio for mortality for patients with baseline plasma HIV RNA ≥ 100 000 c/ml was 0.89 (95% CI: 0.53–1.49; P = 0.656) in univariate analyses. In a logistic regression model constructed identical to the Cox model shown in Table 2, plasma HIV RNA ≥ 100 000 copies/ml was not associated with a greater odds of mortality among patients with baseline CD4 cell counts ≥ 200 cells/μl (odds ratio, 1.32; 95% CI: 0.94–1.83; P = 0.186).
Lastly, it was recognized that the primary mechanism through which non-adherence would be associated with elevated mortality would be through lower rates of suppression of plasma HIV RNA. In the group of 516 non-adherent patients, 175 (33.9%) with baseline CD4 cell counts ≥ 200 cells/μl had a plasma HIV RNA < 500 copies/ml during the first year of HAART, whereas 579 (89.1%) of the 650 adherent patients with baseline CD4 cell counts ≥ 200 cells/μl experienced plasma HIV RNA suppression during the first year of HAART (P < 0.001).
The present analyses confirmed that there is an independent effect of plasma HIV RNA ≥ 100 000 copies/ml on mortality when all HIV-infected patients were considered. However, when our analyses were restricted to patients with a baseline CD4 cell count ≥ 200 cells/μl, we only observed a statistical association between plasma HIV RNA ≥ 100 000 copies/ml and elevated mortality when the analyses were restricted to non-adherent patients. Plasma HIV RNA ≥ 100 000 copies/ml was also not statistically associated with mortality among patients with a CD4 cell count ≥ 200 cells/μl in multivariate analyses that adjusted for adherence.
While there has emerged considerable agreement across international consensus guidelines regarding the impact of baseline CD4 cell count on survival after the initiation of HAART [6–8], the importance of baseline plasma HIV RNA on survival after the initiation of therapy remains among the most controversial issues in the treatment of HIV infection [16,24–26]. As noted in some therapeutic guidelines , many clinicians favour immediate initiation of HAART if the plasma HIV RNA level rises above 100 000 copies/ml, regardless of the CD4 cell count. However, initiating HAART at a CD4 cell count well above 200 cells/μl (e.g. > 350 cells/μl) because the plasma HIV RNA level is > 100 000 copies/ml is a critical decision given that it would likely represent an additional 3–5 years of additional antiretroviral exposure in most patients . As earlier initiation of HAART does not appear to protect against the deleterious effects of non-adherence , our findings indicated that HAART should be delayed in favour of adherence-readiness interventions in these patients , as the plasma HIV RNA level is of negligible prognostic value among adherent patients with baseline CD4 cell counts ≥ 200 cells/μl.
The observation that plasma HIV RNA ≥ 100 000 copies/ml is only associated with mortality among non-adherent patients if the CD4 cell count is ≥ 200 cells/μl is not surprising in view of what is known about disease progression among treated and untreated HIV-infected individuals [17,18]. Specifically, while antiretroviral therapy would be expected to suppress plasma HIV RNA levels and preserve CD4 cell counts among most adherent patients regardless of the baseline plasma HIV RNA level [14,29], among non-adherent patients with limited or no plasma HIV RNA response , more rapid disease progression would be expected among those individuals with higher plasma HIV RNA levels at baseline [17,18]. This observation has significance for developed world settings where plasma HIV RNA measurement has become routine, but it also has major significance for developing world settings. Specifically, although our findings should ideally be replicated in the context of a developed world setting, they do indicate that the decision regarding when to initiate HAART can be primarily driven by the more inexpensive measures of CD4 cell count rather than the combined measures of plasma HIV RNA and CD4 cell count [31,32].
It is important to stress that these data arose in a setting where all HIV/AIDS care, antiretroviral drugs and laboratory monitoring are available free of charge, and where previous studies have shown that virtually all patients acquire antiretroviral drugs through a single centralized source [3,19,33]. In addition, the centralized death registry enabled complete population-level data on HIV/AIDS deaths for the entire province. Finally, since the British Columbia Centre maintains complete prospective records of antiretroviral drugs dispensed, it was possible to determine each individual's level of prescription refill compliance. Although using refill compliance as a surrogate for adherence has been previously validated [3,34–36], there is likely a strong conservative bias operating in our study because patients could have been less than optimally adherent to daily treatment intake despite maintaining a high level of refill compliance during the first year of therapy. In addition, as we have previously discussed , a limitation of the present study is that patients were defined as adherent based on their behaviour during the first year of therapy and then assigned to adherent strata for an analysis of baseline characteristics. Like all studies of patients treated in observational cohorts, unmeasured differences may exist among study populations and for this reason caution is warranted.
In summary, the present study demonstrates that plasma HIV RNA ≥ 100 000 copies/ml is only associated with mortality when the CD4 cell count is ≥ 200 cells/μl among patients who are non-adherent. Plasma HIV RNA ≥ 100 000 copies/ml was not associated with mortality among patients with a CD4 cell count ≥ 200 cells/μl when adherent patients were considered in stratified analyses or in multivariate analyses that adjusted for adherence. These findings are likely explained by elevated HIV disease progression among patients who have a worse baseline plasma HIV RNA profile and who are non-adherent, and they are consistent with natural history studies from prior to the advent of HAART [17,18]. These findings should be useful for clinicians facing the existing uncertainty regarding the timing of HAART among patients with a high CD4 cell count but a plasma HIV RNA level ≥ 100 000 copies/ml.
We thank Bonnie Devlin, Elizabeth Ferris, Nada Gataric, Kelly Hsu, Myrna Reginaldo, Jennifer Adachi, Deborah Graham and Peter Vann for their research and administrative assistance. Particular thanks goes to Gerome Asselin, Kathy Li, Kevin Craib and Martin Schechter for their advice on the statistical methods.
Sponsorship: This work was supported by the Michael Smith Foundation for Health Research through a Senior Scholar award to Robert Hogg.
1. Hammer SM, Squires KE, Hughes MD, Grimes JM, Demeter LM, Currier JS, 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.
2. Montaner JS, Hogg R, Raboud J, Harrigan R, O'Shaughnessy M. Antiretroviral treatment in. Lancet 1998; 352:1919–1922.
3. Wood E, Hogg RS, Yip B, Harrigan PR, O'Shaughnessy MV, Montaner JS. Effect of medication adherence on survival of HIV-infected adults who start highly active antiretroviral therapy when the CD4+
cell count is 0.200 to 0.350 × 109
cells/L. Ann Intern Med 2003; 139:810–816.
4. Sterling TR, Chaisson RE, Moore RD. Initiation of highly active antiretroviral therapy at CD4+
T lymphocyte counts of > 350 cells/mm3
: disease progression, treatment durability, and drug toxicity. Clin Infect Dis 2003; 36:812–815.
5. Cote HC, Brumme ZL, Craib KJ, Alexander CS, Wynhoven B, Ting L, et al
. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. N Engl J Med 2002; 346:811–820.
6. Yeni PG, Hammer SM, Hirsch MS, Saag MS, Schechter M, Carpenter CC, et al
. Treatment for adult HIV infection: 2004 recommendations of the International AIDS Society-USA Panel. JAMA 2004; 292:251–265.
7. BHIVA Writing Committee. British HIV Association (BHIVA) Guidelines for the Treatment of HIV-infected Adults with Antiretroviral Therapy
. London: British HIV Association Writing Committee; July 2003. Accessed January 2005: http://bhiva.org
8. National Institutes of Health. Guidelines for the Use of Antiretroviral Agents in HIV-Infected Adults and Adolescents
. Bethesda, MD: National Institutes of Health Updated October 29, 2004 (accessed January, 2005): http://www.aidsinfo.nih.gov/
9. Deeks SG. Treatment of antiretroviral-drug-resistant HIV-1 infection. Lancet 2003; 362:2002–2011.
10. Hogg RS, Yip B, Chan KJ, Wood E, Craib KJ, 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.
11. Cozzi Lepri A, Phillips AN, d'Arminio Monforte A, Castelli F, Antinori A, de Luca A, et al
. When to start highly active antiretroviral therapy in chronically HIV-infected patients: evidence from the ICONA study. AIDS 2001; 15:983–990.
12. 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.
13. Wood E, Hogg RS, Yip B, Quercia R, Harrigan PR, O'Shaughnessy MV, et al
. Higher baseline plasma HIV-1 RNA levels are associated with increased mortality after the initiation of triple drug antiretroviral therapy. J Infect Dis 2003; 188:1421–1425.
14. Wood E, Hogg RS, Yip B, Harrigan PR, Montaner JS. Why are baseline HIV RNA levels 100 000 copies/mL or greater associated with mortality after the initiation of antiretroviral therapy? J Acquir Immune Defic Syndr 2005; 38:289–295.
15. Bangsberg DR, Perry S, Charlebois ED, Clark RA, Roberston M, Zolopa AR, et al
. Non-adherence to highly active antiretroviral therapy predicts progression to AIDS. AIDS 2001; 15:1181–1183.
16. Wood E, Hogg RS, Harrigan PR, Montaner JS. When to initiate antiretroviral therapy in HIV-infected adults: a review for clinicians and patients. Lancet Infect Dis 2005; 5:407–414.
17. Mellors JW, Munoz A, Giorgi JV, Margolick JB, Tassoni CJ, Gupta P, et al
. Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection. Ann Intern Med 1997; 126:946–954.
18. Mellors JW, Rinaldo CR Jr, Gupta P, White RM, Todd JA, Kingsley LA. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 1996; 272:1167–1170.
19. Strathdee SA, Palepu A, Cornelisse PG, Yip B, O'Shaughnessy MV, Montaner JS, et al
. Barriers to use of free antiretroviral therapy in injection drug users. JAMA 1998; 280:547–549.
20. Hogg RS, Heath KV, Bangsberg D, Yip B, Press N, O'Shaughnessy MV, et al
. Intermittent use of triple combination therapy is predictive of mortality at baseline and after one year of follow-up. AIDS 2002; 16:1051–1058.
21. Low-Beer S, Yip B, O'Shaughnessy MV, Hogg RS, Montaner JS. Adherence to triple therapy and viral load response. J Acquir Immune Defic Syndr 2000; 23:360–361.
22. Wood E, Hogg RS, Yip B, Harrigan PR, O'Shaughnessy MV, Montaner JS. Is there a baseline CD4 cell count that precludes a survival response to modern antiretroviral therapy? AIDS 2003; 17:711–720.
23. Carpenter CC, Fischl MA, Hammer SM, Hirsch MS, Jacobsen DM, Katzenstein DA, et al
. Antiretroviral therapy for HIV infection in 1997. Updated recommendations of the International AIDS Society-USA panel. JAMA 1997; 277:1962–1969.
24. Schechter M. Therapy for early HIV infection: how far back should the pendulum swing? J Infect Dis 2004; 190:1043–1045.
25. Lane HC, Neaton JD. When to start therapy for HIV infection: a swinging pendulum in search of data. Ann Intern Med 2003; 138:680–681.
26. Holmberg SD, Palella FJ Jr, Lichtenstein KA, Havlir DV. The case for earlier treatment of HIV infection. Clin Infect Dis 2004; 39:1699–1704.
27. Phillips AN, Lepri AC, Lampe F, Johnson M, Sabin CA. When should antiretroviral therapy be started for HIV infection? Interpreting the evidence from observational studies. AIDS 2003; 17:1863–1869.
28. Wood E, Hogg RS, Yip B, Harrigan PR, Montaner JS. Earlier initiation of highly active antiretroviral therapy does not protect against the deleterious effects of non-adherence. AIDS 2004; 18:2432–2434.
29. Gross R, Bilker WB, Friedman HM, Strom BL. Effect of adherence to newly initiated antiretroviral therapy on plasma viral load. AIDS 2001; 15:2109–2117.
30. Chene G, Sterne JA, May M, Costagliola D, Ledergerber B, Phillips AN, et al
. Prognostic importance of initial response in HIV-1 infected patients starting potent antiretroviral therapy: analysis of prospective studies. Lancet 2003; 362:679–686.
31. van der Ryst E, Kotze M, Joubert G, Steyn M, Pieters H, van der Westhuizen M, et al
. Correlation among total lymphocyte count, absolute CD4+ count, and CD4+ percentage in a group of HIV-1-infected South African patients. J Acquir Immune Defic Syndr Hum Retrovirol 1998; 19:238–244.
32. Hogg RS, Weber AE, Craib KJ, Anis AH, O'Shaughnessy MV, Schechter MT, et al
. One world, one hope: the cost of providing antiretroviral therapy to all nations. AIDS 1998; 12:2203–2209.
33. Wood E, Hogg RS, Bonner S, Kerr T, Li K, Palepu A, et al
. Staging for antiretroviral therapy among HIV-infected drug users. JAMA 2004; 292:1175–1177.
34. Steiner JF, Prochazka AV. The assessment of refill compliance using pharmacy records: methods, validity, and applications. J Clin Epidemiol 1997; 50:105–116.
35. Grossberg R, Zhang Y, Gross R. A time-to-prescription-refill measure of antiretroviral adherence predicted changes in viral load in HIV. J Clin Epidemiol 2004; 57:1107–1110.
36. Gross R, Zhang Y, Grossberg R. Medication refill logistics and refill adherence in HIV. Pharmacoepidemiol Drug Saf 2005; 14:789–793.