Determinants of Interruption and Discontinuation of HAART
There were 2397 person-visit pairs (514 men) in the continuing HAART group, 197 person-visit pairs (127 men) in the interrupted HAART group, and 109 person-visit pairs (100 men) in the discontinued HAART group who had complete data on all potential predictors at Vi (Table 2). Univariately, younger age (P = 0.001), black race (P < 0.001), lower CD4 cell count (P = 0.011), higher HIV RNA level (P < 0.001), CES-D score ≥16 (P < 0.001), shorter time on HAART (P = 0.004), and not taking 3TC (P = 0.004) predicted interrupting HAART. Interruption differed by geographic location as well; those in Baltimore and Chicago were slightly more at risk for interrupting HAART than those in Pittsburgh and Los Angeles. Site did not modify the effects of other covariates of interruption (data not shown). With the exception of CD4 cell count, these factors remained independent predictors of interrupting HAART (see Table 2). After 1998, when adherence data were ascertained, men with lower adherence (<100%) at Vi were more likely to interrupt HAART in the following 6 months (odds ratio [OR] = 3.52, 95% confidence interval (CI): 2.18, 5.68; from the multivariate model). When adherence was included in the model, younger age and shorter time on HAART were no longer statistically significant, because younger age was associated with nonadherence (P = 0.004) and the restriction to after 1998 removed those with shorter times on HAART.
Among those interrupting HAART from April 1999 onward, 44.7% stopped all antiretrovirals for <4 days the last time, 25.0% stopped for 4 to 7 days, and 30.3% stopped for >7 days. Compared with continuing HAART, the independent predictors for interrupting HAART for ≤7 days (92 person-visit pairs) were younger age (OR = 1.30, 95% CI: 1.02, 1.65), black race (OR = 2.42, 95% CI: 1.23, 4.74), CES-D score ≥16 (OR = 2.17, 95% CI: 1.26, 3.73), and lower adherence to HAART (OR = 4.53, 95% CI: 2.66, 7.72). Those whose HAART regimen contained 3TC were less likely to interrupt therapy for ≤7 days (OR = 0.44, 95% CI: 0.25, 0.77). Predictors of missing all antiretrovirals for >7 days (40 person-visit pairs) were lower CD4 cell count (OR = 1.72, 95% CI: 1.08, 2.72), higher HIV RNA level (OR = 1.64, 95% CI: 1.22, 2.21), and CES-D score ≥16 (OR = 2.27, 95% CI: 1.03, 4.98)
Information on physician concordance for interruptions was collected since April 2000. Overall, 32.6% of the group who interrupted HAART reported that the interruption was prescribed by their physician. The extent of physician concordance differed by the length of the interruption: 20.3% of interruptions for ≤7 days were prescribed by physicians compared with 58.6% of interruptions for >7 days (P < 0.001).
As shown in Table 2, the factors univariately associated with discontinuing HAART were younger age (P = 0.003); lower CD4 cell count (P = 0.026); higher HIV RNA level (P < 0.001); CES-D score ≥16 (P < 0.001); and use of a regimen containing abacavir (P = 0.017), amprenavir (P = 0.018), or lopinavir (P = 0.005). Multivariately, CD4 cell count and using a regimen containing amprenavir did not independently predict discontinuing HAART. Restricting the analysis to data collected after 1998, <100% adherence predicted discontinuing HAART univariately (OR = 2.02, 95% CI: 1.20, 3.39). After controlling for other determinants, the effect of lower adherence was slightly diminished and was not statistically significant (OR = 1.62, 95% CI: 0.97, 2.71).
We also examined the self-reported reasons given by interrupters and discontinuers for discontinuing specific antiretrovirals. HAART discontinuers were significantly more likely to report stopping because of side effects than HAART interrupters: 58% versus 34%, respectively (P < 0.001). Also, 50.5% of the discontinuers reported that stopping was a result of their physician's recommendation compared with 32.6% of interrupters (P < 0.001). The proportion of discontinuers reporting discontinuation of HAART as a personal decision was similar to that of interrupters (33% and 30%, respectively, P = 0.505), however.
Consequences of Interruption and Discontinuation of HAART
Among individuals with ≤1000 HIV RNA copies/mL at Vi, 5.2% of the men who continued HAART use had a ≥1 log10 HIV RNA increase in the next 6 months. Such increases were more likely in those interrupting or discontinuing HAART than in those continuing HAART (Table 3). The increase in HIV RNA among those interrupting HAART was mostly driven by those who stopped antiretroviral therapy for >7 days (35.7% had a ≥1 log10 HIV RNA increase; OR = 10.07, P < 0.001 compared with continuing HAART). Only 5.4% of those interrupting HAART for <7 days had such HIV RNA increases.
Among those with 1001 to 10,000 HIV RNA copies/mL at Vi, less than 3% of those interrupting or continuing HAART had a ≥1 log10 HIV RNA increase. Only those who discontinued HAART were significantly (P < 0.001) more likely to have such an HIV RNA increase (see Table 3). Few HAART users with >10,000 HIV RNA copies/mL had a ≥1 log10 increase in HIV RNA at Vi+1.
The average percent increases in CD4 cell count over 6 months for the continuing HAART group were 7.9%, 5.3%, and 5.5% for those with HIV RNA levels ≤1000, 1001 to 10,000, and >10,000 copies/mL at Vi, respectively. Average percent increases in CD4 cell count for those interrupting HAART overall did not significantly differ from those for the HAART continuers (Table 4). The mean percent changes in CD4 cell count for those interrupting HAART for ≤7 days were 13.7% (P = 0.152 compared with continuing HAART), 4.3% (P = 0.893), and −7.4% (P = 0.106) in the 3 HIV RNA strata, respectively. Those who interrupted HAART for >7 days also had a similar mean percentage of CD4 changes compared with the continuing HAART group when starting with ≤1000 HIV RNA copies/mL (13.3%; P = 0.552) and 1001 to 10,000 HIV RNA copies/ml (4.2%; P = 0.897). The mean percent change in CD4 for those who interrupted HAART for >7 days was −11.5% (P = 0.01) for those who had an HIV RNA level >10,000 copies/mL at Vi, however. Those who discontinued HAART also had changes in CD4 cell count of −13.2% (P < 0.001), −5.3% (P = 0.10), and −11.7% (P = 0.028) in the 3 HIV RNA strata, respectively, representing significant differences from those who continued HAART (see Table 4). The consequences of interrupting and discontinuing HAART did not change when adjusted for age and race (data not shown).
To assess whether discontinuation had longer consequences, we compared the 78% of the interrupters and the 42% of the discontinuers who resumed HAART from Vi+1 to Vi+2 and for whom we had biomarker data at Vi+2 with those who continuously used HAART. Although few of those who discontinued and resumed HAART had any subsequent increase in HIV RNA levels from Vi+1 to Vi+2 (data not shown), the discontinuers still were more likely to have a ≥1 log10 HIV RNA increase at Vi+2 as compared with that measured at Vi (compared with continuous HAART users; OR = 2.03, 95% CI: 0.72, 5.72). Similarly, although their percent change in CD4 cell count from Vi+1 to Vi+2 (ie, after resuming use of HAART) was similar to that of those who continuously used HAART, it did not compensate for the decrement observed from Vi to Vi+1. Whereas the continuous HAART users had an annual increase of 10.6% (95% CI: 9.2, 12.0) on average by Vi+2, the discontinuers overall had an annualized loss of 1.3% of their Vi CD4 cell count at Vi+2, representing a significant (P = 0.02) difference in the change of CD4 cell count. Compared with those who continuously used HAART, there were no significant differences in the change of HIV RNA level and CD4 cell count at Vi+2 for those who interrupted therapy between Vi and Vi+1.
There has been great interest in the interruption and discontinuation of HAART in clinical practice because of drug resistance, severe toxicity, and the high cost of HAART. In this study, the proportion of HAART users interrupting therapy decreased since 1997 but increased after 1999. The low proportion discontinuing therapy has increased over time. The increasing trend may reflect changes in standard treatment guidelines, suggesting a more conservative approach, difficulty in maintaining continuous use of HAART, and/or the impact of alternative interruption strategies recently introduced in clinical trials6-9 as well as in the news. The increasing trend of discontinuation of HAART in this cohort was consistent with a report from the Women's Interagency HIV Study (WIHS).20
Recommended interruptions (ie, those made in concordance with physicians) also may have resulted from practical needs. Physicians may prefer that all medications are halted rather than using partial suboptimal regimens. Little is known about the consequences of interrupting HAART by the clinicians or patients outside of the STI clinical trials.6-10 Therefore, we studied not only the consequences but the determinants of HAART interruption to understand better why interruptions occur in the general HIV-1-infected population. Whereas our analysis of independently ascertained markers of HIV disease stage (low CD4 cell count and high HIV RNA level) predicted interrupting HAART for >7 days, host characteristics (eg, age, race, lower adherence) were associated with short interruption of HAART (≤7 days). These factors, which indicate personal decision making, were consistent with the finding that those with short interruptions were less likely to have their interruptions prescribed by their physicians, indicating that the short interruptions may be a compliance issue. The finding that taking a 3TC-containing regimen was protective against short-term interruption might be explained by the fact that 3TC has lower toxicity than other antiretroviral drugs.17
Future analyses need to differentiate those who discontinue HAART for personal reasons, side effects, and as a result of physician recommendation, because these causes relate to different lengths of time off HAART and may have different consequences.
A pilot study reported that HIV RNA was suppressed to low levels and CD4 cell count returned to preinterruption levels after resumption of HAART,7 consistent with our findings of no significant consequences of short-term HAART interruption (≤7 days) when compared with uninterrupted use of HAART. Similarly, although a greater proportion of those interrupting for >7 days among those with <1000 HIV RNA copies/mL had HIV RNA increases, the CD4 cell count remained stable. These data suggest that among individuals whose HIV RNA level is relatively low, HAART interruptions may not significantly affect immunologic outcomes at 6 months. We could not examine how long the person was back on HAART when the HIV RNA level and CD4 cell count were measured at Vi+1, but there was no effect on the HIV RNA level and CD4 cell count measured at Vi+2 at least 6 months after the interruption occurred. Results from other studies have been inconsistent. Taffe et al21 showed that occasional interruptions of 1 to 3 months did not worsen clinical disease outcome in 3 to 4 years when the HIV RNA level is low and the CD4 cell count is high, supporting the finding that with resumption of HAART, the occasional interruption may not be detrimental among these people. Yet, frequently reported short interruptions have been shown to lead to virologic failure.22 Data from clinical trials such as Swiss-Thai-Australia Treatment Interruption Trial (STACCATO),22 in which interruptions are driven by CD4 levels, should offer additional information on the effects of the longer interruptions that are anticipated in these individuals.
The observation that higher HIV RNA level and lower CD4 cell count predicted discontinuation of HAART is consistent with the findings of other studies.20,23-25 We also found that depression and lower adherence to HAART were associated with subsequent discontinuation of HAART. Depressive symptoms have been associated with reduced adherence to HAART.26,27 In addition, we found that younger age and taking a regimen containing abacavir or lopinavir predicted discontinuing HAART. Although these regimens may represent salvage therapy, use of antiretroviral therapy before HAART initiation, length of time on HAART, and number of prior HAART regimens were not independently associated with discontinuation.
The increase in HIV RNA level and decrease in CD4 cell count observed after a 6-month interval among HAART discontinuers were similar to results in other studies.6,7,20,28 Although these individuals demonstrated an increase in CD4 cell count on subsequent resumption of HAART, they still had a significant decrement at the next visit (at least 6 months after the discontinuation) compared with those who continuously used HAART. The observed detrimental effect on HIV markers from prolonged treatment discontinuation is consistent with the poor prognosis associated with nonadherence.29-33 Because discontinuers are more likely to have a history of lower adherence, they may be primed for a worse prognosis than those who continue HAART. Although we stratified by HIV RNA level at Vi, thus accounting, in part, for the effect of a history of lower adherence, there still may be some residual effect (ie, lower adherers may have higher viral loads within the strata).
Although these data were from a large cohort study, there are limitations. Our cohort does not include women and children, and most of the men have been infected for many years and had experience with antiretroviral therapy when starting HAART; thus, our findings might not be generalizable to all HIV-infected persons. Also, the numbers of observations in subgroups defined by missing all antiretrovirals for ≤7 days or >7 days and an HIV RNA level >1000 copies/mL at Vi are relatively small; therefore, inferences should be drawn with caution. In addition, the information on medication use, including adherence, is based on self-report; thus, the proportions interrupting HAART may be underestimates. These data were collected independent of health care setting (ie, not by the primary care physician or in a clinical trial setting), however, and thus should minimize the overreporting of continuous HAART use one may expect from patients. Another advantage is that the characteristics assessed as predictors were determined before the outcome.
In summary, our findings indicate that interruption and discontinuation of HAART are gradually increasing in the population. Short-term HAART interruption (≤7 days) among chronically infected HIV patients with an HIV RNA level <10,000 copies/mL did not seem to affect the treatment benefit. Thus, it is possible that short interruptions may be associated with reductions in cost and drug-related toxicity, although these reductions are probably minimal. This potentially viable treatment strategy would need to be tested in large controlled trials. We also could not assess the effect of repeated interruptions; thus, the results need to be viewed with caution. Longer interruptions, as shown here and in a recently reported study,34 seem to have negative effects on HIV disease. In a recent presentation,35 longer interruptions were associated with the development of drug-resistant virus, thus limiting the choice of antiretrovirals on resumption of HAART. Long-term effects of short- and long-term interruptions need to be studied, not only on immune reconstitution but on viral drug resistance and viral replicative capacity. Among those with more advanced HIV disease, patient education, particularly among younger individuals, is needed to emphasize the importance of minimizing unnecessary treatment interruptions. Given the relation to depression, offering mental health services may enhance continuous use of HAART.
The MACS includes the following institutions and individuals: The Johns Hopkins University Bloomberg School of Public Health (Baltimore): Joseph B. Margolick (Principal Investigator), Haroutune Armenian, Barbara Crain, Adrian Dobs, Homayoon Farzadegan, Shenghan Lai, and Justin McArthur; Howard Brown Health Center (Chicago), The Feinberg School of Medicine, Northwestern University (Chicago), and Cook County Bureau of Health Services (Chicago): John P. Phair (Principal Investigator), Joan S. Chmiel (Co-Principal Investigator), Sheila Badri, Bruce Cohen, Craig Conover, Maurice O'Gorman, Frank Pallela, Daina Variakojis, and Steven M. Wolinsky; University of California, University of California at Los Angeles Schools of Public Health and Medicine (Los Angeles): Roger Detels and Beth Jamieson (Principal Investigators), Barbara R. Visscher (Co-Principal Investigator), Anthony Butch, John Fahey, Otoniel Martínez-Maza, Eric N. Miller, John Oishi, Paul Satz, Elyse Singer, Harry Vinters, Otto Yang, and Stephen Young; University of Pittsburgh, Graduate School of Public Health (Pittsburgh): Charles R. Rinaldo (Principal Investigator), Lawrence Kingsley (Co-Principal Investigator), James T. Becker, Phalguni Gupta, John Mellors, Sharon Riddler, and Anthony Silvestre; Data Coordinating Center: The Johns Hopkins University Bloomberg School of Public Health (Baltimore): Lisa P. Jacobson (Principal Investigator), Alvaro Muñoz (Co-Principal Investigator), Haitao Chu, Stephen R. Cole, Stephen J. Gange, Janet Schollenberger, Eric Seaberg, Sol Su, and Xiuhong Li; National Institutes of Health, National Institute of Allergy and Infectious Diseases (Bethesda): Carolyn Williams and Robin Huebner; and National Institutes of Health, National Cancer Institute (Bethesda): Jodi Black. Web site located athttp://www.statepi.jhsph.edu/macs/macs.html.
1. Gulick RM, Mellors JW, Havlir D, et al. Treatment with indinavir, zi-dovudine, and lamivudine in adults with human immunodeficiency virus infection and prior antiretroviral therapy. N Engl J Med
2. 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
3. Detels R, Muñoz A, McFarlane G, et al. Effectiveness of potent antiret-roviral therapy on time to AIDS and death in men with known HIV
infection duration. Multicenter AIDS Cohort Study Investigators. JAMA
4. Yamashita TE, Phair JP, Muñoz A, et al. Immunologic and virologic response to highly active antiretroviral therapy
in the Multicenter AIDS Cohort Study. AIDS
5. Detels R, Tarwater P, Phair JP, et al. Effectiveness of potent antiretroviral therapies on the incidence of opportunistic infections before and after AIDS diagnosis. AIDS
6. Garcia F, Plana M, Ortiz GM, et al. The virological and immunological consequences of structured treatment interruptions in chronic HIV
-1 infection. AIDS
7. Ortiz GM, Wellons M, Brancato J, et al. Structured antiretroviral treatment interruptions in chronically HIV
-1-infected subjects. Proc Natl Acad Sci USA
8. Dybul M, Chun TW, Yoder C, et al. Short-cycle structured intermittent treatment of chronic HIV
infection with highly active antiretroviral therapy
: effects on virologic, immunologic, and toxicity parameters. Proc Natl Acad Sci USA
9. Oxenius A, Price DA, Gunthard HF, et al. Stimulation of HIV
-specific cellular immunity by structured treatment interruption
fails to enhance viral control in chronic HIV
infection. Proc Natl Acad Sci USA
10. Fischer M, Hafner R, Schneider C, et al. HIV RNA
in plasma rebounds within days during structured treatment interruptions. AIDS
11. Kaslow RA, Ostrow DG, Detels R, et al. The Multicenter AIDS Cohort Study: rationale, organization, and selected characteristics of the participants. Am J Epidemiol
12. Chmiel JS, Detels R, Kaslow RA, et al. Factors associated with prevalent human immunodeficiency virus (HIV
) infection in the Multicenter AIDS Cohort Study. Am J Epidemiol
13. Detels R, Phair JP, Saah AJ, et al, for the Multicenter AIDS Cohort Study. Recent scientific contributions to understanding HIV
/AIDS from the Multicenter AIDS Cohort Study. J Epidemiol
14. Dudley J, Jin S, Hoover D, et al. The Multicenter AIDS Cohort Study: retention after 9 1/2 years. Am J Epidemiol
15. Giorgi JV, Cheng HL, Margolick JB, et al. Quality control in the flow cytometric measurement of T-lymphocyte subsets: the Multicenter AIDS Cohort Study experience. The Multicenter AIDS Cohort Study Experience. Clin Immunol Immunopathol
16. Schenker EL, Hultin LE, Bauer KD, et al. Evaluation of a dual-color flow cytometry immunophenotyping panel in a multicenter quality assurance program. Cytometry
17. Guidelines for using antiretroviral agents among HIV
-infected adults and adolescents. Recommendations of the Panel on Clinical Practices for Treatment of HIV
. MMWR Recomm Rep
18. Davey RT, Jr, Bhat N, Yoder C, et al. HIV
-1 and T cell dynamics after interruption
of highly active antiretroviral therapy
(HAART) in patients with a history of sustained viral suppression. Proc Natl Acad Sci USA
19. Kleeberger CA, Phair JP, Strathdee SA, et al. Determinants of heterogeneous adherence to HIV
-antiretroviral therapies in the Multicenter AIDS Cohort Study. J Acquir Immune Defic Syndr
20. Ahdieh Grant L, Silverberg MJ, Palacio H, et al. Discontinuation
of potent antiretroviral therapy: predictive value of and impact on CD4 cell counts and HIV RNA
21. Taffe P, Rickenbach M, Hirschel B, et al. Impact of occasional short interruptions of HAART on the progression of HIV
infection: results from a cohort study. AIDS
22. Ananworanich J, Nuesch R, Le Braz M, et al. Failures of 1 week on, 1 week off antiretroviral therapies in a randomized trial. AIDS
23. Bassetti S, Battegay M, Furrer H, et al. Why is highly active antiretroviral therapy
(HAART) not prescribed or discontinued? Swiss HIV
Cohort Study. J Acquir Immune Defic Syndr
24. d'Arminio Monforte A, Lepri AC, Rezza G, et al. Insights into the reasons for discontinuation
of the first highly active antiretroviral therapy
(HAART) regimen in a cohort of antiretroviral naive patients. I.CO.N.A. Study Group. Italian Cohort of Antiretroviral-Naive Patients. AIDS
25. van Roon EN, Verzijl JM, Juttmann JR, et al. Incidence of discontinuation
of highly active antiretroviral combination therapy (HAART) and its determinants. J Acquir Immune Defic Syndr
26. Cook JA, Cohen MH, Burke J, et al. Effects of depressive symptoms and mental health quality of life on use of highly active antiretroviral therapy
-seropositive women. J Acquir Immune Defic Syndr
27. Starace F, Ammassari A, Trotta MP, et al. Depression is a risk factor for suboptimal adherence to highly active antiretroviral therapy
. J Acquir Immune Defic Syndr
. 2002;31(Suppl 3):S136-S139.
28. Tebas P, Henry K, Mondy K, et al. Effect of prolonged discontinuation
of successful antiretroviral therapy on CD4+ T cell decline in human immu-nodeficiency virus-infected patients: implications for intermittent therapeutic strategies. J Infect Dis
29. Paterson DL, Swindells S, Mohr J, et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV
infection. Ann Intern Med
30. Bangsberg DR, Perry S, Charlebois ED, et al. Non-adherence to highly active antiretroviral therapy
predicts progression to AIDS. AIDS
31. Hogg RS, Heath K, Bangsberg D, et al. Intermittent use of triple-combination therapy is predictive of mortality at baseline and after 1 year of follow-up. AIDS
32. Press N, Tyndall MW, Wood E, et al. Virologic and immunologic response, clinical progression, and highly active antiretroviral therapy
adherence. J Acquir Immune Defic Syndr
. 2002;31(Suppl 3):S112-S117.
33. Garcia de Olalla P, Knobel H, Carmona A, et al. Impact of adherence and highly active antiretroviral therapy
on survival in HIV
-infected patients. J Acquir Immune Defic Syndr
34. Vella S, Giuliano M, Palmisano L, et al. A prospective, randomized, multi-center clinical trial of intermittent therapy in HIV
+ subjects with persistent suppression of viral replication [Abstract 66]. Presented at the 10th Conference on Retroviruses and Opportunistic Infections, Boston, 2003.
35. Dybul M, Nies-Kraske E, Daucher M, et al. A randomized, controlled trial of long cycle structured intermittent versus continuous ARV therapy for chronic HIV
infection [Abstract 68Ib]. Presented at the 10th Conference on Retroviruses and Opportunistic Infections, Boston, 2003.
Keywords:© 2005 Lippincott Williams & Wilkins, Inc.
HIV; highly active antiretroviral therapy; interruption; discontinuation; HIV RNA; CD4 cell count