Abstract
Background
Understanding the impact of hepatitis B virus (HBV) coinfection on HIV outcomes in the HAART era continues to be a critical priority given the high prevalence of coinfection and the potential for impaired immunologic, virologic and clinical recovery.
Methods
Participants from the U.S. Military HIV Natural History Study with an HIV diagnosis, on HAART and serologically confirmed HBV infection status at HAART initiation (HI) were classified into four HBV infection (HB) groups. HIV virologic, immunologic, and clinical outcomes were evaluated by HB status.
Results
Of 2536 HIV positive HAART recipients, with HBV testing results available to determine HB status in the HI window, HB status at HI was classified as HB negative (n=1505; 66%), resolved HB (n=518; 23%); isolated HBcAb (n=139; 6%) or; chronic HB (n=131; 6%). HIV virologic suppression and failure at 6 months or 1 year were not significantly different by HB status. A significantly faster rate of increase in CD4 cell count during the period between 4 and 12 years was observed for chronic HB relative to HB negative.
Chronic and resolved HB were associated with an increased risk of AIDS/death compared to HB negative individuals (chronic HB; HR=1.68; 95% CI 1.05–2.68, resolved HB; HR=1.61; 95% CI 1.15–2.25).
Conclusions
HB status did not have a significant impact on HIV virologic outcomes, however, CD4 cell count reconstitution post HI and the risk of an AIDS event or death post HI may be associated with HB status.
Keywords: Hepatitis B virus, chronic hepatitis B, human immunodeficiency virus, highly active antiretroviral therapy
INTRODUCTION
Hepatitis B virus (HBV) is more common in HIV-infected individuals than in the general population due to similar routes of transmission for viral acquisition [1–3]. Current evidence suggests that HIV infection has an adverse impact on HBV-related liver disease progression, with an increase in HBV replication, reduction in the rate of clearance of serum hepatitis B e antigen, and increased risk of cirrhosis, liver-related mortality and hepatocellular carcinoma at lower CD4+ T cell counts [1,4–6]. Coinfection rates are estimated between 5–10%, with up to 40% of immunocompromised patients developing chronic infection [6].
Clinical studies prior to the general availability of HAART evaluating the impact of HB on HIV progression have been mixed [7–9]. Some studies found no differences in HIV progression between those with and without chronic HB [1,8,10], while other studies have shown that chronic HB may negatively impact HIV progression with a significant increased risk of AIDS or death [11,12,13].
Studies evaluating the influence of HIV-HBV coinfection on HIV RNA suppression, immunologic CD4 cell count recovery and clinical outcomes in individuals on HAART have been limited and conflicting, with several studies finding no difference [14–18]. Several studies from HBV endemic countries have also found no difference on HIV outcomes [19–22]. Law et al showed a smaller early increase in CD4 response post HI, however this was not sustained [23]. Hawkins et al showed coinfected compared to mono-infected individuals had significantly lower CD4+ counts throughout the period of recovery [24]. Other studies have shown coinfected individuals are less likely to achieve virologic suppression (VS) as compared to HIV monoinfected individuals [25–26]. Recent studies have for the most part failed to show a substantial impact of HBV coinfection on immunologic or HIV virologic responses to ART [21,24,27]. Idoko et al, however, found a lower proportion of HBeAg positive individual s achieving HIV virologic suppression at 24 weeks as compared to HBeAg negative or HIV mono-infected individuals, but the findings were not seen at 48 weeks [22]. One recent study evaluating HIV outcomes during the first three years of ART found impaired CD4 cell recovery in HBsAg-positive and anti-HBc patients as compared to HBV-uninfected patients [28]. Heterogeneity of available data warrant further evaluation of this key question.
Limitations and heterogeneity of results from other studies may be attributed to the small number of individuals analyzed, defining patients with chronic HB sometimes with only one positive test for HBsAg, or limited follow up post HI. We sought to evaluate the impact of HBV infection in HIV coinfected HAART recipients in a large cohort with known and limited duration of HIV infection, free access to health care, racial diversity, minimal injection drug use (IDU), and long term follow-up.
METHODS
Study Participants and Definitions
The U.S. Military HIV Natural History Study (NHS) is a prospective multicenter continuous enrollment observational cohort of HIV-infected active duty military personnel and other beneficiaries (spouses, adult dependents, and retired military personnel), with over 5400 HIV-infected participants from the Army, Navy/Marines and Air Force enrolled since 1986. Participants are followed at five military medical centers in the United States. Demographics, medical and medication histories, and standard laboratory studies are collected biannually as previously described [13]. In the NHS, dates of death are collected through the review of death certificates and medical records by study staff as well as by searching the Social Security Death Index and National Death Index databases annually. Although not captured in the NHS database, IDU has been reported to be very rare in this cohort [29]. All NHS participants provided informed consent, and approval for this research was obtained from the institutional review board at each participating site.
NHS participants with a documented HIV diagnosis who ever received HAART and had a categorizable HB status were considered for these analyses. HAART was defined according to guidelines from the DHHS/Kaiser Panel and was defined as: (a) two or more NRTIs in combination with at least one PI or one NNRTI (88% of observations classified as HAART); (b) one NRTI in combination with at least one PI and at least one NNRTI (5%); (c) a regimen containing ritonavir and saquinavir in combination with one NRTI and no NNRTIs (1%); and (d) an abacavir or tenofovir containing regimen of three or more NRTIs in the absence of both PIs and NNRTIs (6%), except for the three-NRTI regimens consisting of: abacavir + tenofovir + lamivudine or didanosine + tenofovir + lamivudine [30]. HBV-active HAART was defined as HAART containing lamivudine, emtricitabine or tenofovir disoproxil fumarate.
Screening for HB was performed in accordance with clinical standards of care and practice guidelines at the time, and included hepatitis B surface antigen (HBsAg), total antibody to hepatitis B core antigen (HBcAb), and hepatitis B surface antibody (HBsAb). Those whose HB status could not be categorized were excluded from these analyses. The remaining participants were classified into one of four mutually exclusive groups determined by HB status in the HAART initiation window (HIW, −396 to +90 days from HAART initiation), defined as follows: 1) Chronic HB: HBsAg reactivity on two or more separate occasions at least 6 months apart with at least one positive HBsAg within the HIW or one positive HBsAg before HI and one after HI; 2) Isolated HBcAb: HBcAb reactivity on two or more occasions without any other reactive HBV marker; 3) Resolved HB: a panel with markers reactive for HBcAb and HBsAb concurrently closest to HI, or before and after the HIW and HBsAg negative; or 4) HB negative: both HBsAg and HBcAb negative in the HIW or after the HIW with no prior positive panels.
Hepatitis C virus (HCV) infection was defined as having at least one positive HCV antibody test prior to or in the HIW without any negative HCV antibody thereafter. For those not classified as positive for HCV infection, a negative HCV antibody anytime after HI was used to classify individuals as HCV antibody negative at HI. An HIV seroconverter was defined as having a documented negative and positive HIV date.
HIV virologic suppression (VS) after HI was defined as having at least one VL assay <400 copies/ mL within 6 months or 1 year after HI. Among subjects having ever suppressed VL, virologic failure (VF) was defined as having two consecutive VL assays ≥400 copies/mL.
AIDS-defining illnesses were defined using the 1993 CDC classification excluding an isolated CD4 count < 200 cells/µL [31]. Individuals whose visit occurred after HI were included in this model.
Participants were followed from the time of HI for the composite endpoint of an AIDS-defining illness or death from any cause through 18 months after the last study visit. For those without an event, data were censored 18 months after the last study visit.
Statistical analysis
The four groups, defined by baseline HB status, were summarized with descriptive statistics. All continuous covariates were tested for normality using Shapiro-Wilks; covariates with a P value < 0.05 were considered not normally distributed. All covariates were non-normal, so medians were compared using Kruskal-Wallis test for continuous data. Categorical data were evaluated with either Fisher’s exact test when the cell contained small counts, or the Chi-squared test.
Logistic regression was used to estimate the adjusted odds ratios for ever achieving VS within six months and one year of HI. Subjects were excluded from analyses if no VL assays within either 6 months or 1 year of HI or any model covariates were missing. Sensitivity analyses were performed on all models to assess the effect of missing data on outcomes utilizing multiple imputation (MI) with predictive mean matching for missing data [32]. Backwards selection methods, based on significance levels in the full mode (P<0.1) and other clinically relevant covariates, were later added. The final adjusted models (within 6 and 12 months of HI) were adjusted for age at HI, gender, race, era of HI, HCV status, VL and CD4 cell count at HI, ART use prior to HI and number of VL assays during 6 or 12 months of HI. Odds ratios (OR) are given with 95% confidence intervals (CI). A Cox proportional hazard model was used to estimate the adjusted hazard rate of VF. Only those who ever achieved virologic suppression were included. Time to VF was defined as the number of years from the first suppressed VL after HI to the first VF after HI. Backward step-wise selection was used to choose statistically significant covariates then other clinically relevant covariates were added to the model. Multiple imputations were used as a sensitivity analysis to assess the effect of missing data on parameter estimation. The final adjusted model was adjusted for age at HI, race, gender, HCV status at HI, year of HI, HIV VL at HI, CD4 count at HI, ART prior to HI, AIDS prior to HI, years from HI to VS. Hazard ratios are given with 95% confidence intervals.
To quantify the effect of covariates at HI on the rate of increase in CD4 cell count, analysis of immunologic outcomes was evaluated using linear spline mixed models to account for within-subject repeated CD4 measures and to assess the effect of HB status and other factors on CD4 trajectories. Knots for the linear spline model were selected by inspection of LOWESS curves. Backwards selection methods were used to select covariates. Mean rates of increase in CD4 were allowed to differ by CD4 at HI, HB status at HI and the era of HI. P values <0.05 were considered statistically significant. Given the Q-Q-plot comparing the standardized residuals to theoretical normal quantiles showed a lack of fit in the tails, a regression technique using the multivariate t distribution and bootstrapping was used as a sensitivity analysis with respect to normality and outlier to improve model fit and compare parameters with the standard mixed model approach. R programming 2.13.2 was used for all other analyses [33].
The number of events, person-years (PY) at risk, and unadjusted rates per 100 PY were calculated for the four baseline groups; proportional hazard ratios were estimated using Cox regression models.
RESULTS
Of 5403 HIV-positive participants enrolled in the NHS with a documented HIV positive date, 2797 were ever on HAART (after 1996), 2536 (81%) had HBV testing results available to determine HB status in the HIW and were included in these analyses. Individuals whose HB status were not classified differed from those who were in that they were older at HAART start, had a greater percentage of HCV seropositivity, had a longer duration of HIV prior to HAART initiation, were more often diagnosed in the pre HAART era and had lower CD4 nadirs (data not shown). HB status at HI was classified as HB negative (n=1505; 66%), resolved HB (n=518; 23%); isolated HBcAb (n=139; 6%) or chronic HB (n=131; 6%). Characteristics at time of HI by HB status are shown in Table 1. Age at HIV diagnosis and HI (overall median 29 years; IQR 24–35 and 33 years; IQR 28–39, respectively) were significantly different among the four HB groups, with HB negative individuals diagnosed with HIV in the later calendar years, initiating HAART at the youngest age, having the lowest number of individuals with prior AIDS, having the least number with prior ART use, and having the highest median CD4 count at HI. Isolated core subjects had the highest proportion of HCV (15%). The chronic HB group had the lowest median CD4 at HI (median 278; IQR 146–410), and highest alanine aminotransferase (ALT) values (median 59 IU/L; IQR 33–102). HIV RNA levels at the time of HI were available for 2005 (87%) participants, and were similar among the groups (median 4.5 log10 copies/ml; IQR 3.8–5). Median follow-up was 9.2 years; IQR 4.4 – 16. Of all subjects, 93% received HBV-active HAART regimens at HI. Of individuals initiating HAART between 1996 and 1999, the majority were on unboosted PIs (78%); however, from 2000 to 2012, 68% were on an NNRTI regimen (data not shown).
Table 1.
Baseline Characteristics
| Total | Negative | Chronic | Isolated Core |
Resolved | P value | |
|---|---|---|---|---|---|---|
| Subjects | 2293 | 1505 | 131 | 139 | 518 | |
| Age at HIV Dx, | <0.001 | |||||
| years | 29 (24,35) | 28 (24,34) | 29 (25,33) | 29 (25,37) | 31 (26,37) | |
| Age at HI, | <0.001 | |||||
| years | 33 (28,39) | 31 (26,37) | 37 (33,42) | 38 (32,43) | 37 (32,41) | |
| Gender | <0.001 | |||||
| Male | 2118 (92%) | 1345 (89%) | 127 (97%) | 134 (96%) | 512 (99%) | |
| Female | 175 (8%) | 160 (11%) | 4 (3%) | 5 (4%) | 6 (1%) | |
| Race | <0.001 | |||||
| Caucasian | 978 (43%) | 645 (43%) | 48 (37%) | 46 (33%) | 239 (46%) | |
| African-American | 983 (43%) | 618 (41%) | 69 (53%) | 77 (55%) | 219 (42%) | |
| Hispanic/Other | 332 (14%) | 242 (16%) | 14 (11%) | 16 (12%) | 60 (12%) | |
| HCV at HI | <0.001 | |||||
| Negative | 2187 (96%) | 1458 (97%) | 119 (91%) | 118 (85%) | 492 (95%) | |
| Positive | 97 (4%) | 39 (3%) | 12 (9%) | 21 (15%) | 25 (5%) | |
| Year of HIV Dx, | <0.001 | |||||
| year | 1997 (1991,2003) |
1999 (1994,2005) |
1989 (1986,1992) |
1992 (1988,1997) |
1992 (1988,1998) |
|
| HIV Dx to HI, | <0.001 | |||||
| months | 27 (3.4,79) | 14 (2.5,51) | 100 (52,123) | 79 (29,116) | 58 (9.7,109) | |
| Year of HI | <0.001 | |||||
| 1996 – 1999 | 1259 (55%) | 691 (46%) | 107 (82%) | 93 (67%) | 368 (71%) | |
| 2000 – 2012 | 1034 (45%) | 814 (54%) | 24 (18%) | 46 (33%) | 150 (29%) | |
| Seroconverter | <0.001 | |||||
| No | 450 (20%) | 210 (14%) | 54 (41%) | 47 (34%) | 139 (27%) | |
| Yes | 1843 (80%) | 1295 (86%) | 77 (59%) | 92 (66%) | 379 (73%) | |
| HBV Vaccine before HI | <0.001 | |||||
| No | 348 (15%) | 289 (19%) | 6 (5%) | 17 (12%) | 36 (7%) | |
| Yes | 1945 (85%) | 1216 (81%) | 125 (95%) | 122 (88%) | 482 (93%) | |
| AIDS before HI | <0.001 | |||||
| No | 2111 (92%) | 1414 (94%) | 108 (82%) | 118 (85%) | 471 (91%) | |
| Yes | 182 (8%) | 91 (6%) | 23 (18%) | 21 (15%) | 47 (9%) | |
| HIV VL at HI Available | 0.363 | |||||
| No | 288 (13%) | 179 (12%) | 22 (17%) | 17 (12%) | 70 (14%) | |
| Yes | 2005 (87%) | 1326 (88%) | 109 (83%) | 122 (88%) | 448 (86%) | |
| HIV VL at HI | 0.619 | |||||
| log10(copies/mL) | 4.5 (3.8,5) | 4.5 (3.9,5) | 4.5 (3.6,4.9) | 4.4 (3.5,5) | 4.5 (3.8,4.9) | |
| CD4 at HI Available | 0.44 | |||||
| No | 278 (12%) | 184 (12%) | 21 (16%) | 14 (10%) | 59 (11%) | |
| Yes | 2015 (88%) | 1321 (88%) | 110 (84%) | 125 (90%) | 459 (89%) | |
| CD4 at HI | <0.001 | |||||
| Cells/mm3 | 338 (224,460) | 350 (243,471) | 278 (146,410) | 280 (122,422) | 332 (217,454) | |
| ART before HI | <0.001 | |||||
| No | 1334 (58%) | 1046 (70%) | 26 (20%) | 52 (37%) | 210 (41%) | |
| Yes | 959 (42%) | 459 (30%) | 105 (80%) | 87 (63%) | 308 (59%) | |
| ALT at HI | <0.001 | |||||
| UI/L | 33 (24,48) | 31 (23,45) | 59 (33,102) | 32 (24,52) | 33 (24,49) | |
| Initial Regimen | <0.001 | |||||
| 3 NRTI | 110 (5%) | 84 (6%) | 4 (3%) | 3 (2%) | 19 (4%) | |
| Boosted PI | 234 (10%) | 168 (11%) | 10 (8%) | 13 (9%) | 43 (8%) | |
| NNRTI | 826 (36%) | 636 (42%) | 24 (18%) | 36 (26%) | 130 (25%) | |
| PI+NNRTI+NRTI | 69 (3%) | 33 (2%) | 10 (8%) | 6 (4%) | 20 (4%) | |
| Unboosted PI | 1054 (46%) | 584 (39%) | 83 (63%) | 81 (58%) | 306 (59%) | |
| HBV-active Regimen | <0.001 | |||||
| No | 170 (7%) | 76 (5%) | 12 (9%) | 14 (10%) | 68 (13%) | |
| Yes | ||||||
| Dual HBV-active ART | 627 (27%) | 519 (34%) | 15 (11%) | 25 (18%) | 68 (13%) | |
| Mono HBV-active ART | 1496 (65%) | 910 (60%) | 104 (79%) | 100 (72%) | 382 (74%) | |
| HIV Dx to last Med History, | <0.001 | |||||
| years | 9.2 (4.4,16) | 7.4 (3.5,13) | 16 (9.9,21) | 13 (8.6,20) | 13 (7.2,20) | |
| Regimen Change in Follow-up | <0.001 | |||||
| No | 773 (35%) | 583 (40%) | 26 (20%) | 42 (31%) | 122 (24%) | |
| Yes | 1460 (65%) | 878 (60%) | 102 (80%) | 92 (69%) | 388 (76%) | |
| HI to last Med Hx | <0.001 | |||||
| Years | 5.5 (2,12) | 4.5 (1.9,10) | 8 (2,14) | 7.4 (2.5,14) | 8.5 (2.6,14) |
All data are No. (%) for categorical data and median (IQR) for continuous data.
Abbreviations: Negative, negative HBV infection; Chronic, chronic HBV infection; Isolated Core, isolated core HBV infection; Resolved, resolved HBV infection; Dx, diagnosis; HI, HAART initiation, HIV, human immunodeficiency virus; HCV, hepatitis C virus; VL, viral load; ART, antiretroviral therapy; ALT, alanine aminotransferase.
HIV Virologic Suppression
In unadjusted analyses by HB status, HB status was associated with HIV VS in the first 6 months and 1 year (data not shown). Within 6 months, VS rates for 1015 subjects was 53% overall, with HB negative, chronic HB, isolate core and resolved HB achieving rates of 58%, 35%, 44% and 45%, respectively, P<0.001. Within 1 year, VS rates for 1250 subjects was 59% overall, with HB negative, chronic HB, isolate core and resolved HB achieving rates of 65%, 40%, 53% and 47%, respectively, P<0.001. Table 2 presents the adjusted odds ratios for achieving HIV VS within 6 months and one year of HI for the four HB groups. The number of subjects included in model 1 (6 months) and 2 (12 months) were 1922 and 2130 subjects, respectively. The adjusted odds ratio for achieving VS was not significantly different by HB status; while age, year of HI, VL at HI, prior ART and HIV VL assay count were significantly associated with the odds ratios for achieving VS within 6 months and 12 months of HI, and African-American race was significantly associated with achieving VS within 12 months of HI. HB status at HI, gender, HCV status at HI, CD4 count at HI, AIDS prior to HI were not predictive of VS.
Table 2.
Predictors of HIV Virologic Suppression (Logistic Regression)
| Six Months | One Year | |||
|---|---|---|---|---|
| OR (95% CI) | P value | OR (95% CI) | P value | |
| HBV Status at HI | ||||
| Negative | 1 | (ref) | 1 | (ref) |
| Chronic | 1.06 (0.59,1.88) | 0.848 | 0.89 (0.51,1.53) | 0.682 |
| Isolated Core | 0.81 (0.46,1.4) | 0.448 | 1.06 (0.63,1.79) | 0.828 |
| Resolved | 1.02 (0.74,1.4) | 0.911 | 0.77 (0.56,1.04) | 0.089 |
| Age at HI, | ||||
| per 10 years | 1.39 (1.19,1.63) | <0.001 | 1.37 (1.17,1.6) | <0.001 |
| Race | ||||
| Caucasian | 1 | (ref) | 1 | (ref) |
| African-American | 0.9 (0.69,1.17) | 0.43 | 0.74 (0.56,0.96) | 0.023 |
| Hispanic | 1.28 (0.88,1.89) | 0.2 | 1.28 (0.86,1.91) | 0.217 |
| Gender | ||||
| Male | 1 | (ref) | 1 | (ref) |
| Female | 0.88 (0.54,1.44) | 0.62 | 0.89 (0.56,1.42) | 0.634 |
| HCV Status at HI | ||||
| Negative | 1 | (ref) | 1 | (ref) |
| Positive | 0.91 (0.49,1.67) | 0.753 | 1.11 (0.61,2.01) | 0.733 |
| HAART Initiation | ||||
| 1996 – 1999 | 1 | (ref) | 1 | (ref) |
| 2000 – 2012 | 13.25 (9.89,17.92) | <0.001 | 16.27 (12,22.28) | <0.001 |
| VL at HI | ||||
| log10 (copies/mL) | 0.69 (0.6,0.8) | <0.001 | 0.73 (0.63,0.84) | <0.001 |
| CD4 Count at HI | ||||
| <200 | 1.27 (0.88,1.84) | 0.196 | 1.19 (0.83,1.72) | 0.337 |
| 200 – 350 | 1.14 (0.86,1.51) | 0.349 | 1.27 (0.95,1.69) | 0.102 |
| >350 | 1 | (ref) | 1 | (ref) |
| ART prior to HI | ||||
| No | 1 | (ref) | 1 | (ref) |
| Yes | 0.61 (0.44,0.85) | 0.003 | 0.52 (0.38,0.7) | <0.001 |
| AIDS prior to HI | ||||
| No | 1 | (ref) | 1 | (ref) |
| Yes | 0.78 (0.45,1.32) | 0.355 | 0.87 (0.53,1.42) | 0.582 |
| HIV VL Assay Count | ||||
| log(n) | 2.2 (1.77,2.74) | <0.001 | 2.24 (1.84,2.74) | <0.001 |
HIV virologic suppression defined as having at least one HIV VL assay <400 copies/mL within 6 months or 1 year after HI.
Abbreviations: Negative, negative HBV infection; Chronic, chronic HBV infection; Isolated Core, isolated core HBV infection; Resolved, resolved HBV infection; HI, HAART initiation; ART, antiretroviral therapy.
In an unadjusted model, over 5533 years of follow-up, there were 863 events of virologic failure. The rate of VF was 15.6 events/100 person-years; 95% CI 14.6–16.7 overall. VF by HBV status was higher in the chronic HB status group (23.5 events/ 100 person-years; 95% CI 17.9–30.3) over the years of follow up as compared to the isolated, resolved and HB negative status groups (18.8 events per 100 person-years; 95% CI 14.4–24, 18.5 events per 100 person-years; 95% CI 16.3–21, and 13.7 events per 100 person-years; 95% CI 12.6–15, respectively) (data not shown). Predictors of VF in adjusted analyses however, did not show a significant difference by HB status, suggesting the association was confounded by other factors (see Table 3, Supplemental Digital Content 1, which demonstrates predictors of VF). Younger age, African-American race, HI in the period 1996–1999, and prior ART were significantly associated with an increased risk of VF. Higher VL at HI showed a trend towards increased risk of VF (HR=1.09; 95% CI 1.00–1.18).
Table 3.
Predictors of HIV Virologic Failure (Cox Regression)
| HR (95% CI) | P value | |
|---|---|---|
| HBV Status at HI | ||
| Negative | 1 | (ref) |
| Chronic | 1.1 (0.81,1.51) | 0.53 |
| Isolated Core | 0.98 (0.73,1.31) | 0.873 |
| Resolved | 1.1 (0.91,1.31) | 0.32 |
| Age at HI | ||
| per 10 years | 0.85 (0.77,0.94) | 0.002 |
| Race | ||
| Caucasian | 1 | (ref) |
| African-American | 1.24 (1.06,1.45) | 0.008 |
| Hispanic | 0.89 (0.7,1.15) | 0.386 |
| Gender | ||
| Male | 1 | (ref) |
| Female | 1.04 (0.78,1.37) | 0.803 |
| HCV Status at HI | ||
| Negative | 1 | (ref) |
| Positive | 1.02 (0.73,1.44) | 0.901 |
| HAART Initiation | ||
| 1996 – 1999 | 1 | (ref) |
| 2000 – 2012 | 0.2 (0.16,0.25) | <0.001 |
| VL at HI | ||
| log10(copies/mL) | 1.09 (1,1.18) | 0.056 |
| CD4 Count at HI | ||
| <200 | 0.99 (0.8,1.23) | 0.949 |
| 200 – 350 | 0.84 (0.71,1.01) | 0.058 |
| >350 | 1 | (ref) |
| ART prior to HI | ||
| No | 1 | (ref) |
| Yes | 1.67 (1.39,2.01) | <0.001 |
| AIDS prior to HI | ||
| No | 1 | (ref) |
| Yes | 1.04 (0.78,1.38) | 0.809 |
| Time to Suppression, | ||
| years | 0.86 (0.82,0.9) | <0.001 |
HIV virologic failure defined as two consecutive HIV VL assays ≥ 400 copies/mL.
Abbreviations: Negative, negative HBV infection; Chronic, chronic HBV infection; Isolated Core, isolated core HBV infection; Resolved, resolved HBV infection; HI, HAART initiation; VL, viral load; ART, antiretroviral therapy; HR, hazard ratio; CI, confidence interval.
CD4 Recovery
Unadjusted trends in CD4 reconstitution by HBV status are shown in Figure 1. CD4 slope did not significantly differ by HB status during the first six months after HI (chronic: P=0.838; isolated core: P=0.270; resolved HB: P=0.239) (Table 4). The rate increase in CD4 cell count in chronic HB compared to HB negatives during the period between six months and 4 years after HI was also not significantly different (chronic: P=0.069; isolated core: P= 0.289; resolved HB: P=0.099). A significantly greater rate increase in CD4 cell count in chronic HB compared to HB negative, however, was seen during the period from 4 years to 12 years after HI (chronic: P<0.001; isolated core: P= 0.841; resolved HB: P=0.071), after adjustment for age at HI, gender, race, HCV status at HI, AIDS prior to HI, CD4 cell count at HI, HIV VL at HI, ART prior to HI and year of HI.
Table 4.
CD4 Trajectory (Linear Spline Mixed Regression)
| HBV Status at HI | Difference in Cells/mm3 per year (95% CI) |
P value |
|---|---|---|
| 0 – 0.5 years after HI | ||
| Negative | 0 | (ref) |
| Chronic | −12.28 (−130.14,105.59) | 0.838 |
| Isolated Core | −54.62 (−151.67,42.44) | 0.27 |
| Resolved | −33.77 (−89.98,22.45) | 0.239 |
| 0.5 – 4 years after HI | ||
| Negative | 0 | (ref) |
| Chronic | −7.92 (−16.46,0.61) | 0.069 |
| Isolated Core | 4.02 (−3.4,11.43) | 0.289 |
| Resolved | −3.66 (−8.01,0.69) | 0.099 |
| 4 – 12 years after HI | ||
| Negative | 0 | (ref) |
| Chronic | 9.76 (5.68,13.83) | <0.001 |
| Isolated Core | 0.39 (−3.39,4.17) | 0.841 |
| Resolved | 2.16 (−0.18,4.49) | 0.071 |
| CD4 Count at HI | ||
| 0 – 0.5 years after HI | ||
| <200 | −102.38 (−164.89,−39.88) | 0.001 |
| 200 – 350 | −84.85 (−138.97,−30.72) | 0.002 |
| >350 | 0 | (ref) |
| 0.5 – 4 years after HI | ||
| <200 | 13.17 (8.23,18.11) | <0.001 |
| 200 – 350 | 14.31 (9.89,18.73) | <0.001 |
| >350 | 0 | (ref) |
| 4 – 12 years after HI | ||
| <200 | 17.34 (14.77,19.92) | <0.001 |
| 200 – 350 | 6.27 (3.78,8.77) | <0.001 |
| >350 | 0 | (ref) |
| HAART initiation | ||
| 0 – 0.5 years after HI | ||
| 1993 – 1999 | 0 | (ref) |
| 2000 – 2010 | 112.82 (63.04,162.61) | <0.001 |
| 0.5 – 4 years after HI | ||
| 1993 – 1999 | 0 | (ref) |
| 2000 – 2010 | 2.31 (−2.05,6.67) | 0.3 |
| 4 – 12 years after HI | ||
| 1993 – 1999 | 0 | (ref) |
| 2000 – 2010 | 8.28 (4.33,12.23) | <0.001 |
Adjusted for age at HI, gender, race, HCV status at HI, AIDS prior to HI, CD4 count at HI, HIV VL at HI, ART prior to HI and year of HI at baseline.
Abbreviations: Negative, negative HBV infection; Chronic, chronic HBV infection; Isolated core, isolated core HBV infection; Resolved, resolved HBV infection; HI, HAART initiation; VL, viral load.
Deaths and AIDS Defining Illnesses
During 14,458 person years of follow-up (range Jan 1996– Feb 2010), 269 participants experienced an AIDS event or death after HI. AIDS events occurred at a rate of 1.86 per 100 person-years; 95% CI 1.64–2.10. In multivariate analysis, chronic HB status and resolved HB status were significantly associated with an increased risk of AIDS or death (chronic HB; HR=1.68; 95% CI 1.05–2.68, isolated HB; HR=1.51; 95% CI 0.93–2.44, resolved HB; HR=1.61; 95% CI 1.15–2.25). The HR comparing individuals with isolated HBcAb to those who were HB negative was not significant (data not shown). Other covariates that were associated with a significantly increased risk of an AIDS or death included subjects with a positive HCV status (HR=1.99; 95% CI 11.26–3.16), CD4 cell count at HI <200 (HR 3.70; 95% CI2.49 – 5.50) and 200–350 (HR=1.35; 95% CI 0.85–2.05), as compared with a CD4 cell count ≥350, and prior ART (HR=2.06; 95% CI 1.39 – 3.07). All models were stratified by AIDS prior to HI. The interaction between HBV and HCV was not significant.
DISCUSSION
We evaluated HIV outcomes in HBV coinfected HAART recipients and found that in agreement with other studies, HB status did not have a significant impact on HIV virologic outcomes. To adjust for the disparity in ART between HBV infected and HBV negative groups, we used era of HAART initiation in all models as well as ran sensitivity analyses (data not shown) to adjust for regimen potency (unboosted PI vs boosted PI, 3 NRTI, and NNRTI-based regimens, excluding PI+NNRTI+NRTI). Expectedly, regimen potency was significant, while P values and odd ratios for HBV status did not greatly change, indicating that while disparity in ART regimen between HBV groups affect virologic and immunologic outcomes, it is accurately adjusted for in our analysis. Older age, HI in the era 2000–2010, lower HIV RNA, and absence of ART use prior to HI were associated with improved odds of achieving VS in our study in agreement with findings observed from other studies [14,18]. Of interest, our cohort demonstrated 53% of subjects achieving HIV VS at 6 months and 59% at 12 months, which is lower than a recent study evaluating patients in the Swiss HIV Cohort Study (SHCS) (85.3–90.1% at 12 months), as well as previously published results from NHS participants starting HAART after 2000 (81% VS at 12 months) [28,34]. This likely reflects differences in ARV potency as well as treatment practices and prior mono- and dual-NRTI exposure. While HB status was not significantly associated with VS, VL at HI was significantly associated with the odds of achieving VS within 6 months and 12 months of HI. We also found no difference in time to VF by HB status at HI.
Prior studies evaluating T cell recovery in HBV infection have been mixed, with several showing no difference or early lower median CD4 cell count gains in HIV-HBV coinfection with no differences by 48 weeks [14,16,23,24]. The SHCS, however, which evaluated CD4 count recovery over a 36 month period, showed impaired CD4 recovery in patients with both HBsAg and anti-HBc alone suggesting direct impairment of immunologic recovery in HBV-coinfected patients [28]. Results from our study show that HB status may be associated with CD4 cell count reconstitution post HI, as seen with a trend for slower CD4 cell count trajectories in individuals with chronic HB and resolved HB between 6 months and 4 years after HI. However, any trend in impaired immunologic recovery was not sustained as we observed a significantly greater rate increase in CD4 cell count in chronic HB compared to HB negative patients during the period from 4 years to 12 years after HI (P<0.001). Our study, with long median times of patient observation, is the first to demonstrate significantly greater long term CD4 cell count gains in chronic HB individuals. The explanation for this difference is not clear, but one possibility is a positive impact on CD4 regenerative capacity in individuals with chronic HB when placed on HAART as compared with ART regimens that were not as robust as HAART. We also reviewed ART regimen changes over follow-up. Initially, only 11% (Table 1) of the CHB group was on dual HBV-active therapy; however, we found that 47% of the CHB initially on non-dual HBV-active ART, were switched to dual HBV-active ART at a median of 7.2 years after HAART initiation (data not shown). This may account for the significant CD4 cell count increase in the CHB group to similar levels as the other groups. Despite individuals with CHB having lower CD4 cell counts and older age at HI, which would negatively impact CD4 cell count rise, long term increases were observed. Similar to other studies, we show that individuals initiating HAART in the later years have greater CD4 slope changes in CD4 cell count as compared with individuals who started HAART in the early years of HAART availability, with VL not affecting CD4 trajectory [35]. Factors associated with poorer immunologic outcomes in our study, such as AIDS prior to HI, prior ART, lower CD4 count at HI are well-established determinants of HIV disease progression and risk of death [36].
Similar to other studies, we found an increase in AIDS events or death in patients with chronic HB, but also found an increased risk in resolved HB compared to HB negative individuals, which has not been previously described. These findings in individuals with resolved HB were not expected, given there were no significant differences in virologic and immunologic responses to ART over time between individuals with resolved HB as compared to individuals who were HB negative.
Study limitations include the relatively small number of individuals with chronic HB and isolated HB in this cohort in comparison to individuals with resolved and HB negative status. The population whose HB status was not classified differed from those who were classified in that they were more often diagnosed in the pre HAART era, had lower CD4 nadirs, were older at HAART start, had a greater percentage of HCV seropositivity, and had a longer duration of HIV diagnosis prior to HAART initiation. This difference would likely contribute to worse outcomes in individuals not HB classified, therefore HB unclassified individuals were excluded from analyses. Whether the CD4 cell count recovery findings observed in our study reflect survivor bias in individuals with chronic HB is possible, however, our cohort had significantly longer follow up for the chronic HB compared to other groups. The interpretation of data regarding HCV status is limited as confirmatory HCV RNA results were unavailable for the majority of those who were anti-HCV positive. The quantification of the contribution of liver-related causes to death is not possible in our cohort. Findings from this study may be relevant to groups similar to our cohort, and may not be applicable to groups with high rates of IDU, women or those with limited access to care.
The NHS cohort provides a unique opportunity to understand the impact of HBV coinfection on HIV outcomes in HAART recipients. The low use of injection drugs in this cohort, as well as open access to medical care, vaccinations and medications in the military health system also help reduce potential confounding. Additonally, the long median ART duration and person years of observation in our cohort allow for the observation of potential associations otherwise not seen in other studies. However, some of the cohort’s features may limit external generalizability. In addition, given the limited number of women in our cohort, study findings may not be generalizable to women. Therefore, in our cohort, and likely for other HIV-infected persons with similar characteristics, HB status does not appear to be associated with worse immunologic or virologic HIV outcomes.
In conclusion, our study adds further insight into the understanding of the complex interactions between HBV- and HIV-infection, and demonstrates the possible association between HB status at the time of HI and subsequent HIV immunologic and clinical outcomes. Further evaluation of the impact of HBV on cellular immunity and clinical outcomes should be prioritized. Findings from this study add further support to the recommendations to expand HBV screening to potentially minimize the morbidity and mortality associated with HIV/HBV coinfection.
Figure 1.
CD4 Count Reconstitution by HBV Status at HI
CD4 trends determined using LOWESS (locally weighted scatter plot smoothing).
Abbreviations: Negative, negative HBV infection; Chronic, chronic HBV infection; Isolated Core, isolated core HBV infection; Resolved, resolved HBV infection; HI, HAART initiation.
Acknowledgments
The investigators would like to thank our patients for their enormous contributions over the years. In addition to authors listed above, the IDCRP HIV Working Group includes: Susan Banks, RN, Nancy Crum-Cianflone, MD, MPH, Cathy Decker, MD, Conner Eggleston, LTC Tomas Ferguson, MD, COL Susan Fraser, MD, MAJ Joshua Hartzell, MD, MAJ Joshua Hawley, MD, LTC Gunther Hsue, MD, Arthur Johnson, MD, COL Mark Kortepeter, MD, MPH, Tahaniyat Lalani, MD, Robbin Lockhart, MS, Scott Merritt, LTC Robert O’Connell, MD, Maj Jason Okulicz, MD, Michael Polis, MD, John Powers, MD, MAJ Roseanne Ressner, MD, COL(ret) Edmund Tramont, MD, LT Tyler Warkentien, MD, CDR Timothy Whitman, MD, and COL Michael Zapor, MD.
Funding:
This work was supported by the Infectious Diseases Clinical Research Program (IDCRP, www.idcrp.org), a DoD program executed through Uniformed Services University of the Health Sciences. This project has been funded in whole, or in part, with federal funds from the National Institute of Allergy and Infectious Diseases (NIAID), National Institutes of Health (NIH), under Inter-Agency Agreement Y1-AI-5072. The IDCRP reviewed the study design, collected the data and provided salary support to investigators (MLL, ACW, AG, and BKA). The analyses, conclusions and decision to submit the manuscript are the independent work and decision of the authors.
The content of this publication is the sole responsibility of the authors and does not necessarily reflect the views or policies of the NIH or the Department of Health and Human Services, the DoD or the Departments of the Army, Navy or Air Force. Mention of trade names, commercial products, or organizations does not imply endorsement by the U.S. Government.
Footnotes
Potential conflicts of interest: none reported.
References
- 1.Gilson RJC, Hawkins AE, Beecham MR, et al. Interactions between HIV and hepatitis B virus in homosexual men: Effects on the natural history of infection. AIDS. 1997;11:597–606. doi: 10.1097/00002030-199705000-00007. [DOI] [PubMed] [Google Scholar]
- 2.Kellerman SE, Hanson DL, McNaghten AD, Fleming PL. Prevalence of chronic hepatitis B and incidence of acute hepatitis B infection in human immunodeficiency virus-infected subjects. Journal of Infectious Diseases. 2003;188(4):571–577. doi: 10.1086/377135. [DOI] [PubMed] [Google Scholar]
- 3.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(15):1632–1641. doi: 10.1001/archinte.166.15.1632. [DOI] [PubMed] [Google Scholar]
- 4.Mathews G, Bhagani S. The epidemiology and natural history of HIV/HBV and HCV coinfections. J HIV Ther. 2003;8(4):77–84. [PubMed] [Google Scholar]
- 5.Thio CL, Seaberg EC, Skolasky R, Jr, et al. HIV-1, hepatitis B virus, and risk of liver-related mortality in the Multicenter Cohort Study (MACS) Lancet. 2002;360(9349):1921–1926. doi: 10.1016/s0140-6736(02)11913-1. [DOI] [PubMed] [Google Scholar]
- 6.Thio CL. Hepatitis B and human immunodeficiency virus coinfection. Hepatology. 2009;49(5 Suppl):S138–S145. doi: 10.1002/hep.22883. [DOI] [PubMed] [Google Scholar]
- 7.Eskild A, Magnus P, Petersen G, et al. Hepatitis B antibodies in HIV-infected homosexual men are associated with more rapid progression to AIDS. AIDS. 1992;6(6):571–574. doi: 10.1097/00002030-199206000-00008. [DOI] [PubMed] [Google Scholar]
- 8.Scharschmidt BF, Held MJ, Hollander HH, et al. Hepatitis B in patients with HIV infection: relationship to AIDS and patient survival. Ann Intern Med. 1992;117:837–838. doi: 10.7326/0003-4819-117-10-837. [DOI] [PubMed] [Google Scholar]
- 9.Sinicco A, Raiteri R, Sciandra M, et al. Coinfection and superinfection of hepatitis B virus in patients infected with human immunodeficiency virus: no evidence of faster progression to AIDS. Scand J Infect Dis. 1997;29:111–115. doi: 10.3109/00365549709035869. [DOI] [PubMed] [Google Scholar]
- 10.Solomon RE, Van Raden M, Kaslow RA, et al. Association of hepatitis B surface antigen and core antibody with acquisition and manifestations of human immunodeficiency virus type 1 HIV-1infection. Am J Public Health. 1990;80:1475–1478. doi: 10.2105/ajph.80.12.1475. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Greenspan D, Greenspan JS, Overby G, et al. Risk factors for rapid progression from hairy leukoplakia to AIDS: a nested case-control study. J Acquir Immune Defic Syndr. 1991;4:652–658. [PubMed] [Google Scholar]
- 12.Ockenga J, Tillmann HL, Trautwein C, Stoll M, Manns MP, Schmidt RE. Hepatitis B and C in HIV-infected patients. Prevalence and prognostic value. J Hepatol. 1997;27:18–24. doi: 10.1016/s0168-8278(97)80274-7. [DOI] [PubMed] [Google Scholar]
- 13.Chun HM, Fieberg AM, Hullsiek KH, et al. The epidemiology of hepatitis B virus infection in a U.S. cohort of HIV-infected individuals during the last 20 years. Clin Infect Dis. 2010;50:426–436. doi: 10.1086/649885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.de Luca, Bugarinin R, Lepri AC, et al. Coinfection with hepatitis viruses and outcome of initial antiretroviral regimens in previously naïve HIV-infected subjects. Arch Intern med. 2002;162:2125–2132. doi: 10.1001/archinte.162.18.2125. [DOI] [PubMed] [Google Scholar]
- 15.Omland N, Skinhøj P, Laursen A, et al. Impact of hepatitis B virus coinfection on response to highly active antiretroviral treatment and outcome in HIV-infected individuals: a nationwide cohort study. HIV Med. 2008;9(5):300–306. doi: 10.1111/j.1468-1293.2008.00564.x. [DOI] [PubMed] [Google Scholar]
- 16.Konopnicki D, Mocroft A, De WS, et al. Hepatitis B and HIV: prevalence, AIDS progression, response to highly active antiretroviral therapy and increased mortality in the EuroSIDA cohort. AIDS. 2005;19:593–601. doi: 10.1097/01.aids.0000163936.99401.fe. [DOI] [PubMed] [Google Scholar]
- 17.Hoffmann CJ, Seaberg EC, Young S, et al. Hepatitis B and long-term HIV outcomes in coinfected HAART recipients. AIDS. 2009;23(14):1881–1889. doi: 10.1097/QAD.0b013e32832e463a. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Lincoln D, Petoumenos K, Dore GJ. HIV/HBV and HIV/HCV coinfection, and outcomes following highly active antiretroviral therapy. HIV Med. 2003;4(3):241–249. doi: 10.1046/j.1468-1293.2003.00152.x. [DOI] [PubMed] [Google Scholar]
- 19.Hoffmann CJ, Charalambous S, Martin DJ, et al. Hepatitis B virus infection and response to antiretroviral therapy (ART) in a South African ART program. Clin Infect Dis. 2008;47(11):1479–1485. doi: 10.1086/593104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.lsa SE, Gwamzhi LN, Akolo C, Giyan J. A prospective cohort study of immunologic and virologic outcomes in patients with HIV/AIDS and hepatitis virus coinfection in Jos, Nigeria. Niger J Med. 2010 Jul-Sep;19(3):279–285. doi: 10.4314/njm.v19i3.60188. [DOI] [PubMed] [Google Scholar]
- 21.Moore E, Beadsworth MB, Chaponda M, et al. Favourable one-year ART outcomes in adult Malawians with hepatitis B and C coinfection. J Infect. 2010;61(2):155–163. doi: 10.1016/j.jinf.2010.04.009. Epub 2010 May 12. [DOI] [PubMed] [Google Scholar]
- 22.Idoko J, Meloni S, Muazu, et al. Impact of hepatitis B virus infection on HIV response to antiretroviral therapy in Nigeria. Clin Infect Dis. 2009;49(8):1268–1273. doi: 10.1086/605675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Law WP, Duncombe CJ, Mahanontharit A, et al. Impact of viral hepatitis coinfection on response to antiretroviral therapy and HIV disease progression in the HIV-NAT cohort. AIDS. 2004;18(8):1169–1177. doi: 10.1097/00002030-200405210-00010. [DOI] [PubMed] [Google Scholar]
- 24.Hawkins C, Christian B, Ye J, et al. Prevalence of Hepatitis B coinfection and response to antiretroviral therapy among HIV-infected patients in urban Tanzania. AIDS. 2012 Nov 28; doi: 10.1097/QAD.0b013e32835cb9c8. [Epub ahead of print]. [DOI] [PubMed] [Google Scholar]
- 25.Sheng WH, Chen MY, Hsieh SM, et al. Impact of chronic hepatitis B virus (HBV) infection on outcomes of patients infected with HIV in an area where HBV infection is hyperendemic. Clin Infect Dis. 2004;38(10):1471–1477. doi: 10.1086/420744. Epub 2004 Apr 28. [DOI] [PubMed] [Google Scholar]
- 26.Sheng WH, Kao JH, Chen PJ, et al. Evaluation of hepatitis B serological markers in HIV-infected patients receiving highly active antiretroviral therapy. Clin Infect Dis. 2007;45:1221–1229. doi: 10.1086/522173. [DOI] [PubMed] [Google Scholar]
- 27.Thio CL, Smeaton L, Saulynas M, et al. Characterization of HIV-HBV coinfection in a multinational HIV-infected cohort. AIDS. 2013;27:191–201. doi: 10.1097/QAD.0b013e32835a9984. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Wandeler G, Gsponer T, Bihl F, et al. Hepatitis B virus infection is associated with impaired immunological recovery during antiretroviral therapy in the Swiss HIV Cohort Study. J Infect Dis. 2013 doi: 10.1093/infdis/jit351. Epub Jul30. [DOI] [PubMed] [Google Scholar]
- 29.Brodine SK, Starkey MJ, Shaffer RA, et al. Diverse HIV-1 subtypes and clinical, laboratory and behavioral factors in a recently infected US military cohort. AIDS. 2003;17:2521–2527. doi: 10.1097/00002030-200311210-00016. [DOI] [PubMed] [Google Scholar]
- 30.DHHS/Henry J. Kaiser Family Foundation Panel on Clinical Practices for the Treatment of HIV infection. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. 2008 Oct; revision. Available at http://aidsinfo.nih.gov/ContentFiles/AboutHIVTreatmentGuidelines FS en.pdf.
- 31.Centers for Disease Control and Prevention. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. [Accessed Jan 4, 2008];MMWR. 1992 41 at http://www.cdc.gov/mmwr/preview/mmwrhtml/00018871.htm. [PubMed] [Google Scholar]
- 32.Harrell Frank E., Jr . Regression Modeling Strategies with applications to linear models, logistic regression, and survival analysis. Springer; 2001. [Google Scholar]
- 33.R Development Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing. Vienna: 2010. [Accessed 1 November 2011]. Available from http://www.R-project.org. [Google Scholar]
- 34.Marconi VC, Grandits GA, Weintrob AC, et al. Outcomes of highly active antiretroviral therapy in the context of universal access to healthcare: the U.S. Military HIV Natural History Study. AIDS Res Ther. 2010;7:14. doi: 10.1186/1742-6405-7-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Lifson AR, Krantz EM, Eberly LE, et al. Long-term CD4+ lymphocyte response following HAART initiation in a U.S. Military prospective cohort. AIDS Res Ther. 2011;8(1):2. doi: 10.1186/1742-6405-8-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.SMART Study Group. El-Sadr WM, Grund B, Neuhaus J, et al. Risk for opportunistic disease and death after reinitiating continuous antiretroviral therapy in patients with HIV previously receiving episodic therapy: a randomized trial. Ann intern Med. 2008;149(5):289–299. doi: 10.7326/0003-4819-149-5-200809020-00003. [DOI] [PubMed] [Google Scholar]