|
|
|
|
|
Session 127
Poster Abstracts Diagnostics: Measuring the Consequences of Antiretroviral Therapy Friday, 1:30 - 3:30 pm Hall A |
Background: Poor adherence to ART is the most important
factor contributing to virologic failure and HIV resistance. Using biochemical
markers like mean corpuscular volume (MCV) to monitor adherence is simple and
inexpensive. Increased MCV has a sensitivity of 80% and specificity of 44% for
adherence to zidovudine- (AZT) or stavudine- (d4T) based regimens, but many
newer once-daily regimens lack AZT or d4T, limiting use of MCV as a marker of
adherence. Atazanavir (TAZ) raises bilirubin in 83% of patients. We explored
the use of bilirubin as a marker of adherence in TAZ-containing regimens.
Methods: Retrospective chart review of 92 patients on TAZ-based
ART to determine bilirubin, CD4, and viral load before start of ART and at
subsequent follow-ups. Durability of bilirubin rise at subsequent follow-up and
effect of ritonavir on bilirubin were also studied.
Results: Of the 92 patients, 78 met criteria for inclusion. Mean age was 39; 67% were male;
baseline CD4 was 284; and median viral load was 10,000 copies/mL. Median
baseline bilirubin was 0.5. Patients had taken a median of 8 ART agents
previously. Median first follow-up occurred at 57 days. There was a mean
decrease in viral load at first follow-up of 1.4 log (95% CI 0.8 to 2.0). Bilirubin
increased by a median of 0.4. Using a 2-log drop as a marker for successful
suppression, 41% of patients were suppressed at first follow-up. Association of
bilirubin increase with successful viral suppression showed a statistically
significant difference in bilirubin rise among responders compared to
non-responders (2-sample Wilcoxon rank-sum, p
= 0.001). Median bilirubin increase (IQR) for patients achieving successful
viral suppression was 1.0 (0.4 to 1.9), compared with 0.2 (0.1 to 0.9) for
those not achieving viral suppression. With a 2-log drop in viral load, bilirubin
levels had various sensitivity and specificities as shown in the table.
|
Cut-point |
Sensitivity |
95% C.I. |
Specificity |
95% C.I. |
|
≥ 0.2 |
84 |
67–95 |
48 |
33–63 |
|
≥ 0.3 |
81 |
64–93 |
61 |
45–75 |
|
≥ 0.4 |
75 |
57–89 |
65 |
50–79 |
|
≥ 0.5 |
66 |
47–81 |
67 |
52-80 |
Conclusions: Our results suggest that bilirubin increase ³. 0.3 may be a good marker for adherence to TAZ (sensitivity
81%, specificity 61%). Further studies are required including a larger
population cohort, longer follow-up time and determination of confounders such
as Gilbert’s disease to confirm the use of bilirubin increase to measure TAZ
adherence.
Keywords: adherence; atazanavir; bilirubin
 |