icon-folder.gif   Conference Reports for NATAP  
  10th International Congress on Drug Therapy in HIV Infection
November 7-11, 2010
Back grey_arrow_rt.gif
Anemia in SMART Study Tied to Risk of non-AIDS Disease and Death
  Tenth International Congress on Drug Therapy in HIV Infection, November 7-11, 2010, Glasgow
Mark Mascolini
SMART trial participants who interrupted antiretroviral therapy ran a higher risk of new or worsening anemia, according to results of a 2248-person analysis [1]. New anemia during follow-up was linked to increased incidence of AIDS, non-AIDS diseases, and death.
Earlier research shows that combination antiretroviral therapy boosts hemoglobin levels and thereby lowers anemia rates. SMART investigators analyzed study participants to determine whether antiretroviral interruptions in this randomized trial affected hemoglobin levels and the risk of new or worsening anemia. The researchers defined anemia as a hemoglobin reading below 12 mg/dL for women and below 14 mg/dL for men.
This analysis focused on 2248 of 5472 SMART participants (41%) from study sites with adequate hemoglobin data. Median age stood at 43 years (interquartile range [IQR] 38 to 50). Nearly three quarters of those analyzed (72%) were men, 26.4% were black, 14% had HCV infection, 22% had an AIDS diagnosis, and only 6% were antiretroviral naive when they started SMART. About half of the people in this analysis got randomized to SMART's treatment interruption arm.
Median hemoglobin levels when SMART began stood at 14.8 mg/dL (IQR 12.1 to 24.2) in people without anemia and at 12.9 mg/dL (IQR 5.1 to 14.0) in people with anemia. Study participants had a median of 11 hemoglobin levels (IQR 9 to 13) measured during follow-up. In the steady-therapy group, the proportion of people with anemia rose a few percentage points in the first month of the trial, then fell to below the baseline level and remained lower through 24 months of follow-up. The proportion of people with anemia in the treatment-interruption arm dipped a few percentage points in the first month of SMART, then rose smartly, climbing from 25% at month 1 to about 32% at month 24.
Anemia developed or worsened in 759 people during the trial, including 420 of 1106 (38%) in the treatment-interruption group and 339 of 1127 (30%) in the steady-therapy arm, a highly significant difference (P < 0.0001). In SMART's first 4 months, risk of new or worsening anemia did not differ significantly between the two study arms (adjusted relative hazard [ARH] 1.02, 95% confidence interval [CI] 0.82 to 1.25). But after the first 4 months, people in the antiretroviral interruption group had more than a 50% higher risk of new or worsening anemia (ARH 1.56, 95% CI 1.28 to 1.89, P < 0.0001).
Among people who died or had a new AIDS or non-AIDS diagnosis during follow-up and had hemoglobin measured within 28 days before the new diagnosis or death, 25 of 55 in the interruption group (45.5%) and 24 of 47 in the steady-therapy group (51.1%) had anemia, a nonsignificant difference (P = 0.57). Compared with people who did not have anemia, those with anemia had a doubled risk of dying (incidence rate ratio [IRR] 2.19, 95% CI 1.23 to 3.87, P = 0.0073), a doubled risk of AIDS (IRR 2.31, 95% CI 1.34 to 3.98, P = 0.0027), and a tripled risk of a non-AIDS diagnosis (IRR 2.98, 95% CI 2.01 to 4.40, P < 0.0001).
Amanda Mocroft and SMART colleagues said it is not clear whether the link between anemia and progressive disease "is causal or a consequence of emerging subclinical disease." But they believe the findings suggest "that anemia, or drop in haemoglobin, might be of use as a pre-clinical marker of disease."
The investigators noted that these results add to evidence that interrupting antiretroviral therapy is dangerous. They proposed that "close attention should be paid to haemoglobin and anaemia in both patients on and off antiretrovirals as low and decreasing levels are associated with increased likelihood of a wide spectrum of clinical disease."
This poster is online at
1. Mocroft A, Lifson AR, Touloumi G et al. Hemoglobin and anemia in the SMART study. Tenth International Congress on Drug Therapy in HIV Infection. November 7-11, 2010. Glasgow. Abstract P144.