In many countries injection drug users (IDU) constitute a ‘key population’ for HIV/AIDS. There may in some contexts be substantial overlap with other groups – where, for example, the IDU are also sex-workers, or prisoners. In reality, IDU may be exposed to multiple risk factors, which can be hard to isolate from each other. However, there is one issue specific to drug use: namely, how substance use itself might independently contribute to pathophysiologic processes associated with HIV infection.

Various papers published in this journal illustrate the difficulty of answering this question without evidence of the biological mechanisms. Ti & Wood (STI), for example, consider virological suppression in a longitudinal cohort of ART-exposed Canadian sex-workers who use drugs. What is the evidence, they ask, of the impact of being an IDU sex-worker on disease progression and transmission as measured by plasma RNA viral load? They do discover an increased risk of disease progression (AOR 0.66); but adding adherence to their multivariate model entirely eliminated the effect. However, Tran & Zhang (STI), investigating the determinants of 6.5% virological failure amongst 3,500 people living with HIV in Vietnam, found an enhanced risk of viraemia (OR=1.32) with drug-use.

The difficulty of isolating any specifically pathophysiologic effect in such studies explains the interest of the recent paper of Carrico & Pahwa. This study examines, not failure of virological suppression, but known factors predictive of mortality in virologically suppressed HIV – notably plasma biomarkers of monocyte activation and intestinal barrier integrity. The researchers investigated the association between these biomarkers and reactive urine toxicology results for stimulants (methamphetamine or cocaine). Of the sample of 84 virally suppressed MSM, those who tested positive for stimulants showed soluble CD14 levels of 2,087 vs. 1,801 ng/ml (p=0.009), and higher sCD14. Both results remained significant after adjusting for co-variates. The authors conclude that monocyte activation may be a mechanism by which drug-use impacts on HIV progression.

So it would seem that, even with good adherence, and responsible risk-limiting strategies, drug-use exacts its toll on the health HIV-positive drug-users.