The prognostic value of interleukin-6 and D-dimer levels in primary HIV infection

Gareth Hardy, HIV i-Base

Inflammatory and thrombotic markers are attracting increasing attention for their potential as predictive markers of HIV disease progression, prompted by results from the SMART study in 2008, which addressed the efficacy of structured interruptions in ART, and raised considerable interest in two of these markers. [1]

In SMART, levels of interleukin-6 (IL-6) and D-dimer in plasma at study entry were highly predictive of disease progression or subsequent death. This relationship with mortality included any cause of death, including AIDS and non AIDS-related causes such as cardiovascular disease, hepatic and renal disease or substance abuse. The striking association with all-cause mortality generated huge interest in both their prognostic significance and their potential mechanistic role in disease pathogenesis. In other studies, plasma IL-6 and D-dimer have been associated with non-AIDS related co-morbidities of HIV infection, even during effective ART. [2]

IL-6 is an inflammatory cytokine. D-dimer are small protein subunits that result from the degradation of fibrin following blood clots. Such clots form in blood vessels during cardiovascular disease and may be facilitated by inflammatory cytokines.

New results by researchers investigating the predictive value of these markers during primary HIV infection have been published in 27 March 2014 edition of AIDS.  Elizabeth Hamlyn and colleagues looked at patients taking part in the SPARTAC trial to assess whether various factors were associated with levels of IL-6 or D-dimers in plasma at seroconversion, and whether plasma IL-6 and D-dimers were predictive of time to the study’s primary endpoint. [3]

The SPARTAC trial assessed whether 12 weeks ART, 48 weeks ART or no ART, initiated during primary HIV infection, had any impact on the subsequent time to initiate ART or reach a CD4 count below 350 cells/mm3. Primary HIV infection was defined as being within the first 6-months of infection.

Hamlyn and colleagues assessed IL-6 and D-dimer levels in the plasma of 200 participants from the SPARTAC trial at study baseline (before initiation of ART). There was a strong correlation between the plasma levels of IL-6 and D-dimers (r = 0.31, p<0.001) at this time point. The levels of both markers in plasma increased with older age. For IL-6 this was equivalent a 19% increase per 10 year increase of age and for D-dimer a 16% increase per 10 years of age. D-dimer was also strongly associated with viral load at baseline (p<0.001), equivalent to a 10% increase per log(10) increase in viral load.

There was no relationship between plasma IL-6 and viral load at baseline. There was also no baseline association between either marker and CD4 count, time since seroconversion, sex/risk group, or body mass index.

Longitudinal analysis of IL-6 and D-dimer levels were then conducted for 73 individuals who were randomised to the “no ART” trial arm, for a median of 225 weeks. During this time, 48 of these patients reached the trial primary endpoint: initiation of long-term ART or a CD4 count below 350 cells/mm3 blood. While baseline D-dimer and IL-6 levels were both associated with a shorter time to the primary endpoint in univariate analysis (hazard ratio 1.41 for IL-6 and 3.12 for D-dimer), the association between baseline D-dimer and time to end point was lost in multivariate analysis, after adjusting for baseline age, viral load and IL-6 levels. Only baseline IL-6 levels were independently associated with time to endpoint (hazard ratio 1.38).

This study demonstrates that IL-6 levels during primary HIV infection independently predict disease progression, after controlling for age, viral load and CD4 count at baseline. It should be noted that a possible association between D-dimer levels during primary infection and disease progression should not be completely ruled out, as the numbers of subjects in longitudinal analysis in this study were very small.


  1. Kuller LH et al. Inflammatory and coagulation biomarkers and mortality in patients with HIV infection. PLoS Med, 2008, Volume 5. e203.
  2. Duprez DA et al. Inflammation, coagulation and cardiovascular disease in HIV-infected individuals. PLoS One, 2012, Volume 7. e44454.
  3. Hamlyn E et al. Interleukin-6 and D-dimer levels at seroconversion as predictors of HIV-1 disease progression. AIDS (2014), 28 (6): 869–874.

Links to other websites are current at date of posting but not maintained.