Aside from the biological factors which influence drug absorption discussed above, a central factor in ensuring that blood levels remain within the therapeutic range is treatment adherence - always remembering to take your pills on time and within any recommended food guidelines.

Data from the French Trilege study has been important in raising debate about the relationship between adherence, inadequate drug levels and a poor response to treatment. Participants in Trilege started treatment with indinavir/AZT/3TC and after three months some de-intensified to either AZT/indinavir or AZT/3TC. This induction-maintenance strategy was unsuccessful - those people who continued on the triple drug combination were more likely to maintain undetectable viral load than those who switched to two drugs.

In an analysis of all participants whose viral load rebounded whilst taking indinavir, low blood levels of indinavir were seen frequently. The researchers assessed treatment adherence by looking at clinical records and performing pill counts. Poor adherence was documented for all those with viral load rebound on triple therapy and half of those on indinavir/AZT. This poor adherence correlated with poor indinavir blood levels, leading the study group to conclude that treatment failed because it wasn't being taken (Brun-Vezinet; Descamps 2000).

There are some data to suggest that good adherence to a protease inhibitor-containing regimen may paradoxically lead to the very thing which it is meant to prevent - high level resistance to protease inhibitors. Drug plasma levels, genotypic and phenotypic resistance to protease inhibitors and adherence behaviour was examined in a group of prospectively recruited patients for whom either nelfinavir or indinavir was failing.

Plasma drug levels were measured 2, 4 and 6 hours after observed dosing. Plasma drug level testing showed a significant association between viral genotypic resistance and a lower nelfinavir trough level. These patients may have poor drug absorption or rapid elimination; coupled with good adherence this would be expected to lead to frequently sub-optimal drug levels that would select for drug resistant virus (Walsh).

For references, see Key research on testing for drug levels in Anti-HIV therapy: Testing drug levels.