Duong et al developed an algorithm for evaluating a patient with HIV who shows virologic failure while taking antiretroviral (ARV) therapy. It separates patients who are nonadherent to the drug regimen from those who may have regimen resistant virus. The authors are from the University Hospital in Dijon and Bichat Hospital in Paris, France.


Patients with HIV who have been treated with a regimen containing a protease inhibitor for >= 4 weeks have their plasma HIV RNA levels determined.

Patient Medication Exposure

Positive Virologic Response in Plasma HIV RNA Levels

first time use of protease inhibitors

decrease >= 2 LOG10 (copies per mL)

history of previous protease inhibitor therapy

decrease >= 1 LOG10(copies per mL)



• A >= 2 LOG10 decrease indicates <= 1% of pretreatment level.


Failure to have a positive virologic response is considered virologic failure.


If a patient is identified with virologic failure:


Step 1: Screen the patient for drug regimen adherence. The authors used an adherence score, which is given in the chapter on clinical pharmacology.


Step 2: If the person indicates possible nonadherence and if the drug regimen includes zidovudine (Zdv) or stavudine (d4t), then determine the mean corpuscular volume (MCV). The presence of macrocytosis (increased MCV) correlates with adherence.


Step 3: If the person indicates possible nonadherence and:

(3a) is not taking zidovudine or stavudine, or

(3b) is taking zidovudine or stavudine but does not show macrocytosis

then determine plasma drug levels.


Criteria for adherence: one or more of the following:

(1) Adherent by screening questionnaire: Adherence score: in chapter on clinical pharmacology

(2) For patients on a drug regimen includes zidovudine (Zdv) or stavudine (d4t), macrocytosis (increased mean corpuscular volume).

(3) Plasma drug level analysis shows adequate drug level.


If the patient does not meet the criteria for adherence, then s/he is presumed to be drug nonadherent.


Management of presumed drug adherence:

(1) Consider change of regimen if first-line therapy.

(2) Consider genotypic test for viral resistance.


Management of presumed drug nonadherence:

(1) Consider interventions to improve adherence.


To read more or access our algorithms and calculators, please log in or register.