Background: Studies have shown that transmission of HIV strains resistant to antiretroviral drugs may increase as therapy becomes more widespread. 

Methods: We prospectively evaluated the prevalence of antiretroviral drug resistance (ARVDR) among recently diagnosed untreated individuals with HIV (not AIDS) sequentially enrolled in 10 U.S. cities.  Plasma specimens were analyzed using conventional sequencing.  To identify persons infected within 4-6 months, specimens were tested with a modification of an HIV-1 enzyme immunoassay (3A11, Abbott).

Results:  Of 923 persons enrolled in 1998-2000, 74.7% were male, 45.4% were African Americans, 23.6% were Hispanic, 44.0% were men who have sex with men, 10.6% were heterosexual injecting drug users, and 45.4% reported only heterosexual exposure.  Mutations associated with ARVDR were seen in 8.3% overall.  Of the 695 specimens so far tested with 3A11, 12.5% had evidence of recent infection.  The prevalence of ARVDR mutations in the recently infected group was 8.1%.  When analysis was restricted to the 5 cities participating in all 3 years (n=723), ARVDR prevalence increased significantly between 1998 and1999 (p=0.01 by Fisher’s exact test), but did not change significantly between 1999-2000 (see Table).  A similar (non-significant) trend was seen among the recently infected (data not shown).   No significant changes in proportions of gender, race, risk group, or recently infected persons were seen among these years.  A non-significant increase in ARVDR prevalence between 1998 and 1999 was seen in each of the 3 drug classes.

                                                                                                               

Presence of ARVDR mutations to:                                   1998          1999             2000

                                                                                           (N=238)    (N=240)        (N=245)

Any drug class                                                                  9 (3.8%)   24 (10.0%)  22  (9.0%)

Nucleoside reverse transcriptase inhibitors                8 (3.4%)   20 (8.3%)    17 (6.9%)

Non-nucleoside reverse transcriptase inhibitors        1 (0.4%)     5  (2.1%)      3 (1.2%)   

Protease inhibitors                                                          0 (0%)       4  (1.7%)       5 (2.0%)

More than one drug class                                               0 (0%)       4  (1.7%)        3 (1.2%)           

 

Conclusions: An increase in ARVDR prevalence among persons with newly diagnosed HIV was observed between 1998 and 1999 in 5 cities.  Transmission of ARVDR-resistant strains may increase as treatment becomes more common, although prevalence will also vary depending on treatment outcomes and the success of risk reduction measures.  Further multiyear studies in the same geographic areas are needed to evaluate whether the increase will continue or stabilize.