Determinants of sustainable CD4 lymphocyte count increases in response to antiretroviral therapy

Objective: HIV-induced CD4 lymphocyte depletion is partially reversed by an tiretroviral therapy but it is unclear if the degree to which the CD4 count rises depends on viral suppression (if so, the extent of viral suppression required to achieve a maximal CD4 count rise), whether the rise is sustain able and whether it occurs in patients with CD4 count <10 x 10(6) cells/l. We aimed to address these Issues. Methods: We studied CD4 count and plasma HIV RNA values every 4 weeks for 7 2 weeks in 154 patients starting indinavir-containing regimens. Results: Mean baseline HIV RNA and CD4 count were 4.8 log(10) copies/ml and 180 x 10(6) cells/l, respectively. Overall, there was a mean increase in C D4 count of 143 x 10(6) cells/l by 72 weeks. The adjusted mean increase (ad justed for initial viral load, CD4 count and age) was strongly related to t he mean viral suppression over the follow-up period (P < 0.0001). Important ly, there was a highly significant difference (P = 0.0004) in the rise in C D4 count between those with 2-3 log suppression (161 x 10(6) cells/l) and t hose with > 3 log suppression (314 x 10(6) cells/l; mean 3.6 log suppressio n in this group), suggesting that with even greater suppression the rise in CD4 lymphocytes may be still larger. We also studied whether CD4 counts we re still rising after 72 weeks in patients with sustained suppression of at least 3 log in viral load. There was a significant (P = 0.004; paired t-te st) rise in count of 43 x 10(6) cells/l between weeks 64 and 72 in these pa tients, suggesting that regeneration continues at least up to 72 weeks afte r therapy, provided virus replication continues to be suppressed. Patients with initial CD4 counts < 10 x 10(6) cells/l experienced no smaller rises t han those at higher levels, even after adjustment for other factors. Conclusion: These results strongly support a direct causal relationship bet ween HIV replication and CD4 lymphocyte count depletion. The rise in those with > 3 log suppression provides the best available indicator of the poten tial for natural CD4 regeneration in HIV-infected patients. However, since stilt greater CD4 count rises may be seen with more suppressive regimens, i t may not be possible to study the intrinsic CD4 regenerative capacity unti l such regimens are available. (C) 1999 Lippincott Williams & Wilkins.