Palella Frank J, Armon Carl, Buchacz Kate, Cole Stephen R, Chmiel Joan S, Novak Richard M, Wood Kathleen, Moorman Anne C, Brooks John T
Division of Infectious Diseases, Northwestern University Feinberg School of Medicine, Chicago, Illinois 60611, USA.
Ann Intern Med. 2009 Jul 21;151(2):73-84. doi: 10.7326/0003-4819-151-2-200907210-00003.
HIV-1 genotypic and phenotypic susceptibility testing (GPT) optimizes antiretroviral selection, but its effect on survival is unknown.
To evaluate the association between GPT and survival.
Cohort study.
10 U.S. HIV clinics.
2699 HIV-infected patients eligible for GPT (plasma HIV RNA level >1000 copies/mL) seen from 1999 through 2005.
Demographic characteristics, clinical factors, GPT use, all-cause mortality, and crude and adjusted hazard ratios (HRs) for the association of GPT with survival.
Patients were followed for a median of 3.3 years; 915 (34%) had GPT. Patients who had GPT had lower mortality rates than those who did not (2.0 vs. 2.7 deaths per 100 person-years). In standard Cox models, GPT was associated with improved survival (adjusted HR, 0.69 [95% CI, 0.51 to 0.94]; P = 0.017) after controlling for demographic characteristics, CD4+ cell count, HIV RNA level, and intensity of clinical follow-up. In subgroup analyses, GPT was associated with improved survival for the 2107 highly active antiretroviral therapy (HAART)-experienced patients (2.2 vs. 3.2 deaths per 100 person-years for patients who had GPT vs. those who did not have GPT; adjusted HR, 0.60 [CI, 0.43 to 0.82]; P = 0.002) and for the 921 triple antiretroviral class-experienced patients (2.1 vs. 3.1 deaths per 100 person-years; adjusted HR, 0.61 [CI 0.40 to 0.93]; P = 0.022). Marginal structural models supported associations between GPT and improved survival in the overall cohort (adjusted HR, 0.54; P = 0.001) and in the HAART-experienced group (adjusted HR, 0.56; P = 0.003).
Use of GPT was not randomized. Residual confounding may exist.
Use of GPT was independently associated with improved survival among HAART-experienced patients.
Centers for Disease Control and Prevention.
HIV-1基因分型和表型药敏试验（GPT）可优化抗逆转录病毒药物的选择，但其对生存率的影响尚不清楚。
评估GPT与生存率之间的关联。
队列研究。
美国10家HIV诊所。
1999年至2005年期间就诊的2699例符合GPT条件（血浆HIV RNA水平>1000拷贝/mL）的HIV感染患者。
人口统计学特征、临床因素、GPT的使用情况、全因死亡率以及GPT与生存率关联的粗危险比和校正危险比（HR）。
患者的中位随访时间为3.3年；915例（34%）进行了GPT检测。进行GPT检测的患者死亡率低于未进行检测的患者（每100人年2.0例死亡对2.7例死亡）。在标准Cox模型中，在控制了人口统计学特征、CD4+细胞计数、HIV RNA水平和临床随访强度后，GPT与生存率提高相关（校正HR，0.69[95%CI，0.51至0.94]；P = 0.017）。在亚组分析中，对于2107例有高效抗逆转录病毒治疗（HAART）经验的患者，GPT与生存率提高相关（进行GPT检测的患者与未进行检测的患者相比，每100人年2.2例死亡对3.2例死亡；校正HR，0.60[CI，0.43至0.82]；P = 0.002），对于921例有三种抗逆转录病毒药物治疗经验的患者也如此（每100人年2.1例死亡对3.1例死亡；校正HR，0.61[CI 0.40至0.93]；P = 0.022）。边际结构模型支持GPT与总体队列生存率提高之间的关联（校正HR，0.54；P = 0.001）以及在有HAART经验的组中（校正HR，0.56；P = 0.003）。
GPT的使用并非随机进行。可能存在残余混杂因素。
在有HAART经验的患者中，使用GPT与生存率提高独立相关。
疾病控制和预防中心。
