Development and validation of a prognostic nomogram for HIV/AIDS patients with antiretroviral therapy
-
摘要:
目的 针对接受抗逆转录病毒治疗的HIV/AIDS患者建立有效的列线图并验证,准确预测HIV/AIDS患者的死亡风险。 方法 所有数据均来自中国艾滋病防治信息系统新疆维吾尔自治区2006 — 2019年的数据。通过基于建模组的单因素和多因素Cox比例风险回归分析来确定列线图中纳入的因子,在建模组(3272人)与验证组(1636人)中均使用受试者工作特征曲线下面积(AUC-ROC)和校准曲线评估列线图的预测准确性,使用决策曲线分析(DCA)、x-tile分析和Kaplan-Meier曲线评估列线图的临床实用性。 结果 根据多因素Cox比例风险回归确定独立的预后因素包括血红蛋白、体质指数(BMI)、性别、谷草转氨酶、WHO分期、延迟时间和CD4细胞,以此建立了列线图模型。列线图模型在建模组的AUC-ROC为0.781(95 % CI = 0.703~0.861),在验证组的AUC-ROC为0.829(95 % CI = 0.758~0.896),在预测生存率的校准曲线中可以观察到列线图预测与实际观察之间一致性良好。此外,根据列线图得分,将研究对象分为3个不同的(低、中、高)风险组,以预测不同组患者生存率。 结论 列线图可以为接受抗逆转录病毒疗法的HIV/AIDS患者提供准确和有利的预测。 Abstract:Objective To establish and verify a nomogram for accurately predicting the mortality risk of human immunodeficency virus/acquired immunodeficiency syndrome (HIV/AIDS) patients receiving antiretroviral therapy (ART). Methods From China Information System for AIDS Prevention and Control, we extracted the data on HIV/AIDS patients registered during the period from 2006 through 2019 in Xinjiang Uygur Autonomous Region. Univariate and multivariate Cox proportional hazard regression analysis were carried out for the data on 3 272 HIV/AIDS patients of modeling group to determine the factors to be included in the nomogram. The area under the receiver-operating characteristics curve (AUC-ROC) and calibration curve were adopted to assess the prediction accuracy of the established nomogram for the data of modeling group and verifying group (1 636 HIV/AIDS patients). Decision curve analysis (DCA), x-tile analysis and Kaplan-Meier curve were used to evaluate the clinical utility of the established nomogram. Results Based on the results of multivariate Cox proportional hazard regression analysis, the established nomogram model included following independent factors for predicting the prognosis of HIV/AIDS patients with ART: hemoglobin, body mass index, gender, aspartate aminotransferase, disease staging according to criterions proposed by World Health Organization, delay time between HIV infection and ART, and CD4 cell count. The AUC-ROC of the established nomogram is 0.781 (95% confidence interval [95% CI]: 0.703 – 0.861) for the modeling group and that for the verification group is 0.829 (95% CI: 0.758 – 0.896). The calibration curve for the survival of the patients during the 3-year period demonstrated a good consistency between the survival rate predicted by the nomogram and that observed actually. The survival rate was predicted based on the established nomogram for the patients at different (low, moderate and high) risk of mortality assessed according to nomogram scoring resulted from the analysis. Conclusion The study established a nomogram which could provide accurate and meaningful predictions for the survival of the HIV/AIDS patients with antiretroviral therapy. -
Key words:
- AIDS /
- nomogram /
- antiretroviral therapy /
- prognosis
-
表 1 建模组和验证组患者的基线人口统计学和临床特征
变量 建模组(N = 3272) 百分比(%) 验证组(N = 1636) 百分比(%) χ2 值 P 值 性别 0.625 0.429 男性 1 946 0.59 953 0.58 女性 1326 0.41 683 0.42 WHO分期 0.315 0.957 Ⅰ期 1135 0.35 574 0.35 Ⅱ期 1739 0.53 858 0.52 Ⅲ期 328 0.10 170 0.10 Ⅳ期 70 0.02 34 0.02 基线CD4(个/μL) 2.209 0.531 < 150 594 0.18 272 0.17 150~ 1363 0.42 687 0.42 300~ 706 0.21 354 0.22 500~ 609 0.19 323 0.19 BMI(kg/m2) 0.588 0.899 < 18.5 321 0.10 161 0.10 18.5~ 1838 0.56 936 0.57 24~ 832 0.25 404 0.25 28~ 281 0.09 135 0.08 是否出现相关疾病 0.001 0.999 是 281 0.09 141 0.08 否 2991 0.91 1495 0.92 乙肝 0.001 0.999 是 185 0.06 93 0.06 否 3087 0.94 1543 0.94 传播途径 2.145 0.543 血液 669 0.20 345 0.21 性 1514 0.46 777 0.48 吸毒 996 0.30 465 0.28 其他 93 0.04 49 0.03 表 2 建模组和验证组患者的基线人口统计学和临床特征
变量 建模组 验证组 t 值 P 值 白细胞(109个/L) 5.71 ± 1.97 5.71 ± 1.95 – 0.041 0.921 血红蛋白(g/L) 135.24 ± 21.92 134.85 ± 20.37 0.606 0.545 血小板(109个/L) 183.43 ± 73.62 184.65 ± 76.04 – 0.539 0.589 血肌酐(μmol/L) 68.43 ± 21.46 69.15 ± 20.52 – 1.123 0.262 血糖(mmol/L) 5.24 ± 1.33 5.26 ± 1.36 – 0.413 0.680 谷草转氨酶及95 % CI(U/L) 28.01(21.65~42.72) 28.24(21.82~42.27) 0.792 谷丙转氨酶及95 % CI(U/L) 29.54(19.72~52.42) 28.46(18.93~50.82) 0.125 胆红素及95 % CI(μmol/L) 11.42(8.37~14.87) 10.86(8.19~14.34) 0.871 表 3 建模组患者的单因素与多因素比例风险回归分析
变量 单因素分析 多因素分析 HR 95 % CI P 值 HR 95 % CI P 值 WHO分期 Ⅰ 1.00 1.00 Ⅱ 1.67 1.31~2.13 < 0.001 1.31 1.06~1.75 0.034 Ⅲ 3.63 2.75~4.80 < 0.001 1.96 1.43~2.85 < 0.001 Ⅳ 3.04 1.91~4.82 < 0.001 1.60 0.98~2.73 0.073 基线CD4(个/μL) < 150 1.00 1.00 150~ 0.46 0.38~0.56 < 0.001 0.60 0.50~0.76 < 0.001 300~ 0.33 0.25~0.43 < 0.001 0.52 0.39~0.72 < 0.001 500~ 0.14 0.09~0.21 < 0.001 0.22 0.14~0.36 < 0.001 BMI(kg/m2) < 18.5 1.00 1.00 18.5~ 0.73 0.58~0.93 0.009 0.93 0.74~1.23 0.546 24~ 0.44 0.32~0.60 < 0.001 0.72 0.54~1.05 0.056 28~ 0.22 0.12~0.40 < 0.001 0.46 0.26~0.89 0.014 途径 血液传播 1.00 1.00 经性传播 0.67 0.51~0.87 0.002 0.78 0.57~1.04 0.109 吸毒 1.79 1.41~2.27 < 0.001 1.14 0.88~1.57 0.368 其他 0.21 0.05~0.83 0.027 0.27 0.06~1.05 0.067 性别 0.43 0.35~0.53 < 0.001 0.59 0.43~0.79 < 0.001 延迟时间 1.10 1.06~1.14 < 0.001 1.08 1.04~1.12 < 0.001 血红蛋白(g/L) 0.99 0.98~0.99 < 0.001 0.99 0.98~0.99 < 0.001 谷草转氨酶(U/L) 1.01 1.01~1.02 < 0.001 1.01 1.01~1.02 0.012 丙肝 0.51 0.43~0.60 < 0.001 0.87 0.68~1.07 0.199 是否出现相关疾病 0.50 0.40~0.63 < 0.001 1.06 0.81~1.41 0.679 胆红(μmol/L) 1.01 1.00~1.03 0.056 血糖(mmol/L) 0.96 0.89~1.03 0.284 乙肝 0.72 0.50~1.02 0.067 -
[1] Ford N, Ball A, Baggaley R, et al. The WHO public health approach to HIV treatment and care: looking back and looking ahead[J]. Lancet Infectious Diseases, 2018, 18(3): E76 – E86. [2] Ji YJ, Wang ZY, Shen JY, et al. Trends and characteristics of all-cause mortality among HIV-infected inpatients during the HAART era (2006 – 2015) in Shanghai, China[J]. Bioscience Trends, 2017, 11(1): 62 – 68. [3] Ortblad KF, Baeten JM, Cherutich P, et al. The arc of HIV epidemics in sub-Saharan Africa: new challenges with concen-trating epidemics in the era of 90 – 90 – 90[J]. Current Opinion in HIV and AIDS, 2019, 14(5): 354 – 365. [4] Wang SD, Yang L, Ci B, et al. Development and validation of a nomogram prognostic model for SCLC patients[J]. Journal of Thoracic Oncology, 2018, 13(9): 1338 – 1348. [5] Lee CK, Goldstein D, Gibbs E, et al. Development and validation of prognostic nomograms for metastatic gastrointestinal stromal tumour treated with imatinib[J]. European Journal of Cancer, 2015, 51(7): 852 – 860. [6] Kagaayi J, Gray RH, Whalen C, et al. Indices to measure risk of HIV acquisition in Rakai, Uganda[J]. PLoS One, 2014, 9(4): e92015. [7] McNairy ML, Jannat-Khah D, Pape JW, et al. Predicting death and lost to follow-up among adults initiating antiretroviral therapy in resource-limited settings: derivation and external validation of a risk score in Haiti[J]. PLoS One, 2018, 13(8): e0201945. [8] 杨爱学, 陈涛, 刘伟. 新疆伊犁州直1996 — 2015年艾滋病病例死亡情况分析[J]. 中国艾滋病性病, 2016, 22(10): 788 – 790. [9] Byrd KK, Hou JG, Bush T, et al. Adherence and viral suppression among participants of the patient-centered Human Immunode-ficiency Virus (HIV) care model project: a collaboration between community-based pharmacists and HIV clinical providers[J]. Clinical Infectious Diseases, 2020, 70(5): 789 – 797. [10] Lei ZQ, Li J, Wu D, et al. Nomogram for preoperative estimation of microvascular invasion risk in hepatitis B virus-related hepatocellular carcinoma within the Milan criteria[J]. JAMA Surgery, 2016, 151(4): 356 – 363. [11] Justice AC, Erlandson KM, Hunt PW, et al. Can biomarkers advance HIV research and care in the antiretroviral therapy era?[J]. The Journal of Infectious Diseases, 2018, 217(4): 521 – 528. [12] 陈洋, 宋晓甜, 姚永明, 等. 贵州省1996 — 2015年注射吸毒人群HIV/AIDS死亡率及其影响因素分析[J]. 中华流行病学杂志, 2019, 40(7): 765 – 769. doi: 10.3760/cma.j.issn.0254-6450.2019.07.006 [13] Bajpai R, Chaturvedi H, Jayaseelan L, et al. Effects of antiretroviral therapy on the survival of human immunodeficiency virus-positive adult patients in Andhra Pradesh, India: a retrospective cohort study, 2007 – 2013[J]. Journal of Preventive Medicine and Public Health, 2016, 49(6): 394 – 405. [14] Shoko C, Chikobvu D. A superiority of viral load over CD4 cell count when predicting mortality in HIV patients on therapy[J]. BMC Infectious Diseases, 2019, 19(1): 169. [15] Quiros-Roldan E, Castelli F, Lanza P, et al. The impact of antiretroviral therapy on iron homeostasis and inflammation markers in HIV-infected patients with mild anemia[J]. Journal of Translational Medicine, 2017, 15(1): 256. [16] Cai J, Osikowicz M, Sebastiani G. Clinical significance of elevated liver transaminases in HIV-infected patients[J]. AIDS, 2019, 33(8): 1267 – 1282. [17] Lodi S, Günthard HF, Dunn D, et al. Effect of immediate initiation of antiretroviral treatment on the risk of acquired HIV drug resistance[J]. AIDS, 2018, 32(3): 327 – 335. [18] 张娜, 朱晓艳, 王国永, 等. 山东省抗病毒治疗HIV/AIDS生存状况及影响因素分析[J]. 中华流行病学杂志, 2019, 40(1): 74 – 78. doi: 10.3760/cma.j.issn.0254-6450.2019.01.015 [19] Sharma S, Schlusser KE, De La Torre P, et al. The benefit of immediate compared with deferred antiretroviral therapy on CD4+ cell count recovery in early HIV infection[J]. AIDS, 2019, 33(8): 1335 – 1344. [20] Damtew B, Mengistie B, Alemayehu T. Survival and determinants of mortality in adult HIV/AIDS patients initiating antiretroviral therapy in Somali Region, Eastern Ethiopia[J]. The Pan African Medical Journal, 2015, 22: 138. [21] Naidoo K, Yende-Zuma N, Augustine S. A retrospective cohort study of body mass index and survival in HIV infected patients with and without TB co-infection[J]. Infectious Diseases of Poverty, 2018, 7(1): 35. [22] Ingle SM, May MT, Gill MJ, et al. Impact of risk factors for specific causes of death in the first and subsequent years of antiretroviral therapy among HIV-infected patients[J]. Clinical Infectious Diseases, 2014, 59(2): 287 – 297.