论文部分内容阅读
背景:在撒哈拉以南非洲地区,采用可靠的临床程序,对开始接受或已经接受抗病毒治疗的艾滋病病人进行活动性结核病筛查,可减少对诊断程序的需要。方法:我们对6个与结核病相关的症状和体征单独或组合在一起的效用进行了估计,并与乌干达卫生部对现患结核病(在评估抗病毒治疗资格时,作为基线)、抗病毒治疗早期(抗病毒治疗开始后的3个月内)结核病和新发结核病(抗病毒治疗3个月以后)的诊断指南进行了比较。结果:在1 995例接受抗病毒治疗资格筛查的HIV感染者中,71例(3.6%)被诊断患有结核病。对于预测活动性结核病而言,下列任一症状或体征:咳嗽≥3周,发热≥4周,淋巴结肿大或基线体质量指数≥18 kg/m~2,其灵敏度为99%(95%可信区间为96～100),特异度66%(95%可信区间为64～68),阴性预测值为100%(95%可信区间为99～100)。在抗病毒治疗随访期间,结核病发病率为2.4/100人年(95%可信区间为1.6～3.4)。出现咳嗽≥3周或全身乏力症状或体征的灵敏度为100%(95%可信区间为99～100),特异度为66%(95%可信区间为59～74),阴性预测值为100%(95%可信区间为99～100)。结论:在资源贫乏的非洲地区,应用简单的结核病筛查流程,可以准确地辨别出哪些HIV感染者需要接受进一步的结核诊断检查。
Background. In sub-Saharan Africa, using reliable clinical procedures to screen for active tuberculosis among HIV-positive patients who have started or who have already received antiviral therapy reduces the need for diagnostic procedures. METHODS: We assessed the utility of 6 tuberculosis-related symptoms and signs, either alone or in combination, with the Ugandan Ministry of Health for tuberculosis in emergencies (as a baseline in eligibility for antiviral treatment), early antiviral treatment (Within 3 months after the start of ART) diagnostic guidelines for tuberculosis and new tuberculosis (3 months after ART). Results: Of the 1 995 HIV-infected patients eligible for antiviral therapy 71 (3.6%) were diagnosed with tuberculosis. For the prediction of active tuberculosis, one of the following symptoms or signs: cough ≥ 3 weeks, fever ≥ 4 weeks, lymph node enlargement or baseline body mass index ≥ 18 kg / m ~ 2, the sensitivity was 99% (95% The confidence interval ranged from 96 to 100) with a specificity of 66% (95% confidence interval 64 to 68) and a negative predictive value of 100% (95% confidence interval 99 to 100). The incidence of tuberculosis was 2.4 / 100 person-years (95% confidence interval, 1.6 to 3.4) during ART follow-up. Sensitivity of coughing ≥3 weeks or generalized fatigue symptoms or signs was 100% (95% confidence interval 99 to 100), specificity 66% (95% confidence interval 59-74), negative predictive value 100 % (95% confidence interval 99 ~ 100). Conclusion: In resource-poor Africa, a simple tuberculosis screening process can be used to accurately identify which HIV-infected individuals need further tuberculosis diagnostic tests.