<sup>*</sup>University of Cape Town, Cape Town, South Africa;
<sup>†</sup>Oxford University Clinical Research Unit, Hospital for Tropical Diseases, Ho Chi Minh City, Viet Nam.
Int J Tuberc Lung Dis. 2015 Mar;19(3):276-7. doi: 10.5588/ijtld.14.0805.
We disagree with the recommendation by the World Health Organization to use Xpert(®) MTB/RIF on cerebrospinal fluid for the initial diagnosis of tuberculous meningitis (TBM). TBM is a devastating disease requiring empirical treatment even when the probability of disease is low. We suggest that a useful TBM diagnostic test needs a negative predictive value (NPV) of ⩾ 99% so that empirical treatment can be stopped safely. The NPV of Xpert is around 84%, making a negative test of limited value. While better tests are awaited, a composite score, possibly combining Xpert with clinical variables and with high NPV, should be constructed and validated prospectively.
我们不同意世界卫生组织推荐使用 Xpert(®) MTB/RIF 检测脑脊液来进行结核性脑膜炎(TBM)的初始诊断。TBM 是一种破坏性疾病,即使疾病的可能性较低,也需要进行经验性治疗。我们认为,有用的 TBM 诊断测试需要阴性预测值(NPV) ⩾ 99%,以便能够安全地停止经验性治疗。Xpert 的 NPV 约为 84%,因此阴性检测的价值有限。在等待更好的检测方法的同时,应该构建和前瞻性验证一个综合评分,可能将 Xpert 与临床变量和高 NPV 结合起来。