Dalmat Ronit R, Visek Caitlin, Budiawan Elvira, Stein Gabrielle, Nalutaaya Annet, Mukiibi James, Nantale Mariam, Biché Patrick, Sung Joowhan, Magcaba Zanele, Ngcobo Nompumelelo, Morton Jennifer F, Lenn Meena, Aucock Samantha, Shapiro Adrienne E, Steadman Amy, Draper Robin, Ganguloo Amanda, Wilson Douglas, Katamba Achilles, Kendall Emily A, Drain Paul K
Department of Global Health, University of Washington, Seattle, Washington, USA.
Division of Infectious Diseases, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA.
medRxiv. 2025 Jul 15:2025.07.14.25331384. doi: 10.1101/2025.07.14.25331384.
Highly sensitive molecular tests, like Xpert Ultra, are reshaping TB diagnosis-detecting paucibacillary TB but sometimes creating uncertainty when they detect DNA in extremely low quantities that may not signal disease. This ambiguity also complicates the evaluation of novel diagnostic strategies. We sought to monitor adults with a 'Trace' result on an Xpert Ultra test to estimate the risk of tuberculosis disease up to 24 months later.
We conducted a multi-site clinical observational study in South Africa and Uganda, where we enrolled ambulatory participants aged >=15 years with a sputum Xpert Ultra Trace result who had not yet initiated TB treatment. All participants underwent comprehensive clinical evaluation and repeated, standard sputum TB testing. Clinicians deferred treatment recommendations if TB status remained uncertain after evaluation. Untreated participants were followed regularly until TB diagnosis and/or treatment initiation. We estimated cumulative incidence of TB disease, defined by three reference standards.
Of 311 participants with Trace results (50% male, 57% PLHIV, 37% treated for TB within the last 5 years), 24% were positive for TB within 12 months by culture, 37% by sputum Xpert or culture, and 54% by culture or clinical diagnosis. After excluding those diagnosed with TB at baseline, patients identified at baseline as having recent TB history, abnormal chest x-ray, or positive tongue swab, had higher risk of TB diagnosis (hazard ratios: 2.6, 2.4, 4.5, respectively) during follow-up. This hazard was highest in the first three months after the negative baseline evaluation (0.22 [95% CI: 0.19-0.26] per person-month) and decreased to 0.01 [95% CI: 0-0.02] per person-month in both the 3-6 and 6-12-month intervals.
Approximately half of adults and adolescents with a sputum Trace result were diagnosed with TB disease within twelve months. Although most TB diagnoses were made within 3 months, risk remained higher than estimated population incidence rates through the follow-up period. Individuals with sputum Trace results should receive close clinical monitoring, despite initial clinical treatment decision.
高灵敏度分子检测,如Xpert Ultra,正在重塑结核病诊断——能检测出少菌型结核病,但有时当它们检测到极少量可能不表明患病的DNA时会产生不确定性。这种模糊性也使新型诊断策略的评估变得复杂。我们试图对Xpert Ultra检测结果为“痕量”的成年人进行监测,以评估直至24个月后发生结核病的风险。
我们在南非和乌干达开展了一项多中心临床观察性研究,纳入年龄≥15岁、痰Xpert Ultra检测结果为痕量且尚未开始抗结核治疗的门诊参与者。所有参与者均接受了全面的临床评估,并进行了重复的标准痰结核菌检测。如果评估后结核病状态仍不确定,临床医生会推迟治疗建议。未治疗的参与者被定期随访,直至结核病诊断和/或开始治疗。我们根据三种参考标准估计了结核病的累积发病率。
在311名检测结果为痕量的参与者中(50%为男性,57%为艾滋病毒感染者,37%在过去5年内接受过结核病治疗),12个月内痰培养结核病阳性率为24%,痰Xpert或培养结核病阳性率为37%,培养或临床诊断结核病阳性率为54%。在排除基线时诊断为结核病的患者后,基线时被确定有近期结核病病史、胸部X光异常或舌拭子阳性的患者,在随访期间结核病诊断风险更高(风险比分别为:2.6、2.4、4.5)。这种风险在基线阴性评估后的前三个月最高(每人每月0.22 [95%置信区间:0.19 - 0.26]),在3 - 6个月和6 - 12个月期间降至每人每月0.01 [95%置信区间:0 - 0.02]。
约一半痰检测结果为痕量的成人和青少年在12个月内被诊断为结核病。尽管大多数结核病诊断在3个月内做出,但在随访期间风险仍高于估计的人群发病率。痰检测结果为痕量的个体应接受密切临床监测,无论最初的临床治疗决定如何。