VISAVET Health Surveillance Centre, Universidad Complutense de Madrid, Spain.
VISAVET Health Surveillance Centre, Universidad Complutense de Madrid, Spain; MAEVA SERVET, S.L., Alameda del Valle, Madrid, Spain.
Prev Vet Med. 2019 Dec 1;173:104800. doi: 10.1016/j.prevetmed.2019.104800. Epub 2019 Oct 18.
In spite of the efforts invested to eradicate bovine tuberculosis (bTB) in cattle, the disease is still present in several developed countries, including Spain. Among the factors related with disease persistence in high prevalence areas, the lack of sensitivity of the screening test for detection of infected herds [single intradermal tuberculin (SIT) test] can play a major role. Here, a scenario tree model mimicking the diagnostic test scheme in place in the region of Castilla y Leon (Spain) was developed to estimate the probability of detecting bTB in an infected-non detected officially tuberculosis free (OTF) herd (herd sensitivity, HSe). In order to do so the probability of detecting at least one positive animal in the SIT test with/without post-mortem (detection of lesions and culture) confirmation in an infected herd was estimated using Monte Carlo simulation through @RISK (Palisade Co, NY, USA). Uncertainty on the accuracy of the diagnostic tests was introduced in the model using distributions based on the literature. The performance of the model was evaluated by comparing the predicted number of SIT/post-mortem positive animals in infected herds with those observed in newly detected bTB-infected herds in the region in 2011-2015. The estimated HSe of the SIT test was 76.2% (95% probability interval: 19.8-97.6). According to the model, bTB infection would be then confirmed through culture in 65.3% (95% PI: 50.0-82.3) of the herds detected through the SIT test, so that overall the proportion of infected-non detected OTF herds in which the infection could be confirmed after the initial SIT test was 49.6% (95% PI: 9.75-80.3). The predicted HSe of both SIT test and culture was directly correlated with herd size. Results from the model suggest a moderate but highly variable HSe of the current surveillance system in place for bTB detection in OTF herds located in high prevalence areas, that could be maximized by performing multiple tests within a year as indicated in the Spanish eradication program (with a median SIT HSe of 87% when two consecutive tests were considered). In addition, these results highlight the usefulness of performing subsequent SIT tests to rule out infection in SIT-positive herds even when the causative agent cannot be isolated.
尽管在根除牛结核病（bTB）方面投入了大量努力，但该病仍在包括西班牙在内的几个发达国家存在。在高流行地区疾病持续存在的相关因素中，检测感染牛群的筛查试验（单皮内结核菌素（SIT）试验）缺乏敏感性可能是一个主要因素。在这里，我们开发了一个情景树模型，模拟了卡斯蒂利亚-莱昂（西班牙）地区现行的诊断测试方案，以估计在官方无结核病（OTF）牛群中检测 bTB 的概率（ herd sensitivity，HSe）。为此，使用蒙特卡罗模拟通过 @RISK（Palisade Co，NY，USA）估算了在感染牛群中，使用/不使用尸检（检测病变和培养）确认，在 SIT 试验中至少检测到一个阳性动物的概率。使用基于文献的分布，在模型中引入了对诊断测试准确性的不确定性。通过将模型预测的感染牛群中 SIT/尸检阳性动物数量与 2011-2015 年该地区新发现的 bTB 感染牛群中的观察值进行比较，评估了模型的性能。SIT 试验的估计 HSe 为 76.2%（95%概率区间：19.8-97.6）。根据该模型，bTB 感染将通过培养在通过 SIT 试验检测到的 65.3%（95%PI：50.0-82.3）的牛群中得到确认，因此，在初始 SIT 试验后，感染而非检测到的 OTF 牛群中可以确认感染的比例为 49.6%（95%PI：9.75-80.3）。模型预测的 SIT 试验和培养的 HSe 与牛群规模直接相关。模型结果表明，在高流行地区对 OTF 牛群进行 bTB 检测的现行监测系统的 HSe 适中，但高度可变，可通过在一年内进行多次检测来最大化该系统（如西班牙根除计划所示，连续两次检测时 SIT 的中位数 HSe 为 87%）。此外，这些结果强调了在 SIT 阳性牛群中进行后续 SIT 检测以排除感染的有用性，即使无法分离病原体也是如此。
