運用基因與空間資料探討多重抗藥性結核在台灣之動態傳播

Abstract

背景 為了減少多重抗藥性結核(MDR-TB)的傳播，本研究整合了流行病學、基因及空間資料，並在地圖上標示MDR-TB的遺傳群聚，以徹底了解其傳播模式。次要目標為鑑定群聚可能的風險因素。 方法 收集台灣2012年1月1日至2015年12月31日所有MDR-TB病例的個案。使用MIRU-VNTR 10和spoligotyping將病例分配到所屬的遺傳群聚。個案的居住地址,戶籍地址，以及職業(活動)地址皆以數字座標為其地理編碼，並將每一個遺傳群聚個案單獨標示在地圖上。運用單變量邏輯斯迴歸和多變量邏輯斯迴歸鑑別群聚的潛在風險因子。 結果 MDR-TB最大的遺傳群聚在台灣的3個地區有地理群聚。7名個案確認有空間連結，其中3個透過流行病學確認其連結，另有1例與2個地理區域相關。 從2012年至2015年，台灣共有474個多重抗藥結核病人， 466（98％）有基因分型。台灣東部地區、年齡35-44歲、結核菌北京株以及住宅地址不同於活動地址都會增加群聚的可能性。 討論與結論 使用不同類型的地址有助於提供額外的資訊，解釋同一群聚分布廣泛的特性。距離和空間連結能夠補充說明流行病學連結所缺失的傳播模式。 年齡介於35到44歲之間為MDR-TB聚集的最大風險因子。

INTRODUCTION To reduce transmitted multidrug-resistant tuberculosis (MDR-TB) and break chains of transmission, a thorough understanding of its transmission pattern is necessary. The primary objective was to integrate epidemiological, genetic, and spatial data into maps of genetic clusters. The secondary objective was to identify risk factors for clustering. METHODS Data was gathered on all MDR-TB cases in Taiwan from January 1, 2012 to December 31, 2015. Cases were assigned to genetic clusters using mycobacterial interspersed repetitive units-variable-number tandem repeats (MIRU-VNTR) 10 and spoligotyping. The residential, permanent, and activity addresses of each case were geo-coded into numerical coordinates. Clusters were mapped individually. Univariate logistic regression and multivariable logistic regression were used to identify potential risk factors for clustering. RESULTS The largest genetic cluster of 20 MDR-TB cases had geographic clusters in 3 regions of Taiwan. 7 spatial links were identified, and 3 of these were confirmed by epidemiological links. One case was linked to two geographical regions. Between 2012 and 2015, there were 474 cases of MDR-TB in Taiwan. 466 (98%) had genotyped isolates. Increased odds of clustering were found in the Taiwan-born, in East area of Taiwan, in ages of 35-44, in Beijing lineage, and in cases reporting different residential and activity addresses. DISCUSSION Using different types of addresses may be useful to explain seemingly non-localized transmission. Spatial links based on distance could potentially explain transmission patterns where epidemiological links are missing. The working age group between the ages 35-44 was the most at risk for clustering of MDR-TB. Future studies should incorporate other addresses available as well as information from whole genome sequencing to increase the understanding of transmission.