Costa Federico, Ribeiro Guilherme S, Felzemburgh Ridalva D M, Santos Norlan, Reis Renato Barbosa, Santos Andreia C, Fraga Deborah Bittencourt Mothe, Araujo Wildo N, Santana Carlos, Childs James E, Reis Mitermayer G, Ko Albert I
Centro de Pesquisas Gonçalo Moniz, Fundação Oswaldo Cruz, Ministério da Saúde, Salvador, Bahia, Brazil; Department of Epidemiology of Microbial Diseases, School of Public Health, Yale University, New Haven, Connecticut, United States of America; Instituto de Saúde Coletiva, Universidade Federal da Bahia, Bahia, Brazil.
Centro de Pesquisas Gonçalo Moniz, Fundação Oswaldo Cruz, Ministério da Saúde, Salvador, Bahia, Brazil; Instituto de Saúde Coletiva, Universidade Federal da Bahia, Bahia, Brazil.
PLoS Negl Trop Dis. 2014 Dec 4;8(12):e3338. doi: 10.1371/journal.pntd.0003338. eCollection 2014 Dec.
The Norway rat (Rattus norvegicus) is the principal reservoir for leptospirosis in many urban settings. Few studies have identified markers for rat infestation in slum environments while none have evaluated the association between household rat infestation and Leptospira infection in humans or the use of infestation markers as a predictive model to stratify risk for leptospirosis.
METHODOLOGY/PRINCIPAL FINDINGS: We enrolled a cohort of 2,003 urban slum residents from Salvador, Brazil in 2004, and followed the cohort during four annual serosurveys to identify serologic evidence for Leptospira infection. In 2007, we performed rodent infestation and environmental surveys of 80 case households, in which resided at least one individual with Leptospira infection, and 109 control households. In the case-control study, signs of rodent infestation were identified in 78% and 42% of the households, respectively. Regression modeling identified the presence of R. norvegicus feces (OR, 4.95; 95% CI, 2.13-11.47), rodent burrows (2.80; 1.06-7.36), access to water (2.79; 1.28-6.09), and un-plastered walls (2.71; 1.21-6.04) as independent risk factors associated with Leptospira infection in a household. We developed a predictive model for infection, based on assigning scores to each of the rodent infestation risk factors. Receiver operating characteristic curve analysis found that the prediction score produced a good/excellent fit based on an area under the curve of 0.78 (0.71-0.84).
CONCLUSIONS/SIGNIFICANCE: Our study found that a high proportion of slum households were infested with R. norvegicus and that rat infestation was significantly associated with the risk of Leptospira infection, indicating that high level transmission occurs among slum households. We developed an easily applicable prediction score based on rat infestation markers, which identified households with highest infection risk. The use of the prediction score in community-based screening may therefore be an effective risk stratification strategy for targeting control measures in slum settings of high leptospirosis transmission.
挪威大鼠(褐家鼠)是许多城市环境中钩端螺旋体病的主要宿主。很少有研究确定贫民窟环境中鼠患的标志物,也没有研究评估家庭鼠患与人类钩端螺旋体感染之间的关联,或者使用鼠患标志物作为预测模型来分层钩端螺旋体病的风险。
方法/主要发现:2004年,我们招募了来自巴西萨尔瓦多的2003名城市贫民窟居民组成一个队列,并在四次年度血清学调查期间对该队列进行跟踪,以确定钩端螺旋体感染的血清学证据。2007年,我们对80户病例家庭(每户至少有一名钩端螺旋体感染个体)和109户对照家庭进行了鼠患和环境调查。在病例对照研究中,分别在78%和42%的家庭中发现了鼠患迹象。回归模型确定,褐家鼠粪便的存在(比值比,4.95;95%置信区间,2.13 - 11.47)、鼠洞(2.80;1.06 - 7.36)、靠近水源(2.79;1.28 - 6.09)以及未抹灰的墙壁(2.71;1.21 - 6.04)是与家庭中钩端螺旋体感染相关的独立危险因素。我们基于为每个鼠患危险因素赋予分数,开发了一种感染预测模型。受试者工作特征曲线分析发现,基于曲线下面积为0.78(0.71 - 0.84),预测分数产生了良好/优异的拟合。
结论/意义:我们的研究发现,很大比例的贫民窟家庭有褐家鼠出没,且鼠患与钩端螺旋体感染风险显著相关,这表明在贫民窟家庭中存在高水平传播。我们基于鼠患标志物开发了一种易于应用的预测分数,可识别感染风险最高的家庭。因此,在基于社区的筛查中使用该预测分数可能是一种有效的风险分层策略,用于在钩端螺旋体病高传播的贫民窟环境中针对性地采取控制措施。