Seropositivity and temporo-spatial clustering of Leptospira microscopic agglutination test (MAT) seroreactivity among dogs in the United States between 2000 and 2007

Canine leptospirosis is a re-emerging infectious disease in the United States (U.S.). Increasing prevalence of canine leptospirosis could have important public health implications. The aims of this study were to determine the seropositivity and potential spatial clustering of seroreactivity to Leptospira in dogs using microscopic agglutination test (MAT) results of a commercial veterinary diagnostic laboratory. Electronic records of 33,119 Leptospira MAT results for dogs from 2000 through 2007 were obtained from IDEXX Laboratories, Inc. A titer of ≥1:1600 to at least one of seven serovars was defined as a positive test. A spatial scan statistic for population rate data (Poisson distribution) incorporated in SaTScan was used to identify statistically significant clusters using a spatial window of 0-20% of the total study area/population at risk and a temporal window of 0-6 months. There were 2,680 samples (8.1%; 95% CI=7.8-8.4) MAT positive from a total of 33,119 submissions from 3,224 zipcodes, representing all states of the U.S. The lowest seropositivity (4.8%; 95% CI=4-5.7) was recorded for the year 2001 and the highest (10.2%; 95% CI=9.5-10.9) for 2007. Two significant spatial clusters (26.3 and 246.5 km radius; P=0.001) were identified. The smaller cluster included Chicago and surrounding zipcodes (232 [14.4%] of 1,612 tests positive; RR=1.95). The larger cluster included the central region of Texas (292 [12.6%] of 2,314 tests positive; RR=1.71). Eight significant space-time clusters were identified (29-335 km radius; P=0.001-0.048 and RR=3.98-24.69) that covered different geographic locations for different time points. Significant space-time clusters of seroreactivity to individual serovars were also identified for six serovars. Three significant space-time clusters of seroreactivity to Icterohaemorrhagiae and Canicola each, 8 each of Grippotyphosa and Pomona, 7 of Bratislava, and 5 of Autumnalis were identified. This study shows that Leptospira seropositivity in dogs does cluster geographically. While space-time clusters for overall seropositivity could be explained by seroreactivity to individual serovars, reasons for clusters of seroreactivity to individual serovars need to be explored in future investigations. Presence of clusters of seroreactivity to individual serovars in different geographic regions suggests potential roles of specific reservoir hosts, environmental risk factors or difference in practice of preventative measures. Further investigation is required to identify the risk factors for clusters of seroreactivity to individual serovars in dogs