Limits...
Temporal and spatial association of Streptococcus suis infection in humans and porcine reproductive and respiratory syndrome outbreaks in pigs in northern Vietnam.

Huong VT, Thanh LV, Phu VD, Trinh DT, Inui K, Tung N, Oanh NT, Trung NV, Hoa NT, Bryant JE, Horby PW, Kinh NV, Wertheim HF - Epidemiol. Infect. (2015)

Bottom Line: Only residency in adjacent districts remained significantly associated with risk of S. suis infection after adjusting for sex, occupation, and eating practices.SaTScan analysis showed a possible cluster of S. suis infection in humans around PRRS confirmed locations during the March-August period.Effective strategies to strengthen control of PRRS in pigs may help reduce transmission of S. suis infection to humans.

View Article: PubMed Central - PubMed

Affiliation: Wellcome Trust Major Overseas Programme,Oxford University Clinical Research Unit,Hanoi,Vietnam.

ABSTRACT
Porcine reproductive and respiratory syndrome (PRRS) outbreaks in pigs are associated with increased susceptibility of pigs to secondary bacterial infections, including Streptococcus suis - an important zoonotic pathogen causing bacterial meningitis in humans. This case-control study examined the association between human S. suis infection and PRRS outbreaks in pigs in northern Vietnam. We included 90 S. suis case-patients and 183 non-S. suis sepsis controls from a referral hospital in Hanoi in 2010, a period of major PRRS epizootics in Vietnam. PRRS exposure was determined using data from the National Centre of Veterinary Diagnosis. By univariate analysis, significantly more S. suis patients were reported residing in or adjacent to a PRRS district compared to controls [odds ratio (OR) 2·82, 95% confidence interval (CI) 1·35-5·89 and OR 3·15, 95% CI 1·62-6·15, respectively]. Only residency in adjacent districts remained significantly associated with risk of S. suis infection after adjusting for sex, occupation, and eating practices. SaTScan analysis showed a possible cluster of S. suis infection in humans around PRRS confirmed locations during the March-August period. The findings indicate an epidemiological association between PRRS in pigs and S. suis infections in humans. Effective strategies to strengthen control of PRRS in pigs may help reduce transmission of S. suis infection to humans.

Show MeSH

Related in: MedlinePlus

Clusters of S. suis cases detected in humans in three SaTScanrunning sets. (a) Set 1: only human cases and controls were used asinput of the Bernoulli model. (b) Set 2: human cases and controlswith porcine reproductive and respiratory syndrome (PRRS) locations as centroid ofthe moving space-time window. (c) Set 3: human cases and controlswith eight groups of covariates by sex, occupation and history of eating high-riskpig dishes. Red dots represent human cases (90 cases); black square representcontrols (183 controls); pig symbols represent locations confirmed with PRRS virus;purple circles represent the possible clusters constructed from SaTScan software.Cluster 1 is the most likely cluster, cluster 2 is the secondary cluster. For eachcluster, P value, time-frame of the cluster detected, loglikelihood ratio (LLR) and relative risk [RR; except panel (c)] areprovided.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4697300&req=5

fig03: Clusters of S. suis cases detected in humans in three SaTScanrunning sets. (a) Set 1: only human cases and controls were used asinput of the Bernoulli model. (b) Set 2: human cases and controlswith porcine reproductive and respiratory syndrome (PRRS) locations as centroid ofthe moving space-time window. (c) Set 3: human cases and controlswith eight groups of covariates by sex, occupation and history of eating high-riskpig dishes. Red dots represent human cases (90 cases); black square representcontrols (183 controls); pig symbols represent locations confirmed with PRRS virus;purple circles represent the possible clusters constructed from SaTScan software.Cluster 1 is the most likely cluster, cluster 2 is the secondary cluster. For eachcluster, P value, time-frame of the cluster detected, loglikelihood ratio (LLR) and relative risk [RR; except panel (c)] areprovided.

Mentions: At the global scale, our bivariate K-function analysis suggested spatialclustering of S. suis human cases occurring at distances of 2–50 km(Supplementary Fig. S1). There was weak evidence of space–time interaction for bothS. suis cases in human and PRRS outbreaks in pigs (Supplementary Fig.S2) at this global level. However, space–time analyses at the local level showed strongclusters of human S. suis cases occurring around locations where PRRSoutbreaks were confirmed. Possible human S. suis clusters were found inall three running sets performed in SaTScan (Table4). In running set 1, the most likely cluster was found within a radius of ~39·4 kmfrom April to October (Fig. 3a).People who lived within the cluster had a higher risk of contracting S.suis infection than people living outside the cluster (RR 2·82). Using PRRSoutbreaks for locating cluster centroids, the second running set found two likelyS. suis clusters. The most likely cluster was diagnosed between Marchand August with a larger radius (53·6 km) (Fig.3b) with a similar RR (2·86). In set 3, which included sex,occupation, and history of eating high-risk dishes as covariates, we only found one humanS. suis cluster also between March and August (Fig. 3c). This cluster contained predominantly thefour male patient groups with and without occupational exposure and history of high-riskconsumption. The greatest risks were in men who worked in swine-related occupations and/orhad a history of eating high-risk pig dishes (RRs from 3·78 to 6·0). Fig. 3.


Temporal and spatial association of Streptococcus suis infection in humans and porcine reproductive and respiratory syndrome outbreaks in pigs in northern Vietnam.

Huong VT, Thanh LV, Phu VD, Trinh DT, Inui K, Tung N, Oanh NT, Trung NV, Hoa NT, Bryant JE, Horby PW, Kinh NV, Wertheim HF - Epidemiol. Infect. (2015)

Clusters of S. suis cases detected in humans in three SaTScanrunning sets. (a) Set 1: only human cases and controls were used asinput of the Bernoulli model. (b) Set 2: human cases and controlswith porcine reproductive and respiratory syndrome (PRRS) locations as centroid ofthe moving space-time window. (c) Set 3: human cases and controlswith eight groups of covariates by sex, occupation and history of eating high-riskpig dishes. Red dots represent human cases (90 cases); black square representcontrols (183 controls); pig symbols represent locations confirmed with PRRS virus;purple circles represent the possible clusters constructed from SaTScan software.Cluster 1 is the most likely cluster, cluster 2 is the secondary cluster. For eachcluster, P value, time-frame of the cluster detected, loglikelihood ratio (LLR) and relative risk [RR; except panel (c)] areprovided.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4697300&req=5

fig03: Clusters of S. suis cases detected in humans in three SaTScanrunning sets. (a) Set 1: only human cases and controls were used asinput of the Bernoulli model. (b) Set 2: human cases and controlswith porcine reproductive and respiratory syndrome (PRRS) locations as centroid ofthe moving space-time window. (c) Set 3: human cases and controlswith eight groups of covariates by sex, occupation and history of eating high-riskpig dishes. Red dots represent human cases (90 cases); black square representcontrols (183 controls); pig symbols represent locations confirmed with PRRS virus;purple circles represent the possible clusters constructed from SaTScan software.Cluster 1 is the most likely cluster, cluster 2 is the secondary cluster. For eachcluster, P value, time-frame of the cluster detected, loglikelihood ratio (LLR) and relative risk [RR; except panel (c)] areprovided.
Mentions: At the global scale, our bivariate K-function analysis suggested spatialclustering of S. suis human cases occurring at distances of 2–50 km(Supplementary Fig. S1). There was weak evidence of space–time interaction for bothS. suis cases in human and PRRS outbreaks in pigs (Supplementary Fig.S2) at this global level. However, space–time analyses at the local level showed strongclusters of human S. suis cases occurring around locations where PRRSoutbreaks were confirmed. Possible human S. suis clusters were found inall three running sets performed in SaTScan (Table4). In running set 1, the most likely cluster was found within a radius of ~39·4 kmfrom April to October (Fig. 3a).People who lived within the cluster had a higher risk of contracting S.suis infection than people living outside the cluster (RR 2·82). Using PRRSoutbreaks for locating cluster centroids, the second running set found two likelyS. suis clusters. The most likely cluster was diagnosed between Marchand August with a larger radius (53·6 km) (Fig.3b) with a similar RR (2·86). In set 3, which included sex,occupation, and history of eating high-risk dishes as covariates, we only found one humanS. suis cluster also between March and August (Fig. 3c). This cluster contained predominantly thefour male patient groups with and without occupational exposure and history of high-riskconsumption. The greatest risks were in men who worked in swine-related occupations and/orhad a history of eating high-risk pig dishes (RRs from 3·78 to 6·0). Fig. 3.

Bottom Line: Only residency in adjacent districts remained significantly associated with risk of S. suis infection after adjusting for sex, occupation, and eating practices.SaTScan analysis showed a possible cluster of S. suis infection in humans around PRRS confirmed locations during the March-August period.Effective strategies to strengthen control of PRRS in pigs may help reduce transmission of S. suis infection to humans.

View Article: PubMed Central - PubMed

Affiliation: Wellcome Trust Major Overseas Programme,Oxford University Clinical Research Unit,Hanoi,Vietnam.

ABSTRACT
Porcine reproductive and respiratory syndrome (PRRS) outbreaks in pigs are associated with increased susceptibility of pigs to secondary bacterial infections, including Streptococcus suis - an important zoonotic pathogen causing bacterial meningitis in humans. This case-control study examined the association between human S. suis infection and PRRS outbreaks in pigs in northern Vietnam. We included 90 S. suis case-patients and 183 non-S. suis sepsis controls from a referral hospital in Hanoi in 2010, a period of major PRRS epizootics in Vietnam. PRRS exposure was determined using data from the National Centre of Veterinary Diagnosis. By univariate analysis, significantly more S. suis patients were reported residing in or adjacent to a PRRS district compared to controls [odds ratio (OR) 2·82, 95% confidence interval (CI) 1·35-5·89 and OR 3·15, 95% CI 1·62-6·15, respectively]. Only residency in adjacent districts remained significantly associated with risk of S. suis infection after adjusting for sex, occupation, and eating practices. SaTScan analysis showed a possible cluster of S. suis infection in humans around PRRS confirmed locations during the March-August period. The findings indicate an epidemiological association between PRRS in pigs and S. suis infections in humans. Effective strategies to strengthen control of PRRS in pigs may help reduce transmission of S. suis infection to humans.

Show MeSH
Related in: MedlinePlus