Porcine reproductive and respiratory syndrome virus (PRRSV) has plagued pig populations worldwide causing severe economical impacts. In order to establish effective strategies for prevention of PRRS, infection patterns on the herd level are primarily evaluated. In the present study, therefore, serological and virological analyses were conducted in 20 pig farms suffering from PRRS. Seroprevalence levels in each farm were grouped into 3 patterns: SN (Stable sow groups/Not infected piglet groups, SI (Stable sow groups and Infected piglet groups), and UI (Unstable sow groups and Infected piglet groups). The rates of each serological pattern were 15% (n=3), 10% (n=2), and 75% (n=15), respectively. In addition, the pattern analysis was extended to virological monitoring on the same farms that further included suckling pig groups. As a result, the infection pattern was classified into 4 categories: SNI (Stable sow groups/Not infected suckler groups/Infected piglet groups), SII (Stable sow groups/Infected suckler groups/Infected piglet groups), UNI (Unstable sow groups/Not infected suckler groups/Infected piglet groups), and UII (Unstable sow groups/Infected suckler groups/Infected piglet groups). The rates of each viroprevalence were estimated at 50% (n=10), 30% (n=6), 10% (n=2), and 10% (n=2), respectively. PRRSV viroprevalence results of suckling pig groups revealed that 8 farms were considered virus positive. In 2 farms among these farms, PRRSV appeared to be transmitted vertically to suckling piglets from their sows. In contrast, piglet-to-piglet horizontal transmission of PRRSV seemed to occur in sucking herds of the remaining farms. Thus, this virological analysis on suckling piglets will provide useful information to understand PRRSV transmission routes during the suckling period and to improve a PRRS control programs. Our seroprevalence and viroprevalence data found that infection patterns between sow and piglet groups are not always coincident in the same farm. Remarkably, 15 farms belonging to the UI seroprevalence pattern showed four distinct viroprevalence patterns (SNI; 7, SII; 4, UNI; 2 and UII; 2). Among these farms, 11 farms with unstable seroprevalence sow groups were further identified as the stable viroprevalence pattern. These results indicated that despite the absence of typical seroconversion, PRRSV infection was detected in several farms, implying the limitation of serological analysis. Taken together, our data strongly suggests that both seroprevalence and viroprevalence should be determined in parallel so that a PRRS control strategies can be efficiently developed on a farm level.