Reduction in finisher pig death rate on a farm in Japan following use of a commercial killed vaccine against porcine reproductive and respiratory syndrome virus

Abstract

PRRS (Porcine Reproductive and Respiratory Syndrome) is a costly disease in the global pig industry. In 2021, Nisseiken Co., Ltd. in Japan launched a PRRS-killed vaccine, PRRS-ME, but its effectiveness on reducing the death rate under field conditions remained unassessed. A longitudinal study on a PRRS-positive farm in Japan (July 2019–July 2022) evaluated its effect on death rates. Piglets on a PRRS-negative farrowing farm were vaccinated at 3 and 8 weeks of age in March 2021. The vaccinated piglets were introduced to the finisher farm, where a PRRS outbreak occurred. Vaccination coverage was not associated with the death rates of growers but was negatively associated with those of finishers. The predicted reduction in the death rate of finishers associated with complete vaccination coverage was 56.8%.

Porcine reproductive and respiratory syndrome (PRRS) is a major swine disease affecting the swine industry worldwide, causing significant economic losses. In the USA, these losses are estimated at 664 million United States Dollars annually [7]. Since first reported in Japan in 1994 [18, 33]. PRRS has severely impacted the Japanese pig industry, with estimated losses of approximately 28 billion Japanese Yen between 2006 and 2008 [39]. PRRS is caused by the PRRS virus (PRRSV), an enveloped, single-stranded, positive-sense RNA virus of the family Arteriviridae, which is divided into two genotypes: PRRSV-1 (European genotype) and PRRSV-2 (North American genotype). In Japan, both PRRSV-1 and PRRSV-2 have been reported, with PRRSV-2 being the predominant genotype circulating in the country [8, 14]. Both PRRSV species cause respiratory disorders and subsequent death in growing pigs, as well as reproductive failure in sows [3, 32]. These clinical manifestations lead to a decline in farm productivity due to reproductive impairments in sows and increased mortality rates among grower-finishers [7, 21, 32].

Vaccination is a key approach for controlling PRRS [3, 27]. Commercial PRRS vaccines include modified-live virus (MLV) and killed virus (KV) vaccines, each of which has pros and cons [3, 19]. MLV vaccines are considered more efficacious but raise safety concerns associated with mutation, recombination [13, 40], and reversion to virulence in the field [11, 22]. KV vaccines are considered safer than MLV vaccines; however, the efficacy of KV vaccines is debated because both insufficient and sufficient immune responses in pigs have been reported [3, 19, 31]. Some studies have reported that KV vaccines are efficacious in certain situations; for example, the efficacy of a KV vaccine is more noticeable when applied to pigs that have been exposed to PRRSV, as evidenced by improved humoral immune responses and reproductive performance [23, 25].

KV vaccines appear to be utilized less frequently compared with MLV vaccines for PRRS. Several commercial MLV vaccines for PRRS have been available globally since the first PRRSV-2 MLV vaccine was introduced in North America in 1994, followed by a PRRSV-1 MLV vaccine in Europe in 2000 [2, 41]. By contrast, choices of major commercial KV vaccines for PRRS are limited and currently licensed only in Europe and Asia [3, 16, 19]. A PRRSV-2 KV vaccine designated “PRRS-ME” (Nisseiken Co., Ltd., Tokyo, Japan) was introduced in 2021. The PRRS-ME vaccine contains inactivated whole virions with mineral oil as an adjuvant [20]. The PRRS-ME vaccine strain is classified as Cluster III of PRRSV-2, according to a previous study [8]. Under experimental conditions, the PRRS-ME vaccine shortened the duration of viremia and reduced viremia levels following challenge with a virulent strain when pigs were vaccinated according to the prescribed method and dosage [24]. Additionally, efficacy of the PRRS-ME vaccine in reducing the duration of viremia was observed up to 4 months after the second vaccination upon challenge with the virulent strain [24].

To our knowledge, the effectiveness of the PRRS-ME vaccine in safeguarding commercial farm productivity has yet to be thoroughly studied. Therefore, the present study evaluated the effectiveness of the PRRS-ME vaccine in reducing death rates as an indicator of productivity in growers and finishers during a PRRS epidemic. We evaluated the association between PRRS-ME vaccination and grower-finisher deaths on a commercial farm in Japan using data recorded on the farm. We also evaluated the effect of the PRRS-ME vaccine in terms of the rate of detecting the virus genome in the blood of vaccinated pigs.

This was a longitudinal study conducted on a commercial pig farm with 700 sows in Aichi Prefecture, Japan, where PRRS-ME vaccination was implemented during part of the study period in response to an outbreak of PRRS. PRRSV antibodies and PRRSV genes were detected in January 2018 on a finishing farm, and respiratory cases and deaths among finishers increased in early 2019 (Fig. 1a). The observation period covered July 2019 to July 2022. The farm consisted of a farrowing farm and a finishing farm located 13.5 km apart. The outbreak was observed only on the finisher farm. Farrowing on the farm occurs weekly, and piglets are reared until approximately 70 days of age before being transferred to the finishing farm (Fig. 1b). Approximately 350 pigs are moved weekly. Based on the feed change, pigs on the finishing farm ranging in age between 71 days and 112 days were defined as growers, which are in the intermediate growth phase, and those aged ≥113 days were defined as finishers, which are in the final phase before market or slaughter. The finishing farm operates an all-in–all-out system on a building basis. Annual routine testing is conducted in January, with testing of blood samples from sows and pigs aged 30, 60, 90, 120, and 150 days for PRRSV, porcine circovirus 2 (PCV2), Actinobacillus pleuropneumoniae, Mycoplasma hyopneumoniae, Erysipelothrix rhusiopathiae, Pasteurella multocida, and Bordetella bronchiseptica.

Fig. 1.

Background information of the farm showing the history of porcine reproductive and respiratory syndrome (PRRS) (a) and vaccination protocol (b). (a) History of PRRS in the farm. No PRRS outbreaks occurred in the farrowing farm from 2018 to 2022. (b) Vaccination protocol. The farm consisted of a PRRS-negative farrowing farm and a PRRS–positive finishing farm. Pigs are typically moved from the farrowing farm to the finishing farm at approximately 70 days of age. Piglets on the farrowing farm were administered the PRRS-ME vaccine at 3 and 8 weeks of age.

The finishing farm had no history of PRRS outbreaks or vaccination until 2017. However, PRRS–positive 150-day-old pigs were identified on the finishing farm in January 2018 based on routine testing. PRRSV-specific antibodies and genes were detected using enzyme-linked immunosorbent assay (ELISA) and reverse transcription–polymerase chain reaction (RT-PCR), respectively. In January 2019 and January 2020, 90-day-old pigs also tested positive for PRRSV by RT-PCR and ELISA. In addition, since 2019, respiratory symptoms with suspected infection of PRRSV and monthly death rates in growers/finishers increase was observed by a veterinarian in charge (Fig. 1a). Vaccination with PRRS-ME began in March 2021 to reduce PRRS-related productivity losses as per the policy of the farm and responsible veterinarian. The farrowing farm remained PRRS test–negative during the observation period. Vaccinations against PCV2, M. hyopneumoniae, and classical swine fever virus were routinely administered to piglets. There were no notable changes in farm management practices during the observation period, except for PRRS-ME vaccination.

Piglets were administered the PRRS-ME vaccine in March 2021. Initially, all piglets on the farrowing farm aged ≥3 weeks were administered the PRRS-ME vaccine on March 30, 2021, followed by a second dose 3 weeks later to quickly establish herd immunity. Subsequently, routine vaccination was started; newborn piglets were vaccinated at 3 and 8 weeks of age (Fig. 1b). All vaccinations were conducted according to the manufacturer’s instructions. Piglets that received two doses were introduced to the finishing farm between April 2021 and July 2022. The PRRS-ME vaccine was provided through an agency by Nisseiken Co., Ltd., where the authors are affiliated. During this period, a veterinarian in charge monitored the pigs and diagnosed and treated any diseased pigs as needed. Blood specimens from piglets on the farrowing farm were sampled 8 times, whereas blood samples from growers/finishers were sampled once or twice each month during the observation period between April 2021 and July 2022 in order to monitor PRRS status on the farms. All samples were stored in the refrigeration room of the farm at approximately 4°C. A total of 190 blood samples from piglets aged 14 to 72 days and 715 samples from growers/finishers aged 71 to 166 days were stored at the farm. All samples were transported to a private company, SMC Co., Ltd. (Atsugi, Japan), at a cool temperature of 0–10°C.

Samples were examined using quantitative reverse transcription–PCR (qRT-PCR) for PRRSV-specific genes. One representative sample was selected from approximately five pigs within each batch introduced into the finishing farm on the same day. In total, 50 samples were sequenced to determine the PRRSV ORF5 gene. Extraction of RNA from all samples, qRT-PCR, sequencing and phylogenetic analysis were outsourced to SMC Co., Ltd. Homology between the farm strains and the vaccine strain of PRRS-ME (Accession Number: AB175721) was evaluated using MEGA11 software (Tamura, Stecher, and Kumar 2021) [35].

Monthly data regarding pig inventory (total number of pigs on the farm), grower and finisher deaths, and the age of pigs when they died for the period 2019 to 2022 were recorded on the farm. These data were obtained and used under permission of the farm. Monthly death rates for both growers and finishers were calculated by dividing the number of deaths by the monthly pig inventory for the respective stages. Annual death rates for growers and finishers were calculated by dividing the annual number of deaths at respective phases by the number of pigs introduced to the finishing farm (note that these indicators show relative intensities of risk by phases which is commonly used in the industry). Monthly vaccination coverage was determined by dividing the number of pigs vaccinated twice by the total number of inventory pigs on the fattening farm at the middle of the month.

Univariable and multivariable analyses were conducted using a generalized linear model (GLM) with negative binomial errors to examine the association between monthly number of deaths and vaccination coverage on the grower and finisher farms, respectively. The negative binomial error was selected based on the Akaike Information Criterion (AIC) using the GLM. For this, univariable analyses were first conducted selecting the monthly number of deaths as the outcome variable and vaccination coverage, season, and year as explanatory variables, setting the monthly inventory as an offset. The year period for the analysis was defined from July of a certain year to June of the following year, resulting in consideration of the periods 2019/2020 (19/20), 2020/2021 (20/21), and 2021/2022 (21/22, which included July 2022 due to the study’s observation end point). Seasons were categorized as spring (March, April, and May), summer (June, July, and August), autumn (September, October, and November), and winter (December, January, and February) based on the definitions from the Japan Meteorological Agency [10]. The significance of each variable was assessed using the Wald test, and variables with P-values <0.20 in univariable analyses were included in the multivariable analysis. The best model was then determined using the AIC. To assess the potential for multicollinearity among the explanatory variables, the variance inflation factor (VIF) for the best models was calculated. Based on the best GLM, a generalized linear mixed-effects model (GLMM) was performed, selecting the average number of monthly deaths as the outcome variable and vaccination coverage as the fixed-effect explanatory variable, with seasonality as a random effect and the monthly inventory as an offset. In the GLMM, the average number of monthly deaths at the 0% and 100% vaccination rates was predicted using the prediction function in R. To measure the predicted vaccination effectiveness, the rate of reduction in monthly deaths by full introduction of vaccination to the totally non-vaccinated herd was calculated as: rate of reduction=(V0-V100)/V0, where V0 represents the predicted monthly deaths at the 0% vaccination rate and V100 represents that at the 100%. The univariable and multivariable statistics were conducted separately for growers and finishers. This stratification is because these two age groups, although both housed on the finishing farm, are exposed to different durations of risk and potentially different transmission dynamics during the PRRS outbreak on that farm. To assess the difference in stage-dependent susceptibility of PRRSV, the proportion of PCR–positive samples collected between April 2021 and July 2022 was compared between growers and finishers using the chi-square test. Additionally, to assess the efficacy of the vaccine in preventing PRRSV infection, the proportion of PCR–positive samples was compared between fully vaccinated (twice-vaccinated) pigs and non–fully vaccinated (unvaccinated or once-vaccinated) pigs, which were expected to have vaccine-induced protection as they were vaccinated according to the manufacturer’s instructions, using the data from April to July 2021, when the finishing farm accommodated all unvaccinated, once-vaccinated, and twice-vaccinated pigs, using the chi-square test. All analyses were conducted using R, version 4.3.3 [29] and RStudio Cloud [28].

During the period from July 2019 to March 2021 (i.e., before vaccination), the median and interquartile range of monthly deaths of growers/finishers were 178 and 154 to 229, respectively. The median and interquartile range of monthly total pig inventory (including both growers and finishers) were 6,443 and 6,067 to 6,761, respectively. The median and interquartile range of monthly death rates for growers were 1.00% and 0.70 to 1.54%, respectively, and those for finishers were 1.83% and 1.33 to 2.15%, respectively. A seasonality pattern, with the peak in monthly death rate occurring in winter—December to February, with the peak in January 2020—was observed in this period for both growers and finishers (Fig. 2).

Fig. 2.

Monthly trends in death rates of growers and finishers. Monthly death rates are shown for growers (black dashed line) and finishers (black solid line). Red vertical solid line shows the introduction of vaccinated pigs to the finishing farm.

After PRRS-ME vaccination was initiated in March 2021, although the monthly death rates in growers and finishers peaked in winter, the pattern greatly declined in 2022 (Fig. 2). For finishers, the seasonal pattern was less pronounced compared with growers. An outbreak of Streptococcus suis infection was reported in winter during the vaccination period, leading to an acute increase in mortality. However, mortality declined after the outbreak compared with the pre-vaccination period. The annual death rate was 12.4%, 10.0%, and 6.1% for growers and finishers overall; 5.0%, 4.0%, and 2.5% for growers; and 7.4%, 6.0%, and 3.6% for finishers during the 19/20, 20/21, and 21/22 periods, respectively. Clinical observations indicated a decrease in chronic cough after vaccination began, as reported by the veterinarian in charge. This seasonality might play a significant role in influencing the rate of death due to PRRSV. The seasonality of PRRS varies across different countries and regions [1, 15, 37]. In the United States, the fall and winter seasons are commonly referred to as the “PRRS season”, described as a period with a high frequency of PRRSV infection [36]. This seasonal pattern has been attributed to climatic factors. For instance, the mechanical transmission of PRRSV via common farm workers is more likely to occur in cold weather [5]. Additionally, lower temperatures contribute to the stability of PRRSV in aerosol [6]. These reports suggest that the likelihood of PRRSV transmission is greater during winter in regions with a cold climate. In the area in which the observation farm is located, winter average temperatures (December–February) range from 5°C to 7°C. This cold climate likely contributed to a seasonal PRRS epidemic on the farm.

In the univariable analysis, all three variables (i.e., vaccine coverage, season, and year) showed P<0.20 in the Wald test for both growers (Table 1) and finishers (Table 2). Using these variables, three multivariable models were constructed: models with (i) vaccination coverage, season, and year (designated G1 in growers and F1 in finishers); (ii) season and year (G2 in growers and F2 in finishers); and (iii) vaccination coverage and season (G3 in growers and F3 in finishers). The smallest AIC values were obtained with model G2 for growers (AIC values in models G1, G2, and G3 for growers were 338.7, 336.7, and 336.9, respectively) (Table 3). In the best model for growers, spring (estimate=−0.44, standard error [SE]=0.18, P=0.015), summer (estimate=−0.59, SE=0.18, P=0.001), and 21/22 (estimate=−0.77, SE=0.15, P<0.001) were associated negatively with death, whereas winter was associated positively (estimate=0.49, SE=0.18, P=0.005) (Table 3). The VIFs for the season and year variables were both calculated to be 1.00. Vaccination coverage was not included in the best model (G2) for growers.

Table 1. Univariable analysis results on deaths in growers (generalized linear model with negative binomial errors).

Variable		Estimate	Standard error	95%CI		P-value
				Lower	Upper
Vaccination coverage		−0.70	0.20	−1.10	−0.31	<0.001
Season	Autumn (reference)
	Spring	−0.39	0.23	−0.85	0.06	0.091
	Summer	−0.55	0.23	−0.99	−0.10	0.016
	Winter	0.54	0.23	0.09	0.99	0.019
Year	19/20 (reference)
	20/21	−0.23	0.23	−0.68	0.23	0.330
	21/22	−0.78	0.23	−1.22	−0.33	<0.001

^a CI: confidence interval.

Table 2. Univariable analysis on deaths in finishers (generalized linear model with negative binomial errors).

Variable		Estimate	Standard error	95%CI		P-value
				Lower	Upper
Vaccination coverage		−0.75	0.19	−1.12	−0.37	<0.001
Season	Autumn (reference)
	Spring	−0.51	0.27	−1.04	0.03	0.06
	Summer	−0.26	0.27	−0.78	0.26	0.32
	Winter	0.20	0.27	−0.33	0.73	0.45
Year	19/20 (reference)
	20/21	−0.22	0.22	−0.66	0.21	0.31
	21/22	−0.80	0.22	−1.22	−0.37	<0.001

^a CI: confidence interval.

Table 3. Multivariable analysis results on deaths in growers (generalized linear model with negative binomial errors).

Model	Variable		AIC	Estimate	Standard error	95%CI		P-value
						Lower	Upper
G1			338.73
	Vaccination coverage			−0.03	0.67	−1.35	1.28	0.961
	Season	Autumn (reference)		-	-	-	-	-
		Spring		−0.44	0.18	−0.80	−0.08	0.015
		Summer		−0.59	0.18	−0.94	−0.23	0.001
		Winter		0.49	0.18	0.15	0.84	0.005
	Year	19/20 (reference)		-	-	-	-	-
		20/21		−0.22	0.16	−0.53	0.10	0.174
		21/22		−0.74	0.68	−2.07	0.59	0.276
G2			336.73
	Season	Autumn (reference)		-	-	-	-	-
		Spring		−0.44	0.18	−0.80	−0.09	0.015
		Summer		−0.59	0.18	−0.94	−0.24	0.001
		Winter		0.49	0.18	0.15	0.84	0.005
	Year	19/20 (reference)		-	-	-	-	-
		20/21		−0.22	0.15	−0.52	0.08	0.155
		21/22		−0.77	0.15	−1.08	−0.47	<0.001
G3			336.91
	Vaccination coverage			−0.68	0.14	−0.96	−0.40	<0.001
	Season	Autumn (reference)		-	-	-	-	-
		Spring		−0.43	0.19	−0.80	−0.07	0.020
		Summer		−0.56	0.18	−0.92	−0.20	0.002
		Winter		0.49	0.18	0.13	0.84	0.008

^a CI: confidence interval, AIC: Akaike information criterion.

For finishers, the best model was F3 (AIC values in models F1, F2, and F3 in finishers were 378.6, 378.2, and 375.8, respectively) (Table 4). In the best finisher model F3, spring (estimate=−0.56, SE=0.21, P=0.008) and increased vaccination coverage (estimate=−0.87, SE=0.16, P<0.001) were associated negatively with deaths (Table 4). The VIFs for the vaccination coverage and season variables were both calculated to be 1.00. The results of the GLMM with negative binomial errors for finisher pigs showed that increased vaccination coverage was associated with a significant reduction in finisher deaths (estimate=−0.84, SE=0.17, P<0.001) (Table 5). The rate of reduction in the monthly deaths of finishers after vaccination had been fully introduced (vaccination rate 100%) to a totally non-vaccinated herd (vaccination rate 0%) was 56.8%. The significant negative association between finisher deaths and PRRS-ME vaccine coverage suggests the vaccine is effective in preventing deaths and supports a published report that a KV vaccine is effective in preventing PRRS-related death [4]. PRRS is known to increase mortality among growers/finishers. For example, Neumann et al. estimated that PRRS increases grower-finisher mortality in the US by 6.1% [21]. Ishizeki et al. reported significantly higher post-weaning mortality on PRRS–positive Japanese farms (6.1%) compared with PRRS-negative farms (3.9%) [9]. Although the impact of PRRS on death rates on our targeted farm was unclear due to a lack of pre-outbreak data, the risk of death presumably increased due to PRRS. As such, the predicted 56.8% reduction in monthly deaths of finishers in association with PRRS-ME vaccination strongly supports the effectiveness of the vaccine. Furthermore, the reduction in annual death rate from 12.4% to 6.1% in growers/finishers in our study was similar to the difference in death rates between PRRS-affected and unaffected farms reported by Neuman et al. [21].

Table 4. Multivariable analysis results on deaths in finishers (generalized linear model with negative binomial errors).

Model	Variable		AIC	Estimate	Standard error	95%CI		P-value
						Lower	Upper
F1			378.60
	Vaccination coverage			−0.95	0.76	−2.43	0.53	0.209
	Season	Autumn (reference)		-	-	-	-	-
		Spring		−0.57	0.21	−0.98	−0.16	0.006
		Summer		−0.28	0.20	−0.67	0.12	0.173
		Winter		0.30	0.20	−0.09	0.70	0.135
	Year	19/20 (reference)		-	-	-	-	-
		20/21		−0.19	0.18	−0.55	0.17	0.306
		21/22		0.01	0.76	−1.49	1.50	0.993
F2			378.22
	Season	Autumn (reference)		-	-	-	-	-
		Spring		−0.59	0.21	−1.01	−0.18	0.005
		Summer		−0.29	0.21	−0.70	0.11	0.153
		Winter		0.30	0.21	−0.11	0.71	0.151
	Year	19/20 (reference)		-	-	-	-	-
		20/21		−0.25	0.18	−0.60	0.11	0.174
		21/22		−0.93	0.18	−1.28	−0.58	<0.001
F3			375.79
	Vaccination coverage			−0.87	0.16	−1.18	−0.55	<0.001
	Season	Autumn (reference)		-	-	-	-	-
		Spring		−0.56	0.21	−0.97	−0.15	0.008
		Summer		−0.29	0.20	−0.69	0.11	0.152
		Winter		0.31	0.21	−0.10	0.71	0.139

^a CI: confidence interval, AIC: Akaike information criterion.

Table 5. Generalized linear mixed-effects model result on deaths in finishers.

Variable	Estimate	Standard error	95%CI		P-value
			Lower	Upper
Intercept	−4.10	0.17	−4.43	−3.76	<0.001
Vaccination coverage	−0.84	0.17	−1.17	−0.51	<0.001

^a CI: confidence interval.

All 50 field strains belonged to PRRSV-2. The median homology among the field strains was 99.7%, ranging from 98.7% to 100%. The degree of homology in ORF5 between the detected strains and the vaccine strain was relatively low, ranging from 86.0% to 88.2%. The strains from this farm were classified into Cluster III, the same as the PRRS-ME vaccine strain. All samples from the farrowing farm tested negative by qRT-PCR. On the finishing farm, the proportion of qRT-PCR–positive samples during the entire study period was significantly higher in finishers (150/376, 39.9%) than growers (41/339, 12.1%, χ²=68.96, df=1, P<0.001). The change in the monthly percentage of qRT-PCR–positive samples over time is summarized in Fig. 3, which revealed an increase between April and July 2021 as well as a higher percentage of positive samples around the winter season. Between April and July 2021, PCR results showed no significant difference in the proportion of positive samples between twice-vaccinated pigs (41/120, 34.1%) and once-vaccinated or unvaccinated pigs (32/85, 37.6%, χ²=0.13, df=1, P=0.72). The results of the present study correspond to reported laboratory results regarding the efficacy of KV vaccines. The protective effect of a vaccine containing PRRSV-specific IgM (IgM-KV) was reported in a laboratory study by Chen et al. [4]. They also reported that viremia levels determined using qRT-PCR did not decrease significantly in pigs vaccinated with the IgM-KV vaccine, whereas mortality was lower in pigs vaccinated with the IgM-KV vaccine compared with pigs vaccinated with the MLV vaccine [4]. The lack of a significant difference in the percentage of positive samples between twice-vaccinated pigs and those that were either once-vaccinated or unvaccinated in the present study was consistent with this result. These laboratory and epidemiologic results suggest that KV vaccines do not prevent the presence of the viral genome in blood but do modulate the progression of symptoms caused by the virus.

Fig. 3.

Monthly trends in the percentage of quantitative reverse transcription PCR (qRT-PCR)-positive samples. Dots represent the estimated percentage of qRT-PCR–positive samples each month. Error bars indicate 95% confidence intervals (CI). Red arrow shows the introduction timing of vaccinated pigs to the finishing farm.

In contrast to previous reports, this study demonstrated recovery in fattening productivity when pigs were administered a KV vaccine before PRRSV exposure. Many studies have suggested that KV vaccines are particularly efficacious in previously infected pigs [3, 23, 25, 26, 30]. By contrast, several reports have highlighted the efficacy of KV vaccines in naïve pigs by demonstrating augmented humoral responses upon PRRSV exposure [12, 17, 34, 38, 42]. The results of the present study support the latter, indicating that vaccine-induced immunity contributed to mitigating the death rate among finishers.

This study has several limitations. First, infectious pathogens other than PRRS that affect the death rate in pigs, such as PCV2, M. hyopneumoniae, A. pleuropneumoniae, and Lawsonia intracellularis, were not evaluated. Therefore, the death rate may have fluctuated due to the impact of these pathogens. However, we conclude that the reduction in death rate was primarily due to PRRS-ME vaccination because decreased coughing was observed during the vaccination period. Second, this study focused on only one farm, thus diminishing generalization of the results. A multi-farm study will be required to confirm our finding regarding the positive association between PRRS-ME vaccination and productivity. Third, this study did not evaluate average daily weight gain or feed requirements. The overall impact on productivity on the farm is thus veiled and should be assessed in future research.

In conclusion, PRRS-ME vaccination significantly reduced the finisher death rate on a commercial farm experiencing a PRRS outbreak. Although KV vaccines are commonly used to increase productivity on PRRS–positive farrowing farms, this study suggests the vaccines are also effective in reducing finisher mortality. Considering previous experimental studies and these findings, the effectiveness of KV vaccines, including the PRRS-ME vaccine, in reducing death rates should be highlighted. The PRRS-ME vaccine is thus a new option for improving productivity during PRRS outbreaks.

CONFLICT OF INTEREST

This study was jointly conducted by Nisseiken Co., Ltd. and Rakuno Gakuen University. The views expressed in this article are those of the authors and not necessarily those of Nisseiken Co., Ltd.