In general, a high prevalence and severity of pneumonia was observed in this study. The percentage of 79% of pigs with lung lesions in this study was comparable to that found in other studies in comparable production systems, which reported 73.9% in Brazil [26]. In western France, [8] reported that pneumonia (69.3% of lungs) and pleurisy (15% of lungs) were the most frequent lesions seen at slaughter. Large variations in pneumonia and pleuritis (41–76% and 2–35%, respectively) were reported in Brittany, France [14, 27]. A similar observation was made for other studies in Italy, which reported prevalences of 59.6% and 61.4% respectively [11, 28]. A study in Phillipines reported that 48% of slaughter pigs had high lung scores, while 22% had pleurisy [29]. In Brazil, a recent study reported that 68.5% of slaughter pigs presented with macroscopic lung lesions [15]. In Belgium, the prevalence of pleuritis and pneumonia was reported to be 20.76% and 23.85%, respectively [30]. In Makurdi, Benue State, Nigeria, it was reported that 36.4% of sampled pigs had lung lesions [31]. Differences in these studies likely reflect differences in production systems, hygiene and health status overall.
The prevalence of pleuritic lesions found in this study is comparable to other studies elsewhere, in which it was found to vary between 20.5% and 33.1% in Sweden [32]. In south eastern Norway, it was reported that pneumonic or pleuritic lesions were found in 84% of the lungs and that of bronchopneumonia was found in 70% of the lungs examined [33]. A study in France reported that pleuritis and cranio-ventral pulmonary consolidation lesions were recorded in 26.8% and 55.7% of slaughter-aged pigs, respectively [9].
In contrast, the high prevalence of pneumonic lesions found in this study is higher than that reported in Eastern Spain, which reported the mean lesion values for CPBP, PLP and pleuritis as 8.4%, 0.4% and 0.3% of farms, repectively [13]. In Ghana, [34] found that only 5% of slaughtered pigs had pneumonia, though the form of pneumonia was not reported. These studies demonstrate a wide variation in prevalence and severity of lung pneumonic lesions, which may be explained by differences in production systems and management factors, which ultimately influence exposure levels to respiratory pathogens. Also, the different study designs, scoring methods used, environmental conditions and production systems in which pigs are raised account for variations in pneumonia prevalence and severity.
Overrall, the odds of detecting CPBP, PLP and pleuritis were higher in pigs that tested seropositive to all the pathogens (ORs 1.2–6.2), except for PRRSv on CPBP. This suggests possible associations between pathogen seropositivity and lesion scores. However, differences in median pneumonic scores between seropositive and seronegative pigs were statistically insignificant. M. hyo is reported to be strongly associated with lung lesions [35] and pulmonary consolidation [36]. In contrast, the odds of detecting pigs with CPBP and PLP increased in pigs that tested positive to APP. The finding that only APP-seropositive pigs resulted in a significantly higher MLS could be due to either infection with pathogenic APP serotypes, or that coinfections with APP and other pathogens produced considerable lung damage. This conforms to documented reports of the gross morphological features of APP as characterised by fibrinonecrotic pleuritis and diffuse heamorrhages [25]. These findings are not unusual, as previous studies have documented associations between lung lesions and serology to respiratory pathogens in pigs [32, 36].
Table 6 showed relationships between pneumonia type and multiple infections. Results of the regression model showed that the odds of detecting multiple pneumonia forms increased in pigs with coinfections and Metastrongylus spp. This corroborates previous studies which documented synergistic or potentiating effects of coinfections between PRRSv and other pathogens [10], PCV2 and other pathogens in the induction of respiratory disease [9, 37, 38].
This study did not find association between CPBP scores and PRRSv seropositivity. The observed PLP and pleuritis lesions could be due to the effect of coinfections between PRRSv and other pathogens, notably PCV2 and M. hyo. The ability of M. hyo infection to potentiate and prolong PRRSv-induced pneumonia clinically and macroscopically has been documented [10]. Our findings that M. hyo seropositivity was more associated with PLP than CPBP (OR 6.2 vs 1.3 respectively) instead contrasts with other studies which found that M. hyo is associated with CPBP [13]. Notwithstanding differences in the study design by [13], which sampled only heavy pigs (100kg), we sampled pigs of varying ages and live weights from predominantly small scale production systems. The disease progression from acute to chronic as pigs grow older, may explain differences in the lesion scores observed.
The finding that approximately a third of the pigs sampled had Metastrongylus spp nematodes, frequently observed in the tips of diaphragmatic lobes, is in agreement with a previous study, which found high prevalences of GIT nematodes in Ugandan pigs [39]. GIT parasites such as Ascaris suum, Metastrongylus spp are known to induce pulmonary tissue damage through their migratory larvae, increasing the susceptibility of pigs to various respiratory pathogens [25].
Apart from two studies in Nigeria and Ghana, no other published study was found in Africa (with comparable production systems) which documented the magnitude of pneumonia in pigs. It is worthy to mention that in Uganda, no other published study or report was found on the magnitude of pneumonia in pigs. Thus, in the context of different pig production systems documented in Uganda [40], our findings can only be extrapolated to the swine population in northern Uganda, with similar husbandry systems. This study showed that a high proportion of pigs brought for slaughter in the region presented with high prevalence and severity of pneumonic lesions, and the association between lesions and serologic status suggests a significant contribution of the studied pathogens in lung pathology.
Limitations of the study
The scoring methodology used in this study may have underestimated the actual magnitude of pneumonia, since some pigs may have suffered early in life and lesions could have resolved. We acknowledge the method used for estimation of surface area grossly affected by pneumonia may have reduced accuracy in estimation of lesion scores. Besides, due to the need to perform the scoring process quickly to match with the slaughter speed, it is probable that some hidden lesions may have been missed.