An epidemiological study of parasites infecting the South African abalone (Hiliotis midae) in Western Cape aquaculture facilities

Abstract

Global growth in aquaculture, referred to as the blue revolution, is seen by many to be the solution to future food scarcity. However, this growth has been accompanied by disease emergence. Disease emergence is inevitable when host populations are concentrated and densities exceed the threshold value for occurrence of outbreaks. Abalone farming is a relatively recent development and diseases of abalone are not well characterised. There have been relatively few systematic surveys of abalone diseases in the world. Much of the available information resulted from investigations of specific disease outbreaks, such as withering syndrome in California. The outstanding example of a formal survey of abalone health was conducted in Australia. A methodical survey of abalone health, encompassing all farms and including wild abalone, has never been done in South Africa. However, South Africa has for many years had a herd health program for abalone producers and this has generated the largest body of data on abalone disease occurrence in the world. Although these data have some shortcomings, it was felt that analysis could provide insights into the epidemiology of parasites in farmed Haliotis midae, as well as informing better surveillance techniques for the future. Data for abalone submitted from nine farms as part of the herd health management program during the period 1 January 2000 to 31 December 2004 were analysed. No wild abalone were included in these data and the only abalone species considered was the South African abalone or perlemoen, Haliotis midae. Data on the age of the abalone and their diet were obtained from the farms. The abalone originated from either flow through or recirculation systems. Each animal was weighed and measured at the laboratory. A standard set of tissue sections was fixed and processed for histology. Presence of parasites was recorded, as well as the sex and degree of gonad development. Once data had been captured in Excel, a series of tables was constructed from counts of infected and non infected abalone for all host and environmental factors contained in the data set. Charts of the tables were drawn. Where host and environmental factors appeared to interact, these data were also tabulated and charted. Statistical analyses of the data in Excel followed. All analyses were performed for sessile ciliates, renal coccidia, gut protozoa, digestive gland PAGE 2 OF 137 protozoa and rickettsia like prokaryotes. The rarity of trematode infections made meaningful analyses difficult. The chi square test, effect sizes and odds ratios were used to seek significant associations. When confounding and interaction were suspected, stratum specific odds ratios were calculated. The summary odds ratio used in this study was the Mantel Haenszel summary odds ratio. The Breslow Day test for interaction was performed when necessary. Confidence intervals were determined using the method of Woolf. The overall prevalence of the various parasites was very variable. Sessile ciliates were the most common, with a prevalence of 68.3%. Rickettsia like prokaryotes were found in 13.1% of animals. The other gut associated parasites were more scarce, with prevalences under five percent. Renal coccidia affected less than two percent of animals. Trematode infections were extremely rare, at a prevalence of 0.05%. The results of the chi square test showed a significant association between age and parasite prevalence for all parasites tested. Odds ratios were calculated comparing animals of 24 months and younger to those older than 24 months. In all cases, except trematode and left kidney coccidian infections, risk of parasite infection tended to increase with increasing age. For left kidney coccidian infections, risk of infection decreased with increasing age. Trends for body mass were similar to those for age, which is expected, as animals generally become larger with increasing age. A significant association between growth rate and parasite prevalence existed for some parasites. The chi square test showed a significant association between condition index and parasite prevalence for all parasites tested. A significant association between sex and parasite prevalence was found for all parasites tested. A significant association also existed between parasite prevalence and gonad development for sessile ciliates, renal coccidia and gut protozoa. There were significant differences in parasite prevalences between farms for all parasites tested. The South and West coasts were next compared using the chi square test. There were significant differences in parasite prevalences between coasts for only three of the parasites tested, namely renal coccidia, gut protozoa and rickettsia. Crude odds ratios showed that, with the exception of left kidney coccidia and trematodes, there was a greater risk of parasite infection on the West than the South coast. For left kidney coccidia, the risk was greater on the PAGE 3 OF 137 South coast. A chi square test was performed to examine the relationship between parasite prevalence in Hermanus and other areas. A significant difference was found for renal coccidia and gut associated parasites. The crude odds ratios for parasite prevalence in other areas compared to Hermanus were calculated. With the exception of sessile ciliates and trematodes, there was a greater risk of parasite infection in areas other than Hermanus. In the case of sessile ciliates, there was a greater risk within Hermanus than in other areas. Unfortunately, it was almost impossible to determine whether a seasonal effect exists for parasite prevalence from the available data. This was shown to be partly due to the effect of prevalence on individual farms. Significant differences in parasite prevalences between diets for all parasites tested were shown using the chi square test. To further test the strength of the association, odds ratios were calculated comparing only kelp and artificial feed. For right kidney coccidia and gut associated parasites, the odds ratios indicated a significantly increased risk of infection in animals receiving kelp compared to those on artificial feed. There was no difference in risk for sessile ciliates and left kidney coccidia. The majority of animals originated in flow through systems. The chi square test showed significant differences in parasite prevalences between systems for sessile ciliates, renal coccidia and rickettsia like prokaryotes. Odds ratios showed a significantly greater risk of sessile ciliate infections, but a smaller risk of left kidney coccidia, in animals in flow through systems when comparing only kelp fed animals. Age is likely to lead to increased prevalences if the risk of infection is constant over time and also if infections are retained. Physiological changes in the animals may also affect their risk of infection. A further important aspect of age in abalone relates to changes in husbandry. Increasing age may be the underlying reason for some of the prevalence patterns seen with mass and sex. The origin of the sample population was considered in terms of farm of origin as well as geographic area. Abalone production in the study area was highly concentrated, with approximately two thirds coming from six farms situated within ten kilometres of each another on the South coast and almost forty percent from Hermanus alone. The present study found no PAGE 4 OF 137 evidence of increased parasite prevalence in areas where abalone farming is concentrated. It was felt that this study could not generate much insight into seasonal occurrence, due to uneven distribution of variables between months. The relationship between diet and parasite prevalence was perhaps the most interesting aspect of this study and possible reasons for the association are explored. Lastly, the results indicated that parasite buildup in recirculation systems was not as problematical as may be expected. It is possible that the increased prevalence of left kidney coccidia in recirculation systems is linked to the resistance of the host population rather than to the dynamics of the actual system. The very low prevalence of sessile ciliates in recirculation systems could not be explained by examination of any other variable considered in this study. Overall, the prevalences of soft tissue parasites in Haliotis midae compared favourably with those found for parasites of other abalone species abroad. Measures which would tend to reduce parasite prevalence include separation of age groups and maintaining a relatively young population on the farm. Culling of underperforming animals is recommended. Kelp should not be used in animals of two years or younger. In older animals, there was still a greater risk associated with kelp than with artificial feed, but it was not as marked. Recirculation systems proved to be less associated with increased parasite prevalence than one may expect. The other major findings of this study did not lend themselves to practical application.