Disease investigation relating to a herpes like virus infection in Abalone
3. Key assumptions
4. Origin of the virus
5. Introduction of the virus
6. Spread of disease on farms
7. Disease in the wild population
8. Disease investigation and response
9. Looking forward
10. Sources of information
The objective of this report is to review and advise on the disease investigation and response relating to a herpes like virus infection of abalone in Victoria. It is considered that the virus is most probably endemic to abalone in Australia. The best fit scenario as described by Paul HardySmith in his report of August 2006 provides a logical explanation for the introduction of the virus. The subsequent spread of the disease within farms is difficult to explain based on current knowledge. Mortalities in the wild population have been relatively modest so far, with the exception of an area approximately 10kms to the west of Port Fairy. This suggests that the wild population may be resistant to infection, which in turn supports the argument that the virus is endemic. However, the outbreak has not yet reached a conclusion. The disease investigation to date has been rapid and thorough. The epidemiological report by Paul HardySmith presents an excellent review of relevant events. Response by the Department of Primary Industries was conservative, which may be considered appropriate given the lack of information on which to base decisions and the likely endemic nature of the disease. Future management will no doubt be facilitated by the development of better diagnostic tests for the herpes like virus.
To review and advise on the disease investigation and response relating to a herpes like virus infection of abalone in Victoria
When considering the epidemiology of a disease outbreak, it is preferable to draw conclusions from fact, rather than forming conjecture based on uncertainties. Lack of information leads to reliance on assumptions which may in time be proven false. The current outbreak of disease in Victoria was first noticed late last year. A herpes like virus was identified as a probable cause during January of this year. Affected farms had depopulated by the middle of June. Reporting on the events surrounding the outbreak occurred by the end of August and a workshop to discuss these events was held in September. At the time of the workshop, the outbreak in the wild abalone population was still ongoing and incompletely described. In addition, the virus held to be responsible for the disease has not yet been fully characterised. The absence of a reliable diagnostic test has precluded determination of prevalence and the identification of a source of infection. Although it is recognised that action cannot wait on complete understanding, it must be stated that a shortfall of certainties does affect the probability of an appropriate response.
This report was commissioned by the Department of Primary Industries for Victoria, to provide an independent review of the disease investigation to date. It relies on the same sources of information that were used to compile the reports drafted by Panaquatic Health Solutions and is thus subject to the same limitations. These sources are listed below, together with certain key assumptions. Should any of the assumptions be incorrect, subsequent conclusions are also likely to be incorrect. A number of explanations exist that fit the current known facts of the outbreak. Some of these may be regarded as more probable than others and this report will focus on most probable or best fit scenarios, at the same time acknowledging that rare events do rarely occur. This report also provides a review of the disease response, as described by various role players, including industry. There is far more known about the outbreak at this time than when the initial response occurred and it is tempting, but ultimately pointless, to speculate on outcomes resulting from decisions which were not taken. As far as possible, this report will evaluate response without using hindsight wisdom.
3. Key assumptions
- The mortalities experienced on the abalone farms in Victoria are due to ganglioneuritis caused by a herpes like virus.
- Mortalities experienced on these farms prior to December 2005 are from causes other than this herpes like virus.
- The disease may be spread by direct contact between infected abalone as well as indirect contact, such as through water.
- Although the possibility of vertical transmission cannot be excluded, this was not seen during the current outbreak and is not considered in this report.
- Latency, giving rise to the existence of a carrier status, is a feature of herpes virus infection and is assumed to be likely for the purposes of this report.
4. Origin of the virus
The origin and ecology of the herpes like virus will to some extent dictate control measures. There are essentially three possible origins of this virus, listed in order of increasing probability:
- it is a virus not naturally occurring in abalone
- it is a virus of abalone which is exotic to Australia and was introduced
- it is a virus of abalone which is endemic to Australia, but previously went unnoticed.
Whereas it is feasible to eradicate an exotic virus, or even one which has newly evolved to infect abalone, this will not be possible should the virus prove to be endemic. The case for each of these options will be considered below.
4.1. non abalone virus
The ability of viruses, which tend to be host specific, to infect new host species, has been well documented. In such cases, infection in the new host is often more serious than in the old, as the new host has not had the opportunity to evolve resistance. Evidence from other abalone species as well as other molluscs suggests that herpes like viruses of molluscs may be able to infect a relatively wide range of hosts. Certainly, in the current outbreak, the virus appeared equally deadly for blacklip and greenlip abalone, and hybrids. Exposure to potential primary hosts of the virus would occur more readily in wild abalone than those on farms, but farmed abalone do come into contact with wild molluscs, such as predatory gastropods and, of course, wild abalone. The importation of wild molluscs, for example frozen limpet meat, for processing, is another potential route of virus introduction. Present diagnostic abilities do not allow investigation of other species for presence of virus and this possibility cannot be excluded from currently known facts. Exposure of candidate species to virus as a result of the ongoing outbreak in the wild population may render future testing inconclusive, so collection of appropriate samples should be conducted now.
4.2. abalone virus exotic to Australia
An outbreak of disease apparently very similar to this one occurred in Asia in recent years. The question has been posed whether the virus responsible for the Asian outbreak is the same as that currently affecting Australian abalone. This will be known in due course when the virus has been better characterised. If one virus proves to be responsible for both outbreaks, did it spread to Australia from Asia? This would seem unlikely for the simple reason that Australia practices strict import control, compared to less rigorous prevention of disease introduction by Asian countries. In addition, there is a constant flow of live abalone from Australia to Asia, as opposed to the reverse. More plausible is that the virus spread from Australia to Asia as a result of abalone movements.
4.3. abalone virus endemic to Australia
This would appear to be the most likely source of the virus, for several reasons. As discussed, importation is possible but improbable and spread of virus to a new host species is, by definition, less likely than a virus infecting its primary host. The strongest evidence for the virus being native comes from the patterns of mortality shown in farmed and wild abalone, together with the transmission studies conducted so far. These studies indicate that the virus is highly infective and capable of causing significant mortalities. When introduced to farmed populations, the resulting epidemic was characterised by rapid spread, ultimately affecting most of the population. In contrast, although the disease has been found in a large area inhabited by wild abalone, mortalities to date have been relatively modest. There are several possible reasons for this, but one is resistance. Resistance of the wild population to infection would be expected if the virus is endemic.
5. Introduction of the virus
The rapid spread and resulting high mortality rate of the disease on the abalone farms suggest that this population was naive, in other words, had never previously been exposed to the disease and possessed no innate resistance. It follows that the herpes like virus was introduced from an external source. This aspect of the epidemiology was investigated by Paul HardySmith and a detailed description is given in his report. He concludes that the best fit scenario is infection of the first farm, farm A, through wild abalone intended for broodstock. Subsequent spread of the disease from farm A to other affected sites was primarily as a result of stock movement. He presents a logical argument for this scenario, taking into account evidence both for and against each assumption.
From examination of the data, including mortality data supplied by farm A, there are three interesting features about the outbreak on farms.
- the hatcheries were not greatly affected. No abnormal mortalities of free swimming or recently settled larvae were seen. Although one report was given of mortalities in older settled larvae, this occurred towards the end of the outbreak and ganglioneuritis was not confirmed in these animals. In contrast, herpes virus of oysters affects primarily early life stages.
- the spread of the disease on farm A is difficult to explain in terms of direct transmission. Tanks became affected in a seemingly random pattern, not related to their proximity to other infected tanks, or to any obvious aspect of their husbandry or inhabitants, such as harvesting, age or density. This is unexpected in light of information resulting from laboratory trials on transmission.
- at farm B, a period of almost two months elapsed between initial mortalities and the development of an epidemic due to ganglioneuritis. If it is assumed that initial mortalities were due to the same disease, the delay in spread is difficult to explain. Reintroduction to tanks, from a source on or close to the farm or from contaminated harvest boxes, is considered likely.
Work presented by Mark Crane demonstrated that infected abalone transmit the disease through the water column and this was also shown on the farms. At farm A, an experiment was conducted where apparently naive abalone were kept in isolation, with a control group placed in a tank where mortalities were ongoing. The control group died. The abalone which were kept in isolation remained healthy, although exposed to the same incoming water as the rest of the farm. The experiment was designed to determine whether the farm was being infected through incoming water with the conclusion that it was not.
6. Spread of disease on farms
Mortality data was provided by farm A, where all mortalities were recorded by tank for the duration of the outbreak. Analysis of this data is complicated by harvesting, but a few interesting patterns emerge from a preliminary examination. Abalone in the one year age class suffered significantly greater mortalities than older animals. However, they are stocked in greater numbers per tank than older animals. On this farm, the younger animals were also predominantly hybrids, whereas the older abalone were mostly greenlip. When comparing mortality curves for younger and older abalone, there are some indications that mortalities in the younger abalone progress more rapidly. This may be largely a function of population size, as disease transmission is facilitated by high host densities. In older animals, mortalities sometimes showed an initial spike, but were generally characterised by persistent, relatively constant deaths eventually resulting in large cumulative losses. In contrast, laboratory trials resulted in complete mortality in treatment groups, usually within less than a week of exposure.
farm A also constructed a map of the farm, showing spread of the disease over time. As mentioned above, no clear pattern emerges, even when considering activities such as harvesting. This is unfortunate, as a better understanding of factors affecting transmission would contribute to development of management practices for use in future outbreaks.
7. Disease in the wild population
At the time of writing, the impact of ganglioneuritis on the wild abalone population is much debated and reports on the extent of mortalities are conflicting. However, some objective information is available in the form of survey reports. Surveys done by Australian Marine Ecology, for the specific purpose of finding diseased abalone and fresh shells, yielded very few moribund animals and modest numbers of shells. Peter Appleford presented a summary of surveillance results at the scientific forum in September, showing apparent spread of the disease from the immediate vicinity of farm B, which was the first site where wild abalone were examined. Subsequently, the disease seems to have left this area but continues to extend its known range. At the same meeting, Harry Gorfine presented new data from an area approximately 10kms to the west of Port Fairy, where the first significant mortalities in the wild population were then occurring.
The behaviour of infectious disease in wild animal populations is difficult to interpret, partly because the problem is subject to the complexities of ecology. Usually, there is also an absence of good data on wild populations. It is telling that, in the current outbreak, the disease was first found in wild abalone outside an abalone farm, because this was the first site which was examined. Surveillance of the wild population was a result of the mortalities suffered on the farms, not a result of unexplained disease in the ocean. Although the question is contentious, events appear to show that the virus may have been present in wild abalone, even causing some deaths, without drawing attention, for an unknown period prior to this outbreak. Those involved with the abalone fishery agree that the population has fluctuated over the years, although not about the causes or extent of this fluctuation. It is conceivable that infectious disease was a contributing factor.
Epidemiological theory suggests some explanations for the outbreak pattern which has been observed in the wild abalone population to date. Two aspects merit discussion:
- population density
- disease resistance.
It should be noted that, whereas the outbreaks on the farms have effectively been terminated and will not be generating new raw data, the situation in the wild is still developing.
7.1. the spread of infectious disease is dependent, amongst other things, on adequate contact between infectious and susceptible individuals to ensure transmission of infection. Contact between infectious and susceptible individuals is obviously facilitated at high population densities. In general, the greater the opportunity for contact, the more rapidly an outbreak can progress, all else being equal. When looking at mortality data from farm A, animals at higher densities did appear to have more acute mortalities than those at lower densities, but unfortunately the animals are not directly comparable in all ways. In the wild population, densities would differ both between areas and within them and may be very high where abalone aggregate. It can be expected that aggregated populations would show mortalities comparable to those on the farms, if adequate contact is considered as the driver of infection.
7.2. disease will only spread if susceptible individuals are exposed. Resistant ones will clearly not be at risk. Therefore, the degree of resistance in a population will substantially affect the development of an epidemic. In a large naive population, even very low probabilities of adequate contact will eventually result in a significant number of infections, unless infected individuals lose their infectivity before contact occurs. Transmission trials suggest that the herpes like virus responsible for this outbreak is highly infective, as well as deadly. It is known that the wild population in some areas has been exposed for several months, but mortalities at most sites have been modest. An example is the area referred to in discussions near Port Fairy. A single abalone affected with ganglioneuritis was found on 27 July, but significant mortalities have not been seen, in spite of high population densities described by commercial divers. This would suggest that the population may be resistant to infection.
In an endemic situation, it is expected that the disease agent will circulate in the general population, but that subsets of this population will vary in their resistance. The presence of the disease will tend to exert some selection pressure for resistance in the population and resistance is also affected by other factors, not least of which is the general health of the animals. Resistance is not absolute and even good resistance will fail when sufficiently challenged. In addition, the possibility that this herpes like virus may be capable of latency implies that the infection may be present in the population without any signs.
During the current outbreak, there has been an assumption that the wild population in the vicinity of farm B became infected from contact with farm effluent. It has also been assumed that the infection spread from this locality to affect other localities along the coastline, including an area approximately 10kms to the west of Port Fairy. Although there is certainly evidence to support these assumptions, this evidence was mostly generated in response to the outbreak on the farms. It is questionable whether most of the affected populations would otherwise have been identified. Furthermore, it can be asked whether they have been affected in the past without attracting notice.
8. Disease investigation and response
Although the first mortalities in farmed abalone are thought to have occurred in early December 2005, samples were not submitted for diagnosis until much later. In spite of this, ganglioneuritis was identified and the probable cause found to be a herpes like virus during January 2006. The first epidemiological investigation was completed and a report generated in that month. Experimental work on transmission complemented the epidemiological investigation. The speed with which this disease, previously unknown in Australia and elsewhere poorly described, was diagnosed is impressive. Immediate focus on the epidemiological aspects is commendable, as this is most likely to yield reliable data. The second, more extensive, epidemiological report by Paul HardySmith of Panaquatic Health Solutions became available in August 2006. It provides a logical and complete evaluation of the outbreak up to that point. All important aspects are covered. Conclusions are derived from facts, with any limitations in the data clearly stated.
The official response to the disease outbreak appears to be based on the assumption that the responsible virus is endemic to Australia. This assumption is reasonable and has enjoyed support from consulting scientists outside the Department of Primary Industries. The virus may very well have occurred in Victoria prior to the current outbreak. However, the exact distribution and host range have not been determined. Eliminating the disease in the wild would require removal of all potentially infected abalone in addition to all other possible hosts, something which is clearly not feasible. The only practical measure is to attempt containment by limiting potential transfer of infective material to unaffected areas. Unfortunately, such measures can be undone by realities of currents and animal movements, as may have happened in this outbreak.
Eliminating a disease from a farm can be achieved by depopulation and disinfection. This was undertaken on a voluntary basis by farms affected by ganglioneuritis. It has been debated whether farms should have been encouraged to close when the diagnosis was first made. In the case of farm B, the presence of a herpes like virus was confirmed during January 2006. Abalone were dying on the farm at this time. Considering what was known about the disease then, an argument can be made that the farm should have depopulated immediately. Abalone farms in general are not designed or managed in a way that permits the control of a highly infectious disease. It is not probable that farm A would have been able to stop the disease from affecting the entire farm and subsequent events demonstrated this. Immediate steps to harvest as many abalone as possible and destroy the remainder would have been the option of least risk.
However, farm A has not been directly implicated in spread of the herpes like virus to wild abalone populations. This claim has been made for farm B. When the presence of the virus was confirmed at farm A, no mortalities were occurring at farm B and there was no conclusive evidence that the farm had ever been infected, as earlier mortalities were not subject to laboratory examination. Therefore, there were no grounds on which to depopulate the farm. The disease resurfaced in the middle of March. Emergency harvest and destocking would have been appropriate then, but it cannot be said whether this would have prevented the development of disease in the wild population outside the farm. It is interesting that wild abalone outside farm A and at the affected cage culture sites have so far not shown signs of disease. Surveillance reports indicate that the population outside farm B has recovered.
Mention was made of mortalities in wild abalone in the Port MacDonald region of South Australia occurring in January 2006. It is telling that, although the existence of mortalities has apparently been confirmed, little has been done to establish the cause. In addition, there does not appear to be any connection to abalone farming in this case. This illustrates that wild abalone do die of disease without human intervention. In addition, human intervention is not always successful in preventing the spread of disease.
9. Looking forward
Development of a reliable diagnostic test which can detect the herpes like virus in the absence of disease will contribute to better understanding of the natural range of the virus, both geographically and in terms of hosts. Unfortunately, unlike in most farmed animals where serology can show where a disease has been, even when it is no longer present, surveillance in abalone will only identify current distribution. For this reason, collection of appropriate samples should be a priority and not wait on a test to become available. It should also be understood that, even if the disease is endemic in the population as a whole, variations in prevalence will be present on a smaller scale. One of the key questions in the current outbreak is whether the herpes like virus was present in Victoria prior to introduction to the farms. This question is unlikely to be answerable, unless by chance archived samples are found.
If the assumption is made that the herpes like virus is endemic to abalone in Australia, it is to be expected that mortalities may occur from time to time in the wild population. In a stable situation, these mortalities are unlikely to impact significantly on the population. However, circumstances may alter in ways that lead to increased disease outbreaks, for example due to climatic change. There has been speculation about the contribution of aquaculture to disease in wild populations of other aquatic animals, such as prawns and Atlantic salmon. Concentration of aquatic animals for culture purposes may create a point source for spread of diseases to wild populations. When diseases are endemic, wild populations are known to act as reservoirs for infection of farmed animals. These issues must be addressed for the future sustainability of both aquaculture and fisheries to be ensured.
A lack of basic biosecurity practices in the abalone industry has been highlighted during the past few months. In turn, this has led to increased awareness of the need for such practices and interest in adopting them. The Department of Primary Industries is using this opportunity to develop and implement guidelines on good biosecurity practices in the abalone industry. Practically, these guidelines cannot be enforced and the continued commitment of all in the abalone industry will be required for success. Participation of processing and relay facilities is as important as cooperation from farms and commercial divers.
In conclusion, the current outbreak in Victoria is essentially no different to countless disease outbreaks which have occurred in farmed animals before. Such outbreaks are inevitable in intensive animal husbandry systems. Simultaneous mortalities in the wild abalone population are all the more unfortunate due to the value of the fishery. The investigation of the outbreak has been rapid and thorough. Response by the Department of Primary Industries was conservative, which may be considered appropriate given the lack of information on which to base decisions and the likely endemic nature of the disease. Future management will no doubt be facilitated by the development of better diagnostic tests for the herpes like virus.
10. Sources of information used in compiling this report
- HardySmith, P. 2006. Report on the events surrounding unusually high mortalities of farmed abalone in Victoria. 26 January 2006. Panaquatic Health Solutions.
- HardySmith, P. 2006. Report on the events surrounding the disease outbreak affecting farmed and wild abalone in Victoria. 29 August 2006. Panaquatic Health Solutions.
- Stewart, K. and Edmunds, M. 2006. Health monitoring of abalone wild stock. July 2006 and August 2006. Australian Marine Ecology.
- Veterinary Diagnostic Services. 2006. Reports on diagnostic examinations of abalone submitted during the outbreak. Primary Industries Research, Attwood, Victoria.
- Additional maps and photographs provided by Western Abalone Divers Association.
- Site visits and interviews
- Licensed abalone fish receiver
- farm B
- farm A
- Meetings and workshops
- Chief Veterinary Officer's unit
- Pathologists from Veterinary Diagnostic Services at Attwood. Also present were Carolyn Friedman, Judith Handlinger, Celia Hooper, Mike Hine and Tristan Renault.
- Department of Primary Industries. Representatives from veterinary services and fisheries.
- Western Abalone Divers Association workshop. Included scientific consultants.
- Scientific forum on the abalone herpes like virus. Attended by government and industry representatives, including from other states, and various scientific consultants, including most of those listed under 3.7.2.
- Other interviews and discussions
- Mehdi Doroudi. Background.
- Peter Lawson. Information on affected cage culture operations.
- Harry Gorfine. Information of wild abalone population.
- Paul HardySmith. Epidemiology of the outbreak.
- Carolyn Friedman. Background on Great Southern Waters and processing plant in Portland.
- Farm data, including mortality figures, provided by farm A.