Strategy for Salmonella Prevention on Egg Laying Farms

In the genus Salmonella (Family Enterobacteriaceae) there are over 2,500 serovars. In this paper we are not dealing with the two avian host specific salmonella nonmotile serovars, S. Pullorum (Pullorum disease) and S. Gallinarum (Fowl typhoid) which are both considered to be exotic to Australian commercial poultry. Similarly not the human specific salmonella called Salmonella Typhi the cause of typhoid disease in man.

The group of salmonella we are dealing with are the motile and non-host adapted serotypes referred to as the paratyphoid salmonellae.

This group of salmonella is found widely distributed in both wild and domestic animals and cold blooded species. Salmonella is so ubiquitous that it is hard not to find it in both animate and contaminated inanimate sources. In poultry these salmonella do not generally cause disease and thus egg layers are commonly asymptomatic carriers of salmonella. Under stressor conditions some salmonella, including Salmonella Typhimurium type 9 and Salmonella Hessarek, can cause significant clinical disease and mortalities particularly in day olds and young layers.  These salmonella generally colonise the intestinal tract with various degrees of persistency and are shed in the faeces thus providing a potential mechanism of contamination of eggs and in some cases with salmonella strains that cause food borne disease in humans. One salmonella of this group that has a slightly different pathogeneses and epidemiology is Salmonella Enteritidis (SE) which can actively colonise the ovary of the layer and thus be deposited into a clean egg free of faecal contamination. Australia, with New Zealand, is very fortunate in that the commercial poultry industries are SE free. In the EU it is one of the predominant serotypes and thus control programs are commonly based around SE.  This should be kept in mind when reading international literature on salmonella control. The colonisation of clean eggs means that the refrigeration of all eggs is a critical control point and vaccination against SE tends to be more efficacious because of the systemic characteristics of its colonisation.

Salmonella as microorganisms are relatively non-particular about their growth media requirements which are simple and while they grow best at 37°C they can grow at temperatures from 5 to 45°C and within a pH range of 4 to 9. They can survive in the environment for years and colonise invertebrates which can act as passive carriers. So overall they are very well adapted microorganisms and thus ubiquitous and hard to control infection in domestic livestock. Salmonella can be destroyed by temperatures over 60°C if heated for in excess of one hour and at temperatures above 90°, such as in pelleting feed, in less than 1 minute.  Thus the normal cooking process will kill salmonella and reduce the food safety risk, which is fortunate for the chicken meat industry, but the egg industry still has the consumer “problem” of the consumption in some food preparation of raw eggs or lightly cooked eggs.

This creates a dilemma regarding where the food safety responsibility rests.

Salmonella after they have been identified by various biochemical tests and cultural characteristics are then further classified by a serological grouping, using the Kauffman and White classification system. This results in the serovars such as S. Typhimurium, S. Agona, S. Dublin, S. Infantis, etc. The next level of typing is using viruses called bacteriophages that are viruses that infect specific strains of bacteria. By using this technology salmonella serovars can be further more finely classified according to their phage types such as S. Typhimurium type 9 (ST9), or ST 44, ST 135, etc. Such typing allows more precise epidemiological studies and tracing.

More recently this has been refined even further to a molecular typing called MLVA (multilocus variable number tandem repeat analysis) and is now more regularly used by government departments for disease outbreak investigations and has the technical rigor to be used for prosecutions where warranted by health departments. 

As previously mentioned Australian is free in commercial poultry of SE phage types that are commonly cause food poisoning. However SE clinical disease in people does occur, with the greater majority of these associated with overseas travel, particularly to Asia. This identifies a serious risk for Australians egg produces who are often worried about avian influenza and overseas visitors but fail to realise the likelihood of recent overseas travel by staff being a higher risk for the inadvertent introduction of SE onto a local flock.

The salmonella that are causally associated with table eggs related food poisoning outbreaks in Australia include a number of ST, including, but not exclusively, phage types 9, 44, 135a and 135, 170 (108). The other less common salmonella serovars associated with food poisonings outbreaks are S. Infantis. S. Virchow 34, S. Singapore and S. Saintpaul.

In recent times the ST 9, 44 and 170 have been the predominant serovars in significant food poisoning outbreaks. All these salmonella have been and continue to be recovered from environmental sampling of poultry sheds. The status of poultry layer genetic stock, which is under regular monitoring programs and high standard preventative programs, is historically free of those salmonellae associated with food safety issues. The serovars that have been historically recovered from day old layer stock included S. Infantis, S. Hessarek. S. Agona and a number which are untypable. Generally either in house or commercially reared pullets are salmonella free at point of lay. This leads to the conclusion that most of the change in status occurs during the production period. While traditional dogma states that the young chick is highly susceptible to salmonella colonisation and the mature layer is relatively resistant the epidemiological field picture identifies most layer flocks become positive in lay. 

Inherently very difficult when one considers the host adaptation capabilities of salmonella and its persistence in the environment. Essentially every horizontal contact that commercial poultry come in contact with could have salmonella including the environment, arthropods, water, feed, vermin, people and fomites of many types. It can carry over from previous flocks and even after aggressive cleanouts can persist deep in the pore on concrete or in the beetles, flies and vermin re-entering the shed. Thus control becomes centred on the value of the livestock and the impact that a change in salmonella status would have, the type of salmonella involved in regards to its importance as a food safety pathogen and the likely way the final product will be handled or consumed.

Importantly salmonella can be controlled and even eradicated but the commercial practicality and viability is compared and offset by the risk assessment to the business as a provider of livestock or a provider of a product to the consumer.

The Australian poultry industry has a relatively high level of positive status at the farm level and essentially no regulatory control over this other than the responsibility to provide a food safe product. This is very different to the EU, and particularly the Northern European countries, where there is tight legislative control on the poultry industry from the production of feed free of salmonella to the compulsory slaughter out of Salmonella Typhimurium (ST) positive flocks where eggs are only to be used a fresh table eggs. In feed additive treatments and flock vaccination are a normal part of the regulated control programs. This outcome has been driven by significant food safety events and then followed by sectors of the industry wanting a product distinction for the consumer. Australia as indicated has no formal or even voluntary code regarding salmonella control and depends on industry best practice management to achieve a food safe product. The dilemma facing the Australian industry is while the majority of the industry has no significant food safety events related to table eggs there are ongoing significant egg related food poisoning outbreaks associated with some producers whose practices are not best practice. In some cases this is clouded by poor food preparation practices and / or the production of high risk foods such as raw egg based mayonnaise, ice cream, mousse, egg butter and salad dressings.

Invariably if the Australian egg industry is to avoid having costly compliance imposed on it through regulation as a result of ongoing egg related food poisoning outbreaks it will need to become more proactive and united in facilitating salmonella control programs and improved egg handling practices throughout the industry.

It is to be noted that there is no one effective control tool for salmonella but it is the accumulation of various strategies that result in a quantitative reduction of the salmonella overall. Only under the strictest procedures such as at a primary breeding unit can a salmonella free status be reliably achieved and maintained. 

The necessary tools and laboratories for monitoring salmonella in the egg industry are readily available and most commonly are based on identification by microbiological culture.

Legislatively in Australia the identification of salmonella from an environmental swab does not require reporting (unless part of the SE Accreditation Program) but recovery from a bird does as by definition this is from a case of salmonellosis. There are a number of private, university and government laboratories that can undertake salmonella culture work, not all are NATA accredited. For finalisation of serovars and typing there are two reference laboratories in Australia. Cost recovery from these reference laboratories is applied but varies depending on the state the samples came from and the nature of the investigation.

In a monitoring program sampling can involve drag swabs of deep litter or slats, cracked and dirty eggs, egg surface washes, manure belts, egg belts and nest boxes and other working areas such as the packing floor, wash water and cool rooms.

In regard to sampling there are some fundamentals that need to be understood:

It is advisable that producers establish a formal monitoring program that outlines the testing requirements and time frames. Flock identity is critically important as is the correct recording for traceability reasons. In conjunction with this, a series of SOP’s should be created covering each of the testing / monitoring methodologies. 

Producers buying day olds should as part of their monitoring program collect chick papers from a representative sample of boxes and send these off for culture. Where chick papers are not available samples of 3 day old brooding papers can be used instead. This procedure provides the producer with confidence of his day old status and also assists the hatchery with its own monoitoring program. Salmonella monitoring is at a point of time and thus there is no absolute certainty of the status of a flock between testing.

Pullets purchased as point of lay birds or reared in house should be drag swabbed (or manure belt tested if cage reared) 2 weeks before transfer. The buying of pullets should be conditional on the provision of the batch summary sheet, serology confirming the efficacy of vaccination and the salmonella status from drag swabs. With the salmonella status known prior to transfer this will allow any strategic decision such as an additive program or change of destination to be made. This decision making may be made prior to the serovars of the salmonella being known.

For commercial layers in both cage and alternate systems it is recommended that salmonella testing by drag swabs is conducted 3 monthly. For those joining the SE Accreditation program it will be necessary to test monthly for 3 consecutive months and on the absence of any detection of SE can move to 3 monthly testing. If for any reason birds are being transferred such as after a moult they should be tested prior to the move.

Where salmonella that are of potential food safety significance are detected in a production flock then it is recommended that the egg pulp from the affected shed or a sample of caked and dirty egg pulp is also tested. This provides a practical way of gaining some idea of the shedding rate of the flock and thus an improved ability to evaluate the risk assessment.

Other areas such as egg belts, nest boxes, packing table and other contact areas should be tested from time to time to ensure the hygiene program is adequate and there are no contaminating points. 3M swabs are useful for this procedure.

For those receiving commercial feed most mills will from time to time under their own quality programs test pelleted feed for salmonella. They will make those results available for the particular producer. The testing of mash feed will on occasions have expected positive results, particularly for those salmonella found in vegetable protein meals. These are not usually related to food safety pathogens.

For in-house feed mills similar testing of finished feed should apply but also include environmental swabs of elevators and heavy contact areas.  The technique of putting a very small hole in the main delivery augers and catching a continuum of finished fed throughout the day is a sensitive sampling procedure. In contrast to the perception in the poultry industry most Australian meat meals from accredited suppliers are negative for salmonella because of the export market requirement compliance requirements. 

The most common feed additives used for salmonella control are organic acids or in particular short chain fatty acids (SCFA). These are considered to have an impact on salmonella at two points, in the feed and within the bird. Firstly through lowering the pH of the feed they inhibit the growth and even kill any potential salmonella. If the salmonella take up these products they also can kill the salmonella by chemically dissociating within the organism. Organic acids are dose responsive and generally in Australia they are used at lower levels than optimal, this being in part for cost consideration. In layer feed the dose becomes more important because of the high level of potentially neutralising limestone.

Within the birds organic acids are believed to work by the salmonella taking up the associated form of the molecule which then dissociates inside the bacteria releasing hydrogen ions and other molecules that result in the death of the salmonellae. The need for the organic acid to be associated is important for it to be effective and this depends on the particular SCFA pKa (that is the pH above which the molecule dissociates).  As the intestinal pH is above this range some companies attempt to overcome this by producing protected organic acids. While there is some debate about the effectiveness of organic acids in the intestine studies demonstrate that they do have an effect on the microflora balance in vivo. As there is always equilibrium in any chemical reaction there invariably will always be some level of associated form of these molecules present in the intestine.

Other additives used include the phosphorylated mannosaccharides which have the property of being able to agglutinate salmonella and passively remove them from the intestine. They also have other intestinal health attributes. There is also a general family of phytobiotic products (plant extracts) and these appear to work indirectly by improving intestinal health and integrity and enhancing the opportunity to clear the salmonella colonisation.

The above family of feed additives are referred to as prebiotics in that they enhance the health of the intestinal microbiota.

Probiotics are referred to as the healthy microbiota and their use is believed to improve the gut environment and make this environment incompatible for the colonisation of salmonella or lead to their displacement. Their use is particularly favoured in assisting young birds to establish microbiota that resist colonisation with salmonella and also in older birds where dietary changes or antibiotic use has compromised the microflora balance.

The collective name for prebiotics and probiotics where they are used in combination is synbiotics.

The use of antibiotics in day olds prophylactically is contraindicated in regards to salmonella as their use compromises the establishment of a balanced microflora and thus present a window of opportunity for salmonella to colonise. Where antibiotics are required therapeutically they should always be used currently or post treatment with synbiotics to maintain intestinal health. 

The ability of salmonella to persistently colonise the intestinal tract of poultry indicates that they have mechanisms where they can avoid or modulate the host’s immune response. By default this means that controlling salmonella by vaccination will have limitations. Thus use of live vaccines that colonise before exposure to the wild type provide some degree of protection and the use of killed vaccines can control the degree of replication of salmonella, especially systemically, and thus aid in reducing shedding.

The other difficulty of making effective universal salmonella vaccine is the number of serovars that need to be potentially covered. For this reason the major commercially available vaccines are either aimed at SE or ST, the serovars that are the significant food causing pathogens. Efficacy with SE based vaccines tends to be better than those with ST because of the partly systemic characteristics of the organism in the host.

In Australia we have currently one registered live salmonella vaccine, Vaxsafe ST which is an attenuated vaccine based on ST 44. This vaccine is registered for administration by drinking water or coarse aerosol spray. Current work is investigating the efficacy of this vaccine after the initial live priming, when mixed and administered in conjunction with other routinely used inactivated poultry vaccines.

The other alternative salmonella vaccines available in Australia are autogenous inactivated adjuvanted vaccines. These vaccines are made from salmonella recovered from the farm and can only be used in poultry that have relatedness to the farm in question under permit conditions following the request of a veterinarian.

The combination of multiple live priming followed by an autogenous killed vaccine is currently the best available option in Australia. 

The ultimate control for salmonella is ensuring that the organism is excluded from the environment in which the poultry are reared and produced.

This can be achieved but is difficult and becoming even more difficult with the increasing use of alternative systems. The simple policy of putting day olds in cleaned and salmonella free sheds, commencing feeding with heat treated crumble and then placing the birds as a single age in a cleaned and disinfected shed for the production cycle provides a good start in salmonella control. Unfortunately the reality is less straightforward than this best practice approach with the extreme being the reuse of litter to rear the birds on and the placement of birds in a multi-age shed that has never been totally depopulated and thus never effectively cleaned. In such sheds after intensive colony selection up to 8 or more different serovars of salmonella can be identified.

Cleaning of sheds after depopulation requires the use of a quality detergent to remove the organic loads prior to disinfection. Most disinfectants are effective against salmonella but their efficacy varies with the organic load and biofilm present. Dirt floors provide another challenge and detergent washing and the use of disinfectants on such a surface has limited efficacy. One effective remedy and it technically exceeds all others in trial work on earth floors is hydrated lime.

The control of vermin such as rats and mice is important as is the control litter beetles and flies. Domestic pets can also be asymptomatic carriers and dogs have been validated as the contaminating source. The use of ruminants on free range farm should always involve checking these animals for salmonella before entry. 

The Australian egg industry currently does not approach salmonella control from a holistic perspective. The approach is more ad hoc in form being influenced by the culture of whose actual responsibility is it, demonstrating some due diligence, recognising the costs involved, retailer requirements and in some cases pressure from technical services.

Where a holistic approach does occur is when a producer is associated through traceability with salmonella food posing outbreaks and thus comes under the jurisdiction or control of the state health department or food authority. It is unfortunate that the industry is not the lead authority showing initiative in this area.

Where such salmonella control programs are enforced and implemented with regulators auditing a reduction in salmonella is evident and very satisfactory outcomes are achieved. Thus it can be done and the implementation of such programs should become routinely incorporated in the Australian egg layer industry as standard preventative programs. 

The technical knowhow, the laboratory testing, the feed additives and the vaccines are all available in Australia to achieve a successful outcome for a salmonella control program.

The impediments that exist to do this are in part historical, cultural and cost sensitivities.

It is important though for the industry to take charge of its own salmonella control program and not allow ongoing food poisoning outbreaks to enforce regulatory control outside the industry. The egg industry through its governing bodies should facilitate the implementation of industry minimum standards for control of salmonella. This should be undertaken in conjunction with the food authorities in regard to the education of consumers and food handlers in safe food procedures.