A Critical Appraisal of Global Testing Protocols for Zoonotic Parasites in Imported Seafood Applied to Seafood Safety in Australia.

It is not suggested that any country is intentionally exporting seafood which does not comply with Codex seafood-safety guidelines/codes/standards. However, with an open access resource such as fisheries, there is vast potential for errors to occur along convoluted supply chains, spanning multiple countries, which may negatively impact the safety of edible seafood products imported into Australia. Australian importation policy and inspection procedures are founded upon a bedrock of trust in the integrity, reliability and safety of the global seafood supply chain. In order for seafood imported to Australia to be considered safe the non-mandatory international health standards, governed by Codex Alimentarius, for seafood must be predicated upon the most efficacious methods and stringently governed by each exporting provenance. Currently, tests for zoonotic parasites are not applied to imported edible seafood products on arrival into Australia. Therefore, this critical analysis is aimed at discussing the effectiveness of current testing protocols for zoonotic parasites in edible seafood advised by Codex Alimentarius which may impact the safety of the product imported into Australia.


Introduction
Australia is one of the few countries globally which seems to have remained free of many zoonotic parasites and pathogens which are endemic to other countries [1]. Australian biosecurity is considered critical in the fight to reduce the risks posed by invasive pests and diseases [2] and as such is an essential national asset [3]. The introduction of white spot disease (WSD) into Australia is a propitious reminder of the severe economic and social consequences of a biosecurity breach. By 2016, the outbreak estimated cost to the Australian prawn industry was $49.5 million [4]. Despite fore-warnings from the global scientific community, of traded WSD infected frozen shrimp and an obvious spatial and temporal global spread [5][6][7], WSD was introduced into Australia. Whilst the introduction of WSD into Australia exposed a vulnerability in the nation's biosecurity it also illustrated how difficult biosecurity defense becomes when exporters are prepared to flaunt international food safety recommendations. Australia has a great reliance on imported edible seafood products. No matter how highly regarded and efficient Australian biosecurity policy/procedures are, in a time or rapidly escalating global change, it is perhaps timely to re-evaluate the current international standards for zoonotic parasites in imported edible seafood in support of human health biosecurity in Australia.
Seafood is considered an important source of essential fatty acids, protein and B group vitamins [8] and is a valuable component of a healthy diet [9,10]. In Australia, the imported seafood product

Codex Alimentarius Non-Mandatory Recommendations
Please see Figure 2 for how the non-mandatory Codex international food safety guidelines/codes of practice/standards are developed.

Candling
Codex Alimentarius 'Code of Practice for Fish and Fishery Products' defines candling for parasites as "passing fillets of fish over a translucent table illuminated from below to detect parasites and other defects" [19]. Section 8.1.6 of the code provides technical advice for candling and recognises "viable parasites" in fish as a potential biological hazard (p. 103). The candling line is recommended to be "continuous and sequential to permit uniform flow without stoppages or slowdowns and removal of waste" (Step 1, p. 103). Codex does not recommend candling in conjunction with pressing in any of the codes or standards which contain a reference to parasites. 'Pressing' involves placing the sample of fish between two thin acrylic sheets and examining under an appropriate light source [20]. Table A1 in Appendix A shows Codex recommendations for candling to control parasites for specific seafood products [19] 'Code of Practice for Fish and Fishery Products' still apply. Codex Alimentarius is the global food safety authority and therefore the advised methods for parasite detection should unquestionably be the most effective available. Codex food safety standards are not mandatory or enforced however it is expected that member nations, such as Australia, of the World Trade Organisation will be responsible in implementing the advised international food safety standards. Australia has a great reliance on the testing procedures of the exporting country to ensure imported seafood products are free of parasites and safe for human consumption. Original figure developed from information at [18].

Candling
Codex Alimentarius 'Code of Practice for Fish and Fishery Products' defines candling for parasites as "passing fillets of fish over a translucent table illuminated from below to detect parasites and other defects" [19]. Section 8.1.6 of the code provides technical advice for candling and recognises "viable parasites" in fish as a potential biological hazard (p. 103). The candling line is recommended to be "continuous and sequential to permit uniform flow without stoppages or slowdowns and removal of waste" (Step 1, p. 103). Codex does not recommend candling in conjunction with pressing in any of the codes or standards which contain a reference to parasites. 'Pressing' involves placing the sample of fish between two thin acrylic sheets and examining under an appropriate light source [20]. Table A1 in Appendix A shows Codex recommendations for candling to control parasites for specific seafood products [19] 'Code of Practice for Fish and Fishery Products' still apply. Codex Alimentarius is the global food safety authority and therefore the advised methods for parasite detection should unquestionably be the most effective available.

Operator Constraints and Candling Accuracy
According to Andreoletti, et al. [21] an experienced 'Candler' can examine up to 300 fillets an hour. In order for an operator to examine 300 fillets per hour, every 12 s a fillet must be checked and parasites removed. If CODEX STAN 165-1989; CODEX STAN 190-1995 and CODEX STAN 311-2013 (See Table A1, Appendix A) [22][23][24] apply then operators along the candling line must additionally decide if there are more than 2 parasites per kilo, if the parasite capsule is >3 mm and if the unencapsulated parasite is >10 mm. A questionnaire distributed to fish processors in Scotland highlighted that few carried out any comprehensive examination of fish for larval nematodes. Only one processor used a candling table and commented that the candling method was limited in thick Codex food safety standards are not mandatory or enforced however it is expected that member nations, such as Australia, of the World Trade Organisation will be responsible in implementing the advised international food safety standards. Australia has a great reliance on the testing procedures of the exporting country to ensure imported seafood products are free of parasites and safe for human consumption. Original figure developed from information at [18].

Operator Constraints and Candling Accuracy
According to Andreoletti, et al. [21] an experienced 'Candler' can examine up to 300 fillets an hour. In order for an operator to examine 300 fillets per hour, every 12 s a fillet must be checked and parasites removed. If CODEX STAN 165-1989; CODEX STAN 190-1995 and CODEX STAN 311-2013 (See Table A1, Appendix A) [22][23][24] apply then operators along the candling line must additionally decide if there are more than 2 parasites per kilo, if the parasite capsule is >3 mm and if the un-encapsulated parasite is >10 mm. A questionnaire distributed to fish processors in Scotland highlighted that few carried out any comprehensive examination of fish for larval nematodes. Only one processor used a candling table and commented that the candling method was limited in thick skin or fleshed fish and another said it was not cost effective [25]. Visual fatigue after prolonged periods of observation has been demonstrated to affect diagnostic accuracy [26]. Wootten and Cann [27] comment that operator eye fatigue is rapid and during extended periods of candling the efficiency of the method may be impaired. It cannot be discounted that operator fatigue may limit the efficacy of parasite identification along the candling line. Candling has been demonstrated to be 15% less efficacious under commercial working conditions [28] which may support observer fatigue as significant within parasite identification and removal.

Nematodes
It has been noted that the efficiency of candling as a technique has limitations due to low penetration into fish muscle of the white light used [28]. As a method is also considered ineffective in Foods 2020, 9, 448 5 of 23 bright light [21]. McGladdery [29] considers the technique effective for detection of Pseudoterranova spp., which are darker, but limited in detecting smaller, white worms, such as Anisakis spp. However, only 31.7% (143/450) Pseudoterranova larvae were identified in monkfish fillets using white light candling [25] which is in contradiction of McGladdery [29]. Candling combined with pressing has been demonstrated to be more efficacious to detect parasites in fish than candling alone [20]. There is a great variability between each inspection method in terms of 'hours of labour' which are required. As a result, white light candling may be based upon convenience rather than safety best practice. According to Codex [19] candling carried out by skilled personal in a suitable location is effective in the control of parasites when implicated species of fish are used (step 3, p. 104). However, in Annex 1 "potential hazards associated with fresh fish, shellfish and other aquatic invertebrates", Section 1.1 it is considered that "candling, trimming belly flaps and physically removing the parasite cysts will also reduce the hazards but may not eliminate them" [19]. The effectivity of candling in the same Codex code of practice [19] has been described as both effective and ineffective in controlling seafood-borne parasites. Inconsistencies also appear in recommendations CAC/RCP 1-1969: "No raw material or ingredient should be accepted by an establishment if it is known to contain parasites", [30] and CAC/RCP 52-2003 [19] "Unless they can be reduced to an acceptable level by normal sorting and/or processing, no fish, shellfish or other aquatic invertebrates should be accepted if they are known to contain parasites". If candling does not completely eliminate the parasite hazard and seafood should not be accepted if it contains parasites then reducing the parasites to an acceptable level appears contradictory. Levsen et al. [31], in a study of fish from the Northeast Atlantic Norwegian spring spawning (NSS) herring (Clupea harengus), blue whiting (Micromesistius poutassou), and mackerel (Scomber scombrus) demonstrated only 7 to 10 percent of the nematode Anisakis larvae present in the fillets of all fish species were detected by candling. In NSS herring and blue whiting the detection efficiency of candling was decreased as fillet thickness increased. In blue whiting, the detection efficiency of candling with UV light was only 10-15% despite the average fillet thickness of 11 mm. Adams, et al. [32] in contrast had relatively high recovery of Anisakis larvae from four types of white fleshed fish; rockfish (Sebastes spp.), arrowtooth flounder (Atheresthas stomias), sole spp. (family Pleuronectidae) and true cod (Gadus macrocephalus); utilising the candling method identified 43% to 76% of the anisakids present. However as one viable L3 larvae can result in human infection the method does not completely eliminate the danger. In fillets of monkfish (Lophius piscatorius) and cod (Gadus morhua) candling was only successful in identifying 16.8% and 33.3% of Anisakis and 31.8% and 53.6% of Pseudoterranova respectively which were present [25]. A time saving method to identify parasites in fish is recommended which candles a representative sample of fillets from a batch [27], however, this assumes that parasitism is equal between fish of the same species from the same location. The number of A. simplex larvae in mackerel and blue whiting fillets was from 0-19 and 0-71 respectively in fish caught in the same location [31].

Trematode Metacercariae
The conventional method for detection of zoonotic trematode metacercariae in fish include microscopic examination of compressed flesh samples which according to Andreoletti et al. [21] is time consuming and lacks sensitivity. However, Murrell and Sohn [33] concluded that this method was economical, time effective and determined the exact location of the metacercariae. Andreoletti et al. [21] comment that individual fish harbour few metacercariae so it is difficult to estimate infection intensities. A total of 113 freshwater fish species, mostly cyprinids, have been recorded as hosts for metacercariae of zoonotic flukes [34]. Species of cyprinid fish commercially available in Laos were identified infected with zoonotic metacercariae (number of fish species infected: intensity range), Opisthorchis viverrini, 6: 1-6980; and Haplorchis yokogawai, 3: 1-1370 [35]. Commercially available fish from a Chinese market were identified infected with zoonotic metacercariae of Haplorchis taichui, 10: 1-485; Haplorchis pumilio, 10: 3-312; Centrocestus formosanus, 5: 1-32 and Metagonimus yokogawai, 11: 1-1836 [36]. The recovery rate of zoonotic metacercariae in tilapia and catfish fillets using the candling method according to Murrell and Sohn [33] is 53% and 68% respectively. Metacercaria in fish range in size; Opisthorchiidae Foods 2020, 9, 448 6 of 23 0.1-0.15 mm; Heteropyhidae and Echinostomatidae 0.14-0.16 mm [33] and in cases of intense infection it may be impossible to remove all infectious metacercariae and discarding the fish the only option. In recent times, and in accordance with Article 9 of the SPS [18], critical control intervention programs have been implemented in some Vietnamese aquaculture facilities. These programs have had some success in lowering the burden of infection metacercariae in cultured fish [33] and are a promising initiative. Although not recommended by Codex Alimentarius [37] the Vietnamese catfish industry, mainly driven by the implementation of western quality standards [38] has also taken the initiative to use the press method of candling for metacercariae.

Tapeworm Plerocercoids
Rozas et al. [39] comments that the press method in conjunction with candling provides more effective detection of plerocercoids in fish muscle than candling alone. Torres and Puga [40] compared three methods of candling to isolate plerocercoids (N = 310) in trout fillets. The candling method as advised by Codex had a 22% detection efficacy. A combined slice and candle had 40.8% and press method combined with candling had 59.2% efficacy. When all of the three candling methods were combined there was 90.9% detection efficacy however the total procedures included sectioning muscle tissue and examination of up to 18 compression plates. The incidence and mean intensity of infection of Diphyllobothrium dendriticum (syn. Dibothriocephalus dendriticus) plerocercoids in fish has been identified at 83.2% and 8.8% [41] and D. latum (syn. D. latus) the mean infection intensity has been identified as low as 1.25 parasites/fish [42].

Myxozoans
Species of seafood-borne zoonotic myxozoa are not included as a human health concern in any of Codex seafood safety guidelines. There is one mention of myxosporidia which may hinder the production of surimi due to myoliquefaction of fish muscle. The same source provides a recommendation of the best method to successfully bind infected fish muscle into surimi for human consumption [43]. Olive flounder from Japanese waters have been identified infected with three myxosporean species; Kudoa septempunctata, K. thyrsites, and K. shiomitsui [44]. Imported farmed olive flounder have been demonstrated infected with K. septempunctata [45]. Yellowfin, Bigeye and Bluefin tuna have been identified infected with the zoonotic K. neothunni [46,47] and K. hexapunctata has been identified in Bluefin [48] and Yellowfin tuna [46]. It should be noted that in samples of Northern Bluefin tuna (Thunnus thynnus), obtained from nine different countries, only the samples of Japanese origin were identified infected with K. hexapunctata. Species K. neothunni and K. septempunctata do not form a cyst or pseudocyst, [45,47] and even if inspected for parasites it is doubtful these parasite species would be detected macroscopically.

Ambiguity in Codex Food Safety Guidelines
CODEX STAN 165-1989; CODEX STAN 190-1995 and CODEX STAN 311-2013 [22][23][24] state there should not be "two or more parasites per kg of the sample unit detected by candling". The three standards apply to frozen and smoked products and as freezing is determined by Codex to eliminate the zoonotic potential of all parasites this inclusion seems irrelevant. There are no other Codex standards which clarify what an 'acceptable' number of parasites per kg may be despite "reduced to an acceptable level" being used in CAC/RCP 1-1969 [30] and CAC/RCP 52-2003 [19] in regard to seafood-borne parasites. In CAC/GL 88-2016 [49] there are six separate references to "acceptable" limits of parasites in fish without any clarification of what an acceptable number may be. CODEX STAN 311-2013 [24] states that viability of, and killing method for parasites may be determined using methods "acceptable to the competent authority having jurisdiction". The guidelines use of "acceptable" may allow a subjective interpretation of what is in essence an unmeasurable amount and an interpretative administration of seafood safety standards which may vary significantly between processors and regions. In Codex standard 244-2004, "Standard for Salted Atlantic Herring and Salted Sprat" [50] Annex III, Point 2, states irrespective of the presence of visible parasites which may be seen in the sample unit (Annex III, Point 1), 'the verification of the presence of parasites in intermediate entire fishery products in bulk intended for further processing could be carried out at a later stage" (p. 8). It is unclear when the later stage may be and 'later stage' seems to be an indefinite term particularly when applied to fish species demonstrated to have high intensity of infection with Anisakids. Baltic herring have a demonstrated infection intensity of 20-50 in larger fish. Further, intensity of infection has shown a rapid 30-40% increase in a five-year period [51].

Ambiguity of Codex Salting and Brining Recommendations
In CODEX STAN 244-2004 [50] 3.1 "Fish flesh shall not be obviously infested by parasites" and "If living nematodes are confirmed, products must not be placed on the market for human consumption before they are treated in conformity with the methods laid down in Annex II". In the most recent version accessed (2018) Annex II states "the adequate combination of salt content and storage time (to be elaborated)-or by other processes with the equivalent effect (to be elaborated)." In CAC/RCP 52-2003 [19] 12.1 "Where appropriate, fresh fish intended for processing salted fish should be checked for visible parasites" and "an adequate combination of salt content and storage time can be used as treatment procedures for killing living parasites". It, again, is unclear where 'appropriate' may be along the food chain or if processors would consider salted fish in need of checking for parasites. Adequate salt concentration or storage time required to kill parasites is not defined at all. In the same standard, Section 2.2.2. includes the categories for salted fish and the percentage of salt required in the muscle of the fish during the water phase. These include 2. >20%. Herring at 15-19% brine were found infected with a number of live Anisakis larvae and 22%-23% brine was required to kill nematode larvae over a period of 7 days which commenced 3-4 days post salting [52]. According to Lubieniecki [52], the salt concentration of herring flesh was influenced by brine salt concentration, but also additionally the gonad maturity stage, lipid content of the flesh, and salting temperature and hence are factors which may contribute to increase the viability of Anisakis larvae. Three subsequent studies conducted by Grabda (1971Grabda ( -1973 confirmed that live Anisakis larvae were able to survive in 15-19% brined herring [as cited by 53]. In a study of fresh Baltic herring, after a week at a 5.6% visceral salinity, 98.2% of Anisakis larvae in herring were motile; 2 weeks at salinity of 9.36%-12.9% no motile larvae were observed however after culture 25/25 of the non-motile larvae became motile again over a three-day period. At three weeks visceral salinity, 11.6%-14.04%, 13/71 larvae identified became motile on Day Two of culture. After four weeks at 12.2%-14.6% salinity no motile larvae were found in the cultures [53]. It appears that under Codex definitions of 'very lightly', 'lightly' and 'medium' salted that it would be after four weeks from the initial date salting commenced that the fish product could be regarded as entirely safe for human consumption. Oh, et al. [54] demonstrated Anisakis larvae were viable after seven days emersion in 5% NaCl (81.7%) and 10% NaCl (26.7%). All larvae were inactivated after seven days in 15% NaCl, and six days in 20% NaCl. Most larvae survived in all NaCl concentrations for 3-12 h. However, in this study larvae were introduced directly into brine. It is possible that larvae in fish musculature, where saline penetrates more slowly, may demonstrate longer inactivation times. The slow inactivation of infectious larvae and the regeneration of moribund larvae presumed dead is concerning. This implies that larvae in salted products may become infectious after consumption. Codex stipulates in CODEX STAN 244-2004 and CODEX STAN 311-2013 [24,50] that a viable larvae is one which clearly demonstrates spontaneous movement after mechanical stimulation. By implication the moribund larvae, in the studies cited, which became viable after incubation would be considered non-viable according to Codex recommendations. In the case of Diphyllobothrium spp. (syn. Dibothriocephalus) in fish, salting in 10% to 20% NaCl solution has been demonstrated to kill the plerocercoids after 1 or 2 h [55]. Freshwater fish Pseudorasbora parva infected with metacercariae of Clonorchis sinensis were treated with a heavy salt (fish/salt = 10 gm/3 gm) and kept at 26 • C for 5-15 days. Metacercariae remained viable and produced infection in rats up to seven days after salting [56].

Human Health Risks Posed by Seafood-Borne Parasites
There are many seafood-borne zoonotic parasites which have been implicated in cases of human infection. For a comprehensive list of seafood-borne zoonotic parasites which may be a human health concern in imported edible seafood please see Shamsi and Sheorey [57]. Cooking and freezing according to the methods described in the relevant Codex standards is sufficient to kill all zoonotic parasites in seafood. At present the nematode A. simplex is the only species known to cause allergic reactions of varying exigency [58], with killed parasites in fish representing an allergen risk to some [59]. The consumption of raw or improperly cooked seafood is an important risk factor for humans acquiring a seafood-borne parasite zoonosis [32].
The World Health Organisation and the Food and Agricultural Organisation (FAO) in a review of parasites within the food trade concluded that the complex life cycle of aquatic parasites allows great potential for contamination of edible seafood. Further, the panel commented that food-borne parasitic diseases were neglected and underreported globally [60]. The WHO Foodborne Disease Epidemiology Reference Group [61] observed that the full human health impact of parasites in food is unknown. During the joint WHO/FAO review 6/24 parasites evaluated pertained to those in edible seafood products [60]; Anisakidae rated four according to trade risk. Sumner and Ross [62] in a 2000 Australian risk assessment awarded a low hazard for 'parasites in sushi/sashimi'; the only pairing relating to parasites in seafood. In 2012, an Australian risk assessment of zoonotic parasites in Australian fish [63] commented that freshwater fish are not used for sushi/sashimi, however, identified anisakidosis/anisakiasis as underreported and/or misdiagnosed within Australia. In a 2015 risk assessment of Australian fish used for sashimi the authors concluded that the low incidence of anisakidosis in Australia may be due to underreporting or elimination of the parasite hazard during processing and preparation [64]. However imported fish and the potential for fish substitution was not included in this risk assessment or in a subsequent Australian risk assessment in 2017 [65]. Warner, et al. [66] reported that 58% of fish samples obtained from sushi venues in the Miami/Fort Lauderdale-area were mislabelled and 100% of Snapper was incorrectly labelled. It is unclear if mislabelling seafood is a significant risk in Australia.
Opisthorchiasis and clonorchiasis have been increasingly reported from non-endemic areas [67]. An outbreak of acute opisthorchiasis in an Israeli family was reported after eating illegally imported raw carp [68]. Opisthorchiasis has also been reported in native Hawaiians after consuming imported fish from endemic areas of infection [67]. Rohela et al. [69] regarding Clonorchis sinensis, and Park [70] regarding Opisthorchis viverrini and Clonorchis sinensis, commented that human infection outside traditional areas occurs as a result of consuming frozen, dried or pickled imported freshwater fish infected with metacercariae. Human infection in Hawaii with Clonorchis sinensis has been attributed to the consumption of dried or pickled fish imported from endemic areas [71]. Infected salted, dried or pickled fish is a significant risk factor in the transmission dynamics of Opisthorchis viverrini [72]. The importation of intensively farmed native fish, cyprinid carp species [60], Catfish/Basa, (Pangasianodon hypothalamus) and Tilapia, (Orechromis niloticus/O. mossambicus) which are susceptible to infection, has great potentiality to cause zoonotic infection in geographic regions outside the normal areas of endemicity [60,73]. Evidence in Vietnamese and Chinese aquaculture according to Murrell et al. [73] would suggest the potential contamination with zoonotic seafood-borne trematodes of seafood destined for international trade. The authors further advise that seafood imported from areas of parasite endemicity, particularly Asia, may be an infection risk to consumers and prevention should be implemented throughout the market chain. The global fish trade is considered an important factor in the alteration of the traditional geographical boundaries associated with Diphyllobothrium spp. (syn. Dibothriocephalus) [74]. The consumption of imported fish has been linked to cases of human infection in Spain [74][75][76], France [77,78], Switzerland [79][80][81] and recently the first two cases in Singapore from D. nihonkaiensis [82] which may indicate a deficit in the inspection processes of the exporting countries. Ogata, et al. [83] considers imported/introduced tilapia, farmed in Mexico since 1964, for the increase in gnathostomiasis cases regionally. In America, live eels imported for human consumption from Asia [84] and Vietnam [85] have been demonstrated heavily infected with encysted and un-encysted Gnathostoma spp. larvae. Imported fish has been implicated in cases of human infection from Capillaria philippinensis in Egypt (Youssef et al., 1989), which has been hypothesized as the entry point of the parasite into Egypt [86]. Infestation of fish for human consumption by anisakid nematodes has increased markedly during the last 20 years [87]. Cooking or freezing does not destroy the allergenic capacity of A. simplex which has been implicated in human reactions to canned fish [88]. A. simplex allergens have been identified in baby food products containing plaice and European hake [87]. No mention is made of the allergenic potential of A. simplex in canned products in any Codex recommendations. Mossali et al. [87] considers the frequent presence of anisakids in processed food reflects a utilisation of poor quality fish which would normally be discarded. The European Food Safety Authority (EFSA) has introduced a requirement for the routine testing of canned fish for anisakids using PCR method [21,89]. Australia imports a significant quantity of canned fish from many European countries [90] and these include species of fish high risk for human anisakiasis [58,[91][92][93][94]. It is unknown if any of these canned products pose an Australian human health biosecurity risk.

Imported Seafood Inspection in Australia
At present, there are no additional tests applied to imported edible seafood, for detection of zoonotic parasites, on entry to Australia. Figures 3 and 4 describe current tests applied to imported edible seafood on entry to Australian.    Representation of the legislative instruments that support the inspection of seafood imported to Australia in grey, the testing regimes for 'surveillance' and 'tightened' inspection for 'risk foods' in green. In yellow is the testing regime for 'normal', 'reduced', 'compliance' and foods which have been classified under the 'Trans-Tasman Mutual Recognition Agreement'. 'Compliance agreements' are entered into voluntarily by the exporting provenance which must show compliance and equivalency with the standard of their food management systems which is audited annually. The original figure was developed from information at [18,95,96].

Figure 4.
Representation of the current tests applied to 'risk' food on entry to Australia. All 'risk' food is inspected at the rate included in Figure 2. 'Risk' food will be examined visually and the label Figure 4. Representation of the current tests applied to 'risk' food on entry to Australia. All 'risk' food is inspected at the rate included in Figure 2. 'Risk' food will be examined visually and the label checked.
The tests applied at present to five groups of 'risk' foods have been included. Currently no additional tests are applied to 'risk' foods for detection of zoonotic parasites in imported edible seafood. Visual inspection may be effective to identify some macroscopic parasites in seafood however as a tool it is extremely limited as most parasites infecting seafood require microscopic inspection by a trained professional. Original figure developed from information at [18,[95][96][97][98][99].
As related in an email from an Australian Government Biosecurity Officer on the 26th August, 2019 "lesions on fish caused by parasites are not considered to be either a biosecurity (regulated by my section) or human health (regulated by the Imported Food Inspection Scheme) risk". Of the 29 Schedules in the 'Australian and New Zealand Food Standards Code' [97] there are none which relate to zoonotic parasite contamination of imported edible seafood. Edible seafood may be subject to label inspection as detailed in the Part 1 and Part 2.2 (2.2.3 Fish and Fish Products) of the ANZFSC [97] and visual inspection as detailed in Sections 3(a)(vii) and 3(b) Imported Food Control Act 1992 [95]. Visual inspection is based upon Section 3(a)(vii) "any other contaminant or constituent that may be dangerous to human health" and 3(b) "it has been manufactured or transported under conditions which render it dangerous or unfit for human consumption". Certainly, visual inspection could be interpreted to include visual detection of zoonotic parasites. However, as a tool this is inadequate to detect parasite contamination [100]. For example, in a 2007 study only 26/185 Anisakis larvae in monkfish fillets [25] were identified using visual inspection. Microscopic examination, candling [32], UV light [31], PCR [89] and pepsin digestion method [101] by a trained professional are all valid methods but are not listed as the tests applied to either 'risk' or 'surveillance food' within the Australian food inspection scheme. There have been six import consignments failed in the time period 2010-2018 based upon visual inspection [102]. None of the fails were as a result of parasites visualised in edible seafood. The Imported Food Control Amendment Bill 2017 [103] was passed by the Australian House of Representatives on 11/9/2018 [104] and amends the Imported Food Control Act 1992. The amendments have been designed to place the onus of responsibility on the exporters to provide documentary evidence of their adherence to internationally recognised food safety controls. An impact statement was circulated by the Government during August 2016 pursuant to the Imported Food Control Amendment Bill 2017 [105]. There was no mention of parasites related to seafood in the impact statement [106]. Under Part 4 (35A [1][2][3][4][5][6][7][8][9][10]) of the Imported Food Control Act 1992 [95] food exporters may voluntarily enter an agreement with Department of Agriculture for a 'Food Import Compliance Agreement.' Imported food under this agreement is not inspected or tested under the Food Inspection Scheme [98]. The exporters documented food management system must comply with Australian and New Zealand Food Standards Code [97] and Australian Standard ISO 22000:2005 (Food safety management systems-requirements for any organization in the food chain) [107]. Australian inspection processes for imported seafood places significant trust in the exporting nations 'equivalency' in testing procedures and adherence to the Sanitary and Phytosanitary Measures Agreement, 1995 [18].

Discussion
During 2017, Australia imported a significant amount of seafood from countries endemic for infection with zoonotic parasites. At the time of writing (15/9/2019) The Australia New Zealand Food Standards Code [108], Standard 3.2.1-'Food Safety Programs', Standard 3.2.2-'Food Safety Practices and General Requirements' make no mention of parasite risk in local or imported fish or standards if fish is to be consumed raw. Standard 4.2.1 'Primary production and processing standard for seafood' (Australia Only) list parasites as a possible contaminant however Standard 1.6.1, table to section S27-4, of ANZFSC which should list the maximum allowable levels of contamination does not contain information for parasites at all [108]. The 'Export Control (Fish and Fish Products) Orders 2005 [109] which guides our export policy makes no mention of parasites. The Food Safety Information Council of Australia comments that there is a slight risk associated with the consumption of raw seafood, sushi and sashimi for example, but these risks can be mitigated by consuming seafood from safe waters, chilling and correctly storing or purchasing from licensed suppliers. No mention is made of parasites in raw fish or freezing before consuming raw [110], however, in a 2005 'Safe Seafood Australia' publication there is a recommendation to freeze fish (Australia only) if intended for raw consumption [111]. 'The Compendium of Microbiological Criteria for Food (2018)' from Food Standards Australia and New Zealand in Appendix I mentions parasites as a possible pathogenic microorganism which can cause foodborne illness however this is mentioned only once in the document [112]. Safefish is funded by the FRDC and is concerned with Australian seafood safety and trade. The 'seafood safety fact sheets' (2015) produced by Safefish make no mention of parasites in seafood [113]. At present, there is a paucity of information regarding zoonotic seafood borne parasites in Australia and it is not surprising that seafood borne parasitic disease is almost unknown. Globally, diseases from food-borne parasites are often neglected by Governmental health authorities and official figures are not reflective of the prevalence or incidence of disease [60]. The WHO Foodborne Disease Epidemiology Reference Group [61] commented that despite food borne diseases being a significant worldwide cause of death and morbidity the full impact of parasites in food is unknown. According to Kirk, et al. [114] within Australia only 28% of people affected with food-borne illness will seek medical attention. Absence of reported cases of seafood-borne parasitic disease have been used as evidence that there is no disease in Australia [115]. However, according to Shamsi and Sheorey [57] misdiagnosis in Australia contributes to lack of evidence regarding the prevalence of seafood-borne zoonoses and reliable parasite focused epidemiological data [116]. In Australia, where zoonotic parasites are largely unrecognised [57], the lack of reported cases of seafood-borne parasitic disease may more clearly reflect lack of diagnostic suspicion rather than absence of disease.

Conclusions
It is unlikely that Australia is immune from seafood-borne parasitic disease which has been widely recognised internationally. The intense cultivation of aquaculture species and the international trade in both farmed and wild caught seafood is a key factor in establishing global food security. However, these same endeavors which bring seafood to all corners of the globe are also high risk for the spread of pathogens and zoonotic parasites. As a member nation of the World Trade Organisation Australia is bound to uphold the three agreements signed at the Uruguay round of talks collectively known as the 'Marrakesh Agreement' [99,117,118] and designed to facilitate between country trade. Articles 3, 5 and Article 10 of the SPS encumbers Australian policy makers from implementing any additional testing procedures for imported edible seafood which would hinder between country trade without robust scientific justification. Support of developing nations to reach international safety and health standards is a requirement in Article 9 of the SPS [18]. Rather than a downregulation of health standards to comply with international trade agreements perhaps increased support of the seafood industry in developing nations to achieve upregulation of food safety compliance may be a positive step forward. Australian biosecurity is considered exemplary. However, as the onus for inspection of seafood imported to Australia is increasingly awarded to exporters the issue is focused away from the strength of Australian food biosecurity towards trust in the international food safety standards.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
Appendix A Table A1. Appendix A Represents all Codex Alimentarius food safety codes of practice/standards and guidelines relating to zoonotic parasites in seafood. In orange are those documents which contain specific advice pertaining to zoonotic parasite and those which do not are shaded in green.

INCOMING MATERIAL REQUIREMENTS
No raw material or ingredient should be accepted by an establishment if it is known to contain parasites, undesirable micro-organisms, pesticides, veterinary drugs or toxic, decomposed or extraneous substances which would not be reduced to an acceptable level by normal sorting and/or processing. Where appropriate, specifications for raw materials should be identified and applied.
CAC/RCP 52-2003 (Last updated 2016) Code of practice for fish and fishery products 2.5 Candling Passing fillets of fish over a translucent Filluminated from below to detect parasites and other defects. Hot is generally sufficient to kill parasites, to destroy non-sporulated bacterial pathogens and to injure spores of human health concern.
5. Unless they can be reduced to an acceptable level by normal sorting and/or processing, no fish, shellfish or other aquatic invertebrates should be accepted if they are known to contain parasites. 5.2 Parasites of public health significance: trematodes, nematodes, cestodes 6.2 Infection with nematode parasites is absent from, or very much reduced in, farmed salmon compared with salmon caught in the wild

Raw, fresh or frozen fish reception Potential hazards: microbiological contamination, viable parasites
Training in species identification and communication in product specification should be provided to fish handlers and appropriate personnel to ensure a safe source of incoming fish where written protocols exist. Warranting special consideration are the reception and sorting of fish species that pose a risk from parasites. 9.1.3. Frozen storage Potential hazards: microbiological contamination, toxins, viable parasites. For killing parasites harmful to human health, the freezing temperature and monitoring of duration of freezing should be combined with good inventory control to ensure sufficient cold treatment. 9.1.6 Filleting, skinning, trimming and candling Potential hazards: viable parasites, Potential defects: parasites. 9.1.6. Candling of skinless fillets by skilled personnel, in a suitable location that optimizes the illuminating effect, is an effective technique in controlling parasites (in fresh fish) and should be employed when implicated fish species are being used. For killing parasites harmful to human health, the freezing temperature and monitoring of duration of freezing should be combined with good inventory control to ensure sufficient cold treatment. 9.4.1 Mincing fish using mechanical separation process, Potential defects: parasites. Candling is recommended for fish suspected of high infestation with parasites 10.1 General considerations of hazards and defects for frozen surimi production. Parasites will not be a hazard as the final product will be cooked or pasteurized. 10.1.2 Myxosporidia is a parasite that is common in marine groundfish such as Pacific whiting. This organism contains protease enzymes that chemically separate proteins that can ultimately affect the gel strength of surimi even at very low incidence. If species are used that are known to contain this parasite, protease inhibitors such as beef plasma protein or egg whites may be needed as additives to attain the necessary gel strength capabilities for kamaboko or crab analogue production.

Smoked fish, smoke-flavoured fish and smoke-dried fish. 13.1 Processing of Smoked Fish:
If raw material likely to contain viable parasites is to be used steps must be taken to eliminate this hazard during processing steps, e.g., freezing, heating or salting the product. Alternatively, the final product should be treated in a way to kill parasites. 13 Processes such as brining or pickling may reduce the parasite hazard if the products are kept in the brine for a sufficient time but may not eliminate it. Candling, trimming belly flaps and physically removing the parasite cysts will also reduce the hazards but may not eliminate them.
Nematodes Many species of nematode are known to occur worldwide and some species of marine fish act as secondary hosts. Among the nematodes of greatest concern are Anisakis spp., Capillaria spp., Gnathostoma spp. and Pseudoteranova spp., which can be found in the liver, belly cavity and flesh of marine fish. An example of a nematode causing disease in human beings is Anisakis simplex; the infective stage of the parasite is killed by heating (60 • C for one minute) and by freezing (−20 • C for 24 h) of the fish core Cestodes are tapeworms and the species of greatest concern associated with the consumption of fish is Dibothriocephalus latus. This parasite occurs worldwide and both fresh and marine fish are intermediate hosts. Similar to other parasitic infections, the foodborne disease occurs through the consumption of raw or under-processed fish. Similar freezing and cooking temperatures as applied to nematodes will kill the infective stages of this parasite.

Procedure for the Detection of Parasites (Type 1 Method) in skinless fillets
The entire sample unit is examined non-destructively by placing appropriate portions of the thawed sample unit on a 5 mm thick acryl sheet with 45% translucency and candled with a light source giving 1500 lux 30 cm above the sheet.

Parasites
The presence of two or more parasites per kg of the sample unit detected by the method described in 7.4 with a capsular diameter greater than 3 mm or a parasite not encapsulated and greater than 10 mm in length.

8.6
Flesh abnormalities A sample unit affected by excessive gelatinous condition of the flesh together with greater than 86% moisture found in any individual fillet or a sample unit with pasty texture resulting from parasitic infestation affecting more than 5% of the sample unit by weight.

"
Hot smoking" is a process in which fish is smoked at an appropriate combination of temperature and time sufficient to cause the complete coagulation of the proteins in the fish flesh. Hot smoking is generally sufficient to kill parasites, to destroy non-sporulated bacterial pathogens and to injure spores of human health concern.
6.3 Parasites Products covered by this Standard shall not contain living parasites and particular attention needs to be paid to cold smoked or smoke-flavoured products, which should be frozen before or after smoking if a parasite hazard is present (see Annex 1). Viability of nematodes, cestodes and trematodes shall be examined according to Section 8.10 and/or 8.11.

Determination of the viability of parasites
Methods used for extracting and testing the viability of parasites could include the method set out in Annex I for nematodes in the Standard for Salted Atlantic Herring and Salted Sprat (CXS 244-2004) or other validated methods for parasites acceptable to the competent authority having jurisdiction.

Determination of visible parasites
The entire sample unit is examined for the presence of parasites non-destructively by placing appropriate portions of the thawed (if necessary) sample unit on a 5 mm thick acryl sheet with 45% translucency and candled with a light source giving 1500 lux 30 cm above the sheet.

Parasites
The presence of two or more visible parasites per kg of the sample unit detected by the method described in 8.11 with a capsular diameter greater than 3 mm or a parasite not encapsulated and greater than 10 mm in length.

ANNEX I Procedures sufficient to kill parasites
A method that is acceptable to the competent authority having jurisdiction shall be used to kill parasites. Where freezing is required to kill parasites (i.e., cold smoked fish and smoke-flavoured fish) the fish must be frozen either before or after processing to a temperature time combination sufficient to kill the living parasites. Examples of freezing processes that may be sufficient to kill some or all parasites are: • Freezing at −20 • C at the thermal centre of the product for 24 h (for Anisakis species and Pseudoterranova decipiens only) 1;

Examination for Parasites
The presence of readily visible parasites in a sample unit detected by normal visual inspection of the scallops.

Parasites
The presence of parasites should not be at an objectionable level.