Raw fish can carry live Anisakis roundworm larvae and Diphyllobothrium tapeworms. That much is well documented, and the usual answer is freezing to FDA or EU parameters before the fish is served.
The tapeworm side gets less attention than it deserves. Diphyllobothrium nihonkaiense caused 86.1% of the 958 cestodiasis (tapeworm infection) cases recorded in Japan from 2001 to 2016, and the infection is closely tied to raw Pacific salmon (Emerging Infectious Diseases, 2018).
Whether a given piece is safe depends on whether it was frozen long and cold enough, and on the fish itself, meaning its species and where it came from.
Which parasites can sushi carry?
Anisakis simplex, Pseudoterranova decipiens and the Diphyllobothrium tapeworms are the ones that matter in practice. The first two are larval roundworms and the last is a tapeworm, and all of them live in the flesh of raw fish until somebody swallows them alive.
| Species | Typical host fish | Illness | Where reported |
|---|---|---|---|
| Anisakis simplex complex | Herring, cod, mackerel, squid | Anisakiasis | Worldwide, highest where raw fish is common |
| Pseudoterranova decipiens | Cod and other cold-water fish | Anisakiasis | North Atlantic, Arctic, Japan, southern Chile |
| Diphyllobothrium nihonkaiense | Wild Pacific salmon | Diphyllobothriasis | North Pacific, Japan, imported cases in Europe |
Table sources: CDC DPDx, Emerging Infectious Diseases (2018) and Parasites and Vectors (2019).
Here is how the roundworms end up in the fillet (CDC DPDx):
- Adults live in marine mammals (whales and dolphins for Anisakis, seals for Pseudoterranova), and their eggs pass into the water.
- Crustaceans eat the free-swimming larvae.
- Fish and squid eat the crustaceans.
- Larvae migrate from the fish gut into the abdominal cavity, then into the muscle.
People are accidental hosts. The larvae cannot survive long in a person, which is why most infections end on their own.
The tapeworm takes a different route. Diphyllobothrium nihonkaiense uses wild Pacific salmon (cherry, chum and pink salmon) as its second intermediate host, and it was the predominant tapeworm in Japanese cestodiasis cases from 2001 to 2016 (Emerging Infectious Diseases, 2018).
Because the larvae sit in the muscle, any raw preparation exposes the person eating it, sliced raw fish served as sashimi included.
Parasite risk by sushi fish, wild-caught and farmed
Wild-caught salmon, mackerel, cod, herring and squid carry the most larvae. Tuna, yellowtail, snapper and flatfish rarely do, and farmed salmon raised on pelleted feed carries essentially none.
The sampling numbers back that up. In Seattle-area restaurants, 10% of 279 salmon sushi samples held parasites, and every parasite was dead (FDA parasitologist Ann Adams, reported by The Seattle Times, 1992). The same sampling found parasites in 0 of 201 tuna sushi samples.
Wild versus farmed is even starker. 50 of 50 wild sockeye salmon carried Anisakis simplex larvae, against 0 of 237 pen-reared salmon (Deardorff and Kent, Journal of Wildlife Diseases, 1989). Smoked fish told the same story, with 10 of 13 smoked wild sockeye samples testing positive for Anisakis simplex larvae and 0 of 270 smoked farmed Atlantic salmon samples doing so (Pardo et al., Italian Journal of Food Safety, 2020).
Salmon, mackerel, cod and squid
Wild salmon, Atlantic and Pacific alike, commonly carry Anisakis larvae (Food Standards Agency Scotland, 2010), and the Pacific fish also host Diphyllobothrium tapeworms.
Mackerel, cod and herring are predatory fish, which is where human-infecting anisakids are most common (CDC DPDx). Squid is no safer, since it carries infective larvae the same way fish does.
Tuna and other low-risk fish
Tuna sits at the low end. In the 1992 FDA sampling of Seattle sushi, no tuna sample held a parasite.
The picture is not perfectly clean, though. The same researcher noted reports from other states of anisakiasis involving tuna, but nobody knew which dish was involved (The Seattle Times, 1992).
Nawa and colleagues, in a 2005 review in Clinical Infectious Diseases, also put tuna, yellowtail, snapper and flatfish in the rarely infected group. Salmon is the exception there, because it can carry Diphyllobothrium tapeworms.
Wild-caught versus farmed
Three independent surveys across nearly three decades found no Anisakis larvae in farmed salmon.
Pen-reared salmon from Puget Sound (1989), 720 farmed salmon in a Food Standards Agency Scotland survey (2009) and smoked farmed Atlantic salmon sold in Europe (sampled in 2016) all came back clean. Wild salmon in the Puget Sound and smoked-salmon comparisons showed high infection rates, and every wild Atlantic salmon in a 2010 Scottish survey carried Anisakis larvae in the muscle.
The reason is the feed. Controlled pellets replace the infected crustaceans and small fish that wild salmon eat.
Farmed fish are not a blanket pass, though. Aquacultured fish fed processing waste, fresh fish or plankton can carry a parasite hazard even when wild fish of the same species do not (FDA Fish and Fishery Products Hazards and Controls Guidance).
Are parasites in sushi fish becoming more common?
Anisakis abundance in marine fish and invertebrates rose 283-fold between 1978 and 2015, while Pseudoterranova did not change (Fiorenza et al., Global Change Biology, 2020).
Coverage of the study quotes different start years. The paper reports a 1962 to 2015 trend for Anisakis and standardizes to 1978 to 2015 so both genera can be compared, and the 283-fold figure comes from that second window.
In the sea, that means more Anisakis larvae per fish, so more fillets start out infected. What reaches the plate depends on whether the fish was frozen first, and the parasites found in the 1992 Seattle salmon sushi samples were all dead.
Rising worm counts and a low rate of live infections can both be true. The trend matters most for fish that reaches the plate unfrozen.
Where raw fish is a staple, infections run into the tens of thousands. Japan averaged 19,737 anisakiasis cases a year in 2018 and 2019, estimated from health insurance claims (Sugiyama et al., Emerging Infectious Diseases, 2022).
Tapeworms spread through the same trade routes. A cluster of Diphyllobothrium nihonkaiense cases in Brittany, France, diagnosed between July 2016 and September 2018, was probably acquired from raw salmon in Japanese food, in a region where this infection is not endemic (Autier et al., Parasites and Vectors, 2019).
What happens if you swallow a live parasite?
Gastric anisakiasis causes abdominal pain, nausea and vomiting within hours of eating live larvae, while a fish tapeworm often causes no symptoms at all (MSD Manual).
Symptoms and onset time
- Gastric anisakiasis brings pain, nausea and vomiting within hours of eating (MSD Manual).
- The intestinal form is slower. An inflammatory mass and Crohn disease-like symptoms show up 1 to 2 weeks later, and perforation is rare (MSD Manual).
- Some people react allergically, with rash and itching and, rarely, anaphylaxis. Antigens left in the fish muscle after the larvae die, including after freezing, can trigger it (CDC DPDx).
- A tapeworm is usually silent. Sometimes there is mild gut discomfort, and occasionally vitamin B12 deficiency (MSD Manual).
The early pain looks like a peptic ulcer, food poisoning or appendicitis (CDC DPDx). Most roundworm infections end on their own, but tissue damage can keep symptoms going longer.
Diagnosis and treatment
A 32-year-old man in Lisbon had severe upper stomach pain, vomiting and low-grade fever for about a week. Only when doctors asked about his diet did he mention that he had recently eaten sushi (BMJ Case Reports, 2017).
Upper endoscopy showed a larva attached to his stomach lining, and his symptoms cleared once it was removed with a net.
- For gastric anisakiasis, gastroscopy confirms the diagnosis and removes the larva. Some patients cough one up and bring it in (CDC DPDx).
- Bowel and other cases are confirmed from tissue taken at biopsy or surgery (CDC DPDx).
- A tapeworm shows up as eggs or segments in stool, and a single oral dose of praziquantel treats it (MSD Manual).
Tell the doctor about the raw fish. The symptoms alone point to ulcers or appendicitis, not parasites.
Freezing conditions that kill parasites in sushi fish
FDA Food Code section 3-402.11 and FDA’s Fish and Fishery Products Hazards and Controls Guidance (chapter 5) set the time and temperature rules for freezing parasites out of sushi fish. The slowest option is 7 days at -4 degrees F (-20 degrees C). This holds as of October 2026, including in the 2026 Food Code released in September 2026.
| Protocol | Freezing step | Hold requirement |
|---|---|---|
| Standard freezer | -4 degrees F (-20 degrees C) or below | 7 days (168 hours) total |
| Blast freeze, cold hold | -31 degrees F (-35 degrees C) or below until solid | 15 hours at -31 degrees F or below |
| Blast freeze, standard hold | -31 degrees F (-35 degrees C) or below until solid | 24 hours at -4 degrees F or below |
FDA says how well freezing works depends on the freezing temperature, the time needed to freeze the tissue, the hold time, the species and source of the fish, and the parasite type. Temperature and hold time carry the most weight, and parasite type counts too.
Tapeworms die more readily in the freezer than roundworms do. Flukes, by contrast, appear to be tougher than roundworms (FDA guidance).
The protocols have limits. Particularly large fish are one, since FDA says the conditions may not be suitable for them. Allergy is the other, because dead larvae still leave antigens in the muscle (CDC DPDx).
Freezing that meets these numbers kills the larvae. It does not remove the allergens they leave behind.
Do wasabi, vinegar or a chef’s eye remove the risk?
No. FDA and EU rules name freezing and heating as the controls, and none of these replaces them.
Vinegar and brine
FDA says brining and pickling may reduce the parasite hazard in a fish but do not eliminate it or bring it to an acceptable level (FDA Fish and Fishery Products Hazards and Controls Guidance).
Lab work on Anisakis larvae points the same way. Undiluted alcohol vinegar killed isolated larvae in under 48 hours, but diluted with water the larvae lasted almost 40 days, and inside anchovy fillets only dry salting destroyed them (Simat and Trumbic, Fishes, 2019).
Sushi rice vinegar seasons the rice. No regulation counts it as a parasite control.
Wasabi
Wasabi gets credit it has not earned.
Researchers tested allyl isothiocyanate (AITC), the pungent compound in wasabi, against Anisakis larvae in liquid media, semisolid media and anchovy marinades. It showed significant activity in liquid media, but in semisolid media only at higher concentrations (Giarratana et al., Journal of Food Protection, 2015).
Wasabi paste served alongside raw fish was never tested. The authors call AITC a candidate for further work in industrial marinating, not a finished control.
Chef inspection
The lead author of the 2017 Lisbon case report, Dr. Joana Carmo, said properly trained sushi chefs can detect anisakis larvae (WXYZ). An FDA parasitologist listed the same belief among the misinformation about sushi parasites in 1992 (The Seattle Times).
The measured numbers disagree with each other. Candling had 31% sensitivity for infected fish parts, using candling plus enzymatic digestion combined as the reference standard (Mercken et al., Scientific Reports, 2022). Visual inspection of anchovy batches, on the other hand, had 93% sensitivity for rejecting unmarketable lots (Guardone et al., Food Analytical Methods, 2016).
The studies measured different things. Mercken checked whether each fish part held at least one larva against that combined reference, while Guardone judged whole batches against a mean-abundance threshold, so a batch can be rejected correctly while single larvae slip through.
Inspection lowers the count and never clears a fillet. FDA lists trimming belly flaps and candling as ways to reduce parasite numbers, and Mercken et al. advise removing or freezing the belly flaps, where infection runs highest.
What the FDA and EU require from sushi suppliers and restaurants
Both require fish served raw to be frozen to kill viable parasites, with exemptions, and they set different hold times at the same temperature.
FDA Food Code section 3-402.11 covers raw, raw-marinated and partially cooked fish before service or sale. EU Regulation (EC) 853/2004, Annex III, as amended by Regulation (EU) No 1276/2011, covers fishery products intended to be eaten raw or almost raw.
| Requirement | FDA Food Code | EU Regulation (EC) 853/2004 |
|---|---|---|
| Standard freeze | -4 degrees F (-20 degrees C) for 7 days | -20 degrees C in all parts of the product for not less than 24 hours |
| Fast freeze | -31 degrees F (-35 degrees C) until solid, then 15 hours | -35 degrees C for not less than 15 hours |
| Exemptions | Six Thunnus tuna species, qualifying farmed fish, rinsed fish eggs, molluscan shellfish, shucked scallop adductor muscle | Heat-treated or long-frozen products, fish from grounds or farms meeting set conditions |
At -20 degrees C the FDA wants 7 days. The EU settles for 24 hours, having rewritten its text in 2011 to reflect EFSA’s 2010 scientific opinion on parasites in fishery products.
A restaurant that freezes its own fish has to keep freezing temperature and time records for 90 calendar days beyond service or sale. A supplier’s written statement that the fish was frozen to a listed protocol can substitute (FDA Food Code section 3-402.12).
Beaver Street Fisheries told customers in a January 2018 letter that it requires parasite-destruction freezing for all species, including those the FDA exempts.
Section 3-402.11 has no sushi grade category at all. Whatever the sushi grade label means on a salmon fillet, it is the seller’s own claim, and the freezing record or supplier letter is the evidence.
When freezing and regulation do not protect you
Freezing rules protect you only when the fish passes through a business that follows them. Several situations fall outside that.
Home freezers are the obvious one. A freezer set to the USDA’s 0 degrees F target misses the FDA’s -4 degrees F by 4 degrees (USDA Food Safety and Inspection Service). Check yours with a thermometer before trusting it.
Some exemptions rest on paperwork. A farmed fish skips freezing when the supplier states it was raised in net pens, ponds or tanks on formulated feed, so the statement is the control.
Tuna is not one blanket exemption either. The Food Code exempts six Thunnus species, and Murata and colleagues reported a probable link between Anisakis in skipjack tuna muscle and human anisakiasis in Tokyo (International Journal of Food Microbiology, 2021). Skipjack is not on that list.
Abroad, rural Japan serves freshwater fish, brackish-water fish and wild animal meats as sushi and sashimi, and these are important sources of zoonotic parasites (Nawa et al., Clinical Infectious Diseases, 2005).
If you cannot name who froze the fish and to what temperature, treat it as unfrozen.
How to lower the risk with restaurant and homemade sushi
Verify that the fish was frozen to a Food Code or EU protocol, and cook it when you cannot.
- Ask the restaurant whether its raw fish was frozen to Food Code parameters or arrives with a supplier freezing statement. The records exist on paper for 90 days.
- Treat a “raw use” label as a claim. Ask the seller who froze the fish and to what temperature.
- Buy pre-frozen fish for home sushi instead of freezing it yourself. Trident Seafoods says in its parasite-control letter that it holds its raw-labelled products at -4 degrees F or colder for at least 7 days before release.
- If you freeze at home, confirm with a thermometer that the freezer holds -4 degrees F or colder for the full 7 days.
- Thaw in the refrigerator, as USDA advises, not on the counter.
- Cut away the belly flaps, where larvae concentrate, and slice thin to look for worms.
- Choose cooked fish when freezing cannot be verified. Heating to 145 degrees F (63 degrees C) kills the parasites (MSD Manual).
Cooked fish removes the question. At home, baking the fish into a sushi bake keeps the flavors without the raw-fish risk.
And if freezing cannot be verified, I’d skip wild salmon, mackerel and squid raw.
FAQ About Parasites in Sushi
Is it safe to eat sushi while pregnant?
No. FoodSafety.gov guidance for pregnant women says to avoid sushi, because raw seafood can carry parasites or bacteria, including Listeria.
Rolls built on fully cooked fish remove the parasite risk. Refrigerated smoked salmon counts as a raw-risk food unless it is cooked to 165 degrees F.
Can fish roe like ikura carry parasites?
Rarely. Parasites can sit in the egg skein of an infected fish but generally not in the eggs themselves (FDA Fish and Fishery Products Hazards and Controls Guidance).
That is why separating and rinsing the roe matters. Roe left in the skein is the form that keeps the hazard.
Do cooked sushi rolls have parasites?
No live ones. Fish cooked to an internal temperature of at least 145 degrees F (63 degrees C) has its parasites killed (CDC).
Heat kills the larvae but not their allergens, which are heat-resistant (Mercken et al., Scientific Reports, 2022).
How common is anisakiasis outside Japan?
Nobody has an exact count. A 2017 risk model estimated 7,700 to 8,320 cases a year needing medical attention in Spain, against 500 a year previously reported for all of Europe (Bao et al., Scientific Reports).
The CDC does not run surveillance for the infection and said rare US cases have been identified (CNN, 2017). Modeled estimates and diagnosed counts are not directly comparable.
Residual Parasite Risk in a Sushi Meal
Fish frozen to FDA or EU parameters leaves a small residual risk. The larger risk sits with fish whose freezing nobody can document.
The ceiling for unfrozen fish is measurable. A 2017 risk model put the chance of anisakiasis at 9.56 per 100,000 untreated raw or marinated anchovy meals in Spain (Bao et al., Scientific Reports).
Live larvae drive that figure, and frozen fish carries dead ones, so a documented frozen meal falls below it.
The trade-off is access, since verifying freezing means asking questions and passing on cheap raw fish where nobody can answer them. This position reflects FDA and EU rules verified in October 2026, and a revised EFSA opinion or a change to the Food Code’s 7-day hold would change it.
The same checks apply when deciding whether kids can eat sushi.
