Sunfish: a lean panfish that nobody has studied on its own

In an open-label randomized trial on 34 Chinese adults with fatty liver disease, 84 days on a diet built on freshwater fish cut liver fat by 4.89 percentage points, against 1.83 when fish alternated with red meat. That is the strongest result near this fish, and it used Chinese freshwater species with no fish-free comparison. We found no study of sunfish or bluegill as food on any human outcome. What the USDA tables show is a lean fish. Per 100 g cooked, pumpkinseed sunfish has 24.9 g protein, 114 calories and 0.9 g fat, which leaves about 0.14 g of EPA plus DHA, from a single record with no sample spread. In a cross-sectional survey of 243 moderate freshwater fish eaters, the amount of locally caught fish showed no link with blood omega-3. Only fatty fish, salmonids in particular, tracked it, and the local species were not named.

The numbers, per 3.5 oz (100 g)

How much is actually in it

• Cooked pumpkinseed sunfish carries 24.9 g protein and 114 calories per 3.5 oz (100 g) with under 1 g fat, so only about 0.14 g EPA plus DHA. — USDA reference sample, cooked dry heat (n = 1 food record) Evidence E source
  Per 100 g cooked: protein 24.87 g, energy 114 calories, fat 0.9 g, EPA 0.047 g, DHA 0.092 g, vitamin B12 2.31 ug, selenium 16.2 ug, potassium 449 mg, sodium 103 mg
• A Tanzanian lake-fish eating population had systolic blood pressure of 120 against 135 mmHg in a vegetarian farming population, and its lake fish carried more DHA than EPA. — 1236 adults in two Tanzanian populations, lake-fish diet vs mainly vegetarian (n = 1236) Evidence C source
  SBP 120 vs 135 mmHg, DBP 70 vs 78 mmHg; lake fish DHA 8-19%, EPA 1.8-4.2% of fatty acids

What your body absorbs

• In 243 moderate freshwater fish eaters, the amount of locally caught fish eaten showed no link with blood omega-3, which tracked only fatty fish such as salmonids. — 243 moderate consumers of freshwater fish, mean intake 58 g/d (n = 243) Evidence C source
  No relation between locally caught fish intake and serum n-3; fatty fish intake vs serum EPA+DHA R2 = 0.41

What cooking and storage change

• By the USDA retention table for lean fish under 5% fat, baking or frying without coating keeps about 90% of vitamin B12 and folate and 85 to 90% of thiamin. — Finfish and Shellfish Products group, USDA R6 (n = 2 cooking methods) Evidence E source
  Baked without drippings: B12 90%, folate 90%, thiamin 90%, riboflavin 95%; fried without coating: B12 90%, folate 90%, thiamin 85%
• Lab cooking experiments pooled from 10 studies of seafood and freshwater fish found cooking cut PFAS concentrations by 29% on average, with very high variation. — 10 experimental cooking studies, seafood and freshwater fish, 512 effect sizes (n = 10 studies) Evidence D source
  Mean PFAS reduction 29% raw to cooked; I2 94.65%; longer cooking and more liquid gave larger reductions

What it does to your health

• In an RCT on 100 male students, eating 3.8 freshwater fish meals a week for 12 weeks lowered serum triglycerides and apolipoprotein B, with no change in cholesterol. — 100 healthy male students, Finland, 12 weeks (n = 100) Evidence B source
  Significant fall in serum triglyceride and apoB at 3.8 meals/week; trend only at 1.5 and 2.3 meals/week; cholesterol unchanged
• In an 84-day RCT on 34 adults with fatty liver disease, a freshwater fish based diet cut liver fat by 4.89 points against 1.83 with fish alternating with red meat. — 34 Chinese adults with NAFLD, hepatic steatosis at least 10%, 84 days (n = 34) Evidence B source
  Absolute hepatic steatosis change -4.89% vs -1.83%, P = 0.032

Safety, limits and interactions

• Montreal sport fishers who ate their catch most often had about three times the hair mercury of infrequent eaters, 0.73 against 0.23 ug/g, still under Health Canada limits. — 40 St. Lawrence River ice fishers, high vs low estimated exposure (n = 40) Evidence C source
  Hair mercury median 0.73 vs 0.23 ug/g; 40 vs 7.3 lb (18.3 vs 3.3 kg) sport fish a year; 2 of 25 high consumers above PCB limit
• For fish caught by family or friends, FDA tells pregnant or breastfeeding women and young children to check local advisories, and without one to eat only one serving and no other fish that week. — people who might become or are pregnant or breastfeeding, children 1-11 years Evidence E source
  No advisory: one serving of locally caught fish and no other fish that week

Who this works differently for

• In a cohort of 1,701 pregnant Chinese women, moderate freshwater fish intake of 17.9 to 30 g a day was linked with half the odds of a small-for-gestational-age baby. — 1701 pregnant women, Tongji Birth cohort, Wuhan 2018-2021 (n = 1701) Evidence C source
  OR 0.50 (95% CI 0.25-0.96) for fourth vs lowest quintile; nonlinear; n-3 intake itself not associated

The bottom line

An older randomized trial on 100 healthy male students in Finland found that 3.8 freshwater fish meals a week for 12 weeks significantly lowered serum triglycerides and apolipoprotein B and left cholesterol unchanged. At 1.5 and 2.3 meals a week there was only a trend. The abstract gives neither the species nor the size of the fall. A comparison of two Tanzanian populations, 1,236 adults in all, found blood pressure of 120/70 mmHg among lake-fish eaters against 135/78 in a mainly vegetarian farming group. The two groups differed in many ways besides fish, so this is an association. Cooking a lean fish keeps most of its B vitamins. The USDA retention table for finfish under 5 percent fat, set on the whole group in 1975 and never on sunfish, has baking and frying without coating keeping about 90 percent of vitamin B12 and folate. Lab experiments pooled from 10 studies found cooking cut PFAS in seafood and freshwater fish by 29 percent on average, with very high variation between them. Those were food chemistry tests with no people involved. Who should be careful. If you are pregnant, might become pregnant, breastfeed, or feed a child aged 1 to 11, FDA advises checking local advisories for fish caught by family or friends. Where there is no advisory, eat one serving and no other fish that week. The FDA page does not name sunfish. Among 40 ice fishers on the St. Lawrence River near Montreal, those who ate the most of their catch, 40 lb (18.3 kg) a year, had a median hair mercury of 0.73 micrograms per gram against 0.23 in light eaters. That stayed under Health Canada limits, but 2 of 25 heavy eaters were above the limit for PCBs, in a small pilot on one river with the species mix not given. We found no measurements of mercury or PFAS in sunfish themselves, and no study of sunfish allergy. In a cohort of 1,701 pregnant women in Wuhan, 0.63 to 1.1 oz (17.9 to 30 g) of freshwater fish a day was linked with half the odds of a small-for-gestational-age baby (odds ratio 0.50, 95% CI 0.25 to 0.96). That is an association resting on one food questionnaire in mid-pregnancy and on Chinese species, and the women's omega-3 intake itself showed no link. It does not override the FDA advice above on locally caught fish.

Sources

1. snfs-r2-01 · reference dataset
   FDC 172003 Fish, sunfish, pumpkin seed, cooked, dry heat, per 100 g: Protein 24.87 g; Energy 114 kcal; Total lipid (fat) 0.9 g; PUFA 20:5 n-3 (EPA) 0.047 g; PUFA 22:6 n-3 (DHA) 0.092 g; Vitamin B-12 2.31 ug; Selenium 16.2 ug; Potassium 449 mg; Sodium 103 mg
   USDA FoodData Central SR Legacy 2018, FDC 172003 Fish, sunfish, pumpkin seed, cooked, dry heat · checked 2026-10-02
2. snfs-r2-07 · population comparison
   Lower blood pressure values were found in the FD subjects (n = 618) vs. VD (n = 618) (systolic blood pressure, SBP, 120 +/- 15 vs. 135 +/- 20, P < 0.01; diastolic blood pressure, DBP, 70 +/- 9 vs. 78 +/- 11, P < 0.01, respectively). ... In fact, in four main species of lake fish, DHA was 8-19%, higher than EPA (1.8-4.2%), in contrast with the situation in cold-water fish, and AA was 5.8-8%, higher than in cold-water fish.
   Lipids, 1996 · checked 2026-10-02
3. snfs-r2-04 · cross-sectional
   Only fatty fish intake, particularly salmonid, and estimated EPA + DHA intake from fatty fish were significantly associated with serum EPA + DHA (R2 = 0.41 and 0.40, respectively). No relation was observed between the quantity of locally caught fish (g/d) consumed or the estimated FA intake from locally caught fish and serum n-3 FAs.
   Am J Clin Nutr, 2006 · checked 2026-10-02
4. snfs-r2-02 · USDA Table of Nutrient Retention Factors, dataset
   USDA-RETN-R6:2301 FINFISH,<5%FAT,BAKED,WO/DRIPPINGS, group Finfish and Shellfish Products: vitamin_b12_ug 90% retained, folate_total_ug 90%, thiamin_mg 90%, riboflavin_mg 95%; USDA-RETN-R6:2309 FINFISH,<5%FAT,FRIED,WO/COATING: vitamin_b12_ug 90% retained, folate_total_ug 90%, thiamin_mg 85%
   USDA Table of Nutrient Retention Factors, Release 6, 2007, Retn_Codes 2301 and 2309 · checked 2026-10-02
5. snfs-r2-03 · meta-analysis of cooking experiments
   These studies exclusively explored changes in PFAS concentrations in cooked seafood and freshwater fish. Our multilevel-meta-analysis has revealed that, on average, cooking reduced PFAS concentrations by 29%, although heterogeneity among effect sizes was very high (I2 = 94.65%).
   Environ Pollut, 2022 · checked 2026-10-02
6. snfs-r2-05 · RCT
   One hundred male students took part and were randomly divided into control and four fish diet groups with different fish diets. ... The serum triglyceride and apolipoprotein B concentration fell significantly in the group eating 3.8 fish containing meals a week for 12 weeks. At two lower doses (1.5 and 2.3 meals/week) the tendency to lower values was already seen. No significant changes were observed in the serum cholesterol, apolipoprotein A1 and vitamin E and A concentrations.
   Ann Med, 1989 · checked 2026-10-02
7. snfs-r2-06 · RCT
   In this open-label, 84-day randomized controlled trial, 34 NAFLD patients with hepatic steatosis ≥10% were randomly assigned to the F group or F/M group in a 1:1 ratio using a computer-generated random number allocation by a researcher not involved in the study. ... At the end of intervention, the absolute reduction of hepatic steatosis was significantly greater in the F group than in the F/M group (-4.89% vs -1.83%, P = 0.032).
   Am J Gastroenterol, 2022 · checked 2026-10-02
8. snfs-r2-08 · exposure pilot study
   Tissue contaminant assessments showed significant (P<0.05) groupwise differences: 0-1 cm hair mercury (median 0.73 microgram/g for the high versus 0.23 microgram/g for the low group), lipid-adjusted plasma PCB congeners (Aroclor 1260: median 0.77 microgram/g versus 0.47 microgram/g), and lipid-adjusted plasma DDE (median 0.35 microgram/g versus 0.26 microgram/g). No participant had a hair mercury or plasma DDE concentration above Health Canada recommendations but 2/25 high-level participants (8%) had plasma Aroclor 1260 concentrations above recommended limits.
   Environ Res, 1999 · checked 2026-10-02
9. snfs-r2-09 · agency advice
   Check for fish and shellfish advisories to tell you how often you can safely eat those fish. If there is no advisory, eat only one serving and no other fish that week.
   U.S. Food and Drug Administration, Advice about Eating Fish, 2024 · checked 2026-10-02
10. snfs-r2-10 · cohort
    Compared with the lowest quintile (0-3.2 g/d), the multivariable-adjusted odds ratio for women in the fourth quintile of freshwater fish intake (17.9 to 30.0 g/d) was 0.50 (95% CI 0.25 to 0.96). We found a nonlinear association between freshwater fish intake and SGA risk (Pnonlinearity = .027). However, maternal n-3 PUFAs intake was not significantly associated with SGA risk, either from total intake or from dietary sources alone.
    J Acad Nutr Diet, 2022 · checked 2026-10-02