Blueberry pie: how much of the berry survives the oven

The best blueberry evidence is about the berry, and no study we found fed anyone pie. A meta-analysis of 14 randomized trials in 632 participants found blueberries raised flow-mediated dilation, a measure of how well arteries widen, by 1.11% (95% CI 0.50 to 1.72). Blood pressure did not change. Doses and forms varied across those trials. Baking takes part of the berry away and, in one small short trial, seemed to leave that one effect. In 10 healthy volunteers, baked goods made with blueberry powder, buns rather than pie, had 42% less anthocyanin than an unprocessed blueberry drink and raised flow-mediated dilation to a similar degree over 6 hours. In a lab study of blueberry-filled pastries, about half the filling's anthocyanins were lost, mostly while the filling was stir-fried, a step common in Chinese pastry making. A regular slice is 150 g in US survey data, about 450 calories and 23 g total sugars, scaled from the per 3.5 oz (100 g) values.

The numbers, per 3.5 oz (100 g)

As people eat it, per 3.5 oz (100 g)

How much is actually in it

• Blueberry pie made from a recipe has about 11.9 g fat, 2.9 g saturated fat and 7.4 ug selenium per 3.5 oz (100 g) in USDA data, and the commercial pie has less of each. — USDA SR Legacy entries Pie, blueberry, prepared from recipe (FDC 172779) and commercially prepared (n = 2 records) Evidence E source
  Fat 10-11.9 g, saturated fat 1.679-2.911 g, selenium 1.4-7.4 ug per 3.5 oz (100 g) across the two records
• Ready blueberry pie filling carries 37.75 g total sugars and 181 calories per 3.5 oz (100 g) in USDA survey data, so a cup of filling holds about 98 g sugars. — USDA FNDDS Blueberry pie filling (fdc_id 2709278) (n = 2 measures) Evidence E source
  1 cup 260 g, about 98 g total sugars and 471 calories

What a real portion looks like

• A regular slice of blueberry pie is 150 g in the US survey database, which works out to about 450 calories and 23 g total sugars. — USDA FNDDS household measures for Pie, blueberry (53304000) (n = 9 measures) Evidence E source
  300 calories and 15.49 g total sugars per 3.5 oz (100 g); mini slice 75 g, regular slice 150 g, large slice 250 g, whole 8-9 inch pie 1200 g

What your body absorbs

• In 10 healthy volunteers, baked products with blueberry had 42% less anthocyanin than an unprocessed blueberry drink, yet raised flow-mediated dilation to a similar degree over 6 hours. — ten healthy volunteers (n = 10) Evidence B source
  Anthocyanins -42% in baked products; FMD increased at 1, 2 and 6 h similarly to unprocessed blueberries
• In 9 healthy adults eating 8.8 oz (250 g) blueberries, the whole fruit left 108 urinary metabolites higher than the same berries juiced, against 8 higher after juice. — Nine healthy subjects (n = 9) Evidence B source
  108 vs 8 ions significantly higher at 2 h

What cooking and storage change

• Lab data suggest making blueberry-filled pastries destroys about half of the filling's anthocyanins, mostly during stir-frying of the filling, which also formed 300 mg/kg HMF. — blueberry filling and pastries, laboratory processing Evidence D source
  Anthocyanins -48% to -53%, flavonols -11% to -16%; HMF 300 mg/kg after stir-frying

What it does to your health

• A meta-analysis of 14 RCTs in 632 participants found blueberries raised flow-mediated dilation by 1.11% but did not change blood pressure. — 14 randomized controlled trials involving 632 participants (n = 632) Evidence A source
  FMD +1.11% (95% CI 0.50 to 1.72); SBP -0.11 mmHg (95% CI -2.82 to 2.60)
• In people with cardiometabolic disease, a meta-analysis of 17 trials found blueberry or cranberry had no significant effect on blood pressure (systolic -0.81 mm Hg). — patients with cardiometabolic diseases, 17 articles (n = 17 studies) Evidence A source
  SBP -0.81 mm Hg (95% CI -2.26 to 0.63); DBP -0.15 mm Hg (95% CI -1.36 to 1.05)
• A meta-analysis of 12 blueberry RCTs found lower total and LDL cholesterol and diastolic pressure, with no change in glucose markers or body measures. — 18 randomized controlled trials reviewed, 12 pooled (n = 12 trials pooled) Evidence A source
  Direction reported, sizes not in abstract

Safety, limits and interactions

• Blueberry juice did not significantly change blood levels of buspirone (ratio 1.39) or flurbiprofen (1.03) in healthy volunteers, unlike grapefruit juice (2.12). — healthy volunteers Evidence C source
  AUC ratio buspirone 1.39 (NS) vs grapefruit juice 2.12; flurbiprofen 1.03
• A New Zealand hepatitis A outbreak was traced to raw blueberries, eaten by 56% of cases against 14% of controls (OR 7.6). — 39 cases and 71 controls, New Zealand (n = 110) Evidence C source
  OR 7.6 (95% CI 2.6-22.4)

What people believe that the data does not support

• A meta-analysis pooling 6 of 14 blueberry trials found no change in mood or attention and a small gain only in two-back test accuracy (WMD 0.08). — Fourteen randomized trials, six pooled (n = 6 trials pooled) Evidence A source
  Mood WMD 0.03 (95% CI -0.80 to 0.87); two-back accuracy WMD 0.08 (95% CI 0.02 to 0.13)

Who this works differently for

• In 61 breastfed infants given up to 10 g a day of freeze-dried blueberry powder from 5 to 12 months, allergy-related symptoms resolved more often and serum IL-13 fell versus placebo. — breastfed U.S. infants, 5 to 12 months of age; 61 completed (n = 61) Evidence B source
  Different symptom trajectories (p = 0.05); IL-13 reduced (p = 0.035)
• Observational data from over 86,000 US adults aged 60 and over found higher blueberry intake, among other flavonoid foods, tended to go with lower risks of frailty, impaired function and poor mental health. — 62,743 females and 23,687 males, all aged 60 and over (n = 86430) Evidence C source
  Highest vs lowest flavodiet score: frailty HR 0.85 (95% CI 0.80-0.90) in women; blueberries tended to lower risk

The bottom line

A pie made from a recipe has about 11.9 g fat and 2.9 g saturated fat per 3.5 oz (100 g) in USDA data, and the commercial pie has less of each, from single records with no spread across samples. Ready-made filling is the sweetest part. It carries 37.75 g total sugars per 3.5 oz (100 g), so a 260 g cup holds about 98 g. These are total sugars, and the tables do not split out added sugar. The same lab study found stir-frying the filling formed 300 mg/kg of HMF, a compound made when sugar is heated. The rest of the berry evidence is mixed. A meta-analysis of 12 blueberry trials reported lower total and LDL cholesterol and diastolic pressure, with no change in glucose markers or body measures, and gave no sizes in its abstract. In people with cardiometabolic disease, 17 trials of blueberry or cranberry found no significant effect on blood pressure, systolic -0.81 mm Hg (95% CI -2.26 to 0.63). The brain claim is weak. Pooling 6 of 14 trials found no change in mood or attention and a small gain in one working memory test, WMD 0.08 (95% CI 0.02 to 0.13). In 9 healthy adults, whole blueberries left 108 urinary metabolites higher than the same berries juiced, against 8 higher after juice, at a single 2 hour point. Two cards concern safety, and neither is about pie. Blueberry juice did not significantly change blood levels of buspirone or flurbiprofen in healthy volunteers, unlike grapefruit juice, in a study that did not report its size. A New Zealand hepatitis A outbreak was traced to raw blueberries, eaten by 56% of cases against 14% of controls (OR 7.6). The berries were raw, and that study does not say what baking does to the virus. We found no data on pregnancy or allergy. Two groups were studied with blueberry itself. In 61 breastfed infants given up to 10 g a day of freeze-dried blueberry powder from 5 to 12 months, parent-reported allergy symptoms resolved more often and serum IL-13 fell compared with placebo. Among 86,430 US adults aged 60 and over, blueberries were one of several flavonoid foods that tended to go with lower risks of frailty and poor mental health, an association that was not measured for blueberries on their own.

Sources

1. bbpi-daily-r1-01 · food composition dataset
   Pie, blueberry, prepared from recipe (FDC 172779, SR Legacy): fat 11.9 g, saturated fat 2.911 g, selenium 7.4 ug per 100 g; across 2 records (recipe, commercial): fat 10-11.9 g, saturated fat 1.679-2.911 g, selenium 1.4-7.4 ug
   USDA FoodData Central, SR Legacy, FDC 172779 (Pie, blueberry, prepared from recipe) · checked 2026-09-28
2. bbpi-daily-r1-03 · reference dataset
   Blueberry pie filling, FNDDS fdc_id 2709278: Total Sugars 37.75 g per 100 g | Energy 181 kcal per 100 g | 1 cup 260 g | Quantity not specified 130 g
   USDA FNDDS 2021-2023, Blueberry pie filling, fdc_id 2709278 · checked 2026-09-28
3. bbpi-daily-r1-02 · reference dataset
   Pie, blueberry, FNDDS 53304000: 1 mini/small slice 75 g | 1 regular slice 150 g | 1 large slice 250 g | 1 individual pie 150 g | 1 regular pie (8-9" dia) 1200 g | 1 surface inch 19 g || Energy 300 kcal per 100 g | Total Sugars 15.49 g per 100 g
   USDA FNDDS 2021-2023, Pie, blueberry, food code 53304000, fdc_id 2707998 · checked 2026-09-28
4. bbpi-daily-r1-04 · randomized crossover trial
   In a randomized, controlled crossover trial, ten healthy volunteers consumed (a) blueberry-containing baked products, (b) an unprocessed blueberry drink containing the same amount of freeze-dried blueberry powder as used in the baked products, and (c) matched control baked products. ... Although processing did not significantly change the total (poly)phenolic amount, the processed products contained significantly less anthocyanins (-42%), more chlorogenic acid (23%), no flavanol nonamers or decamers, and significantly more flavanol dimers and trimers (36% and 28%, respectively). FMD increased after 1, 2, and 6 h consumption of the baked products to a similar degree as the unprocessed blueberries
   Mol Nutr Food Res, 2014 · checked 2026-09-28
5. bbpi-daily-r1-06 · randomized crossover trial
   Nine healthy subjects consumed 250 g of fresh blueberries either as the whole fruit or after juicing, and provided blood and urine samples before and 2 h after intake in a cross-over design. ... In urinary samples, whole blueberry treatment resulted in 108 ions that were significantly higher compared with juiced treatment (positive and negative mode combined), whereas only eight were significantly higher after juiced treatment.
   Br J Nutr, 2018 · checked 2026-09-28
6. bbpi-daily-r1-05 · food chemistry
   Anthocyanins (48%-53% reduction in total) were more susceptible to thermal processing than flavonols (11%-16%). ... A high level of HMF (300 mg/kg) was formed during stir-frying. Except for anthocyanins, baking did not significantly influence HMF formation, flavonol degradation, and antioxidant property in the fillings.
   Yang N et al. Influences of stir-frying and baking on flavonoid profile, antioxidant property, and hydroxymethylfurfural formation during preparation of blueberry-filled pastries. Food Chem, 2019 · checked 2026-09-28
7. bbpi-daily-r1-07 · meta-analysis of RCTs
   we analyzed 14 randomized controlled trials (RCTs) involving 632 participants. Results indicated that blueberry intervention significantly improved flow-mediated dilation (FMD) by 1.11% (95% CI: 0.50, 1.72; I² = 23%), suggesting a positive impact on vascular health. However, there were no significant changes in systolic blood pressure (SBP) (reduction of 0.11 mmHg; 95% CI: -2.82, 2.60; I² = 0%)
   Nutr Res, 2026 · checked 2026-09-28
8. bbpi-daily-r1-08 · meta-analysis of RCTs
   A total of 17 articles were included, from which two found significant results from blueberry and/or cranberry supplementation in reducing BP. Pooled results revealed statistically non-significant reductions of -0.81 mm Hg for SBP (95% confidence interval [CI]: -2.26, 0.63; I2 = 0%) and -0.15 mm Hg for DBP (95% CI: -1.36, 1.05; I2 = 27%).
   Phytother Res, 2024 · checked 2026-09-28
9. bbpi-daily-r1-10 · meta-analysis of RCTs
   In total, 18 randomized controlled trials (RCTs) were included in this systematic review, and 12 studies were selected for meta-analysis. ... Meta-analyses of the data showed that blueberry intervention had a significant effect on lipid levels, decreasing total cholesterol and low-density lipoprotein (LDL) levels. We found no significant differences in the glycemic status markers and anthropometric measurements. Blueberry supplementation significantly decreased diastolic blood pressure.
   Nutr Res, 2021 · checked 2026-09-28
10. bbpi-daily-r1-11 · drug interaction study
    In clinical studies, GFJ significantly increased area under the plasma concentration-time curve (AUC) for the CYP3A substrate buspirone. The geometric mean ratio (GMR = AUC with GFJ divided by AUC with water) was 2.12. In contrast, the effect of BBJ (GMR = 1.39) was not significant. In the study of flurbiprofen (CYP2C9 substrate), the positive control inhibitor fluconazole significantly increased flurbiprofen AUC (GMR = 1.71), but BBJ had no significant effect (GMR = 1.03).
    Br J Clin Pharmacol, 2013 · checked 2026-09-28
11. bbpi-daily-r1-12 · case-control
    A case-control study revealed that 56% of 39 cases had consumed raw blueberries, compared with 14% of 71 controls (odds ratio 7.6; 95% confidence intervals 2.6-22.4).
    Epidemiol Infect, 2003 · checked 2026-09-28
12. bbpi-daily-r1-09 · meta-analysis of RCTs
    Fourteenrandomized trials were included in the quantitative analysis, and six were pooled for statistical analysis. Blueberry intervention resulted in no significant change in mood state score (WMD = 0.03; 95% CI: -0.80 to 0.87, P = 0.16). ... Significant effect was indicated (WMD = 0.08; 95% CI: 0.02 to 0.13, P = 0.005) in two-back test accuracy.
    CNS Neurol Disord Drug Targets, 2023 · checked 2026-09-28
13. bbpi-daily-r1-13 · randomized controlled trial
    In a double-blind, randomized, placebo-controlled trial, infants from the Denver metro area were assigned to receive up to 10 g/day of freeze-dried blueberry powder or an isocaloric placebo from 5 to 12 months of age. ... Of the 76 infants enrolled, 61 completed the study (Blueberry: n = 30; Placebo: n = 31). While more infants in the blueberry group had allergy-related symptoms at baseline, they had significantly different longitudinal symptom trajectories than the placebo (p = 0.05), showing a greater resolution rate of symptoms by study end. Pro-inflammatory serum IL-13 levels were significantly reduced (p = 0.035)
    Nutrients, 2025 · checked 2026-09-28
14. bbpi-daily-r1-14 · prospective cohort
    We followed up 62,743 females and 23,687 males, all aged ≥60 y, from the Nurses' Health Study (1990-2014) and Health Professionals Follow-up Study (2006-2018), respectively. ... Increases in flavodiet scores and both higher intakes and increases in intakes of tea, red wine, apples, blueberries, and oranges tended to be associated with lower risks of all outcomes.
    Am J Clin Nutr, 2025 · checked 2026-09-28