Oranges: the fruit is vitamin C, the evidence is juice

The reputation is earned. Raw navel flesh carries 59.1 mg of vitamin C per 100 g, 66 percent of the Daily Value, and the five USDA records for oranges run from 45 to 71 mg. Almost everything measured beyond the composition was measured on the juice. A meta-analysis of 10 randomised trials found orange juice lowered fasting glucose by 2.92 mg/dl, total cholesterol by 9.84 mg/dl and LDL cholesterol by 9.14 mg/dl against control. Those effects held mainly in the trials where people drank 500 ml a day or more, which is two large glasses and not one, and body composition did not move. The intervals are wide: total cholesterol -15.43 to -4.24 mg/dl, LDL -15.79 to -2.49 mg/dl. A newer randomised crossover gave 330 ml a day to 37 centrally overweight men for six weeks and found nothing: no change in fasting or postprandial flow-mediated dilation, and none in any metabolic or anthropometric measure, against an isoenergetic control drink. Thirty-seven men over six weeks cannot rule out a small or slower effect, or one in women.

The numbers, per 100 g

What your body absorbs

• Daily 330 mL of 100% orange juice delivered citrus flavanones as well as a drink enriched with 210 mg of pure hesperidin, giving comparable plasma metabolite exposure and 24 hour urinary excretion in 37 overweight men. — 37 centrally overweight men, randomised double-blind three-period crossover, 6 weeks per period with washouts (n = 37) Evidence B source
  Phase II hesperetin metabolite AUC over 0 to 6 h and 24 h urinary excretion comparable after orange juice and after 210 mg/day hesperidin, indicating limited food matrix effect

What it does to your health

• Across 10 randomised trials pooled in 2022, orange juice lowered fasting glucose by 2.92 mg/dl, total cholesterol by 9.84 mg/dl and LDL cholesterol by 9.14 mg/dl compared with control, and those effects held mainly where people drank 500 ml a day or more. — 10 randomised controlled trials of orange juice in adults, four databases searched to September 2020 (n = 10 trials) Evidence A source
  Glucose -2.92 mg/dl (95% CI -5.327 to -0.530), insulin -1.229 uU/ml, HOMA-IR -0.464, total cholesterol -9.84 mg/dl (95% CI -15.43 to -4.24), LDL-C -9.14 mg/dl (95% CI -15.79 to -2.49), CRP -0.467 mg/l
• In the same crossover trial 6 weeks of 330 mL orange juice a day left flow-mediated dilation and every metabolic and anthropometric measure unchanged against an isoenergetic control drink. — 37 centrally overweight men, 6 weeks per intervention, flow-mediated dilation as primary endpoint (n = 37) Evidence B source
  No significant change in fasting or postprandial FMD, vascular, metabolic or anthropometric parameters, or oxylipin profiles versus control
• When 26 healthy adults took orange juice worth 20% of their energy needs for 4 weeks, drinking it between meals added 1.0 kg of fat mass, while the same juice taken with meals lowered fat mass by 0.30 kg. — 26 healthy adults, mean age 24.7 years, mean BMI 23.2, 4-week cross-over intervention at ad libitum energy intake (n = 26) Evidence C source
  Fat mass +1.0 kg (SD 1.8) drinking juice between meals versus -0.30 kg (SD 0.65) drinking it with meals, both p below 0.05

Safety, limits and interactions

• A 2025 meta-analysis of 51 human pharmacokinetic studies found grapefruit juice cut the AUC and Cmax of aliskiren and celiprolol by about 80 to 90%, and it reports apple and orange juices having comparable effects on certain medications. — 51 human pharmacokinetic studies of fruit juice with antihypertensive drugs, mostly open-label crossover trials (n = 51 studies) Evidence B source
  Grapefruit juice lowered aliskiren and celiprolol AUC and Cmax by roughly 80 to 90%, with orange and apple juice described as comparable on certain medications
• Pooling 21 in situ randomised trials on enamel, calcium-modified blackcurrant juice caused 2.6 times less enamel loss than orange juice (p 0.0001), and adding calcium to juice drinks reduced enamel loss overall. — 21 in situ randomised controlled trials on human enamel specimens, selected from 869 records (n = 21 trials) Evidence A source
  2.6 times less enamel loss with calcium-modified blackcurrant juice than with orange juice (p 0.0001, I2 89%)
• A 40-year-old man with normal kidneys reached a serum potassium of 9.4 mmol/L and needed intensive care after weeks of drinking large daily volumes of reconstituted dried orange juice powder. — one 40-year-old man with no known past medical history and normal renal function (n = 1) Evidence D source
  Serum potassium up to 9.4 mmol/L with peaked T waves and QRS widening, normalised after the drink was stopped

The bottom line

Only one trial has tested when in the day the juice is drunk, and it used far more juice than a glass. Twenty-six healthy adults, mean age 24.7 and mean BMI 23.2, took orange juice worth 20 percent of their energy needs for four weeks, roughly 600 ml a day. Drinking it between meals added 1.0 kg of fat mass, and the same juice with meals lowered fat mass by 0.30 kg, both p below 0.05. That was a controlled cross-over without stated randomisation, so it sits a level below the pooled juice trials. The juice does deliver its flavanones. In those 37 men, 330 ml a day gave plasma hesperetin metabolite exposure over six hours and 24 hour urinary excretion comparable to a drink enriched with 210 mg of pure hesperidin. That is metabolites reaching the blood, not evidence of anything happening in tissue, and exposure varied widely between individuals. Three cautions are real, and all three belong to the juice rather than the fruit. A 2025 meta-analysis of 51 human pharmacokinetic studies reports that apple and orange juices have comparable effects to grapefruit juice on certain medications. The 80 to 90 percent drop in aliskiren and celiprolol exposure that the review pools belongs to grapefruit, and no pooled figure for orange juice is given. If you take a blood pressure drug, the question of whether yours is affected goes to your prescriber. Pooling 21 in situ randomised trials on enamel specimens worn in the mouth, calcium-modified blackcurrant juice caused 2.6 times less enamel loss than orange juice (p 0.0001, heterogeneity 89 percent), which is specimens over weeks rather than teeth over years. A 40-year-old man with normal kidneys reached a serum potassium of 9.4 mmol/L, with peaked T waves and a widening QRS, and needed intensive care after weeks of drinking large daily volumes of dried orange juice powder reconstituted as a drink. One case, a concentrated powder, not fruit. We found no card on oranges in pregnancy, in children or in kidney disease, and that is evidence we did not find rather than reassurance.

Sources

1. oran-dr4-02 · randomised controlled crossover trial
   Plasma exposure to phase II hesperetin metabolites (AUC0-6 h) and 24 h urinary excretion were comparable after OJ and HESP, indicating effective hesperidin delivery and limited matrix effects on bioavailability.
   Food Funct, 2026 · checked 2026-09-24
2. oran-dr4-01 · meta-analysis of randomised controlled trials
   Pooled results demonstrated a significant effect of orange juice on glucose (WMD: -2.92 mg/dl, 95% CI: -5.327, -0.530, p = 0.017), insulin (WMD: -1.229 μU/ml, 95% CI: -2.083, -0.374, p = 0.005), HOMA-IR (WMD: -0.464, 95% CI: -0.747, -0.181, p = 0.001), total cholesterol (WMD: -9.84 mg/dl, 95% CI: -15.43, -4.24, p = 0.001), LDL-C (WMD: -9.14 mg/dl, 95% CI: -15.79, -2.49, p = 0.007), and CRP (WMD: -0.467 mg/l, 95% CI: -0.815, -0.120, p = 0.008) compared to control group.
   Crit Rev Food Sci Nutr, 2022 · checked 2026-09-24
3. oran-dr4-03 · randomised controlled crossover trial
   Neither intervention significantly affected fasting or postprandial FMD, vascular, metabolic or anthropometric parameters, or oxylipin profiles versus CON.
   Food Funct, 2026 · checked 2026-09-24
4. oran-dr4-06 · controlled clinical trial, cross-over design
   After BM-intervention, fat mass increased (+1.0 ± 1.8 kg; p < 0.05) and postprandial insulin sensitivity tended to decrease (ΔMatsudaISI: -0.89 ± 2.3; p = 0.06). By contrast, after WM-intervention fat mass and gamma-glutamyl transferase (GGT) decreased (-0.30 ± 0.65 kg; -2.50 ± 3.94; both p < 0.05), whereas glucose variability was higher (ΔMAGE: +0.45 ± 0.59, p < 0.05).
   Nutr Diabetes, 2018 · checked 2026-09-24
5. oran-dr4-04 · meta-analysis of crossover pharmacokinetic trials in humans
   Grapefruit juice (GFJ), an inhibitor of organic-anion-transporting polypeptide (OATP) transporters and cytochrome P450 (CYP) 3A4, significantly decreased the AUC and Cmax of aliskiren and celiprolol by approximately 80-90 %...Apple and orange juices have comparable effects on certain medications.
   Complement Ther Med, 2025 · checked 2026-09-24
6. oran-dr4-05 · meta-analysis of in situ randomised controlled trials
   The findings of the studies showed that the addition of calcium in juice drinks led to reduced enamel loss, with blackcurrant juice presenting 2.6 times statistically significant less enamel loss compared to orange juice (p = 0.0001, I2 = 89%).
   Evid Based Dent, 2024 · checked 2026-09-24
7. oran-dr4-07 · case report
   We report a case of a 40-year-old man with no known past medical history who presented with progressive neuromuscular symptoms, palpitations, and syncope and was found to have life-threatening hyperkalemia with serum potassium levels rising to 9.4 mmol/L...Further history revealed excessive daily consumption of large volumes of a dried orange juice powder reconstituted as a beverage over several weeks.
   Cureus, 2026 · checked 2026-09-24