Scientific opinion on the tolerable upper intake level for preformed vitamin A and β‐carotene

View full-text article in PMC

. 2024 Jun 6;22(6):e8814. doi: 10.2903/j.efsa.2024.8814

Reference Study Country Duration Funding	Design	Subject characteristics at baseline	Intervention	Endpoint assessed	Results
Farhangi et al. (2013) Iran 4 months Funding: public	RCT (parallel) Inclusion criteria: 20–52 years of age, BMI 30–39.9 kg/m2 for obese and 18.5–24.9 kg/m2 for non‐obese women. Exclusion criteria: history of diabetes, thyroid abnormalities, liver or renal disease and autoimmune disease, consumption of any dietary vitamin A supplements or treatment with drugs that may interfere with absorption or bioavailability of the supplement. N randomised = 84 N completed = 75 G1: N intervention (obese) = 27 G2: N control (obese) = 23 G3: N intervention (normal weight) = 25	Sex (% female): 100 Age (range): 20–52 years Background Vitamin A intake at baseline, mean (SEM): G1: 566 (150) G2: 444 (232) G3: 399 (173)	Vitamin A as retinyl palmitate Doses (IU/day): G1: 25,000 (~7500 μg RE/day) G2: 0 G3: 25,000 (~7500 μg RE/day) Background nutrient intake (RE/day) at end‐of‐study, mean (SEM): G1: 637 (234) G2: 674 (325) G3: 535 (101) Exposure assessment: three‐day 24‐h dietary recall questionnaire at baseline and end. Compliance: NR	Serum AST and ALT Determined by enzymatic methods. Mean inter‐ and intra‐assay CV for these tests were 4.40, 3.25 and 3.08, 6.22, respectively.	Type of analysis extracted: PP Units: U/L Values: mean (SEM) AST at baseline G1: 20.25 (0.86) G2: 22.46 (1.48) G3: 19.07 (1.47) P = NS AST at end‐of‐study G1: 23.60 (1.11) (P = 0.008) G2: 22.84 (0.76) (P = NS) G3: 26.26 (2.60) (P = 0.001) P = NS ALT at baseline G1: 17.64 (1.49) G2: 17.82 (1.40) G3: 15.75 (1.84) P = NS ALT at end‐of‐study G1: 17.08 (1.80) (P = NS) G2: 16.00 (1.05) (P = NS) G3: 17.73 (3.18) (P = NS) P = NS % change in AST G1: 16.72 (8.75) G2: 2.25 (8.80) G3: 37.06 (11.66) P = 0.036 % change in ALT G1: −3.68 (11.85) G2: −6.19 (7.06) G3: 13.28 (10.11) P = NS
Dougherty et al. (2012) USA 12 months Funding: Public	RCT (parallel) Inclusion criteria: Subjects with SCD‐SS aged 2.0–12.9 years. Exclusion criteria: Chronic transfusion therapy or a transfusion within the past 2 months, hydroxyurea therapy, history of stroke, liver enzymes > 3 times the reference range, height > 2.0 SDs above the age and sex mean (> 98th percentile, CDC 2000 reference standards), participation in another intervention study, pregnancy, and other chronic conditions known to affect growth, dietary intake or nutritional status. In addition, subjects taking daily vitamins or commercial nutritional supplements containing vitamin A were not eligible for the study, unless willing to discontinue supplementation and have a 2‐months washout period. N participants with sub‐optimal vitamin A status randomised/completed/analysed: G1 (vitamin A): 23/18/18 G2 (vitamin A + zinc): 18/15/15 G3 (placebo): 21/19/19	Sex at baseline (as % female) G1: 39 G2: 39 G3: 52 Age at baseline, years, mean (SD) G1: 7.5 (2.9) G2: 7.6 (2.4) G3: 7.8 (3.2) Serum retinol at baseline, μg/dL G1: 17.6 (4.0) G2: 18.2 (3.3) G3: 19.4 (4.9) Serum retinol at end‐of‐study, μg/dL G1: 18.9 (4.5) G2: 19.4 (3.2) G3: 19.0 (4.3)	Vitamin A as retinyl palmitate Vitamin A groups received dose by age: 2.0–3.9 years: 300 μg/day (164 μg RE/day) 4.0–8.9 years: 400 μg/day (218 μg RE/day) 9.0–12.9 years: 600 μg/day (328 μg RE/day) Background nutrient intake, total vitamin A (IU/day), median (range) by serum retinol status: All (N = 96): 1876 (159, 14,729) Retinol ≥ 30 μg/dL (n = 26): 1748 (159, 4362) Retinol 20–29 μg/dL (N = 48): 2100 (481, 14,729) Retinol <20 μg/dL (N = 22): 1689 (317, 4359) Exposure assessment: 24‐h recall at screening and three 24‐h recalls during the supplementation study at baseline and 3 and 12 months. Compliance: 64% took suppl ≥85% of the time; 36% took suppl <85% of the time	Serum gamma‐glutamyl transferase (gGT) gGT was measured by the Clinical Chemistry Laboratory at CHOP	Type of analysis extracted: PP Units: U/L, median (range) gGT at baseline G1: 20 (8, 50) G2: 23 (11, 43) G3: 21 (8, 68) P = NR gGT at end‐of‐study G1: 23 (11, 63) G2: 21 (11, 62) G3: 19 (12, 49) P = NR No significant group differences over time (change score) were detected.
Alberts et al. (2004) USA 12 months Funding: Public	RCT (parallel) Inclusion criteria: Moderate to severe sun damage with or without actinic keratoses on posterior forearms, and potentially experienced nonmelanoma skin cancer, minimum 50 years of age, postmenopausal (for women). Exclusion criteria: History of invasive cancers, CVD, stroke or other serious disease. Participants randomised: N = 129 Participants completed: N = 116 Participants analysed: N = 129 (ITT) N participants randomised/completed/analysed G1: 31/30/31 G2: 32/29/32 G3: 33/27/33 G4: 33/30/33	Sex (as % female) G1: 42% G2: 38% G3: 33% G4: 33% Age males, years (mean) G1: 63.6 G2: 62.9 G3: 64.0 G4: 64.5 Age females, years (mean) G1: 64.3 G2: 61.2 G3: 61.9 G4: 64.3	Vitamin A as retinyl palmitate Doses in IU/day: G1: Placebo G2: 25,000 (~7500 μg RE/day) G3: 50,000 (~15,000 μg RE/day) G4: 75,000 (~22,500 μg RE/day) Background nutrient intake: NR Compliance: NR	AST, ALT, ALP Analysis method: NR Liver scan performed at baseline. According to protocol: second scan for subjects with ASAT or ALAT > 3x upper normal limit, but this applied to none.	Type of analysis extracted: ITT AST, N subjects who experienced toxicity G1: 0 G2: 0 G3: 0 G4: 1 (3%) ALT, N subjects who experienced toxicity G1: 0 G2: 0 G3: 1 (3% G4: 1 (3%) ALP, N subjects who experienced toxicity G1: 0 G2: 0 G3: 0 G4: 0
Bitarafan et al. (2015) Iran 12 months Funding: Public	RCT (Parallel) Inclusion: Multiple sclerosis patients in relapsing–remitting phase, using Interferon beta‐1a (Avonex) as a treatment, aged 20–45 years old. Exclusion criteria: any addiction, alcohol intake, dysphagia, history of myocardial infarction, stroke, other autoimmune diseases, hypersensitivity to vitamin A compounds, liver, pancreatic and biliary disorders N participants randomised/completed: 101/93 G1: 51/47 G2: 50/46	Sex: 74% female Age (range): 20–45 years (mean, SD): G1: 30.4 ± 6.9 G2: 32.3 ± 5.9 Other background variables: BMI (kg/m ²) (mean, SD) G1: 23.9 ± 3.1 G2: 24.5 ± 4.3 Range: 18.5–30 Dietary vitamin A intake (μg/day) (mean, SD): G1: 737.4 ± 483.78 G2: 744.8 ± 541.91	Vitamin A as retinyl palmitate Doses (IU/day) G1: 25,000 (~7500 μg RE for the first 6 months, 10,000 IU/~3000 μg RE for the second 6 months) G2: 0 (placebo) Background vitamin A intake, change (μg/day): G1: −11.2 ± 68.04 G2: +17.8 ± 90.85 (p for change between groups = 0.08) Compliance: NR.	ALT, AST Determined with colorimetric method	Type of analysis: NR ALT, change (mean, SD) G1: 0.57 ± 9.44 G2: −0.85 ± 7.91 p = 0.53 AST, change (mean, SD) G1: 0.15 ± 1.85 G2: 0.24 ± 1.48 p = 0.99

Reference

Study

Country

Duration

Funding

Design

Subject characteristics at baseline

Intervention

Endpoint assessed

Results

Farhangi et al. (2013)

Iran

4 months

Funding: public

RCT (parallel)

Inclusion criteria: 20–52 years of age, BMI 30–39.9 kg/m2 for obese and 18.5–24.9 kg/m2 for non‐obese women.

Exclusion criteria: history of diabetes, thyroid abnormalities, liver or renal disease and autoimmune disease, consumption of any dietary vitamin A supplements or treatment with drugs that may interfere with absorption or bioavailability of the supplement.

N randomised = 84

N completed = 75

G1: N intervention (obese) = 27

G2: N control (obese) = 23

G3: N intervention (normal weight) = 25

Sex (% female): 100

Age (range): 20–52 years

Background Vitamin A intake at baseline, mean (SEM):

G1: 566 (150)

G2: 444 (232)

G3: 399 (173)

Vitamin A as retinyl palmitate

Doses (IU/day):

G1: 25,000 (~7500 μg RE/day)

G2: 0

G3: 25,000 (~7500 μg RE/day)

Background nutrient intake (RE/day) at end‐of‐study, mean (SEM):

G1: 637 (234)

G2: 674 (325)

G3: 535 (101)

Exposure assessment: three‐day 24‐h dietary recall questionnaire at baseline and end.

Compliance: NR

Serum AST and ALT

Determined by enzymatic methods.

Mean inter‐ and intra‐assay CV for these tests were 4.40, 3.25 and 3.08, 6.22, respectively.

Type of analysis extracted: PP

Units: U/L

Values: mean (SEM)

AST at baseline

G1: 20.25 (0.86)

G2: 22.46 (1.48)

G3: 19.07 (1.47)

P = NS

AST at end‐of‐study

G1: 23.60 (1.11) (P = 0.008)

G2: 22.84 (0.76) (P = NS)

G3: 26.26 (2.60) (P = 0.001)

P = NS

ALT at baseline

G1: 17.64 (1.49)

G2: 17.82 (1.40)

G3: 15.75 (1.84)

P = NS

ALT at end‐of‐study

G1: 17.08 (1.80) (P = NS)

G2: 16.00 (1.05) (P = NS)

G3: 17.73 (3.18) (P = NS)

P = NS

% change in AST

G1: 16.72 (8.75)

G2: 2.25 (8.80)

G3: 37.06 (11.66)

P = 0.036

% change in ALT

G1: −3.68 (11.85)

G2: −6.19 (7.06)

G3: 13.28 (10.11)

P = NS

Dougherty et al. (2012)

USA

12 months

Funding: Public

RCT (parallel)

Inclusion criteria: Subjects with SCD‐SS aged 2.0–12.9 years.

Exclusion criteria: Chronic transfusion therapy or a transfusion within the past 2 months, hydroxyurea therapy, history of stroke, liver enzymes > 3 times the reference range, height > 2.0 SDs above the age and sex mean (> 98th percentile, CDC 2000 reference standards), participation in another intervention study, pregnancy, and other chronic conditions known to affect growth, dietary intake or nutritional status. In addition, subjects taking daily vitamins or commercial nutritional supplements containing vitamin A were not eligible for the study, unless willing to discontinue supplementation and have a 2‐months washout period.

N participants with sub‐optimal vitamin A status randomised/completed/analysed:

G1 (vitamin A): 23/18/18

G2 (vitamin A + zinc): 18/15/15

G3 (placebo): 21/19/19

Sex at baseline (as % female)

G1: 39

G2: 39

G3: 52

Age at baseline, years, mean (SD)

G1: 7.5 (2.9)

G2: 7.6 (2.4)

G3: 7.8 (3.2)

Serum retinol at baseline, μg/dL

G1: 17.6 (4.0)

G2: 18.2 (3.3)

G3: 19.4 (4.9)

Serum retinol at end‐of‐study, μg/dL

G1: 18.9 (4.5)

G2: 19.4 (3.2)

G3: 19.0 (4.3)

Vitamin A as retinyl palmitate

Vitamin A groups received dose by age:

2.0–3.9 years: 300 μg/day (164 μg RE/day)

4.0–8.9 years: 400 μg/day (218 μg RE/day)

9.0–12.9 years: 600 μg/day (328 μg RE/day)

Background nutrient intake, total vitamin A (IU/day), median (range) by serum retinol status:

All (N = 96): 1876 (159, 14,729)

Retinol ≥ 30 μg/dL (n = 26): 1748 (159, 4362)

Retinol 20–29 μg/dL (N = 48): 2100 (481, 14,729)

Retinol <20 μg/dL (N = 22): 1689 (317, 4359)

Exposure assessment: 24‐h recall at screening and three 24‐h recalls during the supplementation study at baseline and 3 and 12 months.

Compliance: 64% took suppl ≥85% of the time; 36% took suppl <85% of the time

Serum gamma‐glutamyl transferase (gGT)

gGT was measured by the Clinical Chemistry Laboratory at CHOP

Type of analysis extracted: PP

Units: U/L, median (range)

gGT at baseline

G1: 20 (8, 50)

G2: 23 (11, 43)

G3: 21 (8, 68)

P = NR

gGT at end‐of‐study

G1: 23 (11, 63)

G2: 21 (11, 62)

G3: 19 (12, 49)

P = NR

No significant group differences over time (change score) were detected.

Alberts et al. (2004)

USA

12 months

Funding: Public

RCT (parallel)

Inclusion criteria: Moderate to severe sun damage with or without actinic keratoses on posterior forearms, and potentially experienced nonmelanoma skin cancer, minimum 50 years of age, postmenopausal (for women).

Exclusion criteria: History of invasive cancers, CVD, stroke or other serious disease.

Participants randomised: N = 129

Participants completed: N = 116

Participants analysed: N = 129 (ITT)

N participants randomised/completed/analysed

G1: 31/30/31

G2: 32/29/32

G3: 33/27/33

G4: 33/30/33

Sex (as % female)

G1: 42%

G2: 38%

G3: 33%

G4: 33%

Age males, years (mean)

G1: 63.6

G2: 62.9

G3: 64.0

G4: 64.5

Age females, years (mean)

G1: 64.3

G2: 61.2

G3: 61.9

G4: 64.3

Vitamin A as retinyl palmitate

Doses in IU/day:

G1: Placebo

G2: 25,000 (~7500 μg RE/day)

G3: 50,000 (~15,000 μg RE/day)

G4: 75,000 (~22,500 μg RE/day)

Background nutrient intake: NR

Compliance: NR

AST, ALT, ALP

Analysis method: NR

Liver scan performed at baseline. According to protocol: second scan for subjects with ASAT or ALAT > 3x upper normal limit, but this applied to none.

Type of analysis extracted: ITT

AST, N subjects who experienced toxicity

G1: 0

G2: 0

G3: 0

G4: 1 (3%)

ALT, N subjects who experienced toxicity

G1: 0

G2: 0

G3: 1 (3%

G4: 1 (3%)

ALP, N subjects who experienced toxicity

G1: 0

G2: 0

G3: 0

G4: 0

Bitarafan et al. (2015)

Iran

12 months

Funding: Public

RCT (Parallel)

Inclusion: Multiple sclerosis patients in relapsing–remitting phase, using Interferon beta‐1a (Avonex) as a treatment, aged 20–45 years old.

Exclusion criteria: any addiction, alcohol intake, dysphagia, history of myocardial infarction, stroke, other autoimmune diseases, hypersensitivity to vitamin A compounds, liver, pancreatic and biliary disorders

N participants randomised/completed: 101/93

G1: 51/47

G2: 50/46

Sex: 74% female

Age (range): 20–45 years

(mean, SD):

G1: 30.4 ± 6.9

G2: 32.3 ± 5.9

Other background variables:

BMI (kg/m ²) (mean, SD)

G1: 23.9 ± 3.1

G2: 24.5 ± 4.3

Range: 18.5–30

Dietary vitamin A intake (μg/day) (mean, SD):

G1: 737.4 ± 483.78

G2: 744.8 ± 541.91

Vitamin A as retinyl palmitate

Doses (IU/day)

G1: 25,000 (~7500 μg RE for the first 6 months, 10,000 IU/~3000 μg RE for the second 6 months)

G2: 0 (placebo)

Background vitamin A intake, change (μg/day):

G1: −11.2 ± 68.04

G2: +17.8 ± 90.85

(p for change between groups = 0.08)

Compliance: NR.

ALT, AST

Determined with colorimetric method

Type of analysis: NR

ALT, change (mean, SD)

G1: 0.57 ± 9.44

G2: −0.85 ± 7.91

p = 0.53

AST, change (mean, SD)

G1: 0.15 ± 1.85

G2: 0.24 ± 1.48

p = 0.99

Abreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CT, controlled trial; GGT, Gamma‐glutamyl transferase; ITT: Intention‐to‐treat; NR, Not reported; PP, Per protocol; RCT: randomised controlled trial; SCD, sickle cell disease; SEM, standard error mean.