Review Article

Vitamin E and Bone Structural Changes: An Evidence-Based Review

Table 2

Characteristics of human studies included in the review.

StudyType of study/Vit. E analogSample/populationMethodologyResults
(Vitamin E and osteoporosis only)
Comments or outcomes

Maggio et al., 2003 [59]
Human observational study
Vitamin E
1,100 women recruited for instrumental screening for osteoporosis to the Geriatric Division of Perugia University Hospital
150 women (75 osteoporotic and 75 control) gave their consent form and finally enrolled

Duration: 12 months

Final analysis:
150 women
Study variable included age years, body mass index, self-reported fractures, smoking habits, and other related variables. All pertinent information was collected through a questionnaire that was administrated by trained interviewer.
The bone mineral density was measured using dual energy X-ray absorptiometry densitometer.

Study subjects underwent a fasting blood withdraw in 20 ml heparin tubes on the day of the bone scan. Blood was kept on ice and centrifuged within 30 min. Plasma aliquot was frozen at −80°C until analysis.
150 women
Mean Value:
Age: 70.4 ± 8.5
BMI: 25.3 ± 2.9
Years since menopause: 22.8 ± 9

Mean plasma levels of vitamin E was significantly lower in osteoporotic than controls (P < 0.001).

Antioxidant and MDA plasma level:
Plasma vit E (μmol/liter): 46.7 ± 5
Plasma MDA (μmol/liter): 0.34 ± 0.13

MDA did not show any significant difference between osteoporotic and control subjects.


MDA result did not differ between groups.
Low antioxidant levels cause antioxidant deficiency and give negative impact on bone mass.
Role of vitamin E and osteoporosis needs further investigation.

Macdonald et al., 2004 [60]Human epidemiological longitudinal study
Vitamin E
1064 healthy premenopausal women aged 45–54 who took part in the Aberdeen Prospective Osteoporosis Screening Study.
896 responded to 2nd bone scan and completed questionnaire. 5 excluded (3 women had bisphosphonate therapy, 1 was on wheelchair and another had an outlier dietary calcium intake)

Final analysis: 891 women
Participants chosen from women who took part in the Aberdeen Prospective Osteoporosis Screening Study conducted from 1990 to 1993 and had a bone scan done and a completed food-frequency questionnaire (FFQ). This study was a population-based osteoporotic fracture screening program within a 40 km radius of Aberdeen city, Scotland.
Between 1997 and 1999, participants were recalled and had a second bone scan done and again completed the FFQ. Inclusion criteria: women who did not have any conditions or taking any medications that might affect their bone metabolism.

Anthropometric measurements were taken. Bone mineral density (BMD) measurements of left proximal femur or femoral neck (FN) and lumbar spine (LS) were measured and compared between the first and second measurement.

Dietary intake was assessed using the FFQ. The FFQ contained 98 foods or food groups intake of participants recorded over 7 days. Alcohol intake and dietary supplements were also measured.

Physical activity level was obtained using the Scottish heart health Study questionnaire.
891 women participants.
Mean values:
Age (y): 53.9 ±1.6
BMI (kg/m2): 26.1 ± 4.4
Total Vitamin E intake (mg):
 13.3 ± 32.0
BMD (g/cm2):
 Lumbar spine: 0.998 ± 0.17
 Femoral neck: 0.833 ± 0.12

BMD results:
No evidence of an association between nutrient intake and BMD change.

Vitamin E intake (diet only, not including supplements) was negatively correlated with change in both BMD measurements (P < 0.01).
Total Vitamin E intake was positively correlated with both BMD measurements but was not significant.

Regression analysis:
Vitamin E (diet only, not total) accounts for 0.4% of change on femoral neck BMD (P = 0.018).
Dietary vitamin E appeared to be a negative predictor for FN BMD change.
The authors postulated that this may be due to vit E being a surrogate marker for polyunsaturated fatty acids (PUFA). PUFA anditamin E are highly correlated with PUFA having a negative correlation with FN BMD (P < 0.01).

Wolf et al.,2005 [61]

Human epidemiological cross-sectional study
Vitamin E
11,393 women aged 50–79 yrs were recruited between 1993 and 1997 to participate in the Women’s Health Initiative (WHI) observational study and clinical trial at 3 clinics.
BMD was measured and antioxidant intakes were estimated using self-reported food-frequency questionnaire.

Women taking oral glucocorticoids, bisphosphonates, calcitonin or tamoxifen were excluded.

Final analysis: 11,068 women.
All participants underwent the following data collection.
Questionnaire data:
(1) demographic data
(2) smoking status
(3) alcohol intake
(4) medication history
(5) use of hormone therapy
(6) frequency, duration and intensity (strenuous, moderate and mild) of physical activity.
(7) diet intake. Nutrients were calculated using Minnesota Nutrition Coding Center database
(8) dietary supplements

Clinical measurements:
(1) weight and height. body mass index (BMI)
(2) BMD measurement using dual X-ray absorptiometry; total body, lumbar spine, total hip, femoral neck, and trochanter.

Blood measurements:
(1) serum antioxidant concentrations; retinol, α and β carotene, α and γ-tocopherol, β-cryptoxanthin, lycopene, lutein, and zeaxanthin.
(2) total cholesterol and triacylglycerols.
11,068 participants, 4.8% had osteoporosis
Mean values:
Age (y): 63.2
BMI (kg/m2): 28.3 ± 5.9
Diet vitamin E intake (mg): 7.8 ± 3.8
Total vitamin E intake (mg): 28.9 ± 49.4
Serum total tocopherols (μg/mL): 18.0 ± 6.0
BMD (g/cm2):
 Whole body: 1.0 ± 0.1
 Lumbar spine: 1.0 ± 0.2
 Total hip: 0.9 ± 0.1
 Femoral neck: 0.7 ± 0.1
 Hip trochanter 0.6 ± 0.1

BMD results:
Age-adjusted regression analysis resulted in positive association for dietary vitamin E and femoral neck BMD (P = 0.002), but negative association for total vitamin E (P < 0.0001). However, results after adjusting for multiple important BMD-related covariates (age, BMI, waist circumference, race, education, income, physical activity, etc.) showed no significant association at any BMD sites with vitamin E.

No significant association for serum concentration of tocopherols with any BMD sites.
Authors concluded no significant association between vitamin E and BMD.
Authors noted that most participants had normal range BMD which may influence association between low BMD and antioxidants.