ohpd 2013 01 s0061

ORIGINAL ARTICLE Anti-inflammatory Effect of Vitamin D on Gingivitis: A Dose-Response Randomised Control Trial Vishwanat...

0 downloads 72 Views 88KB Size
ORIGINAL ARTICLE

Anti-inflammatory Effect of Vitamin D on Gingivitis: A Dose-Response Randomised Control Trial Vishwanath P. Hirematha/C. Bhasker Raob/Vijaya Naikc/ Kakrala Veera Venkatesha Prasadd Purpose: To assess the anti-inflammatory effect of vitamin D on gingivitis at various doses. Materials and Methods: In this randomized controlled trial, daily oral vitamin D supplementation was given in doses of 2000 IU for group A, 1000 IU for group B, 500 IU for group C and a placebo for group D over a 3-month period. The changes in gingival scores were measured after the 1st, 2nd and 3rd months. Results: The gingivitis score changed in direct proportion to the dose of vitamin D supplementation. In group A, the mean gingival scores were 2.4 (baseline), 1.7 after the first month, 0.8 after the second month and 0.3 after the third month. The group B mean baseline gingival score of 2.3 decreased to 2.0 in the first month, 1.1 after the second month and 0.5 after the third month. In group C, the baseline gingival scores were 2.2 and 1.9 after one month, 1.4 after two months and 0.8 by the last visit. Comparing baseline gingivitis scores with the later-visit score using the Wilcoxon paired test, the significant anti-inflammatory effect was seen in group A after one month, in group B at two months and in group C at three months after oral vitamin D supplementation (P < 0.0001). However, group D did not show a significant antiinflammatory effect. Conclusion: There is a dose-dependent anti-inflammatory effect of vitamin D on gingivitis. Vitamin D is a safe and effective anti-inflammatory agent in doses ranging from 500 IU to 2000 IU. Results are apparent earlier with the higher dose of 2000 IU. Key words: anti-inflammatory effect, gingivitis, 25 hydroxyvitamin D Oral Health Prev Dent 2013; 11: 61-69

G

ingivitis is the inflammation of gingiva and the mildest form of periodontal disease. It affects 50% to 90% of adults worldwide. Clinically gingivitis is recognised by the signs of inflammation such as redness, swelling, bleeding and exudation, but less frequently by pain. It is ubiquitous and affects males and females, young and old (Cincio, 1986). As defined by gingival bleeding adjacent to ≥1

a

Professor, Department of Public Health Dentistry, Maratha Mandal Dental College Belgaum, Karnataka, India.

b

Director, Sri Dharmasthala Manjunatheshwara Dental College Dharwar, Dharwad Karnataka, India.

c

Professor and Head, Department of Community Medicine, Jawaharlal Nehru Medical College, Belgaum, Karnataka, India.

d

Professor and Head, Department of Public Health Dentistry, Sri Dharmasthala Manjunatheshwara Dental College, Dharwar, India.

Correspondence: Professor V.P. Hiremath, Department of Public Health Dentistry, Maratha Mandal Dental College, Belgaum, Karnataka, India 560010. Tel: +91-098-4512-2965, Fax: +91-083-12479323. Email: [email protected]

Vol 11, No 1, 2013

Submitted for publication: 15.10.11; accepted for publication: 30.04.12

teeth, about half of the US population has gingivitis with the prevalence slightly greater in males than females, in Afroamericans and Latinos than in whites (Stamm, 1986). There is evidence that the prevalence of gingivitis has been decreasing in developed countries over the last few decades. Globally, there appears to be considerable heterogeneity in the prevalence of gingivitis, with higher prevalences reported in certain parts of the world (Stamm, 1986; Armitage, 1999). The cause of gingivitis is primarily bacterial plaque at the gingival margin, but possible nutritional influences on the development and progression of this disease cannot be ignored, as it may lead to periodontitis and tooth loss (Caton, 1999). Gingivitis can be prevented by regular visits to a dental professional for oral prophylaxis, together with adequate oral hygiene performed at home (Hasturk, 2004). However, if that is not possible, for instance, in the physically handicapped, medically compromised or those otherwise unable to do

61

Hiremath et al

so, it is necessary to think about an alternative preventive treatment. Vitamin D could be an alternative supplement which would provide a simple, safe and inexpensive way of reducing gingivitis (Vasquez et al, 2004). Vitamin D is a vital nutrient for bone strength and prevention of osteoporosis, but many people are unaware that vitamin D has many other functions at higher doses. One review suggests that the margin of safety for vitamin D consumption for adults is 10 times greater than current recommended intake (Grant, 2003). The safe upper limit of 2000 IU per day has been recommended by the US Food and Nutrition Board (Grant, 2003). Vitamin D deficiency has been linked to many chronic diseases such as high blood pressure, certain cancers and insulin resistance; it has been found to protect against muscle weakness (Pfeifer et al, 2001; Hypponem et al, 2001; Munger et al, 2004). Recently, one study also linked lower levels of vitamin D to inflammation in the gums, a precursor to gingivitis (Dietrich, 2005). In addition, an anti-inflammatory effect of vitamin D at higher doses is established (Daniells, 2003). Data from in vitro and animal studies also suggest anti-inflammatory effects of vitamin D at higher doses (Bhalla et al, 1984; Rigby et al, 1987; Mathieu et al, 2002). However, whether an increase in 25-hydroxy vitamin D serum concentrations by oral supplementation with vitamin D has an effect on gingivitis has not been investigated. Thus, the purpose of the present study was to estimate the dose-dependent anti-inflammatory effect of vitamin D on gingivitis. The serum concentration of vitamin D which could initiate an anti-inflammatory effect was also assessed.

MATERIALS AND METHODS Study design This is a randomized, double-blind placebo controlled clinical study to assess the anti-inflammatory effect of vitamin D on gingivitis over a period of three months.

Subjects The study was conducted at Maratha Mandal Dental College, Belgaum, India, starting with subject recruitment in June 2010; the follow-up after pa-

62

tient recruitment was from February 15, 2011 to May 15, 2011 for a period of three months. Permission was obtained from the K. L. E University Ethics Committee, Belgaum, India, and also from the local ethics committee of Maratha Mandal Dental College, Belgaum. From a general population aged 18 to 64 years, both male and female residents of Belgaum were considered for the study. Around 452 people were screened and the first 110 who satisfied the inclusion criteria were considered as study subjects. The expected enrollment was calculated to be 96. This number was obtained based on the pilot study results conducted before the start of the study on 12 subjects. The standard deviation of the gingival scores of the groups and mean difference (μd) from the pilot study were considered for calculating the sample size. In addition, the _ error was fixed at 5% and the power of the test (1 – `) at 90%. The statistical software N Master and SPSS version 10 were used for this part of the statistical analysis. Inclusion criteria: Males and females in good general health aged 18 to 64 years diagnosed with different clinical stages of gingivitis and willing to sign the informed consent form, and who demonstrated a willingness to comply with all study procedures and clinical examination schedules were enrolled for the study. A minimum of 20 natural teeth with facial and lingual scorable surfaces with adequate oral hygiene and no signs of oral neglect, and with calculus deposits on not more than one third of the cervical tooth surface were preferred. The selected subjects also had to have periodontal pockets < 3 mm and a gingivitis score of > 1 as measured by the Löe and Silness gingival index (1963). Furthermore, the subject had to demonstrate a serum 25-hydroxyvitamin D concentration of 20 ng/ml to 65 ng/ml. The subjects could not undergo any dental procedure during the study period. Exclusion criteria: Those who showed tooth mobility, a history of active severe periodontal disease with bleeding gums, gross dental caries, severe generalized cervical abrasion and/or enamel abrasion or large fractured or temporary restorations (based on visual examinations) were not included. Persons with fixed or removable orthodontic appliances or removable dental prostheses, a history of dental prophylaxis or treatments in the past month or who required some dental treatment during the study period were excluded from the study. Women who were postmenopausal, pregnant or expecting pregnancy during the period of the trial were not considered for the study. Current or former smok-

Oral Health & Preventive Dentistry

Hiremath et al

ers with cessation 0.05)

Serum vitamin D Gingival scores The serum vitamin D level in the groups increased after oral supplementation with vitamin D. The increase was directly proportional to the increase in dose of vitamin D (Table 1, Fig 1). One-way ANOVA showed significant differences among the 4 groups at baseline (P < 0.0008). Anal-

64

One-way ANOVA showed that there was no significant difference in gingival scores at baseline, as group standardisation was done at the beginning of the study. The gingival scores after oral supplementation with vitamin D decreased, indicating an anti-

Oral Health & Preventive Dentistry

Hiremath et al

60

50

Mean Value

40

30

20

10

0 Group A

Group B

Group C Baseline

Group D

Final visit

Fig 1  Mean serum vitamin D values in the various groups at baseline and final visit.

inflammatory effect of vitamin D on gingivitis (Table 2, Fig 2). Analysis of covariance showed a significant difference in gingivitis scores compared to baseline (P < 0.0001). The comparison of follow-ups with respect to baseline gingival scores within the groups showed the gingival scores for group A decreased significantly by the 2nd visit (30 days) and the gingival scores of group B changed highly significantly by the 3rd visit (60 days) (Table 3; P < 0.0001, Wilcoxon matched-pair test). Group C showed an anti-inflammatory effect at the final visit, when serum vitamin D concentration was 36.81 ng/ml, which is the optimum level of serum vitamin D for initiating an anti-inflammatory effect. However, group D did not exhibit a significant decrease in gingival scores. It could also be predicted (but not tested) that the anti-inflammatory effect in groups A and B would have been initiated when their serum vitamin D levels reached 32–35 ng/ml, which is the projected optimum score of serum vitamin D based on the baseline and 90-day vitamin D concentrations. As group A was given 2000 IU of vitamin D, the anti-inflammatory effect would be evident early

Vol 11, No 1, 2013

compared with the other groups. This also shows the dose-dependent anti-inflammatory effect of vitamin D on gingivitis; in addition, group A – which achieved serum vitamin D levels of > 32 ng/ml early on – showed an early anti-inflammatory effect. When pairwise comparison (Neuman-Keuls multiple post-hoc test) was done with respect to gingival index scores in the 4 groups at the 2nd, 3rd and final visit (Table 4), group A did not show a significant change vs group B, but did demonstrate a significant difference vs group C. Similarly, group B did not show a significant change vs group C. However, the differences in gingival scores were significant (P < 0.0001) when groups A, B and C were compared with group D. This shows that there is little change in the anti-inflammatory effect between groups A and B, as well as between groups B and C. This could also indicate that the group which could achieve an early serum vitamin D concentration of > 32 ng/ml initiated an early anti-inflammatory effect and later stabilised after reaching a level where the anti-inflammatory effect increased only marginally.

65

Hiremath et al

Table 2 Mean (± SD) gingival scores in the four groups at each visit Groups

Baseline gingival score ± SD

30-day gingival score ± SD

60-day gingival score ± SD

90-day gingival score ± SD

A 2000IU

2.4127 ± 0.5435

1.7782 ± 0.6332

0.8982 ± 0.7420

0.3405 ± 0.6046

B 1000IU

2.3973 ± 0.5736

2.0164 ± 0.5844

1.1695 ± 0.7147

0.5532 ± 0.6615

C 500IU

2.2423 ± 0.4602

1.9623 ± 0.4600

1.4309 ± 0.7579

0.8832 ± 0.9859

D placebo

2.2386 ± 0.6125

1.9686 ± 0.6391

1.9005 ± 0.6721

1.8973 ± 0.6406

F-value

0.6587§

2.4755#

9.2629#

20.2771#

P-value

0.5797*

0.0671

0.0000**

0.0000**

§ one-way ANOVA at baseline; # ANCOVA baseline as covariate; *P < 0.05; level of significance 5%.

3.0

2.5

Mean Value

2.0

1.5

1.0

0.5

0 Group A

Group B Baseline

2nd visit

Group C 3rd visit

Group D Final visit

Fig 2  Mean baseline and follow-up gingival scores in different groups at various visits.

DISCUSSION In the present study, the groups which received oral supplementation of vitamin D in doses of 2000, 1000 and 500 IU showed a highly significant gingival anti-inflammatory effect (P < 0.001) when compared with the placebo group. The placebo group did not show any significant improvement in

66

gingivitis. When baseline gingival scores were compared with subsequent visit scores, group A (2000 IU per day) showed a significant change (P < 0.001) in gingival scores after one month (at the 2nd visit) of daily oral vitamin D supplementation when compared with the other groups. In group B (1000 IU vitamin D), the anti-inflammatory effect was evident at the 3rd visit (P < 0.001), i.e. after two months of

Oral Health & Preventive Dentistry

Hiremath et al

Table 3 Percent change in the gingival scores between visits and vs baseline scores (BL) Groups

BL-2nd visit

BL-3rd visit

BL-final visit

2nd-3rd visit

2nd-final visit

3rd-final visit

A 2000IU

0.6345 ± 0.3702

1.5145 ± 0.7533

2.0723 ± 0.7025

0.8800 ± 0.5595

1.4377 ± 0.5758

0.5577 ± 0.6113

B 1000IU

0.3809 ± 0.3152

1.2277 ± 0.9231

1.8441 ± 0.9766

0.8468 ± 0.7992

1.4632 ± 0.8395$

0.6164 ± 0.4591

C 500IU

0.2800 ± 0.4343

0.8114 ± 0.7796

1.3591 ± 0.9968

0.5314 ± 0.5417

1.0791 ± 0.8290

0.5471 ± 0.5283

D placebo

0.2700 ± 0.6566

0.3382 ± 0.6253

0.3414 ± 0.6155

0.0682 ± 0.2202

0.0714 ± 0.1522

0.0032 ± 0.1757

% Change

 

 

 

 

 

 

A

26.30**

62.77**

85.89**

49.49**

80.85**

62.09**

B

15.88*

51.22**

76.92**

41.99*

72.56**

52.70**

C

12.49

36.19*

60.61**

27.08*

54.99*

38.28**

D

12.06

15.10

15.25

3.46

3.62

0.17

2nd visit: 30 days; 3rd visit: 60 days; final visit: 90 days. Wilcoxon paired test, *P < 0.05; ** P < 0.0001; level of significance set at P < 0.05.

Table 4 Pairwise comparison of four groups with respect to to gingival index scores at 2nd (30 days), 3rd (60 days) and final (90 days) visit Follow-up visit

2nd

3rd

Final

Group

A

B

C

D

Mean

1.778182

2.016364

1.962273

1.968636

A

-

B

0.2764

-

C

0.1658

0.9113

-

D

0.3221

0.7180

0.9617

-

Mean

0.8981818

1.169545

1.430909

1.900455

A

-

B

0.2012

-

C

0.0352

0.2181

-

D

0.0002

0.0024

0.0285

-

Mean

0.3404545

0.5531818

0.8831818

1.897273

A

-

B

0.3379

-

C

0.0419

0.1387

-

D

0.0001

0.0001

0.0001

-

Note: P-values 32 ng/ml first, showed an earlier anti-inflammatory effect.

Vol 11, No 1, 2013

Overall, we found an inverse association between 25-hydroxyvitamin-D serum concentrations and severity of gingivitis as measured by the Löe and Silness gingival index. This is in agreement with the earlier studies, for instance, Dietrich et al (2005) found an inverse association between serum concentrations of 25 hydroxyvitamin D and chronic gingivitis as seen by bleeding on probing. That study

67

Hiremath et al

also revealed that a serum concentration of 30–35 ng/ml may reduce susceptibility to gingivitis in humans. Numerous in vitro studies have demonstrated the anti-inflammatory effects of vitamin D. Dihydroxy vitamin D has been shown to inhibit antigeninduced T-cell proliferation and cytokine production, specifically interleukin 2 and interferon (Bhalla et al, 1984; Rigby et al, 1987; Mathieu et al, 2002). Some studies have suggested that vitamin D may have beneficial effects on periodontal disease and tooth loss, possibly because of its anti-inflammatory effects (Krall et al, 1984). However, other studies have shown that the anti-inflammatory effect of vitamin D is evident only at doses of 500 IU and above (Krall et al, 1984; Dixon et al, 1999; Dietrich et al, 2004). The present study also showed similar results, as group C – which received 500 IU vitamin D supplementation per day – demonstrated a significant anti-inflammatory effect at the third month. Furthermore, Stephen (2006) showed for the first time that a daily supplement of 50 μg (2000 IU) of vitamin D for nine months can increase serum concentrations of the anti-inflammatory cytokine IL-10 and prevent an increase in serum concentrations of the pro-inflammatory cytokine TNF-alpha in chronic heart failure patients. In a cross-sectional study of 116 subjects, van den Berghe et al (2003) found that serum concentrations of 25 hydroxyvitamin D were negatively correlated with serum concentrations of C-reactive protein, which is an indicator of inflammation. In a subsample of 24 patients from that study, vitamin D supplementation significantly reduced serum concentrations of C-reactive protein by 23%. There is no accepted definition available for ‘adequate’ or ‘optimal’ vitamin D status or 25 hydroxyvitamin D concentrations. Suppressed serum concentrations of PTH served as a surrogate marker for vitamin D concentration, and therefore the recommendations for vitamin D vary (Dietrich et al, 2005; Doxon et al, 2009). In the study by Dietrich et al (2005), no evidence was found of a threshold serum concentration of 25 hydroxyvitamin D, above which the association with gingival inflammation leveled off. They suggested that the anti-inflammatory effects of vitamin D could possibly extend to serum concentrations in the range of 30–35 ng/ml. In the present study as well, the association of an anti-inflammatory effect of serum vitamin D concentration on the level of gingivitis varied in a range of 30 to 35 ng/ml. Also the group which achieved an early serum vitamin D concentration of > 32 ng/ml

68

early on showed an early anti-inflammatory effect, indicating the serum level at which the anti-inflammatory effect of vitamin D may begin. This was evident when group C (500 IU vitamin D) showed an anti-inflammatory effect at the third visit, when the serum level of vitamin D reached 36.81 ng/ml. This indirectly proves that the minimum requirement of vitamin D to initiate an anti-inflammatory effect is 500 IU, which is in agreement with earlier studies (van den Berghe et al, 2003). In the present study, the subjects with teeth having gingivitis and pockets > 3 mm were excluded; hence the analysis did not describe any effects of vitamin D on bone. Nevertheless, it is well known that chronic marginal gingivitis may eventually lead to periodontal problems in susceptible patients and bone loss. An inverse association between serum concentrations of 25 hydroxyvitamin D and the prevalence of periodontal disease as measured by periodontal attachment loss was shown in the National Health and Nutrition Examination Survey (NHANES) III study (Dietrich et al, 2004). However, this association was independent of bone mineral density. As in Dietrich et al (2004), the results of the present study are consistent with an anti-inflammatory effect of vitamin D on gingival inflammation. This indirectly suggests that this may be an alternative pathway for the prevention of periodontal disease. The study has several limitations. Bacteria and bacterial products of dental plaque are the main cause of gingival inflammation; plaque was an important confounding variable in the present study, as it was not measured. Therefore, the role of plaque cannot be explained here. Further studies combining oral vitamin D supplementation with oral prophylaxis may be necessary to explain the confounding influence of plaque. Moreover, studies may be needed to investigate the effect of oral vitamin D supplementation in medically compromised, systemically ill patients with gingivitis. Further studies are also recommended to elucidate the influence of factors such as gender, age and weight on the antiinflammatory effect of vitamin D on gingivitis.

CONCLUSION Vitamin D has an anti-inflammatory effect in doses ranging from 500 to 2000 IU. People with serum vitamin D levels > 30 ng/ml could benefit from this treatment. Vitamin D can be supplemented orally for 2 to 3 months to achieve the desired results for

Oral Health & Preventive Dentistry

Hiremath et al

patients with gingivitis, although 2000 IU of vitamin D per day can produce the anti-inflammatory effect sooner.

ACKNOWLEDGEMENTS The authors would like to thank Dr Kishore Bhat of Maratha Mandal Dental College for conducting all the biochemical analysis needed for the study. Sincere thanks and gratitude are also expressed to Dr K V V Prasad of S D M Dental College Dharwar for his valuable advice and support through out the study. We also thank Dr C Bhasker Rao, Director S D M Dental College Dharwar, without whom the study could not have been possible. Special thanks to Dr Javali, statistician, of SDM Dental College Dharwar. The staff of the Community Dentistry Department of M .M. Dental College and the participants in the study deserve special thanks. Last but not least, our heartfelt thanks to the Principal and Chairperson of Maratha Mandal Dental College for their support and encouragement.

REFERENCES: 1. Armitage GC. Development of a classification system for periodontal diseases and conditions. Annals Periodontol 1999;4:1–6. 2. Bhalla AK, Amento EP, Serog B, Glimcher LH. 5-Dihydroxyvitamin D3 inhibits antigen-induced T cell activation. J Immunol 1984;133:1748-1754. 3. Carter GD, Carter CR, Gunter E, Jones J, Jones G, Makin HL, Sufi S. Measurement of vitamin D metabolites: an international perspective on methodology and clinical interpretation. J Steroid Biochem Mol Biol 2004;89/90:467–471. 4. Caton J. Biological and measurement issues critical to design of gingivitis trials. J Periodontal Res 1999;27:364–368. 5. Cincio SG. Current status of indices of gingivitis. J Clin Periodontol 1986;13:375–378. 6. Daniells S. High-dose vitamin D supplements act as antiinflammatory. J Am College Cardiol 2003;41:105–112. 7. Dawson-Hughes B, Harris SS, Krall EA, Dallal GE. Effect of calcium and vitamin D supplementation on bone density in men and women aged 65 yrs and above. New Engl J Med 1997;33:610–616. 8. Dietrich T, Joshipura KJ, Dawson-Hughes B, Bischoff-Ferrari H. Association between serum concentrations of 25-hydroxyvitamin D3 and periodontal disease in the US population. Am J Clin Nutr 2004;80:108–113. 9. Dietrich T, Nunn M, Dawson-Hughes B, Bischoff-Ferrari HA. Association between serum concentration of 25-hydroxyvitamin D and gingival inflammation. Am J Clin Nutr 2005;82:575–580.

Vol 11, No 1, 2013

10. Dixon D, Hildebolt CF, Miley C. Calcium and vitamin D use among adults in periodontal disease maintenance programmes: Br Dent J 2009;206,627–631. 11. Grant WB, Holock MF. Benefits and requirements of vitamin D for optimal health. Alternative Med Rev 2005;10:23. 12. Hasturk H, Nunn M, Warbington M, van Dyke TE. Efficacy of a fluoridated hydrogen peroxide-based mouthrinse for the treatment of gingivitis: a randomized clinical trial. J Periodontol 2004;75:57–65. 13. Hathcock JN, Shao A, Vieth R, Heany R. Risk assessment for vitamin D. Am J Clin Nutr 2007;87(1):6–18. 14. Hypponen E, Laara E, Reunanen A, Jarvelin MR, Virtanen SM. Intake of vitamin D and risk of type 1 diabetes: a birthcohort study. Lancet 2001;358:15–33. 15. Krall EA, Wehler C, Garcia RI, Harris SS, Dawson-Hughes B. Calcium and vitamin D supplements reduce tooth loss in the elderly. Am J Med 2001;111:452–466. 16. Mathieu C, Adorini L. The coming of age of 5-dihydroxyvitamin D(3) analogs as immunomodulatory agents. Trends Mol Med 2002;8:174-179. 17. Munger KL, Zhang SM, O’Reilly E. Vitamin D intake and incidence of multiple sclerosis. Neurology 2004;62:605–621. 18. Peterson CA, Heffernan ME. Serum tumor necrosis factoralpha concentrations are negatively correlated with serum 25 (OH)D concentrations in healthy women. J Inflammation 2008;5:10–15. 19. Pfeifer M, Begerow B, Minne HW, Nachtigall D, Hansen C. Effect of a short term Vitamin D 3 and calcium on blood pressure and parathyroid hormone levels levels in elderly women. J Clin Endocrinol 2001;86:633–637. 20. Proskin HM. Some thoughts on the measurent of gingivitis. J Periodont Res 1992;27:369–372. 21. Rigby WF, Denome S, Fanger MW. Regulation of lymphokine production and human T lymphocyteactivation by 5- dihydroxyvitamin D3. Specific inhibition at the level of messenger RNA. J Clin Invest 1987;79:164-165. 22. Stamm JW. Epidemiology of gingivitis. J Clin Periodontol 1986;13:360–366. 23. van den Berghe G, van Roosbroeck D, Vanhove P, Wouters PJ, De Pourcq L, Bouillon R. Bone turnover in prolonged critical illness: effect of vitamin D. J Clin Endocrinol Metab 2003;88:4623-4632. 24. Vasquez A, Manso G, Cannell J. The clinical importance of vitamin D (cholecalciferol): a paradigm shift with implications for all health care providers. Alternative Therapies 2004;10:5–9. 25. Vieth R, Bischoff-Ferrari H, Boucher BJ. The urgent need to recommend an intake of vitamin D that is effective. Am J Clin Nutr 2007;85:649–650.

69