Elsevier

Nutrition

Volume 30, Issue 5, May 2014, Pages 551-556
Nutrition

Applied nutritional investigation
Plasma retinol and total carotenes and fracture risk after long-term supplementation with high doses of retinol

https://doi.org/10.1016/j.nut.2013.10.007Get rights and content

Abstract

Objective

Observational studies suggest that moderate intakes of retinol and increased circulating retinol levels may increase fracture risk. Easy access to supplements, combined with an aging population, makes this a potentially important association. The aim of this study was to investigate plasma retinol and total carotene concentrations in relation to fracture risk after long-term supplementation with retinol and/or beta-carotene in 998 adults between 1990 and 2007.

Methods

Participants were 663 men and 335 women in a cancer prevention program who were initially randomized to a retinol (7.5 mg RE/d) or beta-carotene (30 mg/d) supplement between 1990 and 1996. After 1996, all participants received the retinol supplement only. Plasma retinol and total carotene, medication use and various lifestyle factors were measured at annual clinic visits. Fractures were identified by self-report in 2007. The risk for any fracture or osteoporotic fracture was modeled using Cox proportional hazard models.

Results

Over a median follow-up of 7.8 y, 123 participants with plasma samples reported an incident fracture. Although plasma retinol concentrations were markedly higher than those reported in observational studies, no association was observed between plasma retinol and risk for any fracture (hazard ratio [HR], 0.86 μmol/L; 95% confidence interval [CI], 0.65–1.14) or osteoporotic fracture (HR, 0.97 μmol/L; 95% CI, 0.66–1.43). A lower risk for any fracture was suggested with increasing plasma total carotenes (HR, 0.85 μmol/L; 95% CI, 0.71–1.01).

Conclusions

This study does not support earlier reports of an increased fracture risk associated with increased plasma retinol concentration. The potential for carotenes to prevent fractures deserves further investigation.

Introduction

Retinol is an essential contributor to vitamin A status; however, it is well established that chronic, high doses of retinol have a deleterious effect on bone structure in animals and humans [1]. This may result from excess retinoic acid influencing the genetic expression of osteoblasts and osteoclasts, thus increasing bone turnover and fragility [2]. Interest in the effect of “subclinical hypervitaminosis A,” due to moderate intakes of retinol either through food or supplements, was initiated in 1998 when a Swedish study reported a doubling in hip fracture risk with a dietary retinol intake >1.5 mg/d [3]. Subsequent reports have been inconsistent: The Nurse's Health Study reported increased fracture risk with dietary intakes of retinol >1.0 mg/d [4], whereas several other studies have reported null associations with similar ranges of intake [5], [6], [7]. However, dietary retinol intake is one of the most difficult dietary micronutrients to measure because it is highly concentrated in only some foods and so day-to-day intake can be extremely variable [8]. Furthermore, most previous studies have used food frequency questionnaires to estimate retinol intake, and these rely on multiple assumptions when estimating nutrient intakes. For these reasons, studies of objectively measured retinol biomarkers and fracture risk may be more reliable.

Relatively few studies have examined fracture risk and biomarkers of retinol status, and their results have been inconsistent. Some suggest a positive or U-shaped relationship [9], [10] between serum or plasma retinol and fracture risk, whereas others report no association [11]. The possibility of a relationship between retinol status and fracture risk raises concerns in Western populations where food may be fortified and where retinol-containing supplements are commonly consumed and often encouraged for the elderly [12]. In the same populations, fractures due to osteoporosis impose significant health care costs and reductions in quality of life and longevity. Therefore, it is important to establish whether moderate retinol intakes confer any excess risk for bone fractures.

We conducted a secondary analysis of a cancer prevention study of retinol and beta-carotene supplements between 1990 and 2007 in Perth, Western Australia, to examine whether bone fracture risk was positively associated with plasma retinol concentration. Because some observational studies have suggested a protective relationship between beta-carotene intake and fracture risk [13], [14], we hypothesized that fracture risk would be inversely associated with plasma total carotene concentration, and that plasma levels of alkaline phosphatase (ALP), as a marker of bone turnover, would be positively associated with fracture risk.

Section snippets

Participants

The Vitamin A Program was a cancer prevention program designed to examine the efficacy of high-dose retinol and beta-carotene supplements in reducing the risk for malignant mesothelioma, lung cancer, and other cancers in individuals with previous asbestos exposure [15], [16], [17]. The study methods have been detailed elsewhere [18] and are reported here in brief. Study participants were former workers and former residents of the mining township of Wittenoom, a remote town in the northwest of

Results

The average age ± SD at entry to the study was 52.2 ± 10.1 y for men and 47.5 ± 11.7 y for women, and ranged from 15 to 80 y. Follow-up ranged from 0 to 17 y, with a median of 7.8 y. A total of 123 participants reported one or more incident fractures during follow-up, of which 54 were classified as osteoporotic fractures (Table 1). Previous fractures were reported by 15.4% of respondents.

Self-reported incident fractures occurred more frequently among women, former Wittenoom residents, and

Discussion

This study presented a unique opportunity to examine risk for fracture in relation to plasma retinol and carotene concentrations after long-term supplementation with high doses of retinol and/or beta-carotene supplements in a population-based sample. The study included a larger number of individuals with higher plasma concentrations of these two substances than previous observational studies, and was able to examine a wider range of concentrations (Fig. 1). Despite this, and in contrast to two

Conclusions

No convincing associations between plasma retinol concentration and fracture risk were observed in this study of participants who took 7.5 mg retinol/d for up to 17 y. The possibility that higher plasma carotene concentrations may be associated with lower fracture risk is consistent with previous studies and warrants further study.

Acknowledgments

The authors wish to express their sincere thanks to all participants in the Vitamin A Program. The authors acknowledge Lynne Defrenne, Naomi Hammond, Meralyn Pearce, and Diane Jacoby who collected data for this study; Jan Sleith for data entry and data cleaning; Robin Mina who oversaw database management; and staff from the Perth Chest Clinic.

Project grants from the National Health and Medical Research Council of Australia supported the current analysis (Ambrosini 458608) and follow-up of the

References (38)

  • H. Ahmadieh et al.

    Vitamins and bone health: Beyond calcium and vitamin D

    Nutr Rev

    (2011)
  • L. Hu et al.

    Retinoic acid increases proliferation of human osteoclast progenitors and inhibits RANKL-stimulated osteoclast differentiation by suppressing RANK

    PLoS One

    (2010)
  • H. Melhus et al.

    Excessive dietary intake of vitamin A is associated with reduced bone mineral density and increased risk for hip fracture

    Ann Intern Med

    (1998)
  • D. Feskanich et al.

    Vitamin A intake and hip fractures among postmenopausal women

    JAMA

    (2002)
  • L. Rejnmark et al.

    No effect of vitamin A intake on bone mineral density and fracture risk in perimenopausal women

    Osteopor Int

    (2004)
  • L.S. Lim et al.

    Vitamin A intake and the risk of hip fracture in postmenopausal women: The Iowa Women's Health Study

    Osteopor Int

    (2004)
  • G. Ambrosini et al.

    Agreement between a brief food frequency questionnaire and diet records using two statistical methods

    Public Health Nutr

    (2001)
  • K. Michaelsson et al.

    Serum retinol levels and the risk of fracture

    NEJM

    (2003)
  • M.E. Barker et al.

    Serum retinoids and beta carotene as predictors of hip and other fractures in elderly women

    J Bone Min Res

    (2005)
  • Cited by (17)

    • Higher dietary carotenoid intake associated with lower risk of hip fracture in middle-aged and elderly Chinese: A matched case-control study

      2018, Bone
      Citation Excerpt :

      In the Utah Study of Nutrition and Bone Health including 1215 hip fractures and 1349 controls, higher intakes of β-carotene and other antioxidants (vitamin E and C, selenium) were significantly associated with a lower risk of hip fracture [14]. The favorable associations of blood carotenoids were found with hip fracture risk in men (but not in women) in the EPIC-Norfolk cohort [20], and were observed with risk of any fractures in Western Australian adults [16]. In a recent meta-analysis including five prospective and two case-control studies with 140,265 participants and 4324 cases [21], Xu et al. reported that hip fracture risk both decreased by 28% among participants with higher (vs. lower) dietary consumption of total carotenoids and β-carotene (p < 0.05).

    • Nuclear Receptors in Skeletal Homeostasis

      2017, Current Topics in Developmental Biology
      Citation Excerpt :

      High vitamin A diet and high serum retinol level were reported to be positively correlated with a higher risk of bone fracture in humans and animals (Feskanich, Singh, Willett, & Colditz, 2002; Lind et al., 2013; Melhus et al., 1998; Michaelsson, Lithell, Vessby, & Melhus, 2003; Whiting & Lemke, 1999). However, a number of reports showed no correlation between serum retinol level and fracture risk, or even a protective effect of retinol on bone (Ambrosini et al., 2014; Caire-Juvera, Ritenbaugh, Wactawski-Wende, Snetselaar, & Chen, 2009; Holvik et al., 2015; Ribaya-Mercado & Blumberg, 2007). Thus, the relationship between serum retinol level and fracture risk remains unclear.

    View all citing articles on Scopus

    GLA, DM, AR, AWM and NHdK were responsible for the study design. GLA, NJO, JB were involved with the data collection. HA was responsible for the statistical analysis and all the authors were involved in the data interpretation. GLA drafted the manuscript, and all authors were responsible for revising manuscript content. All authors approved the final version of this manuscript. GA, HA, and NHdK take responsibility for the final version of the manuscript. All authors declare no conflicts of interest.

    View full text