A. L. Darling

Dietary protein and bone health: a systematic review and meta-analysis

BACKGROUND There has been a resurgence of interest in the controversial relation between dietary protein and bone health. OBJECTIVE This article reports on the first systematic review and meta-analysis of the relation between protein and bone health in healthy human adults. DESIGN The MEDLINE (January 1966 to September 2007) and EMBASE (1974 to July 2008) databases were electronically searched for all relevant studies of healthy adults; studies of calcium excretion or calcium balance were excluded. RESULTS In cross-sectional surveys, all pooled r values for the relation between protein intake and bone mineral density (BMD) or bone mineral content at the main clinically relevant sites were significant and positive; protein intake explained 1-2% of BMD. A meta-analysis of randomized placebo-controlled trials indicated a significant positive influence of all protein supplementation on lumbar spine BMD but showed no association with relative risk of hip fractures. No significant effects were identified for soy protein or milk basic protein on lumbar spine BMD. CONCLUSIONS A small positive effect of protein supplementation on lumbar spine BMD in randomized placebo-controlled trials supports the positive association between protein intake and bone health found in cross-sectional surveys. However, these results were not supported by cohort study findings for hip fracture risk. Any effects found were small and had 95% CIs that were close to zero. Therefore, there is a small benefit of protein on bone health, but the benefit may not necessarily translate into reduced fracture risk in the long term.

Seasonal 25-hydroxyvitamin D changes in British postmenopausal women at 57°N and 51°N: A longitudinal study

The UK has insufficient intensity of sunlight at wavelengths 290-315 nm to enable cutaneous synthesis of vitamin D from October to April. There are regional differences in UVB strength throughout the UK but whether this translates to differences in vitamin D status is not known. We have reported seasonal variations in a cross-sectional study of over 3000 Scottish women in Aberdeen. The aim of this longitudinal study was to compare the seasonal variation of serum 25-hydroxyvitamin D [25(OH)D] in postmenopausal women residing in Aberdeen (57 degrees N) and Surrey (51 degrees N). Women attended 3-monthly visits over 12 months, starting summer 2006. In Aberdeen, 338 Caucasian women (mean age+/-SD, 61.7+/-1.5 years); and at Surrey, 138 Caucasian women (61.4+/-4.5 years) and 35 Asian women (59.9+/-6.4 years) had serum 25(OH)D measured by IDS enzyme immunoassay. In winter/spring none of the Caucasian women living in Surrey had 25(OH)D<20 nmol/L, but nearly a quarter of women in Aberdeen were vitamin D-deficient. This number decreased to 4.2% in summer/autumn. For the Asian women 17.1% were vitamin D-deficient in summer, increasing to 58.1% in winter. Using higher 25(OH)D deficiency cut-offs, the percentage of women affected was much higher. These longitudinal data show clear differences in vitamin D status between the north and south of the UK, and marked ethnic differences. They are consistent with our previous data and with cross-sectional data from the 1958 birth cohort. The low vitamin D status may have implications for bone health and other health outcomes, which is currently being investigated in this publication group. The extent of vitamin D deficiency in Asian women residing in the South of England is of concern.

Seasonal 25-hydroxyvitamin D changes in British postmenopausal women at 57 degrees N and 51 degrees N: a longitudinal study.

The UK has insufficient intensity of sunlight at wavelengths 290-315 nm to enable cutaneous synthesis of vitamin D from October to April. There are regional differences in UVB strength throughout the UK but whether this translates to differences in vitamin D status is not known. We have reported seasonal variations in a cross-sectional study of over 3000 Scottish women in Aberdeen. The aim of this longitudinal study was to compare the seasonal variation of serum 25-hydroxyvitamin D [25(OH)D] in postmenopausal women residing in Aberdeen (57 degrees N) and Surrey (51 degrees N). Women attended 3-monthly visits over 12 months, starting summer 2006. In Aberdeen, 338 Caucasian women (mean age+/-SD, 61.7+/-1.5 years); and at Surrey, 138 Caucasian women (61.4+/-4.5 years) and 35 Asian women (59.9+/-6.4 years) had serum 25(OH)D measured by IDS enzyme immunoassay. In winter/spring none of the Caucasian women living in Surrey had 25(OH)D<20 nmol/L, but nearly a quarter of women in Aberdeen were vitamin D-deficient. This number decreased to 4.2% in summer/autumn. For the Asian women 17.1% were vitamin D-deficient in summer, increasing to 58.1% in winter. Using higher 25(OH)D deficiency cut-offs, the percentage of women affected was much higher. These longitudinal data show clear differences in vitamin D status between the north and south of the UK, and marked ethnic differences. They are consistent with our previous data and with cross-sectional data from the 1958 birth cohort. The low vitamin D status may have implications for bone health and other health outcomes, which is currently being investigated in this publication group. The extent of vitamin D deficiency in Asian women residing in the South of England is of concern.

The direction of shift-work rotation impacts metabolic risk independent of chronotype and social jetlag - An exploratory pilot study

The aim of this pilot study was to explore the risk of metabolic abnormalities in steel workers employed in different shift-work rotations. Male workers in a steel factory [16 employed in a fast clockwise rotation (CW), 18 in slow counterclockwise rotation (CC), 9 day workers (DW); mean age 43.3 ± SD 6.8 years] with at least 5 years experience in their current work schedule participated. All workers provided fasting blood samples between 06:00 and 08:00 h for plasma glucose, insulin, apo-lipoproteins A and B (ApoA, ApoB), high- and low-density lipoproteins (HDL and LDL), total cholesterol (tCH), triglycerides (TG), minimally oxidized (mox) LDL, C-reactive protein (CRP), interleukin-8 (IL-8) and serum 25-hydroxyvitamin D (25(OH)D). HOMA index (homeostatic model assessment) was calculated to evaluate insulin resistance, beta cell function and risk of diabetes. Information on demographics, health, stimulants, sleep, social and work life, chronotype (phase of entrainment) and social jetlag (difference between mid-sleep on workdays and free days) as a surrogate for circadian disruption was collected by questionnaire. Neither chronotype nor social jetlag was associated with any of the metabolic risk blood markers. There were no significant differences in 25(OH)D, ApoA, ApoB, CRP, HDL, IL-8, insulin, LDL, mox-LDL, mox-LDL/ApoB ratio, tCH and TG levels between the three work groups. Although we did observe absolute differences in some of these markers, the small sample size of our study population might prevent these differences being statistically significant. Fasting glucose and HOMA index were significantly lower in CW compared to DW and CC, indicating lower metabolic risk. Reasons for the lower fasting glucose and HOMA index in CW workers remains to be clarified. Future studies of workers in different shift rotations are warranted to understand better the differential effects of shift-work on individual workers and their health indices.

Protein intake and bone health: a systematic review and meta-analysis

The impact of dietary protein on bone mass density (BMD), bone turnover and fracture risk has been widely researched. Theoretically, protein should benefit bone as a result of its influence on the anabolic hormone insulin-like growth factor 1. However, dietary protein also increases acid load to the body, thus encouraging calciuria, which may be detrimental to bone health. There is a considerable lack of consensus as to whether protein has a beneficial or detrimental effect on bone. Despite the importance of this topic, no systematic review and meta-analysis of studies has been undertaken to date to examine the relationship between protein intake and bone health. The aim of the present study was to conduct a systematic review and meta-analysis of the effect of dietary protein on BMD, bone turnover and fracture risk. MEDLINE (January 1966–September 2007) was searched for all cross-sectional, longitudinal and experimental studies of healthy participants. A total of twenty-one studies were included in the systematic review and six intervention studies in the meta-analysis. All chosen studies were checked by a least two researchers. As shown in the Table, the systematic review showed a positive correlation between total dietary protein and BMD in adults, with the variation in low BMD that was attributable to protein intake being approximately 2–4%. In the longitudinal studies the mean relative risk for any fracture in the highest quartile of total protein intake was 0.7. Some intervention studies showed an increase in markers of bone formation and a decrease in markers of bone resorption with protein supplementation and some found a benefit on BMD. No significant effect was found in the meta-analysis for lumbar spine BMD (P = 0.17, Z 1.36, weighted mean difference 0.05 (95% CI 0.02, 0.12)). The only significant effect found in the meta-analysis was for the effect of soybean protein on bone-specific alkaline phosphatase (BSAP), favouring the control group (P = 0.04, Z 2.08, weighted mean difference 0.03 (95% CI 0.00, 0.07)).

Higher bone resorption excretion in South Asian women vs White Caucasians and increased bone loss with higher seasonal cycling of vitamin D: results from the D-FINES cohort study

Few data exist on bone turnover in South Asian women and it is not well elucidated as to whether Western dwelling South Asian women have different bone resorption levels to that of women from European ethnic backgrounds. This study assessed bone resorption levels in UK dwelling South Asian and Caucasian women as well as evaluating whether seasonal variation in 25-hydroxyvitamin D [25(OH)D] is associated with bone resorption in either ethnic group. Data for seasonal measures of urinary N-telopeptide of collagen (uNTX) and serum 25(OH)D were analysed from n=373 women (four groups; South Asian postmenopausal n=44, South Asian premenopausal n=50, Caucasian postmenopausal n=144, Caucasian premenopausal n=135) (mean (±SD) age 48 (14) years; age range 18-79years) who participated in the longitudinal D-FINES (Diet, Food Intake, Nutrition and Exposure to the Sun in Southern England) cohort study (2006-2007). A mixed between-within subjects ANOVA (n=192) showed a between subjects effect of the four groups (P<0.001) on uNTX concentration, but no significant main effect of season (P=0.163). Bonferroni adjusted Post hoc tests (P≤0.008) suggested that there was no significant difference between the postmenopausal Asian and premenopausal Asian groups. Season specific age-matched-pairs analyses showed that in winter (P=0.04) and spring (P=0.007), premenopausal Asian women had a 16 to 20nmolBCE/mmol Cr higher uNTX than premenopausal Caucasian women. The (amplitude/mesor) ratio (i.e. seasonal change) for 25(OH)D was predictive of uNTX, with estimate (SD)=0.213 (0.015) and 95% CI (0.182, 0.245; P<0.001) in a non-linear mixed model (n=154). This showed that individuals with a higher seasonal change in 25(OH)D, adjusted for overall 25(OH)D concentration, showed increased levels of uNTX. Although the effect size was smaller than for the amplitude/mesor ratio, the mesor for 25(OH)D concentration was also predictive of uNTX, with estimate (SD)=-0.035 (0.004), and 95% CI (-0.043, -0.028; P<0.001). This study demonstrates higher levels of uNTX in premenopausal South Asian women than would be expected for their age, being greater than same-age Caucasian women, and similar to postmenopausal Asian women. This highlights potentially higher than expected bone resorption levels in premenopausal South Asian women which, if not offset by concurrent increased bone formation, may have future clinical and public health implications which warrant further investigation. Individuals with a larger seasonal change in 25(OH)D concentration showed an increased bone resorption, an association which was larger than that of the 25(OH)D yearly average, suggesting it may be as important clinically to ensure a stable and steady 25(OH)D concentration, as well as one that is high enough to be optimal for bone health.

Food Groups and Bone Health

It is important to study the effects of food groups and whole foods on bone health. This is because nutrients may interact in a synergistic manner to influence bone health and osteoporosis risk. This whole diet and food based approach has yielded many insights into the relationship between nutrition and bone health. Cohort and cross-sectional data suggest that diets that are higher in fruit, vegetables, milk and cereal are associated with increased bone mass as compared with diets high in processed and snack foods. Consumption of milk and other dairy products appears to have beneficial effects on building bone mass in childhood and adolescence, and may also help offset bone loss after the menopause. However, more research is required to assess whether milk and dairy product consumption can prevent fractures in later life. The effects of veganism and vegetarianism on bone health, as compared with omnivorous diets, are not yet clear, with conflicting results being found from different research studies. Some research suggests that diets rich in fruit and vegetables may benefit bone health via increased physiological alkalinity. However, conflicting results have been found from recent intervention trials that have attempted to assess the effect of fruit and vegetable supplementation on bone. Alcohol, caffeine and soda intakes have the potential to influence bone health. Currently there is evidence that alcohol may be beneficial to bone in moderation, but toxic to bone at higher doses. There is also concern about the potential negative effects of soda on bone health. However, data are difficult to interpret due to the strong interactions between soda intake and lifestyle factors that are detrimental to bone health.

Relative contributions of different food groups to vitamin D intake in Asian and Caucasian UK women: results from the Vitamin D, Food Intake, Nutrition and Exposure to Sunlight in Southern England (D-FINES) Study

There has been a lack of research into the contribution of different food groups to vitamin D intake in varying age and ethnic groups in theUK. The 2006Ð7 D-FINES Study was undertaken in 373 Surrey premenopausal (PRM) and post-menopausal (POM) Caucasian (C) andAsian (A) women. They were assessed in four seasons of the year for vitamin D (25OHD) status, vitamin D dietary intake and UVBexposure. Relative contributions from different food groups for vitamin D intake were calculated for A and C women in each of the fourseasons.As can be seen in the Table, the top three sources of vitamin D in Asians were meat, cereals and eggs. In Caucasians the top sourceswere meat, Þsh and cereals. Analysis by ANOVA indicated that A women obtained a signiÞcantly higher percentage of vitamin D fromeggs and egg products, meat products and from vegetable foods than the C women. However, C women had a signiÞcantly higher absolutevitamin D intake ( P < 0.001). When the women were divided into four groups by ethnicity and menopausal status, signiÞcant differenceswere found in vitamin D intake from cereals, egg products, vegetables and snacks between the groups but no differences were found formilk, Þsh, meat and fats. These differences were found to be to be signiÞcant ( post hoc tests) only between the PRM A and POM Cgroups. For absolute intakes, there was a signiÞcant difference between the four groups ( P < 0.001). POM C had the highest intake ( posthoc tests), followed in order by PRM C, PRM A and POM A, although signiÞcant differences were only found between intakes of POM Aand POM C ( P < 0.001), and between POM A and PRM C ( P = 0.026).

Associations between serum 25-hydroxyvitamin D and sleep, as estimated by actigraphy and the Pittsburgh Sleep Quality Index (PSQI)

It is unknown whether vitamin D status affects sleep health, but recent studies suggest vitamin D deficiency is associated with shorter sleep duration and lower sleep efficiency. This study investigated whether there is a relationship between vitamin D status and sleepwake cycles in UK dwelling South Asian (SA) and Caucasian (C) women, using ambulatory actigraphic data and self-reported sleep quality data from the D-FINES II (Vitamin D, Food Intake, Nutrition and Exposure to Sunlight in Southern England II) study. In June-August 2010, serum 25-hydroxyvitamin D [25(OH)D] and data on self-reported musculoskeletal pain were collected from n = 47 women. In September-October 2010, participants wore Actiwatch-L (AWL, Cambridge Neurotechnology) monitors on their wrists for 24 h/day, over 14 consecutive days to measure sleep-wake activity as well as completing the PSQI (self-reported sleep quality) once. A subset of n = 37 women also wore an AWL on a neckband during the daylight hours to measure environmental light exposure. Each subject’s actigraphic data (including light exposure) were eligible to be included in the statistical analysis if they had ⩾ 7 days of valid data and a 25(OH)D measurement. Relationships between 25(OH)D and actigraphic measures were analysed by Pearson’s bivariate correlations, as well as by partial correlations to control for potential confounders. PSQI scores are ordinal data so relationships were analysed by Spearman’s correlations only. There was a significant negative relationship between 25(OH)D concentration and actigraphic sleep latency in SA (r =−0·562, P = 0·036), and a significant positive relationship between 25(OH)D and both overall PSQI score (r = 0·385 P = 0·047) and PSQI sleep latency subscale (r = 0·439, P = 0·02) in C (see Table). Partial correlations controlling for bone pain (n = 23 C, n = 11 SA) found a statistically significant positive relationship between 25(OH)D and actigraphic sleep latency (r = 0·426, P = 0·048, n = 23) in C only. However, when adjusting for muscle pain (n = 21 C, n = 8 SA), there were no significant associations between 25(OH)D and actigraphic sleep parameters in either ethnic group (P > 0·05). Finally, there were no significant correlations between 25(OH)D and actigraphic sleep parameters when adjusting for outdoor light exposure (mins/d > 1000 lux) (P > 0·05, n = 20 C, n = 8 SA).

Comparison of vitamin D status in spring 2007 and spring 2008 among Caucasian and Asian women living in the South of England

An individual’s vitamin D status depends on sunlight (UVB) exposure and dietary intake of vitamin D. 25(OH)D status throughout the year is thought to some extent to be dependent on sunlight exposure in the preceding summer. Theoretically, a cloudy and rainy summer with low temperatures could result in low vitamin D levels in the subsequent spring. Additionally, in low temperatures and in wet climates, people tend to wear heavier clothing and spend more time indoors, adding further to the reduced synthesis of vitamin D. In the summer of 2006, Southern England was less cloudy than in the summer of 2007, with a higher number of sunshine hours per day, less rainfall and a higher mean temperature. Due to the adverse weather conditions experienced in summer 2007, we hypothesised that in spring 2008 women would have lower 25(OH)D levels than they had earlier in spring 2007. From summer 2006 to spring 2007 participants were assessed in each season for serum 25(OH)D, UVB exposure and vitamin D intake as part of the main D-FINES (Vitamin D, Food Intake, Nutrition and Exposure to Sunlight in Southern England) study. After the completion of this study in spring 2007, participants were allowed to recommence on vitamin D supplements. Subjects were re-assessed for serum 25(OH)D and UVB exposure in spring 2008. For the current analysis, in order to exclude the impact of vitamin D supplementation on serum 25(OH)D levels, we only compared participants in spring 2007 to participants in spring 2008 who were not taking any vitamin D supplements in both seasons. Initially, Asian and Caucasian women were analysed irrespective of menopausal status, with a paired-samples t-test conducted to assess differences in 25(OH)D between the two paired spring measurements. There was no significant difference in 25(OH)D between the two spring measurements (Table 1). In the Caucasian participants only, results were further analysed based on menopausal subgroups. A paired-samples t-test showed no significant difference in 25(OH)D between the two spring measurements in either of the menopausal subgroups. Asian subgroups were not analysed due to small subject numbers (n= 5).