Home

Abstract (Summary)

Discussion

Over the 12-month intervention period certain favourable dietary changes have been observed for the DG and to a lesser extent for the CaG (Table 2). More specifically both DG and CaG increased daily calcium intake to levels exceeding 1000 mg/d, reaching the recommended adequate intake level of 1200 mg/d (Institute of Medicine Food & Nutrition Board, 2001). Furthermore, increased intakes of protein, magnesium, phosphorus and vitamin D were observed for the DG compared to the two other groups. The changes in the aforementioned nutrients were mainly delivered by the increased consumption of fortified dairy products provided to the DG subjects. These favourable dietary changes are indicative of the effectiveness of the 'Health and Nutrition Education' component of the programme in increasing the self-efficacy of the population under study to comply with the given dietary instructions. This is also reflected in their total energy intake, which was not differentiated over the intervention period compared to the other two groups. Similar to the present findings, other intervention studies conducted with middle-aged women, including nutrition education as an integral part of their intervention scheme, also reported high adherence to the dietary guidelines provided (Bowen et al. 2002). On the contrary, compliance was considerably lower in intervention programmes not followed by regular nutritional education sessions (Storm et al. 1998).

Regarding physical activity, no differences were observed among groups (Table 3), despite the fact that subjects in the DG were instructed and encouraged to increase their exercise levels. This observation is common in other intervention studies where similar difficulties in motivating middle-aged women already having a sedentary lifestyle to become more active were also confronted (Prince et al. 1995; Wolff et al. 1999). Nonetheless, the lack of significant changes in both physical activity and energy intake levels supports the lack of significant differences among groups with respect to the changes observed in BMI.

Based on the data derived from DXA measurements, the current study showed that ensuring calcium intake of approximately 1200 mg/d in the CaG did not manage to induce any favourable bone mass changes in all skeletal sites examined compared to the CG. Similarly to the current study, Prince et al . (1995) observed no BMD changes after providing 1000 mg calcium/d for 1 year to Caucasian women. However, data from other intervention studies conducted with Caucasian cohorts using calcium supplementation as a primary mode of treatment showed that higher doses of 1600 mg/d were adequate to prevent bone loss from the total body, lumbar spine, femoral neck and greater trochanter (Riggs et al. 1998; Storm et al. 1998).

Contrary to the CaG, significant favourable effects on bone mass (total body, total spine and pelvis BMD) were only apparent for the DG, while no differences were observed for the upper and lower body extremes (Table 4). Similarly bone loss prevention from lumbar spine but no BMD changes at the femoral neck were obtained from a dairy intervention study on Caucasian women with comparable nutrient intake to the current study (Storm et al. 1998). When a similar intervention scheme with fortified dairy products was applied to Asian women (Lau et al. 2002; Chee et al. 2003), comparable favourable bone mass changes in lumbar spine and total body were observed, while additional favourable bone mass changes were also observed for the total hip and femoral neck.

From intervention studies using supplements of comparable doses of calcium but considerably higher doses of vitamin D than those administered in the studies presented earlier, including the present one, limited or no BMD changes were obtained at the examined skeletal sites (Dawson-Hughes et al. 1997; Baeksgaard et al. 1998; Hunter et al. 2000; Chapuy et al. 2002; Meier et al. 2004). From these five studies only, three (Dawson-Hughes et al. 1997; Baeksgaard et al. 1998; Meier et al. 2004) reported favourable bone mass changes at the lumbar spine, while none of them reported prevention of bone loss either for total body or at forearm and femoral neck skeletal sites.

Summarizing the findings of the studies presented earlier, supplementation of dairy products fortified with calcium and vitamin D 3 consistently showed favourable BMD changes for total body as well as at several skeletal sites both in Caucasian and Asian cohorts. Still there seems to be a racially different response to treatment, since dairy interventions implemented on Caucasian women, including the present study, consistently reported preservation of bone mass at the lumbar spine but no changes at the lower body extremes. The opposite was observed for Asian women where favourable bone loss retardation was observed in lumbar spine but also bone preservation at the total hip and femoral neck. The different response to treatment between Caucasian and Asian women could be explained by racial differences with respect to spinal and hip bone properties (i.e. density, architecture, turnover and mineralization) (Wu et al. 2003).

The favourable BMD changes observed for subjects in the DG were also followed by increases in QUS parameters of the calcaneus; still the changes observed did not reach a significant level (Table 5). To our knowledge there are no other studies available in the literature examining changes in both BMD and QUS parameters induced by the implemented dietary intervention programme. The limited intervention studies testing the aforementioned hypothesis have used either supplements or pharmaceutical regimens in their treatment protocol and have reported inconsistent findings. More specifically, two studies (Hunter et al. 2000; Chapuy et al. 2002) found no changes in BMD at several skeletal sites (i.e. lumbar spine, femoral neck and total body), as well as in QUS parameters of the calcaneus (i.e. BUA and SOS) after the completion of 1 year of intervention providing supplements of calcium and vitamin D 3 . Only when supplementation was also followed by supervised weekly exercise sessions were positive changes in lumbar spine BMD and SOS of the heel observed (Engelke et al. 2006). Similar findings to Engelke et al . (2006) have been also reported by intervention studies examining skeletal response to treatment using pharmaceutical regimens (i.e. hormone replacement therapy, biphosphonates, calcitonin), where favourable changes were recorded both for BMD (regional skeletal sites and total body) as well as for QUS parameters of the calcaneus after 1 or more years of treatment (Rosenthall et al. 1999; Frost et al. 2001 a ). Finally, in a study conducted by Krieg et al . (1999), after providing 1000 mg calcium/d and 20 [mu]g vitamin D 3 /d, no changes in heel BUA were observed at 12 months, while a significant increase was observed at 24 months.

Based on the aforementioned observations, it has been suggested that more time is required for significant response in QUS parameters to be detected, compared to DXA (Rosenthall et al. 1999). The exact reasons for this are still speculative. To some extent this could be attributed to the fact that QUS parameters also reflect non-mass properties of bone (Hans et al. 1999), which have been suggested to respond more slowly to treatment than BMD (Hans et al. 1998; Gluer, 1999; Rosenthall et al. 1999). Furthermore, the ability of QUS to detect treatment response is not yet well established, mainly due to its lower long-term precision and longitudinal sensitivity (i.e. the ability to monitor skeletal changes over time induced by a specific treatment) compared to DXA (Gluer, 1999). This is also confirmed in the present study, since similarly to other studies (Sahota et al. 2000; Frost et al. 2001 a ), the long-term precision for QUS parameters was found to range from 1·5 to two times that of legs BMD and from two to three times that of lumbar spine BMD. The present observations indicate that although QUS has a potential role in long-term monitoring of skeletal status changes, the period of time required to follow individual subjects remains 1·5 to three times that for conventional DXA measurements (Sahota et al. 2000).

The current study revealed that the application of a holistic approach combining dietary intervention and consumption of fortified dairy products for a period of 12 months can induce favourable changes in pelvis, total spine and total body BMD of postmenopausal women but not in QUS parameters, which probably require longer intervention periods to reach a significant level. Contrary, no such favourable changes either on BMD or on QUS parameters were obtained in the CaG after supplementation of the recommended calcium dose. The favourable BMD changes observed for the DG might not be solely attributed to the increased intake of calcium and vitamin D but also to other less studied ingredients of dairy products. Recent research has highlighted the important role of potassium, magnesium, vitamin A and other micronutrients (Weinsier & Krumdieck, 2000) but also of milk protein on bone metabolism (Aoe et al. 2005). It has been suggested that the effect of dairy products on bone health may be more than can be accounted for by any single constituent and that milk ingredients as a whole may be more effective than the sum of its individual parts (Weinsier & Krumdieck, 2000).