| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Dairy Science,
2 Department of Animal and Poultry Sciences,
3 Department of Human Nutrition, Foods and Exercise, and
4 Department of Biological Systems Engineering, Virginia Polytechnic Institute and State University, Blacksburg 24061
5 Department of Animal Sciences, University of Florida, Gainesville 32611
6 Department of Animal Science, Cornell University, Ithaca, NY 14853
Corresponding author: K. F. Knowlton; e-mail: Knowlton{at}vt.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS and METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
4) stage of lactation (Stage 1: <90 DIM, Stage 2: 90 to 150 DIM, Stage 3: 151 to 250 DIM, and Stage 4: >250 DIM). Samples were analyzed by radiographic photometry and dual-energy x-ray absorptiometry and then analyzed for mineral content chemically. In both experiments, the relationship between mineral content estimated via the imaging techniques and mineral content measured chemically was poor, likely because of the relative maturity of animals in the sample set and lack of variation in mineral content. Ash content was higher in the metacarpal than in the caudal vertebrae, as were concentrations of Mg (expressed as a proportion of bone ash). No effects of stage of lactation were observed on bone mineral in the caudal vertebrae, but in the metacarpal, P content (proportion of total mineral) was highest in second lactation cows. Total bone mineral content (ash) was not affected by parity in the metacarpal or caudal vertebra, but Ca and P content of the metacarpal increased with parity. Noninvasive imaging techniques are not sufficiently sensitive to detect changes in mineral content or composition of mature cows, and only modest changes in bone mineral were observed with stage of lactation and parity.
Key Words: bone mineral content • radiographic photometry • dual-energy x-ray absorptiometry
Abbreviation key: BMC = bone mineral content, BMD = bone mineral density, DXA = dual-energy x-ray absorptiometry, RBAE = radiographic bone aluminum equivalents, RP = radiographic photometry
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS and METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
All bone consists of mineral, primarily hydroxyapatite (Ca10 (PO4)6 (OH)2) deposited in an organic matrix, of which collagen is the major constituent. Bone mineral is deposited in large numbers of small crystals, giving bone a large surface area. The surface area of 1 g of bone mineral is estimated to be >100 m2 (Omnell, 1957); this surface area allows for a rapid interchange of ions between interstitial fluid and bone.
Bone mineral content (BMC) is a result of the balance between bone formation and resorption, and it is influenced by diet, age, and physiological state. In adult mammals, approximately 20% of the bone wet weight is water, 45% is ash, and 35% is OM (Carter and Spengler, 1978). Calcium contributes 37% of the ash content, and P contributes 18.5%. On a dry weight basis, mineral content is 65 to 70%, and OM is 30 to 35%. Bone loss occurs when resorption exceeds bone formation, with osteoporosis in humans as a classic example of bone imbalance in part attributable to diet. This chronic condition results in fractures and loss of bone mass.
In ruminants, changes in BMC have been reported or inferred during lactation (Benzie et al., 1959; Braithwaite, 1983; Knowlton and Herbein, 2002) and with diet. Shupe et al. (1988) reported signs of osteoporosis in beef cattle fed low amounts of P (6 to 12 g/d) for 2 yr.
Bone P content has been used as an indicator of P deficiency in lactating cows (Wu et al., 2001), and such an indicator would be useful in the field. The link between overfeeding P and increased P excretion is well established (Morse et al., 1992; Wu et al., 2000; Knowlton and Herbein, 2002), but the variability in P content of feeds and its uncertain bioavailability creates risk associated with underfeeding (impaired growth or milk yield). An on-farm, noninvasive tool to assess the P status of cows would allow dairy producers and their advisors to revise dietary P to reduce the overfeeding common in the field (Sink et al., 2000). Direct chemical assessment of BMC requires either bone biopsy (Little, 1972), surgical removal of bone sections (Wu et al., 2001), or postmortem assessment, none of which is practical in a production setting.
In addition to their use as a tool to allow refined diet formulation, noninvasive measures would also be useful to monitor changes in BMC during lactation. Two imaging techniques, dual-energy x-ray absorptiometry (DXA) and radiographic photometry (RP), have been used as noninvasive predictors of BMC in other species.
The DXA method assesses BMC, bone mineral density (BMD), and bone area with low radiation (Mazess et al., 1990; Adams, 1997) and low precision error (<1%; Mazess et al., 1990). The method is used extensively in human bone research, and it is considered the "gold standard" method to assess the potential for bone fractures in humans (Mazess et al., 1990). Some research has been conducted with DXA in swine (Mitchell et al., 1996; Yang et al., 1998; Mitchell et al., 2001) and chickens (Mitchell et al., 1997), but its use in livestock is limited because of its high cost and low weight limit (136 kg).
Radiographic photometry has been used to estimate BMC and evaluate the effect of nutrition and condition on BMC in horses (Meakim et al., 1981; Porr et al., 1998; Hoffman et al., 1999) and beef cattle (Williams et al., 1991). Radiographic photometry utilizes a portable x-ray unit, x-ray film, scanning equipment, and imaging software to estimate BMC. X-rays are absorbed by Al to the same degree as bone mineral (Ekman et al., 1970), allowing use of an Al step wedge as a reference standard. The BMC estimated via RP is expressed in radiographic bone aluminum equivalents (RBAE) in millimeters of Al. There are no published reports of the use of DXA or RP to estimate BMC in dairy cattle.
The objectives of the study were to evaluate RP and DXA as predictors of BMC in mature dairy cows and to evaluate the effects of parity, stage of lactation, and site of measurement (fused third and fourth metacarpal bone vs. caudal vertebrae 14 and 15) on BMC.
|
MATERIALS and METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS and METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Experiment 2.
Caudal vertebrae and right front metacarpals (sample pairs) from Holstein cows (n = 107) were excised and collected at slaughter (Taylor Packing) and stored frozen. Information on parity, age, and DIM was obtained for donor animals for 43 sample pairs. Within this subset, sample pairs were grouped by DIM and parity to allow evaluation of the effects of these factors.
Sample Analysis
RP (Experiments 1 and 2)
Dorsopalmer and anterior-posterior radiographs were taken of caudal vertebrae 14 and 15 using a high frequency portable x-ray unit (MinXRay HF-80; MinXRay, Northbrook, IL). Radiographs of caudal vertebrae (experiments 1 and 2) were taken (75 KV, 0.04 s, 66-cm focal distance) on Kodak TMGRA radiographic film in a Kodak Lenex regular screen (18 x43 cm). Radiographs of the metacarpal (experiment 2 only) were taken (80 KV, 0.06 s, 66-cm focal distance) on Kodak TMGRA radiographic film in a Kodak Lenex regular screen (24 x30 cm). For all radiographs, an 11-step Al step wedge was taped to each screen and exposed simultaneously in each radiograph as a reference standard. Film was developed using an automatic processor (Kodak RP X-Omat; model M-6).
X-ray images were scanned (Epson Expression 1680 scanner) and transferred to imaging software (Image-Pro Plus for Windows, version 4.5, Media Cybernetics, Silver Springs, MD). For each film, a calibration curve was developed using the 11-step standard Al step wedge. In the metacarpal, a location 10 cm from the nutrient foramen was evaluated. Observations of RBAE recorded were as follows: 1) peak medial RBAE, mm Al, which is the point of highest absorbance (greatest mineralization) on the medial side of the bone; 2) peak lateral RBAE, mm Al, which is the point of highest absorbance on the lateral side of the bone; 3) bone width, mm, which is the width of the widest part of the bone; 4) cortical width, mm, which is the width of bone less the width of the medullary cavity; 5) bone length, mm, which is the total length of bone (caudal vertebrae only).
DXA (Experiment 2 only).
Samples were thawed, and BMC and BMD were measured on each metacarpal and caudal vertebrae 14 and 15 using DXA (QDR 4500 A; Hologic, Inc., Bedford, MA) with the standard lumbar spine (L1–L4) and forearm protocols (Beiseigel and Nickols-Richardson, 2002). Caudal vertebrae and metacarpals were placed prone on the table, and anterior-posterior scans were performed. Both BMD (g/cm2) and BMC (g) were estimated for caudal vertebrae 14 and 15 and the metacarpal.
BMC (Experiment 1).
Bone samples were dried in an oven at 105°C for 16 h, and fat was removed by petroleum-ether extraction. The bones were again dried at 105°C overnight, weighed, and ashed in a muffle furnace at 600°C overnight. Ash weight was recorded. Ash was solubilized in HCl (Miles et al., 2001), samples were analyzed for Ca and Mg by flame atomic absorption spectrophotometry (Perkin-Elmer 5000; Perkin-Elmer Corp., Norwalk, CT), and P was determined colorimetrically (Harris and Popat, 1954). Macro-elements were expressed as a percentage of ash on a dry, fat-free basis.
BMC (Experiment 2).
After imaging analysis was completed, mechanical tests of breaking strength were conducted (Keene et al., 2004), and a band saw was used to cut the metacarpal into 1-cm cross sections (±0.01 mm) on either side of a mark, indicating the point 10 cm from the nutrient foramen. The caudal vertebrae (complete bones) and metacarpal (2 cm sections) were analyzed for total mineral content by ashing the bone. Samples were dried to a constant weight at 105°C and then ashed in a muffle furnace at 600°C for 12 h (AOAC, 1990). Concentrations of Ca, P, and Mg were assessed via inductively coupled plasma and expressed as a proportion of bone ash.
Statistical Analyses
Experiment 1.
Thirty-three caudal vertebrae samples were used to analyze the relationship between RP and chemical measures of BMC. Regression analysis was conducted using the simple regression model in PROC REG in SAS version 8.0 (SAS Institute, 1999).
 |
where
| Yi | = | response for the dependent variable (all RP measures),
| Xij | = | independent variable (all chemical measures of BMC),
| ß0 | = | intercept estimate,
| ß1 | = | slope estimate (regression of RP measures on chemical measures of BMC), and
| eij | = | residual error.
|
For all statistical analyses, effects were declared significant at P <0.05 and trends at P <0.10.
Experiment 2: Relationship between imaging and chemical measures of BMC
Seventy-eight pairs of caudal vertebrae and metacarpals were used to calculate the relationship between chemical and imaging (RP and DXA) measures of BMC. The sample set was incomplete because 29 of the 107 original metacarpal samples were collected in a way that prevented identification of a reference point 10 cm from the nutrient foramen, an important landmark. Regression analysis was conducted using the simple regression model in PROC REG in SAS version 8.0 (SAS Institute, 1999).
|
|
where
| Yi | = | response for the dependent variable (all RP and DXA measures),
| Xij | = | independent variable (all chemical measures of BMC),
| ß0 | = | intercept estimate,
| ß1 | = | slope estimate (regression of RP and DXA measures on chemical measures of BMC), and
| eij | = | residual error.
|
Experiment 2: Relationship of BMC with site of measurement.
Seventy-eight pairs of caudal vertebrae and metacarpals were used to evaluate the relationship with the site of measurement on BMC. Data were analyzed using PROC Mixed in SAS version 8.0 (SAS Institute, 1999) with the following model:
 |
where
| Yij | = | response of dependent variable (chemical measures of BMC),
| ß | = | mean,
| ai | = | fixed effect of site (metacarpal, caudal vertebrae 14 and 15),
| bj | = | random effect of cow, and
| eij | = | residual error.
|
Using preplanned contrasts, values for the metacarpal were compared with values for the caudal vertebrae, and values for caudal vertebra 14 were compared with values for caudal vertebra 15. Results are presented as least squares means.
Experiment 2: Relationship of BMC with stage of lactation.
Forty-three sample pairs of caudal vertebrae and metacarpals (all samples for which stage of lactation was available) were used to evaluate the effects of stage of lactation on BMC. Sample sets were grouped by stage of lactation (Stage 1: <90 DIM, Stage 2: 90 to 150 DIM, Stage 3: 151 to 250 DIM, Stage 4: >250 DIM). The relationship of chemical measures of BMC with stage of lactation was analyzed using Proc GLM of SAS version 8.0 (SAS Institute, 1999) with the following model:
 |
where
| Yi | = | response (chemical measures of BMC),
| µ | = | mean,
| ai | = | fixed effect of stage of lactation (i = 1 to 4), and
| eij | = | residual error.
|
Experiment 2: Relationship of BMC with parity.
Forty-three sample pairs of caudal vertebrae and metacarpals (all samples for which parity was available) were used to evaluate the effects of parity on BMC. Sample sets were grouped by parity (1, 2, 3, 4). The relationship of parity with BMC was analyzed using Proc GLM of SAS version 8.0 (SAS Institute, 1999) with the following model:
|
|
where
| Yi | = | response (chemical measures of BMC),
| µ | = | mean,
| ai | = | fixed effect of parity (i = 1 to 4 lactations), and
| eij | = | residual error.
|
It would have been preferable to include stage of lactation and parity in the same model, but the number of cows within each subclass was small. Linear and quadratic effects of stage of lactation and parity were assessed using preplanned contrasts. Linear, quadratic, and cubic regressions of BMC and breaking strength on DIM were fit, but the data were unequally distributed, and effects were not significant (results not shown). Data are presented as least square means.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS and METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
) and in caudal vertebrae 14 and 15 and the metacarpal in experiment 2 (Table 1). In experiment 1, the relationship was stronger in the larger caudal vertebra 15 than in caudal vertebra 14. Radiographic bone Al equivalents accounted for 24% of the variation in bone ash in the former but only 5% in the latter. The reason for the much stronger relationship between RBAE and bone ash in the caudal vertebra 15 is unclear. Total bone ash tended to be higher in the smaller caudal vertebra 14, but differences were small (61.6 vs. 61.0%; P <0.10).
|
|
). Although no relationship was large, it appears that in the caudal vertebrae, RP measures reflected total bone ash, whereas in the metacarpal, RP was more predictive of the individual minerals (Ca, P, and Mg) than of total BMC. The observations in the current study fail to reflect the strong predictive relationship between RBAE estimates and chemical measures of BMC observed in the metacarpal of horses (Meakim et al., 1981; Lawrence, 1986; Hoffman et al., 1999) and beef cattle (Williams et al., 1990). In these studies, coefficients of determination of >90% were observed. One possible explanation for this contradiction is the relative maturity of the sample donors in the current studies. In studies reporting strong relationships between BMC measured chemically and estimated via RP, the experimental units were growing animals (Meakim et al., 1981; Lawrence, 1986; Hoffman et al., 1999) with a wide range of observed BMC. In the current experiments, samples were collected from mature cows at slaughter. No data were obtained on the age of the donor animals in experiment 1, but, in experiment 2, the 43 cows for which age information was available ranged in age from 3 to 6 yr. Radiographic photometry may not be sensitive enough to detect small differences in BMC in mature animals.
The relationships between chemical and RP measures of BMC were somewhat stronger in experiment 1 than in experiment 2. The relationship between bone ash and peak lateral RBAE was significant in the CV 15 of the 33 samples in experiment 1, with RBAE accounting for more variation in bone ash than in any bone in experiment 2. The difference may be due to differences in methods of assessing bone ash in the 2 studies. In experiment 1, bone ash was expressed as a proportion of the fat-free DM, whereas fat was not extracted from samples prior to mineral analysis in experiment 2. Even within experiment 1, however, the coefficient of determination was much lower than was observed in the studies focusing on growing beef cattle or horses (Meakim et al., 1981; Lawrence 1986; Hoffman et al., 1999).
Chemical and DXA measures (experiment 2).
Although the weight limit and expense of the equipment prevents its use as a field tool for assessing mineral status, DXA was evaluated in the current study because of its status as the primary means of assessing body content in humans and for its potential as a research tool. As with RP, however, only weak relationships were observed between chemical and DXA measures of BMC (Table 2). In the caudal vertebrae, no relationships between DXA and chemical measures were significant. The strongest relationships between DXA and chemical measures observed were in the metacarpal. Calcium and P content (proportion of ash) accounted for 12.6 and 10.6% of the variation in BMD estimated by DXA and 10.4 and 8.8% of the variation in BMC estimated by DXA. In contrast, several other studies report strong relationships between BMC and BMD estimated with DXA and BMC measured chemically. In the vertebrae of sheep (Pouilles et al., 2000) and in pigs (total body ash; Mitchell et al., 1998), DXA measures accounted for 90% of the variation in bone ash. In 4 experiments with broilers, Mitchell et al. (1997) observed that DXA BMC accounted for 46% of the variation in total body ash of broilers. In that study, prediction of ash content was less precise in larger birds.
|
), whereas Mg content (proportion of ash) was lower in the metacarpal than the caudal vertebrae. Calcium and P content of the metacarpal and caudal vertebrae were similar. Total ash and Ca content of caudal vertebra 14 and caudal vertebra 15 were similar, but P content as a proportion of ash was lower in caudal vertebra 14 than in caudal vertebra 15. Imaging estimates of BMC (peak lateral RBAE, peak medial RBAE, and DXA measures of BMC and BMD) all indicated greater mineral content in the metacarpal than in the caudal vertebrae. These means are reported to demonstrate the sensitivity of RP and DXA relative to chemical measures. The biological importance of these data are limited, however, given the poor relationships observed between imaging and chemical measures of BMC.
|
Because of its porous nature, trabecular bone plays a larger role in metabolic functions, with a more rapid rate of bone turnover than in cortical bone (Albright, 1987). In midlactation ewes, the trabecular bone lost 50% of its mineral compared with controls killed in early gestation (Benzie et al., 1959). In contrast, only with severe Ca or P restriction was significant loss of bone mineral observed in the diaphysis of long bones (Benzie et al., 1955; Benzie et al., 1959). Although subject to more rapid depletion, the trabecular bone of the vertebrae was also repaired more rapidly than the diaphysis of long bone (Benzie et al., 1955; Benzie et al., 1959).
Relationship of Stage of Lactation and Parity with BMC
Bone mineral and stage of lactation.
Total bone mineral and concentration of Ca, P, and Mg were unaffected by stage of lactation in caudal vertebrae 14 and 15 (Table 4). In the metacarpal, total bone mineral and composition of Ca and Mg were similar in cows of different stages of lactation, but bone P composition was highest in cows between 90 and 150 DIM (significant quadratic effect).
|
Also, because of its increased porosity, the caudal vertebrae would be expected to be more responsive to changes in stage of lactation. Instead, we observed no changes in mineral content of the caudal vertebrae with stage of lactation.
Bone mineral and parity.
In this study, effects of both age and parity on BMC were initially assessed, and similar effects were observed. Because parity accounted for a greater proportion of the variability in BMC than did age, only the effects of parity will be discussed.
Total BMC was unaffected by parity in all bones (Table 5). Increasing parity increased the Ca content of the metacarpal linearly, with a similar trend observed for metacarpal P content (P <0.10). Within caudal vertebra 14, a significant effect of parity on Ca content of caudal vertebra 14 was observed, but the effect was quadratic, with a peak observed in parity 2 and the lowest value in parity 4. The Mg content of caudal vertebra 14 decreased with parity (linear effect). In the caudal vertebrae, P content was unaffected by parity.
|
Results observed in our study differed from results of other studies that indicated striking changes in BMC with stage of lactation (Benzie et al., 1959; Braithwaite, 1983; Knowlton and Herbein, 2002). This is likely due to differences in sample collection. Rather than repeatedly measuring mineral balance in live animals over time as is possible with total collection or isotope infusion studies, we measured BMC directly in samples obtained at slaughter. This sample set was not ideally suited for measuring changes in mineral content with time, as animals could not be measured repeatedly throughout lactation.
One technique that allows repeated, direct measures of BMC is serial biopsy of the rib bone. Little (1972) reported that the technique of serial biopsy of rib bone allowed more sensitive assessment of mineral status compared with one-time sampling. In the only other published research assessing the relationship of stage of lactation and measured bone mineral in dairy cows, Beighle (1999) conducted serial rib biopsy and observed no change in P content of cortical rib bone throughout lactation. Calcium content of cortical rib bone (as a proportion of total ash) was higher at parturition and at 30 DIM compared with bone sampled 60 and 120 d postpartum. Cortical bone was thinner at parturition than later in lactation, however, suggesting that concentration of bone mineral may not be an accurate indicator of total bone mineral reserves (Beighle, 1999).
Changes in total bone mass likely overwhelm changes in mineral concentration in determining body mineral reserves. Benzie et al. (1959) weighed individual bones and the total skeleton, obtaining mass measurements in addition to data on mineral concentration. They reported that the total ash weight of individual bones and the total skeleton in sheep changed more dramatically with mineral deficiency and stage of lactation than did bone ash concentration (g/kg of bone). In human cadavers (McCalden et al., 1993), qualitative changes in bone tissue (porosity) changed more with age than did quantitative changes (bone mineral).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS and METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS and METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Current address: Kentucky Equine Research, Versailles, KY 40383.
Received for publication April 1, 2004. Accepted for publication June 22, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS and METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. S. Taylor, K. F. Knowlton, M. L. McGilliard, W. M. Seymour, and J. H. Herbein Blood Mineral, Hormone, and Osteocalcin Responses of Multiparous Jersey Cows to an Oral Dose of 25-Hydroxyvitamin D3 or Vitamin D3 Before Parturition J Dairy Sci, June 1, 2008; 91(6): 2408 - 2416. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
