JOURNAL OF BONE AND MINERAL RESEARCH Volume 5, Number 4,1990 Mary Ann Liebert, Inc., Publishers
Human Lactation: Forearm Trabecular Bone Loss, Increased Bone Turnover, and Renal Conservation of Calcium and Inorganic Phosphate with Recovery of Bone Mass Following Weaning G. NEIL KENT,' ROGER I. PRICE,'.' DONALD H. GUTTERIDGE,' MARGARET SMITH,3 JANET R. ALLEN,' CHOTOO I. BHAGAT,4 MARION P. BARNES,'*6CAROLINE J. HICKLING,' ROBERT W. RETALLACK,' SCOTT G. WILSON,' ROWENA D. DEVLIN,' CHARMIAN DAVIES5 and ANDREW ST. JOHNS
ABSTRACT The calcium (Ca) metabolism of established human lactation was studied in 40 adult women (mean age 32.4 years) who had been breast-feeding for 6 months (Lac) and in 40 age-matched controls (Con)using fasting urine and blood biochemistry and forearm single-photon bone mineral densitometry (BMD). Serial studies were performed up to 6 months after weaning in Lac women and repeated once in Con women. During luctution the significant findings were (1)a selective reduction (7.1%, P < 0.03) in BMD at the ultradistal site containing 60% trabecular bone, but not at two more proximal, chiefly cortical bone sites; (2) increased bone turnover affecting bone resorption [fasting hydroxyproline excretion, Lac 2.22 f 0.12 pmol/liter GF (mean f SEM), Con 1.19 f 0.04, P < 0.0011 and affecting bone formation (plasma alkaline phosphatase, Lac 81.9 f 2.5 IUAiter, Con 53.5 f 2.7, P < 0.001, and serum osteocalcin, Lac 14.0 f 0.7 &liter, Con 7.3 f 0.4, P < 0.001); and (3) renal conservation in the fasting state of both Ca and inorganic phosphate (Pi) with a resultant moderate increase in plasma Pi but not in plasma Ca (total or ionized). There were no differences between the groups in serum parathyroid hormone (PTH, intact and midmolecule assays), 25-hydroxy- and 1,25-dihydroxyvitamin D, nephrogenous cyclic AMP production, or plasma creatinine. In 25 of these Lac women restudied at one or more of the times 2,4, or 6 months after weaning, the findings were (1) an early (2 months) normalization of bone resorption and renal Pi handling with continuing increased bone formation and renal Ca conservation, associated with the onset of increased intact PTH levels; and (2) a recovery, within 4-6 months in ultradistal BMD associated with normalization of bone formation but with persisting renal Ca conservation and elevated intact PTH levels. We conclude that in established adult human lactation there is increased bone turnover with an accompanying loss of trabecular bone, despite renal conservation of Ca and Pi. After weaning, the deficit in trabecular bone is made up during a period of imbalance between a normal bone resorption rate and an elevated bone formation rate. The moderately elevated PTH after weaning may play a role in the recovery of bone mass by maintaining renal Ca conservation and by an anabolic action on bone.
'Department of Endocrinology and Diabetes, Sir Charles Gairdner Hospital, Nedlands 6009. Western Australia. 'Department of Biophysics, Sir Charles Gairdner Hospital, Nedlands 6009, Western Australia. '44 McCourt St., Leederville 6007, Western Australia. 'State Health Laboratories, Nedlands 6009,Western Australia. 'Department of Biochemistry, Royal Perth Hospital, Perth 6001, Western Australia. 6Deceased.
361
KENT ET AL.
362
INTRODUCTION with established lactation, 6-8 mmol calcium is transferred per day into milk.(1)The exact mechanisms by which this extra demand is met have not been determined [reviewed by Garel(2)].The anticipated adaptation mechanisms would include renal Ca conservation, increased dietary Ca, and enhanced efficiency of gut Ca absorption. Urinary Ca (24 h) excretion declines rapidly in early lactation, with a further decline as lactation proc e e d ~ . ‘Lactating ~) women consume diets that are at or near the recommended daily intake for Ca.(‘.’) No data exist on the efficiency of gut Ca absorption in human lactation. The maternal skeleton is another potential source of Ca for milk, but there are no reported controlled studies on bone turnover. Studies in adult women have reported(6-e) or failed to find(g-ll)significant changes in bone mineral density (BMD) during lactation. The absence of controls or small subject numbers made some of the positive reports unconvincing, and the negative studies addressed sites of predominantly cortical bone. In two prospective studies on lactating adolescents,(11.12) there was a small loss of BMD during lactation that was abolished by adequate dietary Ca in one study.(11iRetrospective have suggested that lactation is associated with bone mineral depletion, although parity may be protecti~e.“~) An uncommon, although severe, osteoporotic syndrome may be associated with pregnancy and lactation.(1s) We studied adult women in established lactation and matched control women by measuring forearm bone density at three sites, including an ultradistal site containing predominantly trabecular bone. Fasting biochemical studies were also performed to determine the role of the kidney in Ca and inorganic phosphate (Pi) conservation and to assess other measures of Ca homeostasis, including bone turnover. Of the lactating women, 25 were studied prospectively for up to 6 months postweaning to examine the degree and rate of reversibility of the changes in biochemistry and BMD induced by lactation. Of the controls, 22 were also restudied.
10)
N WOMEN
MATERIALS AND METHODS
Subjects and controls: lactation study Studies were performed on 40 fully breast-feeding white women at 5.8 f 1.3 months (mean + SD, range 4.0-9.6) postpartum and on 40 age-matched nonpregnant, nonlactating controls. At the time of the study only 6 of the lactating women had resumed normal menstrual cycles; 17 of the lactating women were primiparous and 23 were multiparous. Of the controls, 13 were primiparous, 9 were multiparous, and 18 were nulliparous. All parous control women had ceased breast-feeding at least 12 months before inclusion into the study. A total of 18 of the lactating women and 26 of the control women were lifelong nonsmokers, and 4 from each group were current smokers. The average duration of smoking and the peak number of cigarettes smoked per day were the same in both groups.
No subject had a history of surgery, disease, or medications known to affect Ca metabolism. All subjects were ambulant with no history of excessive exercise, period of immobilization, or dietary abnormality. Only 3 of the lactating women and none of the control women were taking Ca supplements of 500-1OOO mg/day. The study protocol was approved by the Human Rights Committee of the University of Western Australia, and informed written consent was obtained from all subjects. Forearm bone mineral density and fasting biochemical studies were performed on all subjects.
Weaning study Of the lactating women, 25 were restudied at one or more of the times 2,4, and 6 months after weaning. Weaning was defined as not more than one period of breastfeeding per day and occurred in these women at 5.9 f 2.9 months (mean f SD, range 1.1-1 1.6 months) after the initial lactation study. For 22 of the lactating women studied at various times after weaning, repeat BMD and fasting biochemical studies were performed on each woman’s paired control at a time equivalent to 2 months postweaning.
Forearm bone mineral density Bone mineral density was measured on the nondominant forearm of each subject by lZsI single-photon absorptiometry using a Mblsgaard Model BMA 1100 Bone Mineral Analyser with scanning and analysis software developed by the authors.(16)Three radioulnar regions of the distal forearm are measured: an ultradistal site comprised of approximately 60% trabecular bone,‘”) a distal site immediately proximal to the 8 mm point of radioulnar bifurcation (16% trabecular), and a shaft site 57.5 mm proximal to the midpoint of the ultradistal site (< 5% trabecular). “Bone mineral density” refers to the mass of mineral per unit scanned area of the bone as projected onto the plane of the rectilinear scan. It is an areal rather than a volumetric density. The units of measurement are mg mineral per cm2 as determined from ashing studies. Repeat measurements of BMD were normalized with respect to the initial scanned bone area at each site, so that serial changes in BMD reflected changes in bone mineral mass at the scan site. The precision of the forearm BMD measurement was determined with data on 13 premenopausal women aged 35.3 f 3.1 years (mean f SD, range 30.4-41.5 years) who had a repeat forearm BMD measurement done 5-12 months after their initial study. The lz5I source was changed at intervals not exceeding 3 months. The coefficients of variation of repeated measurements at the ultradistal, distal, and shaft sites were 1.6, 1.3, and 1.5%, respectively.
Fasting biochemical studies A fasting 2 h urine with midpoint blood sample was collected from each woman after an overnight fast of 10-
363
BONE LOSS AND BONE TURNOVER IN HUMAN LACTATION 12 h.'19) Subjects remained at rest throughout the urine collection period but were allowed water ad libitum throughout the study. Total Ca (Ca,), albumin, creatinine, Pi, and alkaline phosphatase were measured in heparinized plasma using Technicon (Tarrytown, NY) methods on a SMAC I1 analyzer. Car was corrected by 0.02 mmol/liter for each 1 g/liter deviation in plasma albumin concentration from the mean (44g/liter) of the reference range.'2o) Ionized Ca (Caion), pH, and Cqon corrected for pH were measured in serum using a Radiometer ICAl ionized Ca analyzer (Radiometer, Copenhagen, Denmark). Parathyroid hormone (PTH) in serum was measured with both a midmolecule region-specific radioimmunoassay(21)and an intact PTH immunochemiluminometric assay. Reagents for the intact PTH assay were kindly supplied by Dr. Richard Brown and Dr. Stuart Woodhead (University of Wales College of Medicine, Cardiff, UK). Osteocalcin (bone gla protein) in serum was determined using a radioimmunoassay developed by us, based on the method of Price and Nishimoto. 25-Hydroxyvitamin D (25-OHD3) was measured in serum using a competitive protein binding method'24) and 1,25-dihydroxyvitamin D [ 1,ZS-(OH),D,] by a cytoreceptor assay.(25)Vitamin D binding protein (DBP) was measured in serum using a radioimmunoassay developed in this laboratory using antibody kindly supplied by Professor Roger Bouillon (Katholieke Universiteit, Leuven, Belgium). Cyclic AMP (CAMP) in plasma (from blood samples collected on ice in tubes containing EDTA and theophylline) and in urine were measured using a radioimmunoassay based on the method of Steiner et a1.(26iCreatinine, Ca, Pi, and sodium were measured in urine using standard Technicon methods on an RA-1000 analyzer. Hydroxyproline in urine was measured using the manual method of Kivirikko et aI.I2')
tical significance was confirmed by analyzing the data sets after logarithmic transformation. Pearson's correlation coefficient was used to assess the univariate association between normally distributed data sets. Nonnormally distributed data were logarithmically transformed prior to linear regression analysis.
RESULTS
Lactation
There were no significant differences between the 40 lactating women (Lac) and their 40 controls (Con) for age [Lac 32.4 f 0.7 years (mean f SEM); Con 33.8 f 0.81. height (Lac 165.3 f 1.2 cm, Con 164.8 f 0.9). weight 2.5 kg, Con 63.6 f 2.0), or percentage of (Lac 63.3 ideal body weight (Lac 110 f 4%, Con 112 f 3). Results of the cross-sectional study of BMD in lactating and control women are shown in Table 1 as group means for the three forearm sites. Forearm BMD was 7.1 Yo lower (P c 0.03) in the lactating women compared to controls at the (predominantly trabecular) ultradistal site. There were no significant differences in BMD between the groups at the (more cortical) distal or shaft sites. Within each group there was no difference in the forearm BMD at all three sites between smokers and nonsmokers. Forearm BMD at all three sites was not different between primiparous lactating women and multiparous lactating women. Similarly, in the control women there was no effect of parity on forearm BMD at any of the three sites. The results of the cross-sectional study of fasting blood and urine biochemistry are shown in Table 2. All the biochemical indices of bone formation [i.e., plasma alkaline phosphatase and serum osteocalcin) and resorption [i.e., urine hydroxyproline: urine creatinine ratio (UHyp/ UCreat) and fasting H y p ~ ]were significantly elevated (P < 0.001) in the lactating women compared with their conDerived data trols. The logarithmically transformed HypE, In HypE, The renal P, threshold concentration (T,,,,,/GFR) was was correlated with the serum osteocalcin level in lactating calculated from the nomogram of Walton and Bijvoet.'2*) women ( r = 0.49, P < 0.001) but not in control women. Fasting Ca excretion (CaE) was calculated as described by In both groups the correlation between In H y p ~and Nordin'19) and expressed as pmol Ca excreted per liter of plasma alkaline phosphatase was significant ( r = 0.45, P glomerular filtrate using endogenous creatinine clearance. < 0.004 for Lac and r = 0.35, P < 0.03 for Con). Derivations of the fasting hydroxyproline excretion ( H y p ~ Fasting CaE was significantly lower (P < 0.05) in the lacas pmol hydroxyproline per liter glomerular filtrate) and tating women despite the fact that they had increased (P < sodium excretion (NaE) were analogous to that for CaE. 0.01) fasting NaE. The significantly higher fasting serum Pi Total urinary cAMP (UcAMP) excretion and nephrogenous cAMP (NcAMP) production were calculated as described by B r o a d u ~ . ' ~The ~ ) free 1,25-(OH)2D3index was TABLE1. CROSS-SECTIONAL STUDY OF FOREARM BONE calculated as the molar ratio of 1,25-(OH)'D, to DBP as PER cm') IN LACTATING MINERAL DENSITY (mg MINERAL described by Bouillon et a1.'") ( n = 40) AND CONTROL ( n = 40) WOMEN (MEANf SEM)
Statistical analysis The data sets, biochemical and BMD, of the lactating and control groups were compared using the unpaired twotailed t-test. The paired two-tailed t-test was used to compare measurements at peak lactation with repeat measurements on the same subjects after weaning. When data were nonnormally distributed (P < 0.05, Lilliefors test), statis-
*
Forearm site Ultradistal Distal Shaft
Lactating 327.8 468.2 617.9
f
f f
9.lb 8.9 8.8
Controls
352.8 477.6 627.5
f f f
6.6 6.7 8.3
Pa
Human Lactation: Forearm Trabecular Bone Loss, Increased Bone Turnover, and Renal Conservation of Calcium and Inorganic Phosphate with Recovery of Bone Mass Following Weaning G. NEIL KENT,' ROGER I. PRICE,'.' DONALD H. GUTTERIDGE,' MARGARET SMITH,3 JANET R. ALLEN,' CHOTOO I. BHAGAT,4 MARION P. BARNES,'*6CAROLINE J. HICKLING,' ROBERT W. RETALLACK,' SCOTT G. WILSON,' ROWENA D. DEVLIN,' CHARMIAN DAVIES5 and ANDREW ST. JOHNS
ABSTRACT The calcium (Ca) metabolism of established human lactation was studied in 40 adult women (mean age 32.4 years) who had been breast-feeding for 6 months (Lac) and in 40 age-matched controls (Con)using fasting urine and blood biochemistry and forearm single-photon bone mineral densitometry (BMD). Serial studies were performed up to 6 months after weaning in Lac women and repeated once in Con women. During luctution the significant findings were (1)a selective reduction (7.1%, P < 0.03) in BMD at the ultradistal site containing 60% trabecular bone, but not at two more proximal, chiefly cortical bone sites; (2) increased bone turnover affecting bone resorption [fasting hydroxyproline excretion, Lac 2.22 f 0.12 pmol/liter GF (mean f SEM), Con 1.19 f 0.04, P < 0.0011 and affecting bone formation (plasma alkaline phosphatase, Lac 81.9 f 2.5 IUAiter, Con 53.5 f 2.7, P < 0.001, and serum osteocalcin, Lac 14.0 f 0.7 &liter, Con 7.3 f 0.4, P < 0.001); and (3) renal conservation in the fasting state of both Ca and inorganic phosphate (Pi) with a resultant moderate increase in plasma Pi but not in plasma Ca (total or ionized). There were no differences between the groups in serum parathyroid hormone (PTH, intact and midmolecule assays), 25-hydroxy- and 1,25-dihydroxyvitamin D, nephrogenous cyclic AMP production, or plasma creatinine. In 25 of these Lac women restudied at one or more of the times 2,4, or 6 months after weaning, the findings were (1) an early (2 months) normalization of bone resorption and renal Pi handling with continuing increased bone formation and renal Ca conservation, associated with the onset of increased intact PTH levels; and (2) a recovery, within 4-6 months in ultradistal BMD associated with normalization of bone formation but with persisting renal Ca conservation and elevated intact PTH levels. We conclude that in established adult human lactation there is increased bone turnover with an accompanying loss of trabecular bone, despite renal conservation of Ca and Pi. After weaning, the deficit in trabecular bone is made up during a period of imbalance between a normal bone resorption rate and an elevated bone formation rate. The moderately elevated PTH after weaning may play a role in the recovery of bone mass by maintaining renal Ca conservation and by an anabolic action on bone.
'Department of Endocrinology and Diabetes, Sir Charles Gairdner Hospital, Nedlands 6009. Western Australia. 'Department of Biophysics, Sir Charles Gairdner Hospital, Nedlands 6009, Western Australia. '44 McCourt St., Leederville 6007, Western Australia. 'State Health Laboratories, Nedlands 6009,Western Australia. 'Department of Biochemistry, Royal Perth Hospital, Perth 6001, Western Australia. 6Deceased.
361
KENT ET AL.
362
INTRODUCTION with established lactation, 6-8 mmol calcium is transferred per day into milk.(1)The exact mechanisms by which this extra demand is met have not been determined [reviewed by Garel(2)].The anticipated adaptation mechanisms would include renal Ca conservation, increased dietary Ca, and enhanced efficiency of gut Ca absorption. Urinary Ca (24 h) excretion declines rapidly in early lactation, with a further decline as lactation proc e e d ~ . ‘Lactating ~) women consume diets that are at or near the recommended daily intake for Ca.(‘.’) No data exist on the efficiency of gut Ca absorption in human lactation. The maternal skeleton is another potential source of Ca for milk, but there are no reported controlled studies on bone turnover. Studies in adult women have reported(6-e) or failed to find(g-ll)significant changes in bone mineral density (BMD) during lactation. The absence of controls or small subject numbers made some of the positive reports unconvincing, and the negative studies addressed sites of predominantly cortical bone. In two prospective studies on lactating adolescents,(11.12) there was a small loss of BMD during lactation that was abolished by adequate dietary Ca in one study.(11iRetrospective have suggested that lactation is associated with bone mineral depletion, although parity may be protecti~e.“~) An uncommon, although severe, osteoporotic syndrome may be associated with pregnancy and lactation.(1s) We studied adult women in established lactation and matched control women by measuring forearm bone density at three sites, including an ultradistal site containing predominantly trabecular bone. Fasting biochemical studies were also performed to determine the role of the kidney in Ca and inorganic phosphate (Pi) conservation and to assess other measures of Ca homeostasis, including bone turnover. Of the lactating women, 25 were studied prospectively for up to 6 months postweaning to examine the degree and rate of reversibility of the changes in biochemistry and BMD induced by lactation. Of the controls, 22 were also restudied.
10)
N WOMEN
MATERIALS AND METHODS
Subjects and controls: lactation study Studies were performed on 40 fully breast-feeding white women at 5.8 f 1.3 months (mean + SD, range 4.0-9.6) postpartum and on 40 age-matched nonpregnant, nonlactating controls. At the time of the study only 6 of the lactating women had resumed normal menstrual cycles; 17 of the lactating women were primiparous and 23 were multiparous. Of the controls, 13 were primiparous, 9 were multiparous, and 18 were nulliparous. All parous control women had ceased breast-feeding at least 12 months before inclusion into the study. A total of 18 of the lactating women and 26 of the control women were lifelong nonsmokers, and 4 from each group were current smokers. The average duration of smoking and the peak number of cigarettes smoked per day were the same in both groups.
No subject had a history of surgery, disease, or medications known to affect Ca metabolism. All subjects were ambulant with no history of excessive exercise, period of immobilization, or dietary abnormality. Only 3 of the lactating women and none of the control women were taking Ca supplements of 500-1OOO mg/day. The study protocol was approved by the Human Rights Committee of the University of Western Australia, and informed written consent was obtained from all subjects. Forearm bone mineral density and fasting biochemical studies were performed on all subjects.
Weaning study Of the lactating women, 25 were restudied at one or more of the times 2,4, and 6 months after weaning. Weaning was defined as not more than one period of breastfeeding per day and occurred in these women at 5.9 f 2.9 months (mean f SD, range 1.1-1 1.6 months) after the initial lactation study. For 22 of the lactating women studied at various times after weaning, repeat BMD and fasting biochemical studies were performed on each woman’s paired control at a time equivalent to 2 months postweaning.
Forearm bone mineral density Bone mineral density was measured on the nondominant forearm of each subject by lZsI single-photon absorptiometry using a Mblsgaard Model BMA 1100 Bone Mineral Analyser with scanning and analysis software developed by the authors.(16)Three radioulnar regions of the distal forearm are measured: an ultradistal site comprised of approximately 60% trabecular bone,‘”) a distal site immediately proximal to the 8 mm point of radioulnar bifurcation (16% trabecular), and a shaft site 57.5 mm proximal to the midpoint of the ultradistal site (< 5% trabecular). “Bone mineral density” refers to the mass of mineral per unit scanned area of the bone as projected onto the plane of the rectilinear scan. It is an areal rather than a volumetric density. The units of measurement are mg mineral per cm2 as determined from ashing studies. Repeat measurements of BMD were normalized with respect to the initial scanned bone area at each site, so that serial changes in BMD reflected changes in bone mineral mass at the scan site. The precision of the forearm BMD measurement was determined with data on 13 premenopausal women aged 35.3 f 3.1 years (mean f SD, range 30.4-41.5 years) who had a repeat forearm BMD measurement done 5-12 months after their initial study. The lz5I source was changed at intervals not exceeding 3 months. The coefficients of variation of repeated measurements at the ultradistal, distal, and shaft sites were 1.6, 1.3, and 1.5%, respectively.
Fasting biochemical studies A fasting 2 h urine with midpoint blood sample was collected from each woman after an overnight fast of 10-
363
BONE LOSS AND BONE TURNOVER IN HUMAN LACTATION 12 h.'19) Subjects remained at rest throughout the urine collection period but were allowed water ad libitum throughout the study. Total Ca (Ca,), albumin, creatinine, Pi, and alkaline phosphatase were measured in heparinized plasma using Technicon (Tarrytown, NY) methods on a SMAC I1 analyzer. Car was corrected by 0.02 mmol/liter for each 1 g/liter deviation in plasma albumin concentration from the mean (44g/liter) of the reference range.'2o) Ionized Ca (Caion), pH, and Cqon corrected for pH were measured in serum using a Radiometer ICAl ionized Ca analyzer (Radiometer, Copenhagen, Denmark). Parathyroid hormone (PTH) in serum was measured with both a midmolecule region-specific radioimmunoassay(21)and an intact PTH immunochemiluminometric assay. Reagents for the intact PTH assay were kindly supplied by Dr. Richard Brown and Dr. Stuart Woodhead (University of Wales College of Medicine, Cardiff, UK). Osteocalcin (bone gla protein) in serum was determined using a radioimmunoassay developed by us, based on the method of Price and Nishimoto. 25-Hydroxyvitamin D (25-OHD3) was measured in serum using a competitive protein binding method'24) and 1,25-dihydroxyvitamin D [ 1,ZS-(OH),D,] by a cytoreceptor assay.(25)Vitamin D binding protein (DBP) was measured in serum using a radioimmunoassay developed in this laboratory using antibody kindly supplied by Professor Roger Bouillon (Katholieke Universiteit, Leuven, Belgium). Cyclic AMP (CAMP) in plasma (from blood samples collected on ice in tubes containing EDTA and theophylline) and in urine were measured using a radioimmunoassay based on the method of Steiner et a1.(26iCreatinine, Ca, Pi, and sodium were measured in urine using standard Technicon methods on an RA-1000 analyzer. Hydroxyproline in urine was measured using the manual method of Kivirikko et aI.I2')
tical significance was confirmed by analyzing the data sets after logarithmic transformation. Pearson's correlation coefficient was used to assess the univariate association between normally distributed data sets. Nonnormally distributed data were logarithmically transformed prior to linear regression analysis.
RESULTS
Lactation
There were no significant differences between the 40 lactating women (Lac) and their 40 controls (Con) for age [Lac 32.4 f 0.7 years (mean f SEM); Con 33.8 f 0.81. height (Lac 165.3 f 1.2 cm, Con 164.8 f 0.9). weight 2.5 kg, Con 63.6 f 2.0), or percentage of (Lac 63.3 ideal body weight (Lac 110 f 4%, Con 112 f 3). Results of the cross-sectional study of BMD in lactating and control women are shown in Table 1 as group means for the three forearm sites. Forearm BMD was 7.1 Yo lower (P c 0.03) in the lactating women compared to controls at the (predominantly trabecular) ultradistal site. There were no significant differences in BMD between the groups at the (more cortical) distal or shaft sites. Within each group there was no difference in the forearm BMD at all three sites between smokers and nonsmokers. Forearm BMD at all three sites was not different between primiparous lactating women and multiparous lactating women. Similarly, in the control women there was no effect of parity on forearm BMD at any of the three sites. The results of the cross-sectional study of fasting blood and urine biochemistry are shown in Table 2. All the biochemical indices of bone formation [i.e., plasma alkaline phosphatase and serum osteocalcin) and resorption [i.e., urine hydroxyproline: urine creatinine ratio (UHyp/ UCreat) and fasting H y p ~ ]were significantly elevated (P < 0.001) in the lactating women compared with their conDerived data trols. The logarithmically transformed HypE, In HypE, The renal P, threshold concentration (T,,,,,/GFR) was was correlated with the serum osteocalcin level in lactating calculated from the nomogram of Walton and Bijvoet.'2*) women ( r = 0.49, P < 0.001) but not in control women. Fasting Ca excretion (CaE) was calculated as described by In both groups the correlation between In H y p ~and Nordin'19) and expressed as pmol Ca excreted per liter of plasma alkaline phosphatase was significant ( r = 0.45, P glomerular filtrate using endogenous creatinine clearance. < 0.004 for Lac and r = 0.35, P < 0.03 for Con). Derivations of the fasting hydroxyproline excretion ( H y p ~ Fasting CaE was significantly lower (P < 0.05) in the lacas pmol hydroxyproline per liter glomerular filtrate) and tating women despite the fact that they had increased (P < sodium excretion (NaE) were analogous to that for CaE. 0.01) fasting NaE. The significantly higher fasting serum Pi Total urinary cAMP (UcAMP) excretion and nephrogenous cAMP (NcAMP) production were calculated as described by B r o a d u ~ . ' ~The ~ ) free 1,25-(OH)2D3index was TABLE1. CROSS-SECTIONAL STUDY OF FOREARM BONE calculated as the molar ratio of 1,25-(OH)'D, to DBP as PER cm') IN LACTATING MINERAL DENSITY (mg MINERAL described by Bouillon et a1.'") ( n = 40) AND CONTROL ( n = 40) WOMEN (MEANf SEM)
Statistical analysis The data sets, biochemical and BMD, of the lactating and control groups were compared using the unpaired twotailed t-test. The paired two-tailed t-test was used to compare measurements at peak lactation with repeat measurements on the same subjects after weaning. When data were nonnormally distributed (P < 0.05, Lilliefors test), statis-
*
Forearm site Ultradistal Distal Shaft
Lactating 327.8 468.2 617.9
f
f f
9.lb 8.9 8.8
Controls
352.8 477.6 627.5
f f f
6.6 6.7 8.3
Pa
