1976, British Journal of Radiology, 49, 921-925
NOVEMBER
1976
Bone demineralization in renal failure: a longitudinal study of the distal femur using photon absorptiometry By T. R. Overton, Ph.D., M.lnst.P., D. S. Silverberg, M.D., M.Sc, F.R.C.P.(C), M. Grace, B.Sc, B.Ed., A.M. (Statistics), Ph.D., W. M. Rigal, M.B., Ch.B., D.Phil.(Oxon), F.R.C.S., (Ed.), F.R.C.S.(C), M. Higgins, B.Sc., M.B., Ch.B., F.C.R.P.(C), K. B. Bettcher, M.B., B.S., M.R.C.P., F.R.C.P.(C), J. B. Dossetor, B.M., B.Ch., M.R.C.S., L.R.C.P., M.R.C.P., F.R.C.P.(C), Ph.D., F.A.C., F. Harley, B.A., M.D., F.R.C.P.(C), and H. F. DeLuca, Ph.D. Division of Biomedical Engineering and Department of Medicine, Faculty of Medicine, University of Alberta, Edmonton, Canada. (Received April, 1976) ABSTRACT
serial measurements of change in BMC at the lower radius and ulna of patients with chronic renal failure have produced variable results even in the same centre; in some patients, BMC increased, in some it decreased, and in some it remained the same (Catto et al., 1973; Ringe et al., 1974; Schuster, 1974; Griffiths et al., 1973; Meema et al, 1970). Measurements of mineral content in chronic renal failure have also been made for the os calcis (Banzer and Schneider, 1974; Banzer et al., 1974) and distal femur (Atkinson et al., 1970; Atkinson et al., 1973); the results were similar to those obtained for the radius/ulna. Using a measurement apparatus constructed in this laboratory (Overton et al., 1974), with Am 241 as the photon source, we studied BMC in the distal femur of 87 chronic renal failure patients comprising; 21 pre-dialysis patients (CRF), 45 chronic haemodialysis patients (CHD), and 21 renal transplant patients (RT).
The bone mineral content (BMC) of the lower end of the femur was measured by photon absorptiometry in 87 patients with chronic renal failure. The y-ray photon source was Am241. Serial measurements were obtained for up to two years. The mean BMC of the adult patients, comprising: 18 pre-dialysis (CRF), 41 chronic haemodialysis (CHD) and 19 renal transplant (RT) patients were all significantly lower than controls with the exception of the male CRF group. Two adults and one child on chronic haemodialysis showed a significant rate of bone loss (>2% per year). In one of these adults the addition of daily oral 1 a hydroxycholecalciferol was associated with no further reduction in BMC. Two children and one adult on chronic haemodialysis showed a significant rate of increase in BMC ( > 2 % per year). This adult had had a tendency to loss of BMC on standard CHD treatment but after receiving parenteral 1,25 dihydroxycholecalciferol three times weekly showed a significant rate of increase in BMC. No CRF or RT patient had a significant rate of loss of BMC on serial measurement. Two adults and one child with CRF had a significant rate of increase in BMC on standard treatment.
Progressive bone demineralization is a welldocumented complication of chronic renal disease (David, 1970). Routine radiological evaluation of skeletal status is of limited value for patient management in this condition since it is commonly found that some 30 to 50% of bone mineral must be lost before signs of bone demineralization appear on standard X rays (Lachman, 1955; Meema and Meema, 1963). Recently developed photon absorptiometry techniques provide the accuracy and sensitivity required for clinically useful measurements of bone mineral content (BMC); changes in BMC between 2 to 4% can be readily detected (Cameron et al., 1968). The usual sites for measurement of BMC by photon absorptiometry are the lower radius and ulna. In chronic haemodialysis patients, however, the rate of change in BMC obtained at these sites has a low degree of correlation with the rate of change of BMC in other areas of the body (Dalen and Alvestrand, 1973), with progressive pathological changes in the iliac crest (Cordy et al., 1974), or with changes in the calcium content of the hand as measured by neutron activation analysis (Catto et al., 1973). In addition,
METHODS
Bone mineral content measurements are made for all patients referred to the renal unit at the University of Alberta Hospital. Patient selection for inclusion in this report was based upon the availability of at least five BMC measurements for all patients in each group. Volunteers, aged 18-68 years, with no history of diseases involving bone metabolism, provided the "normal" data for this study. These consisted of 60 women and 63 men (Table 1). Pre-dialysis patients (serum creatinine 3 to 14 mg%) and CHD patients received 0.125 mg dihydrotachysterol (HytakerolR) twice daily and sufficient aluminium hydroxide in either the liquid or tablet form to maintain a normal serum phosphorus concentration. Two CHD patients received 0.5 to 2 fxg of 1,25 dihydroxycholecalciferol (1,25 DHCC) parenterally three times weekly while on
921
VOL.
49, No. 587 T. R. Overton et al. TABLE I MEAN BMC
(G/CM) AT DISTAL FEMUR SITE FOR NORMAL AND RENAL DISEASE GROUPS
FEMALES ]Vormal
Age (years) mean ± sd range 31.7 + 18.4 18-68
(Number in group) * Mean BMC g/cm
Group
Height (cm)
165±6
Weight (kg) 58.3 ±7.8
Normal (60) 3.54±0.48
CRF (9) 3.17±0.64
CHD (13) 3.15±0.75
RT (9) 2.98 ±0.72
(9) 4.23 ±0.68
(28) 4.11 ±0.99
(10) 4.04 ±0.71
MALES 29.1±11.3 19-68
178±7
72.4±9.7
(63) 4.59 ±0.74
*Excluding nine children (seven females, two males)
dialysis, and two received 0.5 to 2 jug daily of oral 1 a hydroxycholecalciferol ( l a HCC). Chronic haemodialysis patients received dialysis for three to eight hours, two to three times weekly. The dialysis fluid was fluoride-free; it had a calcium concentration of 6.5 mg% and a magnesium concentration of 1.5 mg%. Either 17 micron Gambro Lundia or Model 4 Cordis Dow dialysers were used. The diet was restricted in salt and water as necessary and had a protein content of 1 to 1.25 g/kg body weight per day. Patients who received a renal transplant were treated with Prednisone and Azathioprine in the following doses: Prednisone, 100 mg daily, commencing immediately following surgery. This dosage was reduced gradually, during a three month period, to 20 mg daily at which level it was maintained for most patients. Azathioprine, average 1.5 mg/kg body weight with a range 0.5 to 3.0 mg/kg body weight. Bone mineral content measurements were made in the distal femur at a site approximately one fifth of the shaft length from the lateral tubercle. Measurements were made at about two month intervals over an average period of 12 months (range 8 to 24 months). Group comparisons of mean BMC using a 2-tailed "t" test were made after determining that no significant differences existed in age, height or weight between groups. For the longitudinal study a weighted linear regression (BMC v. Time in days) was made for eight normals (five males, three females, each with a minimum of five measurements in the study period) and for each individual patient. In all cases the slope of the regression line was tested for a difference from zero slope. In the reported cases where patients changed groups or type of treatment a
comparison was also made between the regression slopes before and after the change. We used data obtained in long-term studies of BMC measurement precision to define a BMC "trend" threshold in terms of the slope of the regression line BMC v. Time for normal subjects. This threshold value was 0.0005 g/cm/day which corresponds to a rate of change in BMC between 1 and 2% per year depending upon the particular group (normal or patient) considered.
RESULTS
Mean BMC values for male and female "normal" groups and for males and females with renal disease are shown in Table I. In all BMC comparisons each patient group had a lower value than the normal group. This difference was significant (p
NOVEMBER
1976
Bone demineralization in renal failure: a longitudinal study of the distal femur using photon absorptiometry By T. R. Overton, Ph.D., M.lnst.P., D. S. Silverberg, M.D., M.Sc, F.R.C.P.(C), M. Grace, B.Sc, B.Ed., A.M. (Statistics), Ph.D., W. M. Rigal, M.B., Ch.B., D.Phil.(Oxon), F.R.C.S., (Ed.), F.R.C.S.(C), M. Higgins, B.Sc., M.B., Ch.B., F.C.R.P.(C), K. B. Bettcher, M.B., B.S., M.R.C.P., F.R.C.P.(C), J. B. Dossetor, B.M., B.Ch., M.R.C.S., L.R.C.P., M.R.C.P., F.R.C.P.(C), Ph.D., F.A.C., F. Harley, B.A., M.D., F.R.C.P.(C), and H. F. DeLuca, Ph.D. Division of Biomedical Engineering and Department of Medicine, Faculty of Medicine, University of Alberta, Edmonton, Canada. (Received April, 1976) ABSTRACT
serial measurements of change in BMC at the lower radius and ulna of patients with chronic renal failure have produced variable results even in the same centre; in some patients, BMC increased, in some it decreased, and in some it remained the same (Catto et al., 1973; Ringe et al., 1974; Schuster, 1974; Griffiths et al., 1973; Meema et al, 1970). Measurements of mineral content in chronic renal failure have also been made for the os calcis (Banzer and Schneider, 1974; Banzer et al., 1974) and distal femur (Atkinson et al., 1970; Atkinson et al., 1973); the results were similar to those obtained for the radius/ulna. Using a measurement apparatus constructed in this laboratory (Overton et al., 1974), with Am 241 as the photon source, we studied BMC in the distal femur of 87 chronic renal failure patients comprising; 21 pre-dialysis patients (CRF), 45 chronic haemodialysis patients (CHD), and 21 renal transplant patients (RT).
The bone mineral content (BMC) of the lower end of the femur was measured by photon absorptiometry in 87 patients with chronic renal failure. The y-ray photon source was Am241. Serial measurements were obtained for up to two years. The mean BMC of the adult patients, comprising: 18 pre-dialysis (CRF), 41 chronic haemodialysis (CHD) and 19 renal transplant (RT) patients were all significantly lower than controls with the exception of the male CRF group. Two adults and one child on chronic haemodialysis showed a significant rate of bone loss (>2% per year). In one of these adults the addition of daily oral 1 a hydroxycholecalciferol was associated with no further reduction in BMC. Two children and one adult on chronic haemodialysis showed a significant rate of increase in BMC ( > 2 % per year). This adult had had a tendency to loss of BMC on standard CHD treatment but after receiving parenteral 1,25 dihydroxycholecalciferol three times weekly showed a significant rate of increase in BMC. No CRF or RT patient had a significant rate of loss of BMC on serial measurement. Two adults and one child with CRF had a significant rate of increase in BMC on standard treatment.
Progressive bone demineralization is a welldocumented complication of chronic renal disease (David, 1970). Routine radiological evaluation of skeletal status is of limited value for patient management in this condition since it is commonly found that some 30 to 50% of bone mineral must be lost before signs of bone demineralization appear on standard X rays (Lachman, 1955; Meema and Meema, 1963). Recently developed photon absorptiometry techniques provide the accuracy and sensitivity required for clinically useful measurements of bone mineral content (BMC); changes in BMC between 2 to 4% can be readily detected (Cameron et al., 1968). The usual sites for measurement of BMC by photon absorptiometry are the lower radius and ulna. In chronic haemodialysis patients, however, the rate of change in BMC obtained at these sites has a low degree of correlation with the rate of change of BMC in other areas of the body (Dalen and Alvestrand, 1973), with progressive pathological changes in the iliac crest (Cordy et al., 1974), or with changes in the calcium content of the hand as measured by neutron activation analysis (Catto et al., 1973). In addition,
METHODS
Bone mineral content measurements are made for all patients referred to the renal unit at the University of Alberta Hospital. Patient selection for inclusion in this report was based upon the availability of at least five BMC measurements for all patients in each group. Volunteers, aged 18-68 years, with no history of diseases involving bone metabolism, provided the "normal" data for this study. These consisted of 60 women and 63 men (Table 1). Pre-dialysis patients (serum creatinine 3 to 14 mg%) and CHD patients received 0.125 mg dihydrotachysterol (HytakerolR) twice daily and sufficient aluminium hydroxide in either the liquid or tablet form to maintain a normal serum phosphorus concentration. Two CHD patients received 0.5 to 2 fxg of 1,25 dihydroxycholecalciferol (1,25 DHCC) parenterally three times weekly while on
921
VOL.
49, No. 587 T. R. Overton et al. TABLE I MEAN BMC
(G/CM) AT DISTAL FEMUR SITE FOR NORMAL AND RENAL DISEASE GROUPS
FEMALES ]Vormal
Age (years) mean ± sd range 31.7 + 18.4 18-68
(Number in group) * Mean BMC g/cm
Group
Height (cm)
165±6
Weight (kg) 58.3 ±7.8
Normal (60) 3.54±0.48
CRF (9) 3.17±0.64
CHD (13) 3.15±0.75
RT (9) 2.98 ±0.72
(9) 4.23 ±0.68
(28) 4.11 ±0.99
(10) 4.04 ±0.71
MALES 29.1±11.3 19-68
178±7
72.4±9.7
(63) 4.59 ±0.74
*Excluding nine children (seven females, two males)
dialysis, and two received 0.5 to 2 jug daily of oral 1 a hydroxycholecalciferol ( l a HCC). Chronic haemodialysis patients received dialysis for three to eight hours, two to three times weekly. The dialysis fluid was fluoride-free; it had a calcium concentration of 6.5 mg% and a magnesium concentration of 1.5 mg%. Either 17 micron Gambro Lundia or Model 4 Cordis Dow dialysers were used. The diet was restricted in salt and water as necessary and had a protein content of 1 to 1.25 g/kg body weight per day. Patients who received a renal transplant were treated with Prednisone and Azathioprine in the following doses: Prednisone, 100 mg daily, commencing immediately following surgery. This dosage was reduced gradually, during a three month period, to 20 mg daily at which level it was maintained for most patients. Azathioprine, average 1.5 mg/kg body weight with a range 0.5 to 3.0 mg/kg body weight. Bone mineral content measurements were made in the distal femur at a site approximately one fifth of the shaft length from the lateral tubercle. Measurements were made at about two month intervals over an average period of 12 months (range 8 to 24 months). Group comparisons of mean BMC using a 2-tailed "t" test were made after determining that no significant differences existed in age, height or weight between groups. For the longitudinal study a weighted linear regression (BMC v. Time in days) was made for eight normals (five males, three females, each with a minimum of five measurements in the study period) and for each individual patient. In all cases the slope of the regression line was tested for a difference from zero slope. In the reported cases where patients changed groups or type of treatment a
comparison was also made between the regression slopes before and after the change. We used data obtained in long-term studies of BMC measurement precision to define a BMC "trend" threshold in terms of the slope of the regression line BMC v. Time for normal subjects. This threshold value was 0.0005 g/cm/day which corresponds to a rate of change in BMC between 1 and 2% per year depending upon the particular group (normal or patient) considered.
RESULTS
Mean BMC values for male and female "normal" groups and for males and females with renal disease are shown in Table I. In all BMC comparisons each patient group had a lower value than the normal group. This difference was significant (p
