- Email: [email protected]
Characteristics of trabecular bone resorption cavities in patients with chronic renal failure
Bone and Mineral, 16 (1992) 139-147
139
EIsevier BAM 00421
Peter I. Croucher’, Christopher D.P. Wright2, Nigel J. Garrahan’, ana Kudlac3, Andy 9. Williams” and Juliet E. Compston2 ‘Department of Pa(hology, University of Wales College of Medicine, Cardiffi 2Department of Medicine, Universityof Cambridge Clinical School, Addenbrooke’s Hospital, Cambridge, and “Department of Nephrology, Morriston Hospital, Swansea, UK
(Received 26 April 1991) (Accepted 27 September 1991)
Summary Using a computerised technique, resorption cavity characteristics in iliac crest trabecular bone were assessed in 30 patients with chronic renal failure and compared with data obtained from healthy subjects. ,. ‘I’he mean and maximum cavity depth were significantiy greater in the patient group (P c u.uwl); in addition, cavity area, the percentage of bone being remodelled, the number of cavities per mm trabecular surface and the percentage eroded surface were all significantly greater than in controls (P C 0.0001). However, the surface length of individual cavities in the patient group did not differ significantly from that of controls. In the patient group, serum intact parathyroid hormone concentrations showed a significant positive correlation with mean resorption cavity depth (r = 0.451, P c 0.05). Our results demonstrate that the increase in bone resorption associated with hyperparathyroidism secondary to chronic renal failure is due to an increase both in the number and depth of cavities, although the surface extent of individual cavities is normal. These findings indicate that factors determining the length of trabecular surface eroded and the depth of individual resorption cavities are controlled by different mechanisms. .-.‘L-
Key words: Renal osteodystrophy; Hyperparathyroidism; rum parathyroid hormone
.
Bone histomorphometry; Resorption; Se-
Secondary hyperparathyroidism is commonly associated with chronic renal failure and leads to metabolic bone disease characterised by increased bone turnover owever., wMst an increase in the extent of trabecular bone surface occu[l-3]. --__
Correspondence to: J.E. Compston, Department of Medicine, Level 5, University of Cambridge Clinical School, Addenbrooke’s Hospital, Cambridge CB2 2QQ, UK. 0169-6009/92/$05.000 1992 Elsevier Science Publishers B.V.
140
pied by resorption cavities is well described in such cases [4-71, it is unknown whether this represents an increased number of cavities, an increase in the surface length of individual cavities, or both. In addition, the depth of resorption cavities, which has important implications for trabecular bone structure and its connectedness, has not been assessed. We have recently described a computerised technique which enables the quantitative assessment of a number of resorption cavity characteristics, including length, area, mean depth and maximum depth [S]. This method has been applied to iliac crest bone from patients with chronic renal failure, in order to define more closely the changes occurring in bone resorption and to investigate the relationship of these to the degree of secondary hyperparathyroidism as assessed by serum intact parathyroid hormone levels.
Materialsand Methods Subjects
Thirty-seven patients with end stage renal failure, 26 males and 11 females, aged 24-72 years (mean 47 years) were studied. These patients were attending the Regional Renal Unit at Morriston Hospital, Swansea and all but 4 were being treated with regular haemodialysis (n = 17) or continuous ambulatory peritoneal dialysis (n = 16) (dialysate calcium 1.65 and 1.6 mmolll respectively). All but one of the patientc ..“...Y
WP~P ..“.W rprdvino I”Wk...“‘b
nraP -a_. nhnanhate p.““Ja”....”
hinderr -....w-LY
(raleinm \‘..“‘....’
pnrhnnatc? _.s- .-IL__._
nr _a duminilrm _._a . . . . -L-1-1 hv“J
droxide) and 18 were receiving lo-hydroxyvitamin D, (length of treatment 1.4 years). All patients gave written informed consent to undergo iliac crest biopsy under local anaesthesia and permission for the study was obtained from the local ethical commit tee. control values were obtained from a group of 50 normal healthy subjects, 23 male, aged 19-78 years (mean 51 years) who gave informed consent to undergo bone biopsy during general anaesthesia for a minor surgical procedure. Details of this study have been described previously [9]. Specimens were fixed in phosphate-buffered formalin and embedded in methylmethacrylate (British Drug House Chemicals Ltd, Poole, Dorset). 8pm undecalcitied sections were cut using a Jung K microtome and stained by the Von Kossa technique using a van Gieson counterstain, or 1% toluidine blue (pH 4.2). Biochemistry
Serum calcium, phosphate, alkaline phosphatase, creatinine and magnesium levels were measured in patients with renal osteodystrophy using standard laboratory techniques. Serum aluminium was measured by atomic absorption spectrometry using a graphite furnace. Serum 25hydroxyvitamin D, was measured by a competitive protein binding assay. Serum intact parathyroid hormone concentration was measured using an immunoradiometric assay (INCStar Ltd, Wokingham, Berkshire). Bone aluminium was measured by drying ,the bone sample overnight followed by digestion in nitric acid for 1 h at 90°C. 4.5 ml water are added prior to
141 assessing aluminium concentration by atomic absorption spectrosco asurement of resorption cavity characteristics
This method has been described in detail previously [8] aving identified a re tion cavity the image is stored in the memory of an IB II image analyser ( CK)and displayed on a television monitor. The two end points of the cavity are identified and entered into the image analyser using the screen cursor. Two circles, with a diameter equal to half the linear distance between the two resorption points, are drawn on the screen. The position of the intersection between the circle and the trabecular surface are also entered. A smooth continuous curve, known as a cubic spline, is drawn through the defined points. To ensure that the smooth continuity of the bone surface on each side of the cavity is maintained, additional points at maximum deviation may be added to refine the curve fit. This line is then used as a baseline for the assessment of cavity characteristics ( The following indices were obtai ed by direct measurement or calculation: Cavity count/BS (/mm) Cavity count/TA (/mm*) Wximum depth (urn) Mean depth Cum)
-
Reconstructed length (urn) Eroded length @m)
-
Cavity area @m*) Reconstructed surface (%) Eroded surface (%)
-
Bone remodelled (%)
-
Number of cavities per mm of trabecuiar surface. Number of cavities per mm*of medullary area. Maximum cavity depth. Mean cavity depth, from four approximately equidistant points. Length of resorption cavity, as measured along the cubic spiine (Pig. i j. Length of resorption cavity, as measured along the crenated surface (Fig. 1). Area of the resorption cavity. Percentage of trabecular surface occupied by resorption cavities (reconstructed length). Percentage of trabecular surface occupied by resorption cavities (eroded length). Percentage bone remodelled, calculated as the cavity area plus bone area and multiplied by 100.
Fig. 1. Diagrammaticrepresentation of a resorption cavity. Position A and B are the two end points of the resorption cavity. The dashed line between A and B is the reconstructed length and the solid crenated line is the eroded length.
142
Resorption cavities were identified using polarised light at magnifications of either x375 or occasionally x750. All crenated surfaces with lamellae whose end points terminated at the bone surface were included. A minimum of 20 cavities from between 1 and 5 sections were used in the 30 biopsies taken from patients with renal osteodystrophy, the mean number + SD of cavities in each biopsy being 24.8 + 4.1. Between 2 and 13 sections were examined in the 41 control subjects, the mean number f SD of cavities in each biopsy being 25.2 f 2.3. Measurement of trabecuiar bone volume and osteoid characteristics
Measurements were made using a MOP videoplan (Kontron, FRG). Images were projected from the microscope onto a digitising tablet using a drawing arm attachment. Both the mineralised bone surface and osteoid surface were traced using the digitiser cursor and measurements were collected and evaluated using the VI plan software. Osteoid width was measured directly, using the digitiser cursor, and taken as the mean of four approximately equidistant points. A minimum of 20 seams were measured from between 2 and 6 sections in each biopsy. Seams were identified and measured at a magnification of x 350. Statistics
Log transformation was performed to normalise non-normally distributed data. Linear regression analysis was used to analyse the relationship between biochemical and histomorphometric variables. Differences between histomorphometric .
C!SUlts
Some biochemical data from the patients are summarised in Table 1, results being expressed as the percentage above or below the reference range. Nearly 85% had elevated serum intact parathyroid hormone concentrations and one had low serum 25-hydroxyvitamin D, levels. Serum alumiiniumconcentrations were above normal Table 1
Biochemical data in the patient group %Outsidereferencerange
Serumcalcium(mmoV1)
Serumphosphate(mmolll) Serumalkalinephosphatase(U/I) Serumcreatinine+mol/l) Serumaluminium@mol/l) SerumintactPTH(ngA) Serum25-hydroxyvitamin D, (@ml)
Lower
Upper
2.9 0 0 0 0 0 3.33
26.5 82.9 23.5 100.0 55.6 84.4 3.33
Referencerange
2.15-2.5s 0.80-1.40 30.0-130.0 50.0-112.0 0.10-0.60 0.00-50.0 3.60-30.0
143 le 2
Resorption cavity characteristics in control subjects and patients with renal osteodystrophy Cavity characteristics
Subject group Control (n ‘- 41)
Maximum depth @m) Mean depth @m) Reconstructed length bm) Eroded length bm) Cavity area km*) Reconstructed surface (%) Eroded surface (%) Number cavities/mm* Number cavities/mm Bone remodellcd (%) a Geometricmean.
34.6= 23.8-50Sb 20.8 14.5-29.8 209.0 146.2-298.8 169.1-340.1 239.8 3447 1871-6422 1.68 0.65-4.32 1.94 0.76-4.93 0.23 0.10-0.56 0.072 0.031-0.169 0.24-1.40 0.58
Significance Renal osteodystrophy (n = 30) 47.9
35.0-65.6 30.4 22.3-41.6 220.2 156.7-309.3 277.5 208.4-369.5 5414 3145-9320 3.58 1.05-12.23 4.46 1.33-14.98 0.40 0.12-1.27 0.148 0.050-0.440 1.34 0.33-5.51
P < 0.0001 P < 0.0091
NS P c 0.0005 P
b 95% range.
in 56% of patients. Serum calcium levels were elevated in 10 of the patients (range 1.74-3.14 mmol/l); of these 7 patients were receiving la-hydroxyvitamin D,. istologically, 24 patients had an increased surface extent of reso normal values being found in the remaining 6 biopsies. Three patients (all with increased eroded surface) had an increased mean osteoid seam width (>12 pm), the values being 39.2, 19.2, and 15.0 pm. Relative osteoid surface in the 30 patients ranged from 1.62-91.1%, the geometric mean being 26.1% and 27% of patients having values above the normal range. Trabecutar bone volume was not significant1Ydifferent from control subjects. 1.9 ,
pc0.0001
p<0.0001
1.8 1.7 1.6 1.6 -
I
aximum depth
an depth
+ 2SD in normal subjects is indiFig. 2. Mean and maximum resorption depth in the patients. The mean _ cated by the horizontal bars.
144 pcO.0~1
3.2 1
Renalostsodystrophy
Control
Fig.3. Resorption cavity area in the patients and controls. The mean is indicated for each group.
Because of either poor quality biopsies, insufficient number of sections, or trabecular shattering only 30 of the 37 biopsies obtained from the patient group were of sufficient quality to enable quantitative assessment of resorption cavities, whilst 41 of the 50 control biopsies were adequate. Resorption cavity characteristics in the two groups are shown in Table 2. Both the mean and maximum cavity depth (Fig. 2) were significantly greater in the patient group (P < 0.0001); the cavity area (Fig. 3), the percentage of bone being remodelled, the number of cavities/mm trabecular surface and the eroded surface were also significantly greater (P < 0.0001). However, the surface length of individual cavities in patients was not significantly different from that of controls. In the patient group, serum intact parathyroid hormone concentrations showed a significant positive correlation with resorption cavity depth (r = 0.451, P < 0.05; as-
3.6 -
3.4 -
3.2 *
1
2 Log
3
4
parathyrold
5
6
hormone
7
9
9
concentration
Fig. 4. The relationship between mean resorption cavity depth and serum intact parathyroid hormone concentration in the patient group. The regressionequation is log mean depth = 3.09 + 0.0631 log PTH.
145 Fig. 4) and osteoid surface (r = 0.429, P < 0.05). Serum alkaline phosphatase levels showed significant correlations with the cavity count/mm trabecular surface (r = 0.474, k c 0.01) and the percentage of bone being remodelled (a = 0.404, P < 0.05). Serum aluminium was not significantly correlated with any of the resorption cavity characteristics. The bone aluminium content showed significant correlations with the eroded surface (r = 0.516, B < 0.05), cavity number/mm2 (r = 0.453, P < 0.05) and the percent bone remodelled (r = 0.588, P c 0.01). No significant correlations were found between bone aluminium content and any of the other resorption cavity characteristics or histomorphometric variables.
iscussio Renal osteodystrophy is a complex disorder in which varying combinations of hyperparathyroidism and osteomalacia may occur 131.Of the patients included in the present study, 80% had evidence of secondary hyperparathyroidism as manifested by an increase in percentage eroded trabecular surface and 3 of these patients also showed some evidence of a mineralisation defect. There was thus a wide spectrum of histological appearances, eroded surface varying from just over 1 to 15% and osteoid surface between 1.6 and 91.9% of the trabecular bone surface. In keeping with the histological appearances, 84.4% of patients had biochemical evidence of secondary hyperparathyroidism with elevated serum intact parathyroid hormone concentrations. The raised serum calcium in 26.5% of patients probably reflects treatment with la-hydroxyvitamin D,, although tertiary hyperparathyroidism or co-existing primary hyperparathyroidism were not specifically excluded, The results of the present study demonstrate that the increase in bone resorption which occurs in patients wi
146
is in keeping with the positive effect of parathyroid hormone on activation frequenhyperparathyroidism in cy and hence bone turnover [ 111.Suppression of cacondary us patients with chronic renal failure with the use of phosphate binders and dietary phosphate restriction results in a reduction of bone turnover [12], although the effects of such treatment on the depth and other morphological characteristics of resorption cavities have not been established. The increased resorption depth in our patients contrasts with data obtained from patients with primary hyperparathyroidism. Using a technique for measurement of resorption depth based on the number of bone lamellae eroded beneath the mineralised bone surface [13] Eriksen [14] reported reduced resorption depth in a group of 19 patients with primary hyperparathyroidism. This difference may be related to the higher serum levels of parathyroid hormone in secondary hyperparathyroidism [15] or might be the consequence of one or more of the many other metabolic disturbances, including uraemia, acid base imbalance or aluminium toxicity, which accompany chronic renal failure; in this context, the relationship between bone aluminium and some resorption cavity characteristics is of interest. The reduced resorption depth demonstrated in patients with primary hyperparathyroidism is consistent with the preservation of trabecular bone volume and structure in iliac crest bone obtained from such patients recently reported by Silverberg et al. 1161.I-Iowever, in secondary hyperparathyroidism, the combination of increased resorption depth and high bone turnover is likely to result in trabecular penetration and loss of connectedness and although treatment of the bone disease may be successful in reducing bone turnover it is unlikely to restore trabecular architecture once this has been disrupted [17]. These considerations emphasise the importance of preventing phosphate retention in chronic renal failure since the adverse mechanical effects associated with loss of trabecular bone structure may result in a higher fracture risk in later life. Until recently, assessment of bone resorption has been unsatisfactory and limited to measurement of the extent of trabecular bone surface occupied by resorption cavities and the number of osteoclasts present. The development of methods which enable direct assessment of resorption depth, based either on the counting of eroded lamellae as described by Eriksen et al. [13] or on computerised techniques as used in the present study [8] should provide valuable information about the contribution of changes in resorption depth to remodelling imbalance in different disease states; in addition, such techniques can be applied to the study of the effects of drugs on resorption cavity dimensions.
Acknowledgements We are grateful to the Welsh Office and the National Kidney Research Fund for generous financial support. Dr J.E. Compston is supported by the Wellcome Trust. We thank Dr I. Mainsworth for biochemical assessment of bone aluminium.
1 Stanbury SW. Bone disease in uraemia. Amer J Med 1%8;44:714-724. 2 Massry SG, Coburn JW, Popovtzer MM, Schinaberger JH, Maxwell MH, Kleeman CR. Secondary hyperparathyroidism in chronic renal failure. The clinical spectrum in uremia, during haemodialy& and after renal transplantation. Arch Intern Med 1969;124:431-441. 3 Slatopolsky E, Caglan S, Pennell JP, Taggart DQ, Canterbury JM, Reiss E, Bricker NS. On the pathogenesis of hyperparathyroidism in chronic experimental renal insufficiency in the dog. J. Clin Invest 1971;50:492-499. 4 Eastwood JB, Bordier PJ, de Wardener HE. Some biochemical, histological, radiological and clinical features of renal osteodystrophy. Kidney Int 1973;4:128-140. S Ritz E, Prager P, Krempien B, Bommer J, Malluche HH, Schmidt-Gayk H. Skeletal K-ray findings and bone histology in patients on haemodialysis. Kidney Int 1978;13:316-323. 6 Bordier PJ, Marie PJ, Arnaud CD. Evolution of renal osteodystrophy: correlation of bone histomorphometry and serum mineral and immunoreactive parathyroid hormone values before and after treatment with calcium carbonate or 25hydroxycholecalcificerol. Kidney Int Suppl1975;2:102-112. 7 Frost HM, Griffith DL, Jee WSS, Kimmel DB, McCandlis RP, Teitelbaum SL. Histomorphometric changes in trabecular bone of renal failure patients treated with calcifediol. Metab Bone Dis Rel Res 1981;2:28S-295. 8 Garrahan NJ, Croucher PI, Compston JE. A computerised technique for the quantitative assessment of resorption cavities in trabecular bone. Bone 1990;11:241-246. 9 Vedi S, Compston JE, Webb A, Tighe JR. Histomorphometric analysis of bone biopsies from the iliac crest of normal British subjects. Metab Bone Dis Rel Res 1982;4:231-236. 10 Delaisse JM, Boyde A, Maconnachie E, Ali NN, Sear CHJ, Eeckout Y, Vaes G, Jones SJ. The effects of inhibitors of cysteine-proteinases and collagenase on the resorptive activity of isolated osteoclasts. Bone 1987;8:305-313. 11 Parfitt AM, Gallagher JC, Heaney RP, Johnston CC, Neer R, Whedon GD. Vitamin D and bone health in the elderly. Amer J Clin Nutr 1982;36:1014-1031. 12 Turner C, Compston J, Mak RHK, Vedi S, Mellish RWE, Haycock GB, Chantler C. Bone turnover and 1,25dihydroxycholecalciferol during treatment with phosphate binders. Kidney Int 1988;33: 989-995. 13 Eriksen EF, Melsen F, Mosekilde L. Reconstruction of the resorptive site in iliac trabecular bone: a kinetic model for bone resorption in 20 normal individuals. Metab Bone Dis Rel Res 1984;5:23S-242. 14 Eriksen EF. Normal and pathological remodelling of human trabecular bone: three dimensional reconstruction of the remodelling sequence in normals and in metabolic bone disease. Endocr Rev 1986;7:379-408. 19 Brown RC, Aston JP, Weeks I, Woodhead JS. Circulating intact parathyroid hormone measured by a two-site immunochemiluminometric assay. J Clin Endocrinol Metab 1987;65:407-414. 16 Silverman SJ, Shane E, De La Crux L, Dempster DW, Feldman F, Seldin D, Jacobs TP, Siris ES, Cafferty M, Parisien MV, Lindsay R, Clemens TL, Bilezikian JP. Skeletal disease in primary hyperparathyroidism. J Bone Miner Res 1989;4:283-291. 17 Compston JE, Mellish RWE, Croucher PI, Newcombe R and Garrahan NJ. Structural mechanisms of trabecular bone loss in man. Bone Miner 1989;6:339-350.
139
EIsevier BAM 00421
Peter I. Croucher’, Christopher D.P. Wright2, Nigel J. Garrahan’, ana Kudlac3, Andy 9. Williams” and Juliet E. Compston2 ‘Department of Pa(hology, University of Wales College of Medicine, Cardiffi 2Department of Medicine, Universityof Cambridge Clinical School, Addenbrooke’s Hospital, Cambridge, and “Department of Nephrology, Morriston Hospital, Swansea, UK
(Received 26 April 1991) (Accepted 27 September 1991)
Summary Using a computerised technique, resorption cavity characteristics in iliac crest trabecular bone were assessed in 30 patients with chronic renal failure and compared with data obtained from healthy subjects. ,. ‘I’he mean and maximum cavity depth were significantiy greater in the patient group (P c u.uwl); in addition, cavity area, the percentage of bone being remodelled, the number of cavities per mm trabecular surface and the percentage eroded surface were all significantly greater than in controls (P C 0.0001). However, the surface length of individual cavities in the patient group did not differ significantly from that of controls. In the patient group, serum intact parathyroid hormone concentrations showed a significant positive correlation with mean resorption cavity depth (r = 0.451, P c 0.05). Our results demonstrate that the increase in bone resorption associated with hyperparathyroidism secondary to chronic renal failure is due to an increase both in the number and depth of cavities, although the surface extent of individual cavities is normal. These findings indicate that factors determining the length of trabecular surface eroded and the depth of individual resorption cavities are controlled by different mechanisms. .-.‘L-
Key words: Renal osteodystrophy; Hyperparathyroidism; rum parathyroid hormone
.
Bone histomorphometry; Resorption; Se-
Secondary hyperparathyroidism is commonly associated with chronic renal failure and leads to metabolic bone disease characterised by increased bone turnover owever., wMst an increase in the extent of trabecular bone surface occu[l-3]. --__
Correspondence to: J.E. Compston, Department of Medicine, Level 5, University of Cambridge Clinical School, Addenbrooke’s Hospital, Cambridge CB2 2QQ, UK. 0169-6009/92/$05.000 1992 Elsevier Science Publishers B.V.
140
pied by resorption cavities is well described in such cases [4-71, it is unknown whether this represents an increased number of cavities, an increase in the surface length of individual cavities, or both. In addition, the depth of resorption cavities, which has important implications for trabecular bone structure and its connectedness, has not been assessed. We have recently described a computerised technique which enables the quantitative assessment of a number of resorption cavity characteristics, including length, area, mean depth and maximum depth [S]. This method has been applied to iliac crest bone from patients with chronic renal failure, in order to define more closely the changes occurring in bone resorption and to investigate the relationship of these to the degree of secondary hyperparathyroidism as assessed by serum intact parathyroid hormone levels.
Materialsand Methods Subjects
Thirty-seven patients with end stage renal failure, 26 males and 11 females, aged 24-72 years (mean 47 years) were studied. These patients were attending the Regional Renal Unit at Morriston Hospital, Swansea and all but 4 were being treated with regular haemodialysis (n = 17) or continuous ambulatory peritoneal dialysis (n = 16) (dialysate calcium 1.65 and 1.6 mmolll respectively). All but one of the patientc ..“...Y
WP~P ..“.W rprdvino I”Wk...“‘b
nraP -a_. nhnanhate p.““Ja”....”
hinderr -....w-LY
(raleinm \‘..“‘....’
pnrhnnatc? _.s- .-IL__._
nr _a duminilrm _._a . . . . -L-1-1 hv“J
droxide) and 18 were receiving lo-hydroxyvitamin D, (length of treatment 1.4 years). All patients gave written informed consent to undergo iliac crest biopsy under local anaesthesia and permission for the study was obtained from the local ethical commit tee. control values were obtained from a group of 50 normal healthy subjects, 23 male, aged 19-78 years (mean 51 years) who gave informed consent to undergo bone biopsy during general anaesthesia for a minor surgical procedure. Details of this study have been described previously [9]. Specimens were fixed in phosphate-buffered formalin and embedded in methylmethacrylate (British Drug House Chemicals Ltd, Poole, Dorset). 8pm undecalcitied sections were cut using a Jung K microtome and stained by the Von Kossa technique using a van Gieson counterstain, or 1% toluidine blue (pH 4.2). Biochemistry
Serum calcium, phosphate, alkaline phosphatase, creatinine and magnesium levels were measured in patients with renal osteodystrophy using standard laboratory techniques. Serum aluminium was measured by atomic absorption spectrometry using a graphite furnace. Serum 25hydroxyvitamin D, was measured by a competitive protein binding assay. Serum intact parathyroid hormone concentration was measured using an immunoradiometric assay (INCStar Ltd, Wokingham, Berkshire). Bone aluminium was measured by drying ,the bone sample overnight followed by digestion in nitric acid for 1 h at 90°C. 4.5 ml water are added prior to
141 assessing aluminium concentration by atomic absorption spectrosco asurement of resorption cavity characteristics
This method has been described in detail previously [8] aving identified a re tion cavity the image is stored in the memory of an IB II image analyser ( CK)and displayed on a television monitor. The two end points of the cavity are identified and entered into the image analyser using the screen cursor. Two circles, with a diameter equal to half the linear distance between the two resorption points, are drawn on the screen. The position of the intersection between the circle and the trabecular surface are also entered. A smooth continuous curve, known as a cubic spline, is drawn through the defined points. To ensure that the smooth continuity of the bone surface on each side of the cavity is maintained, additional points at maximum deviation may be added to refine the curve fit. This line is then used as a baseline for the assessment of cavity characteristics ( The following indices were obtai ed by direct measurement or calculation: Cavity count/BS (/mm) Cavity count/TA (/mm*) Wximum depth (urn) Mean depth Cum)
-
Reconstructed length (urn) Eroded length @m)
-
Cavity area @m*) Reconstructed surface (%) Eroded surface (%)
-
Bone remodelled (%)
-
Number of cavities per mm of trabecuiar surface. Number of cavities per mm*of medullary area. Maximum cavity depth. Mean cavity depth, from four approximately equidistant points. Length of resorption cavity, as measured along the cubic spiine (Pig. i j. Length of resorption cavity, as measured along the crenated surface (Fig. 1). Area of the resorption cavity. Percentage of trabecular surface occupied by resorption cavities (reconstructed length). Percentage of trabecular surface occupied by resorption cavities (eroded length). Percentage bone remodelled, calculated as the cavity area plus bone area and multiplied by 100.
Fig. 1. Diagrammaticrepresentation of a resorption cavity. Position A and B are the two end points of the resorption cavity. The dashed line between A and B is the reconstructed length and the solid crenated line is the eroded length.
142
Resorption cavities were identified using polarised light at magnifications of either x375 or occasionally x750. All crenated surfaces with lamellae whose end points terminated at the bone surface were included. A minimum of 20 cavities from between 1 and 5 sections were used in the 30 biopsies taken from patients with renal osteodystrophy, the mean number + SD of cavities in each biopsy being 24.8 + 4.1. Between 2 and 13 sections were examined in the 41 control subjects, the mean number f SD of cavities in each biopsy being 25.2 f 2.3. Measurement of trabecuiar bone volume and osteoid characteristics
Measurements were made using a MOP videoplan (Kontron, FRG). Images were projected from the microscope onto a digitising tablet using a drawing arm attachment. Both the mineralised bone surface and osteoid surface were traced using the digitiser cursor and measurements were collected and evaluated using the VI plan software. Osteoid width was measured directly, using the digitiser cursor, and taken as the mean of four approximately equidistant points. A minimum of 20 seams were measured from between 2 and 6 sections in each biopsy. Seams were identified and measured at a magnification of x 350. Statistics
Log transformation was performed to normalise non-normally distributed data. Linear regression analysis was used to analyse the relationship between biochemical and histomorphometric variables. Differences between histomorphometric .
C!SUlts
Some biochemical data from the patients are summarised in Table 1, results being expressed as the percentage above or below the reference range. Nearly 85% had elevated serum intact parathyroid hormone concentrations and one had low serum 25-hydroxyvitamin D, levels. Serum alumiiniumconcentrations were above normal Table 1
Biochemical data in the patient group %Outsidereferencerange
Serumcalcium(mmoV1)
Serumphosphate(mmolll) Serumalkalinephosphatase(U/I) Serumcreatinine+mol/l) Serumaluminium@mol/l) SerumintactPTH(ngA) Serum25-hydroxyvitamin D, (@ml)
Lower
Upper
2.9 0 0 0 0 0 3.33
26.5 82.9 23.5 100.0 55.6 84.4 3.33
Referencerange
2.15-2.5s 0.80-1.40 30.0-130.0 50.0-112.0 0.10-0.60 0.00-50.0 3.60-30.0
143 le 2
Resorption cavity characteristics in control subjects and patients with renal osteodystrophy Cavity characteristics
Subject group Control (n ‘- 41)
Maximum depth @m) Mean depth @m) Reconstructed length bm) Eroded length bm) Cavity area km*) Reconstructed surface (%) Eroded surface (%) Number cavities/mm* Number cavities/mm Bone remodellcd (%) a Geometricmean.
34.6= 23.8-50Sb 20.8 14.5-29.8 209.0 146.2-298.8 169.1-340.1 239.8 3447 1871-6422 1.68 0.65-4.32 1.94 0.76-4.93 0.23 0.10-0.56 0.072 0.031-0.169 0.24-1.40 0.58
Significance Renal osteodystrophy (n = 30) 47.9
35.0-65.6 30.4 22.3-41.6 220.2 156.7-309.3 277.5 208.4-369.5 5414 3145-9320 3.58 1.05-12.23 4.46 1.33-14.98 0.40 0.12-1.27 0.148 0.050-0.440 1.34 0.33-5.51
P < 0.0001 P < 0.0091
NS P c 0.0005 P
b 95% range.
in 56% of patients. Serum calcium levels were elevated in 10 of the patients (range 1.74-3.14 mmol/l); of these 7 patients were receiving la-hydroxyvitamin D,. istologically, 24 patients had an increased surface extent of reso normal values being found in the remaining 6 biopsies. Three patients (all with increased eroded surface) had an increased mean osteoid seam width (>12 pm), the values being 39.2, 19.2, and 15.0 pm. Relative osteoid surface in the 30 patients ranged from 1.62-91.1%, the geometric mean being 26.1% and 27% of patients having values above the normal range. Trabecutar bone volume was not significant1Ydifferent from control subjects. 1.9 ,
pc0.0001
p<0.0001
1.8 1.7 1.6 1.6 -
I
aximum depth
an depth
+ 2SD in normal subjects is indiFig. 2. Mean and maximum resorption depth in the patients. The mean _ cated by the horizontal bars.
144 pcO.0~1
3.2 1
Renalostsodystrophy
Control
Fig.3. Resorption cavity area in the patients and controls. The mean is indicated for each group.
Because of either poor quality biopsies, insufficient number of sections, or trabecular shattering only 30 of the 37 biopsies obtained from the patient group were of sufficient quality to enable quantitative assessment of resorption cavities, whilst 41 of the 50 control biopsies were adequate. Resorption cavity characteristics in the two groups are shown in Table 2. Both the mean and maximum cavity depth (Fig. 2) were significantly greater in the patient group (P < 0.0001); the cavity area (Fig. 3), the percentage of bone being remodelled, the number of cavities/mm trabecular surface and the eroded surface were also significantly greater (P < 0.0001). However, the surface length of individual cavities in patients was not significantly different from that of controls. In the patient group, serum intact parathyroid hormone concentrations showed a significant positive correlation with resorption cavity depth (r = 0.451, P < 0.05; as-
3.6 -
3.4 -
3.2 *
1
2 Log
3
4
parathyrold
5
6
hormone
7
9
9
concentration
Fig. 4. The relationship between mean resorption cavity depth and serum intact parathyroid hormone concentration in the patient group. The regressionequation is log mean depth = 3.09 + 0.0631 log PTH.
145 Fig. 4) and osteoid surface (r = 0.429, P < 0.05). Serum alkaline phosphatase levels showed significant correlations with the cavity count/mm trabecular surface (r = 0.474, k c 0.01) and the percentage of bone being remodelled (a = 0.404, P < 0.05). Serum aluminium was not significantly correlated with any of the resorption cavity characteristics. The bone aluminium content showed significant correlations with the eroded surface (r = 0.516, B < 0.05), cavity number/mm2 (r = 0.453, P < 0.05) and the percent bone remodelled (r = 0.588, P c 0.01). No significant correlations were found between bone aluminium content and any of the other resorption cavity characteristics or histomorphometric variables.
iscussio Renal osteodystrophy is a complex disorder in which varying combinations of hyperparathyroidism and osteomalacia may occur 131.Of the patients included in the present study, 80% had evidence of secondary hyperparathyroidism as manifested by an increase in percentage eroded trabecular surface and 3 of these patients also showed some evidence of a mineralisation defect. There was thus a wide spectrum of histological appearances, eroded surface varying from just over 1 to 15% and osteoid surface between 1.6 and 91.9% of the trabecular bone surface. In keeping with the histological appearances, 84.4% of patients had biochemical evidence of secondary hyperparathyroidism with elevated serum intact parathyroid hormone concentrations. The raised serum calcium in 26.5% of patients probably reflects treatment with la-hydroxyvitamin D,, although tertiary hyperparathyroidism or co-existing primary hyperparathyroidism were not specifically excluded, The results of the present study demonstrate that the increase in bone resorption which occurs in patients wi
146
is in keeping with the positive effect of parathyroid hormone on activation frequenhyperparathyroidism in cy and hence bone turnover [ 111.Suppression of cacondary us patients with chronic renal failure with the use of phosphate binders and dietary phosphate restriction results in a reduction of bone turnover [12], although the effects of such treatment on the depth and other morphological characteristics of resorption cavities have not been established. The increased resorption depth in our patients contrasts with data obtained from patients with primary hyperparathyroidism. Using a technique for measurement of resorption depth based on the number of bone lamellae eroded beneath the mineralised bone surface [13] Eriksen [14] reported reduced resorption depth in a group of 19 patients with primary hyperparathyroidism. This difference may be related to the higher serum levels of parathyroid hormone in secondary hyperparathyroidism [15] or might be the consequence of one or more of the many other metabolic disturbances, including uraemia, acid base imbalance or aluminium toxicity, which accompany chronic renal failure; in this context, the relationship between bone aluminium and some resorption cavity characteristics is of interest. The reduced resorption depth demonstrated in patients with primary hyperparathyroidism is consistent with the preservation of trabecular bone volume and structure in iliac crest bone obtained from such patients recently reported by Silverberg et al. 1161.I-Iowever, in secondary hyperparathyroidism, the combination of increased resorption depth and high bone turnover is likely to result in trabecular penetration and loss of connectedness and although treatment of the bone disease may be successful in reducing bone turnover it is unlikely to restore trabecular architecture once this has been disrupted [17]. These considerations emphasise the importance of preventing phosphate retention in chronic renal failure since the adverse mechanical effects associated with loss of trabecular bone structure may result in a higher fracture risk in later life. Until recently, assessment of bone resorption has been unsatisfactory and limited to measurement of the extent of trabecular bone surface occupied by resorption cavities and the number of osteoclasts present. The development of methods which enable direct assessment of resorption depth, based either on the counting of eroded lamellae as described by Eriksen et al. [13] or on computerised techniques as used in the present study [8] should provide valuable information about the contribution of changes in resorption depth to remodelling imbalance in different disease states; in addition, such techniques can be applied to the study of the effects of drugs on resorption cavity dimensions.
Acknowledgements We are grateful to the Welsh Office and the National Kidney Research Fund for generous financial support. Dr J.E. Compston is supported by the Wellcome Trust. We thank Dr I. Mainsworth for biochemical assessment of bone aluminium.
1 Stanbury SW. Bone disease in uraemia. Amer J Med 1%8;44:714-724. 2 Massry SG, Coburn JW, Popovtzer MM, Schinaberger JH, Maxwell MH, Kleeman CR. Secondary hyperparathyroidism in chronic renal failure. The clinical spectrum in uremia, during haemodialy& and after renal transplantation. Arch Intern Med 1969;124:431-441. 3 Slatopolsky E, Caglan S, Pennell JP, Taggart DQ, Canterbury JM, Reiss E, Bricker NS. On the pathogenesis of hyperparathyroidism in chronic experimental renal insufficiency in the dog. J. Clin Invest 1971;50:492-499. 4 Eastwood JB, Bordier PJ, de Wardener HE. Some biochemical, histological, radiological and clinical features of renal osteodystrophy. Kidney Int 1973;4:128-140. S Ritz E, Prager P, Krempien B, Bommer J, Malluche HH, Schmidt-Gayk H. Skeletal K-ray findings and bone histology in patients on haemodialysis. Kidney Int 1978;13:316-323. 6 Bordier PJ, Marie PJ, Arnaud CD. Evolution of renal osteodystrophy: correlation of bone histomorphometry and serum mineral and immunoreactive parathyroid hormone values before and after treatment with calcium carbonate or 25hydroxycholecalcificerol. Kidney Int Suppl1975;2:102-112. 7 Frost HM, Griffith DL, Jee WSS, Kimmel DB, McCandlis RP, Teitelbaum SL. Histomorphometric changes in trabecular bone of renal failure patients treated with calcifediol. Metab Bone Dis Rel Res 1981;2:28S-295. 8 Garrahan NJ, Croucher PI, Compston JE. A computerised technique for the quantitative assessment of resorption cavities in trabecular bone. Bone 1990;11:241-246. 9 Vedi S, Compston JE, Webb A, Tighe JR. Histomorphometric analysis of bone biopsies from the iliac crest of normal British subjects. Metab Bone Dis Rel Res 1982;4:231-236. 10 Delaisse JM, Boyde A, Maconnachie E, Ali NN, Sear CHJ, Eeckout Y, Vaes G, Jones SJ. The effects of inhibitors of cysteine-proteinases and collagenase on the resorptive activity of isolated osteoclasts. Bone 1987;8:305-313. 11 Parfitt AM, Gallagher JC, Heaney RP, Johnston CC, Neer R, Whedon GD. Vitamin D and bone health in the elderly. Amer J Clin Nutr 1982;36:1014-1031. 12 Turner C, Compston J, Mak RHK, Vedi S, Mellish RWE, Haycock GB, Chantler C. Bone turnover and 1,25dihydroxycholecalciferol during treatment with phosphate binders. Kidney Int 1988;33: 989-995. 13 Eriksen EF, Melsen F, Mosekilde L. Reconstruction of the resorptive site in iliac trabecular bone: a kinetic model for bone resorption in 20 normal individuals. Metab Bone Dis Rel Res 1984;5:23S-242. 14 Eriksen EF. Normal and pathological remodelling of human trabecular bone: three dimensional reconstruction of the remodelling sequence in normals and in metabolic bone disease. Endocr Rev 1986;7:379-408. 19 Brown RC, Aston JP, Weeks I, Woodhead JS. Circulating intact parathyroid hormone measured by a two-site immunochemiluminometric assay. J Clin Endocrinol Metab 1987;65:407-414. 16 Silverman SJ, Shane E, De La Crux L, Dempster DW, Feldman F, Seldin D, Jacobs TP, Siris ES, Cafferty M, Parisien MV, Lindsay R, Clemens TL, Bilezikian JP. Skeletal disease in primary hyperparathyroidism. J Bone Miner Res 1989;4:283-291. 17 Compston JE, Mellish RWE, Croucher PI, Newcombe R and Garrahan NJ. Structural mechanisms of trabecular bone loss in man. Bone Miner 1989;6:339-350.