mechanical strength of cortical bone has been established. The contribution of other parameters of microstructure such as osteon dimensions for strength is in ...
Calcif Tissue Int (2001) 68:38-42 DOI: 10.1007/s002230001182
Tissue Cailcified ternational © 2001 Springer-Verlag New York Inc.
Prediction of Cortical Bone Porosity In Vitro by Microcomputed Tomography N. J. Wachter, 1 P. A u g a t , 2 G. D. Krischak, 1 M. Mentzel, 1 L. Kinzl, 1 L. Claes 2
1Department of Traunaatology,Hand- and Reconstructive Surgery, University of Ulna, Ulna, Germany 2 Institute of Orthopaedic Research and Biomechanics, University of Ulm, Helmholzster.14, D-89081 Ulna, Germany Received: 23 February 2000 / Accepted: 14 August 2000 / Online publication: 23 February 2001
Abstract. The high importance of intracortical porosity for
mechanical strength of cortical bone has been established. The contribution of other parameters of microstructure such as osteon dimensions for strength is in discussion. The aim of this study was to evaluate the predictive value of microcomputed tomography (p~CT) for porosity and other microstructural parameters of cortical bone in cortical bone biopsies. Femoral cortical bone specimens from the middiaphysis of 24 patients were harvested during the procedure of total hip replacement at the location where normally one hole ( 0 4.5 mm) for the relief of the intramedullary pressure is placed. In vitro intracortical porosity and bone mineral density (BMD) measurements by IxCT were compared with structural parameters assessed in histological sections of the same specimens. A strong correlation was found between intracortical porosity measured by p~CT and histological porosity (r = 0.95, P < 0.0001). Porosity measured by p~CT was also a strong predictor for other parameters describing dimensions of porous structures. BMD 1 was associated with osteonal area (r = -0.76, P < 0.0001 ). We consider the measurement of porosity by txCT as a very potent procedure for assessing intracortical porosity and parameters related to porous structures of cortical bone nondestructively in vitro. Key words: Cortical bone - - Microstrucmre - - Microcomputed tomography - - Porosity - - Bone mineral density
analysis of histological sections, which is established as a standard procedure for the analysis of microstructure of cortical bone. For cancellous bone, the method of choice for nondestructive evaluation of microstructure is microcomputed tomography 0xCT) [13-16]. The spatial resolution of p~CT systems has been continuously improved and allows the visualization of structures as small as 10 ~m [13]. Systems operating at resolutions of this magnitude might also allow the quantification of the porosity of cortical bone. In addition to porosity, other microstructural parameters are considered to be important for the mechanical strength of cortical bone. It has been stated that in cortical bone there is a decrease in osteonal area, an increase in Haversian canal area, and an increasing number of osteons per unit area with advancing age [17-22]. The aim of our study was to evaluate the predictive value of BMD and intracortical porosity measured in vitro by txCT for the microstructure of cortical bone determined by histological sections. Parameters that were histologically assessed to describe the cortical bone microstructure include porosity, dimension of porous structures, and osteon dimensions. Materials and Methods
Cortical bone significantly contributes to the mechanical strength of bone [1, 2]. Its assessment might therefore be relevant for the prediction of fracture risk or the choice of suitable therapy strategies in orthopedic surgery. Several approaches for the prediction of cortical bone strength have been described, most of them using cortical bone mineral density (BMD) or measures of cortical geometry [3, 4]. However, low correlation coefficients have often been reported between cortical bone strength and BMD [5, 6]. The relationship between intracortical porosity and cortical bone strength has been examined frequently. Strong correlations have been reported and power relationships have been established between these parameters [7-12]. Intracortical porosity has been determined by microscopical
Correspondence to: P. Augat
Extraction of Bone Biopsies
Cortical bone biopsies from the lateral diaphysis of 24 patients (15 female and 9 male, aged 56-88 years; mean 69.1) undergoing surgery for total hip replacement were evaluated. The biopsies were harvested during the implantation of the shaft of the hip prosthesis at approximately middiaphyseal level of the lateral aspect of the femur (Fig. I). Patients' agreement to this procedure was obtained by written consent. A water-cooled diamond extracting system (0 4.5 mm Merck, Germany) was used to harvest the specimens that had a diameter of 3.6 mm and an average length of 4.126 + 1.66 mm. The specimens were immediately frozen at -20°C. CT Measurements
The ixCT-imaging was performed in vitro by a Micro Computed Tomography Device (CT-Microscope, Stratec, Pforzheim, Germany) at a spatial resolution of 30 Ixm. In a first step, a scout view was scanned perpendicular to the axis of the bone cylinders to
N. J. Wachter et al.: Prediction of Cortical Bone Porosity by IxCT
39
Fig, 1. Extracting technique of the cortical bone cylinders from the lateral femur diaphysis.
identify the main direction of the Haversian systems, which was marked manually on the surface of the specimens. Then the bone cylinders were positioned for a second scan parallel to the specimen axis, so that cross-sectional images of the Haversian structures were obtained (Fig. 2). With the software of the txCT-system the BMD was calculated. For the calculation of cortical bone porosity (IxCT-porosity), the scans were converted into a binar?( image and bone area was determined. A threshold of 850 mg/cm ~ was considered optimal for the distinction of cortical bone of the femoral diaphysis. The IxCT-porosity was computed by dividing bone area by section area. After the CT-measurements, the specimens were placed in neutral buffered formalin for 7 days, washed in running water, and dehydrated in successive solutions of 70% and 100% ethanol for 4 days each.
Fig. 2. Position of the ixCT-scan and the histological section relative to the axis of the bone cylinder.
Histological Sections The specimens were stained with a block-staining technique (Paragon), placed in acetone for 4-5 hours, followed by methacrylate monomer infiltration for 7 days. They were embedded in polymethylmethacrylate and allowed to harden. Sections approximately 100-p~m thick were cut from the specimens in the same direction in which the IxCT imaging was performed, so that crosssections of the Haversian systems were obtained (Fig. 2). After grinding and polishing the specimens, microscopy was performed (Zeiss, Germany). The images were recorded by a video imaging system at a magnification of 5x and transferred to an image analysis system (AnalySis, Miinster, Germany). For the evaluation of osteon dimensions, the borders of individual osteons were traced by hand because the distinction between adjacent osteons was not perceptable enough to allow an automatic image analyzing algorithm to discriminate them automatically (Fig. 3). Also, the Haversian canals were traced manually. Empty spaces, which could not be associated with an osteon, including Volkman's canals, were counted as porous structures. The criteria of Barth et al. [8] were applied. For the determination of osteon density the number of osteons per unit area was counted. According to the criteria of Barth et al. [8] some osteons could be counted although osteon dimensions could not be measured. The following parameters were determined: Average Haversian canal area Average osteonal area Average area of pores Total section area Total osteonal Area Total Haversian canal area Total area of pores Intracortical porosity Osteon density Fraction of Osteonal structures Porous structures
Area H a v Area O av Area_P_av A Area O Area_H Area_P (Area_P + Area_H) ! A Number_O / A
[ixm2] [p,m 2] [txm 2] [txm 2] [txm2] [l~m z] [txm2] [%] [cm -2]
Area_O ! A Area P / A
[%] [%]
Fig. 3. Tracing the borders of osteons and Haversian canals in the histological section. Table 1. Dimensions of osteons and porous structures
Max Min Average SD
Area H av [~m 21
Area 0 av [~m 2]
Area P av
31244 665 4156.9 5984.0
63959 26390 41620.5 9235.8
101337 3100 19862.5 24835.8
[b~m2]
Statistical Analysis Correlations were examined by regression analysis using a linear regression model (StatView, Abacus, USA). Results The results for the d i m e n s i o n s of osteons and porous structures o f 24 s p e c i m e n are s h o w n in T a b l e 1. Histological
40
N.J. Wachter et al.: Prediction of Cortical Bone Porosity by ixCT
Table 2. Statistics for histological porosity, osteon density, and osteonal / porous fraction
Max Min Average SD
Hist_Porosity [%]
Number_O / A
26 4 9.1 5.0
22 5 10.6 3.5
[I.zm-2]
Area_O / A
Area P / A
[%]
[%]
CT_Porosity [%]
BMD [mg/cm3]
58 16 41.6 8.8
24 1 5.3 5.1
31 1 10.3 7.0
1345 939 1215.4 101.4
Fig. 4. Comparison of p~CT-scan and histological section (77 years, female), Example for high intracortical porosity (histological: 18.8%; txCT-porosity: 22.3%). Single Haversian canals of the p,CT-scan can be associated with the same canal in the histological section.
porosity, osteon density, osteonal fraction, and porous fraction, as well as the IxCT parameters are presented in Table 2. Significant positive correlations were found between porosity measured in histological sections (intracortical porosity) and porosity measured by p~CT scans (Fig. 5). The histological parameters representing the porous structures also showed significant correlations with the p~CT-porosity (Table 3). Correlations of the parameters describing the porous structures were generally lower for the correlation of BMD -1 than for the correlation with ~zCT porosity (Table 3). For parameters describing osteon and Haversian canal dimensions, only the correlation between average Haversian canal area (Area H_av) and ~xCT-porosity was significant (r = 0.67, P < 0.001) (Table 4). Osteon density correlated significantly with ~CT-porosity and BMD (r = -0.68 / r = 0.58, P < 0.001; Table 4). The only parameter predicted better by BMD -~ than by ~xCT-porosity was the osteonal fraction (Area_O/A, r = 0.76 for BMD -I, r = -0.63 for
ixCT-porosity). A significant positive relationship between lxCT-porosity and BMD -1 could be established (Table 3, Fig. 6).
Discussion
Our data revealed a high predictive value of the porosity calculated from poCT-scans for the histologically determined porosity. In addition, parameters related to the dimensions of porous structures such as fraction of porous structures, average diameter of pores, and average pore area showed strong correlations with histological porosity. All of these parameters were better predicted by IxCT-porosity than by BMD -1. These findings indicate that the nondestructive assessment of intracortical porosity of the femoral diaphysis is possible and that IxCT-porosity and B M D -1 are suitable measures for the estimation of porous structures in biopsies of cortical bone.
41
N. J. Wachter et al.: Prediction of Cortical Bone Porosity by txCT 35-
/
~
30
Table 4. Correlation coefficients between p~CT-Porosity and BMD -1 and parameters describing osteon and Haversian canal dimensions (n = 24, P < 0.0001)
>" 20 '
Number_O A Area_O/A Area O av Area_H av [ c m -2] [%] [>m2] [ D m 2]
~15" 13_ FLIOO
•
•
p~CTPorosity -0.66 a BMD -1 -0.58 ~
5" 0
. . . .
'
....
5
' .... 10
' .... 15
' .... 20
' . . . . -n 25 30
a
Intmcortical Porosity [%] Fig. 5. Regression between histological porosity and IxCTporosity (r = 0.95, n = 24, p < 0.0001).
-0.67 a -0.76
n.s. n.s.
0.67 n.s.
p < 0.05; n.s. = not significant
351
30
•
~2~ Table 3. Correlation coefficients between p,CT-Porosity and
20
BMD -~ and parameters concerning porous structures (n = 24, P < 0.0001) ixCT-Porosity Hist_Porosity Area_P/A Area P av [%]
poCTPorosity BMD -1 0.84
[%]
[%]
[ D m 2] 0
|
. . . .
0
0.95 0.82
0.94 0.84
0.91 0.82
,
5
.
.
.
.
,
10
.
.
.
.
,
15
.
.
.
.
,
.
.
.
.
20
,
25
.
.
.
.
,
30
Intracortical Porosity [%] Fig. 6. Regression between txCT-porosity and BMD -1 (r = 0.84, n = 24, p < 0.0001).
Increased intracortical porosity has been observed as a consequence of aging, disease (hyperparathyroidism, osteoporosis), and pharmacological intervention (thyroid hormone, fluoride, parathyroid hormone, prostaglandins) [23] and has been considered to be important for the mechanical strength of cortical bone [8, 10]. It has also been reported that small changes in porosity or density of compact bone exert a more pronounced influence on its stiffness than would similar changes in trabecular bone [5, 11 ]. The strong correlations of compact bone elastic modulus with bone volume fraction, a measure of intracortical porosity, found by Schaffler and Burr [11], Carter and Hayes [9], and Currey et al. [12] supported this observation. McCalden et al. [24] reported that changes in porosity accounted for 76% of the reduction in the strength of cortical bone. Other microstructural parameters are also considered to be important for the mechanical strength of cortical bone. It has been stated that in cortical bone there is a decrease in osteonal area, an increase in Haversian canal area, and an increase of the diameters of osteons and Haversian canals with advancing age [8, t7-22, 25, 26]. We consider poCT imaging to be a potent method for predicting porosity and visualizing pore dimensions of cortical bone nondestructively. In addition, analogous to cancellous bone, three-dimensional analysis of cortical bone structure is possible [13]. The resolution of the txCTscanner used in our study had a resolution of 30 p~m, which is close to the dimensions of Haversian canals (30-200 txm). Porous structures can be discriminated; even some Haversian canals can clearly be associated with the same canals in the histological section (Fig. 4).
Our data revealed significant correlations for average Haversian canal area with poCT-porosity and B M D (Table 4). Other parameters describing the dimensions of single osteons showed no significant correlations with txCTparameters (Table 2). This could be due to the fact that the resolution of our txCT system was high enough to discriminate Haversian canals, but not the borders of osteons. Comparing ixCT-scans and histological sections, some Haversian canals can clearly be associated with the same canal in the histological section (Fig. 4). However, the osteon borders cannot be identified on the p~CT-sections. Although the predictive value of the IxCT for dimensions of single osteons is low, the average Haversian canal area and osteonal fraction, a parameter associated with the entire area of osteonal structures, was well predicted. The p,CT-imaging therefore seems to be a potent method for measuring the average Haversian canal area in cortical bone. Significant correlations were found between txCTparameters and osteon density. Osteonal fraction (Area_O / A), for which strong correlations with hardness of cortical bone have been reported previously [27], was the only histological parameter predicted better by B M D -~ than by IxCT-porosity (Table 4). The reason for better prediction by density could be that osteonal fraction is a parameter associated with the osteons as mineralized structures and not with porous structures. However, the relevance of BMD measured by Quantitative Computed Tomography for mechanical properties of cortical bone has been discussed because low correlation coefficients have been reported between cortical bone strength and BMD [5, 6].
42 The predictive value of osteon density for fracture risk [8, 26], fracture resistance [7], and mechanical strength [27] has been reported previously. The high relevance of osteon morphology for mechanical properties of femoral cortical bone has been recently confirmed by Yeni et al. [7]. Their results suggested that porosity and osteon density were the best explanatory morphological variables for fracture toughness. The morphological parameters together explained 4 9 68% of the variation in fracture toughness. It has been concluded that, although there must be other factors such as biochemical components [28], osteon morphology has an important influence on fracture resistance of cortical bone. Further studies on the coherence of cortical bone morphology and mechanical properties are worthwhile. We consider p~CT imaging to be a potent method to quantitavely assess porosity and pore dimensions in cortical bone biopsies nondestructively in vitro. However, the p~CTmethod for the prediction of bone microstructure is limited, as the biological background of detected bone deviations cannot be evaluated. Conclusions about the causes of deviations in intracortical porosity and further additional information such as bone turnover require histological sections. On the other hand, IxCT is advantageous over histology in that larger samples of cortical bone biopsies can be automatically analyzed with respect to cortical bone morphometry.
References
1. Spadaro JA, Werner FW, Brenner RA, Fortino MD, Fay LA, Edwards WT (1994) Cortical and trabecular bone contribute strength to the osteopenic distal radius. J Orthop Res 12:211218 2. Augat P, Margevicius K, Simon J, Wolf S, Suger G, Claes L (1998) Local tissue properties in bone healing: influence of size and stability of the osteotomy gap. J Orthop Res 16:475481 3. Augat P, Reeb H, Claes LE (1996) Prediction of fracture load at different skeletal sites by geometric properties of the cortical shell. J Bone Miner Res 11:1356-1363 4. Louis O, Boulpaep F, Willnecker J, Van-den-Winkel P, Osteaux M (1995) Cortical mineral content of the radius assessed by peripheral QCT predicts compressive strength on biomechanical testing. Bone 16:375-379 5. Snyder SM, Schneider E (1991) Estimation of mechanical properties of cortical bone by computed tomography J Orthop Res 9:422-431 6. Stromsoe K, Kok WL, Hoiseth A, Alho A (1993) Holding power of the 4.5 mm AO/ASIF cortex screw in cortical bone in relation to bone mineral. Injury 24:656-659 7. Yeni YN, Brown CU, Wang Z, Norman TL (1997) The influence of bone morphology on fracture toughness of the human femur and tibia. Bone 21:453-459
N. J. Wachter et al.: Prediction of Cortical Bone Porosity by p~CT 8. Barth RW, Williams JL, Kaplan FS (1992) Osteon morphometry in females with femoral neck fractures. Clin Orthop :178186 9. Carter DR, Hayes WC (1977) The compressive behavior of bone as a two-phase porous structure. J Bone Joint Surg Am 59:954-962 10. Martin RB, Ishida J (1989) The relative effects of collagen fiber orientation, porosity, density, and mineralization on bone strength. J Biomech 22:419-426 11. Schaffler MB, Burr DB (1988) Stiffness of compact bone: effects of porosity and density. J Biomech 21:13-16 12. Currey JD (1988) The effect of porosity and mineral content on the Young's modulus of elasticity of compact bone. J Biomech 21:131-139 13. Delling G, Hahn M, Bonse U, Busch F, Gunnewig O, Beckmann F, Uebbing H, Graeff W (1995) New possibilities for structural analysis of bone biopsies using microcomputer tomography (poCT). Pathologe 16:342-347 14. Feldkamp LA, Goldstein SA, Parfitt AM, Jesion G, Kleerekoper M (1989) The direct examination of three-dimensional bone architecture in vitro by computed tomography. J Bone Miner Res 4:3-11 15. Goldstein SA, Goulet R, McCubbrey D (1993) Measurement and significance of three-dimensional architecture to the mechanical integrity of trabecular bone. Calcif Tissue Int 53:127-132 16. Goulet RW, Goldstein SA, Ciarelli MJ, Kuhn JL, Brown MB, Feldkamp LA (1994) The relationship between the structural and orthogonal compressive properties of trabecular bone. J Biomech 27:375-389 17. Black J, Mattson R, Korostoff E (1974) Haversian osteons: size, distribution, internal structure and orientation. J Biomed Mater Res 8:299 18. Currey JD (1964) Some effects of ageing in human haversian systems. J Anat 98:69 19. Evans FB (1976) Mechanical propelties and histology of cortical bone from younger and older men. Anat Rec 185:1 20. Jowsey J (1960) Age changes in human cortical bone. Clin O~hop 17:210 21. Kelsey JL, Hoffman S (1987) Risk factors for hip fractures (editorial). N Engl J Med 316:404 22. Martin RB, Pickett JC, Zinaich S (1980) Studies of skeletal remodeling in aging men. Clin Orthop 149:268-282 23. Sietsema WK (1995) Animal models of cortical porosity. Bone 17:297S-305S 24. McCalden RW, McGeough JA, Barker MB, Court Brown CM (1993) Age-related changes in the tensile properties of cortical bone. The relative importance of changes in porosity, mineralization, and micros~ucmre. J Bone Joint Surg Am 75:1193-1205 25. Brockstedt H, Kassem M, Eriksen EF, Mosekilde L, Melsen F (1993) Age- and sex-related changes in iliac cortical bone mass and remodeling. Bone 14:681-691 26. Squillante RG, Williams JL (1993) Videodensitometry of osteons in females with femoral neck fractures. Calcif Tissue Int 52:273-277 27. Evans FB, Bang S (1967) Differences and relationships between the physical properties and the microscopic structure of human femoral, tibial and cortical bone. Am J Anat 120:79 28. Yeni YN, Brown CU, Norman TL (1998) Influence of bone composition and apparent density on fracture toughness of the human femur and tibia. Bone 22:79-84
