- Email: [email protected]
The morphological association between microcracks and osteocyte lacunae in human cortical bone
Bone 37 (2005) 10 – 15 www.elsevier.com/locate/bone
The morphological association between microcracks and osteocyte lacunae in human cortical bone Shijing Qiua,*, D. Sudhaker Raoa, David P. Fyhrieb, Saroj Palnitkara, A. Michael Parfittc a
Bone and Mineral Research Laboratory, E&R Building 7071, Henry Ford Hospital, 2799 W Grand Boulevard, Detroit, MI 48202, USA b Bone and Joint Center, Henry Ford Hospital, Detroit, MI 48202, USA c Division of Endocrinology and Center for Osteoporosis and Metabolic Bone Disease, University of Arkansas for Medical Sciences, Little Rock, AR 72205, USA Received 5 May 2004; revised 14 December 2004; accepted 11 January 2005 Available online 5 May 2005
Abstract We studied the spatial relationship between the osteocyte lacunar – canalicular network and microdamage accumulation in bone matrix. Rib sections from 9 white women aged 50 – 60 were stained with basic fuchsin and examined using bright-field and fluorescence microscopy. The results showed that the numerical and length density of cracks were 5-fold higher in interstitial bone than in osteons (P < 0.001). Osteocyte lacunar density was 17% lower in interstitial bone than in osteonal bone (P < 0.001). In addition, the osteocyte lacunae in interstitial bone were significantly fewer (by 16%) in the area adjacent to microdamage as compared with the area remote from microdamage (P < 0.001). The proportion of fields with lacunar density less than 728/mm2, the cut-off point calculated from ROC analysis, was 30% in osteonal bone, 55% in interstitial bone remote from microcracks and 83% adjacent to microcracks. The mean values of lacunar density in these bones were 10%, 22% and 27% lower than the cut-off point, respectively. The likelihood of microdamage was 3.8 times higher in bone with osteocyte lacunar density <728/mm2. About 73% of the crack profiles were spatially associated, at least partly, with bone fragments in which osteocyte lacunae were absent. We conclude that microdamage and osteocyte deficiency occur in the same bone regions; there is likely a causal relationship between them but we are unable to say which comes first. D 2005 Elsevier Inc. All rights reserved. Keywords: Interstitial bone; Osteocytes; Microdamage; Bone age
Introduction In the skeleton, microdamage can occur as a result of repetitive mechanical loading and contributes to a reduction in skeletal mechanical properties, specifically elastic modulus and fracture toughness [3,9,16]. There is evidence that fatigue-related microcracks increase with aging and are prone to accumulate in interstitial bone [12,23]. Coincidentally, osteocytes decrease with aging, especially in interstitial bone. Many investigators have suggested that microdamage in bone matrix is associated with loss of osteocytes [5 –7,12,19,24]. More recently, Vashishth et al.
* Corresponding author. Fax: +1 313 916 8201. E-mail address: [email protected] (S. Qiu). 8756-3282/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2005.01.023
[24] found that microdamage was profoundly increased in cortical bone in which mean osteocyte lacunar density was below 600/mm2, but they did not examine the spatial relationship between microdamage and osteocyte deficit. Schaffler et al. [23] and Norman et al. [15] separately reported that approximately 80% of microcracks were located in interstitial bone but did not adjust for the larger area of interstitial bone than osteonal bone. Accordingly, there is no direct evidence that accumulation of microdamage coincides with osteocyte deficiency. Bone microdamage is generally defined as linear cracks detectable by light microscopy. Osteocyte lacunae in bone are regarded by some investigators as stress concentrators and as preferential sites for microdamage initiation [10,14,21]. Reilly et al. [21] provided empirical evidence that microcracks frequently arose from osteocyte lacunae
S. Qiu et al. / Bone 37 (2005) 10 – 15
under fatigue loading, but the load they applied can yield a permanent bone deformation in less than 1000 cycles, indicating that the strain magnitude was greater than 5000 microstrain (Ae) [13]. According to McCreadie’s theoretical calculations [10], the maximum strain surrounding the lacunae is about 1.7 times higher than that applied on bone. Nicollella et al. [14] measured perilacunar bone matrix strain in bovine cortical bone using a digital stereoimaging technique. They found that the average strain around osteocyte lacunae was 1.5– 4.5 times greater than the strain applied on bone, whereas the peak strain in local perilacunar matrix may reach 10 times greater [14]. For normal bone under physiological conditions, the peak strain is about 1500 Ae, and the habitual strains produced in daily life range between 400– 800 Ae [8,22], both of which are considerably lower than the strain applied in Reilly’s experiment. Hence, the damage of osteocyte lacunae in Reilly’s study is likely associated with high strains in perilacunar matrix. There are no data, to our knowledge, on the incidence of damaged osteocyte lacunae in the normal skeleton. The purpose of the present study was to determine the morphological association between microdamage and osteocyte lacunae at the microscopic levels. We attempted to find out whether the accumulation of microdamage was spatially associated with osteocyte deficit.
Materials and methods Archived human rib sections were used for this study. The sections were obtained from 9 healthy white women aged 50 – 60 years at autopsy. The method of sample acquisition and processing has been previously described [7]. In brief, the 3-in. rib segments were placed in 1% basic fuchsin and 40% ethyl alcohol for 4 weeks and then rinsed in a large volume of tap water for 48 h. After hydration, cross and longitudinal sections were cut from each rib, hand ground to a thickness of about 50 Am and mounted on the slide. The cross sections were examined using a Nikon microscope equipped with a CCD video camera (Optronics, Goleta, CA). The microscopic image was imported to a Bioquant NOVA image analysis system (R and M
11
Biometrics Inc., Nashville, TN). For microdamage measurements, only linear cracks were included. Bone area (B.Ar) and the number and length of cracks (Cr.N and Cr.L) were measured using bright-field light microscopy (10 objective). Total bone area excluded all bone cavities except osteocyte lacunae, and intact osteon area excluded the haversian canal. Interstitial bone area was calculated from the subtraction of intact osteon area from total bone area. The mean length (mCr.L), numerical density (Cr.N/B.Ar) and length density (Cr.L/B.Ar) of cracks were separately computed for the entire bone, osteonal bone and interstitial bone. Because of unclear attribution, the cracks that involved both osteon and interstitial bone were only counted in total bone but not in osteon or in interstitial bone. Once confirmed by bright-field microscopy, each crack was examined by fluorescence (UV light) microscopy (20 objective) for determining the morphological relationship between microcracks and the osteocyte lacunar– canalicular network; osteocyte lacunae and canaliculi can be identified in basic fuchsin stained sections by blue epifluorescence [12,24]. Microcrack profiles were classified into two types. In type I profiles, cracks were associated, at least partly, with bone that was lacking osteocyte lacunae (Figs. 1A and B); in type II profiles, cracks were entirely localized within bone with intact lacunar – canalicular network (Fig. 1C). The different crack profiles so defined do not necessarily represent different type of crack, but rather the probability that the section plane will intercept a crack in regions with different lacunar density. The numerical density (Cr.N/B.Ar) was calculated for each type of crack profile. The relationship between osteocyte lacunar density and microdamage was determined only in interstitial bone because very few microcracks were present in osteons. Unlike osteons, which can be defined by the cement line, interstitial bone may not show clear cement lines for compartment definition. Therefore, we developed a special method to determine the osteocyte lacunar density adjacent to and remote from microdamage. The bright-field and fluorescence (UV light) microscopy (20 objective) were performed in interstitial compartments with microcrack(s). The microscopic images were saved in TIFF file. Adobe
Fig. 1. The morphological relationship between microcrack profiles and osteocyte lacunar – canalicular network: Type I: the profiles are partly (A) or entirely (B) associated with the bone that is lacking osteocyte lacunae and canaliculi; type II: the profile is entirely associated with the osteocyte lacunar – canalicular network (C).
12
S. Qiu et al. / Bone 37 (2005) 10 – 15
Photoshop was used for image processing. The bright-field and fluorescence microscopic images obtained from the same field were merged. The cracks, osteon boundary and osteocyte lacunae can be clearly shown in the merged image (Fig. 2). Two concentric circles were laid over the merged image. The area containing one and sometimes several linear cracks was covered only by the inner circle (200 Am in diameter in this case), which enclosed bone adjacent to microdamage. The area between the inner and outer circle (300 Am in diameter in this case) enclosed bone remote from microdamage. The number of osteocyte lacunae was counted within each region occupied by interstitial bone as well as in a neighboring osteon. The areas of measurements in each section depended on the number and distribution of microcracks; in this study, 6 – 17 areas were measured in different subjects. The osteocyte lacunar density was calculated by lacunar number/bone area (Lc.N/B.Ar). Differences between variables were assessed by one way ANOVA and Student’s t test for normally distributed data and by Kruskal – Wallis analysis of variance on ranks and Mann – Whitney rank sum test for nonnormally distributed data. These statistical analyses were performed using Sigmastat software (SPSS, Inc., Chicago, IL). The level of statistical significance was accepted at P < 0.05. The ability of the osteocyte lacunar density to discriminate between cracked and uncracked bone was examined by construction of receiver operating characteristic (ROC) curves. Using the cut-off point of osteocyte lacunar density derived from ROC analysis, the likelihood of bone microdamage associated with osteocyte deficiency was assessed through logistic analysis. Both ROC and logistic analyses were performed using the
Table 1 Microdamage in rib cortical bone 2
Size (mm ) Cr.N/B.Ar (/mm2) Cr.L/B.Ar (Am/mm2) mCr.L (Am)
Osteon
Interstitial bone
P
2.79 0.25 14.9 61.9
12.9 1.32 82.9 63.5
<0.001 <0.001 <0.001 NS
(0.62) (0.26) (15.0) (22.3)
(3.42) (0.63) (38.8) (26.9)
MEDCALC program (MedCalc Software, Mariakerke, Belgium).
Results Histomorphometric data of bone area, numerical and surface density and mean length of cracks are listed in Table 1. Osteons accounted for approximately 18% of the entire bone in cross section. About 3% of the cracks appeared in intact osteons, 80% in interstitial bone and 17% involved both. The numerical and length density of cracks in interstitial bone were significantly (more than 5 times) greater than in osteons (P < 0.001) (Table 1). Fluorescence microscopy showed that 73% of the crack profiles were type I. The numerical density of type I profiles was significantly higher than that of type II (P < 0.001, Fig. 3). In interstitial bone, the osteocyte lacunar density was 25% lower in the area adjacent to a crack than the area remote from a crack (P = 0.009). In addition, the lacunar density in interstitial bone, even in the area remote from crack(s), was 17% lower than in osteons (P < 0.001, Fig. 4). ROC analysis was shown in Fig. 5. Values for sensitivity, specificity and area under the curve were 81.9, 45.7 and 0.681 (95% CI: 0.614 –0.744), respectively. The cut-off point for osteocyte lacunar density was 728/mm2 (Fig. 5). In osteonal bone, 30% of microscopic fields had lacunar density below 728/mm2, but in interstitial bone remote from microdamage, this proportion increased to 55% and in bone adjacent to microdamage to 83%. For the low-density fields in osteonal bone and in interstitial bone remote from and adjacent to microdamage, the mean values of lacunar density were 10%, 22% and 27% lower than the cut-off point, respectively (Fig. 6). The value in osteonal bone was significantly higher than that in interstitial bone both remote from and adjacent to microcracks (P < 0.001), but there was no significant difference between the latter two areas. Logistic analysis demonstrated that the likelihood of bone microdamage was 3.8 times higher in the area with osteocyte lacunae <728/mm2 than in the area >728/mm2 (Odds ratio = 3.81. 95% CI: 2.03– 7.14, P < 0.001).
Discussion Fig. 2. Measurement of osteocyte lacunar density in interstitial bone. The inner and outer circles represent the compartments adjacent to and remote from microdamage, respectively. Osteonal bone (shown by *) in the circles is excluded.
In recent years, the association between osteocytes and microdamage has been investigated using basic fuchsin stained sections [4,12,20,24]. A limitation of
S. Qiu et al. / Bone 37 (2005) 10 – 15
13
Fig. 3. The density of different types of crack profile (Cr.N/B.Ar). Type I is significantly more than type II ( P < 0.001).
these studies is that the osteocytes have not been properly preserved. According to the procedures devised by Frost et al. [7], the rib segments were placed in 40% ethanol that contained 1% basic fuchsin for 4 weeks. Although this method could stain osteocyte lacunae and microcracks, it was not appropriate for bone cell fixation. Consequently, these investigations, including the current study, only observed osteocyte lacunae instead of osteocytes. Since the empty lacunae that remain after osteocyte death will eventually become filled by mineralized osteoid tissue [6,18], the loss of osteocytes may be associated with a decrease in total lacunae, leading to an impairment of the integrity of the osteocyte lacunar– canalicular network. Bone is a tissue that undergoes repeated cyclic loading, so fatigue damage, which is in the form of microcracks, is prone to occur in bone matrix. Previous investigations on microdamage have mainly concerned the density of cracks in whole bone [11,15,24,26]. There is little information on the difference in microdamage between osteons and interstitial bone. We demonstrated for the first time that both numerical and length densities
of cracks were significantly higher in interstitial bone than in osteons. O’Brien et al. [17] observed microcrack accumulation in cortical bone during fatigue testing and found that in the early part of bone fatigue life, (approximately 10% of total life) microcracks initiated in interstitial bone. These lines of evidence suggest that interstitial bone is a preferential place for microdamage accumulation. Up to now, only indirect evidence suggested an association between fatigue damage and osteocyte deficiency. As we and others have shown, fatigue damage tends to occur in interstitial bone [17,23] in which the osteocyte density is lower [6,19], and older subjects with lower osteocyte lacunar density also have more microcracks in bone matrix [24]. Vashishth et al. [24] reported
Fig. 4. The osteocyte lacunar density (Lc.N/B.Ar) in osteon (On) and in interstitial bone adjacent to and remote from microdamage (Ad-IrB and ReIr.B). Lacunar density in On is significantly higher than that in Ad-Ir.B (P < 0.001) and Re-Ir.B (P = 0.001); there is also significant difference between the latter two groups (P = 0.009).
Fig. 6. For the low-density fields (<728/mm2), the osteocyte lacunar density in osteonal bone was significantly higher than that in interstitial bone remote from and adjacent to microdamage (P < 0.001), but there was no significant difference between the latter two areas.
Fig. 5. ROC curve for discrimination between cracked and uncracked interstitial bone by osteocyte lacunar density. The area under the curve is 0.681 (95% CI: 0.614 – 0.744).
14
S. Qiu et al. / Bone 37 (2005) 10 – 15
that a significant increase in microdamage would occur in the subjects whose osteocyte lacunar density was <600/mm2, but their data cannot prove that microdamage is spatially increased in areas with osteocyte deficiency. We developed two novel methods for determining the spatial relationship between microdamage and osteocyte lacunar deficiency. First, we measured the densities of osteocyte lacunae in osteonal bone and in interstitial bone that were adjacent to and remote from microcracks. The results indicated that osteocyte lacunae were significantly fewer in the area adjacent to microcracks than in the area remote from the cracks. ROC analysis showed that osteocyte lacunar density was able to discriminate between damaged and undamaged bone (Fig. 5); the cut-point was 728/mm2. The likelihood of microdamage was 3.8 times higher in an area with <728/mm2 of osteocyte lacunae. Second, we observed the spatial relationship between microcracks and the lacunar– canalicular network. Two types of microcrack profiles were found by using fluorescence microscopy. Type I profiles, in which the crack was wholly or partly associated with bone lacking osteocyte lacunae, were much more frequent than type II profiles, in which the crack was associated with an intact lacunar –canalicular network. We can conclude that the overwhelming majority of microcracks are observed in bone in which the osteocyte lacunar– canalicular network is attenuated. To our knowledge, this is the first demonstration of the spatial coincidence between osteocyte deficiency and microdamage accumulation. Because interstitial bone is composed of the fragments of osteons and other lamellar bone that remains from previous remodeling cycles and is accessed relatively less by bone remodeling due to its deep-seated position, the age of interstitial bone is considerably greater than that of osteons and varies between different regions depending on the local remodeling history [1,23]. Since the likelihood of microdamage and the frequency of osteocyte death both increase with the age of the bone, it is to be expected that both phenomena would be more frequent in interstitial bone than in osteonal bone, but the close spatial relationship suggests a causal relationship, although the direction of causality cannot be inferred from a cross sectional study. One possibility is that microdamage occurs first and leads to osteocyte death in the same region. In favor of this possibility is that local osteocyte death is an early response to microdamage [25] and is normally followed by the origination of a new BMU to remove the microdamaged bone [2]. A failure or long delay of the repair process could account for the spatial relationship we observed. A second possibility is that osteocyte death occurs first and leads to perilacunar hypermineralization and increased bone brittleness, thus predisposing to microdamage [7,18]. Further work will be needed to distinguish these possibilities.
Acknowledgment This work was supported by fund from the National Institutes of Health (RO3-AR 48641, DK43858).
References [1] Akkus O, Polyakova-Akkus A, Adar F, Schaffler MB. Aging of microstructural compartments in human compact bone. J Bone Miner Res 2003;18:1012 – 9. [2] Bentolila V, Boyce TM, Fyhrie DP, Drumb R, Skerry TM, Schaffler MB. Intracortical remodeling in adult rat long bones after fatigue loading. Bone 1998;23:275 – 81. [3] Burr D. Microdamage and bone strength. Osteoporos Int 2003; 14(Suppl 5):67 – 72. [4] Frank JD, Ryan M, Kalscheur VL, Ruaux-Mason CP, Hozak RR, Muir P. Aging and accumulation of microdamage in canine bone. Bone 2002;30:201 – 6. [5] Frost HM. In vivo osteocyte death. J Bone Jt Surg 1960;42A:138 – 43. [6] Frost HM. Micropetrosis. J Bone Jt Surg 1960;42A:144 – 50. [7] Frost HM. Presence of microscopic cracks in vivo in bone. Bull Henry Ford Hosp 1960;8:25 – 35. [8] Martin RB, Burr DB. Fatigue of bone. In: Martin RB, Burr DB, editors. Structure, function, and adaptation of compact bone. New York’ Raven Press; 1989. p. 186 – 213. [9] Mashiba T, Turner CH, Hirano T, Forwood MR, Johnston CC, Burr DB. Effects of suppressed bone turnover by bisphosphonates on microdamage accumulation and biomechanical properties in clinically relevant skeletal sites in beagles. Bone 2001;28: 524 – 31. [10] McCreadie BR, Hollister SJ. Strain concentrations surrounding an ellipsoid model of lacunae and osteocytes. Comput Methods Biomech Biomed Engin 1997;1:61 – 8. [11] Mori S, Burr DB. Increased intracortical remodeling following fatigue damage. Bone 1993;14:103 – 9. [12] Mori S, Harruff R, Ambrosius W, Burr DB. Trabecular bone volume and microdamage accumulation in the femoral heads of women with and without femoral neck fractures. Bone 1997;21:521 – 6. [13] Nicolella DP, Lankford J. Strain concentration effects of osteocyte lacunae. Trans Orthop Res Soc 2002 [27 pp. Paper: 0088. Dallas, Texas]. [14] Nicolella DP, Moravits DM, Lankford J, Bonewald L. Bone matrix strain is amplified at osteocyte lacunae in cortical bone. J Bone Miner Res 2004;19:S72. [15] Norman TL, Wang Z. Microdamage of human cortical bone: incidence and morphology in long bones. Bone 1997;20:375 – 9. [16] Norman TL, Yeni YN, Brown CU, Wang Z. Influence of microdamage on fracture toughness of the human femur and tibia. Bone 1998;23: 303 – 6. [17] O’Brien FJ, Taylor D, Lee TC. Microcrack accumulation at different intervals during fatigue testing of compact bone. J Biomech 2003; 36:973 – 80. [18] Parfitt AM. Bone age, mineral density, and fatigue damage. Calcif Tissue Int 1993;53:S82 – 5 [discussion S85-6]. [19] Qiu S, Rao DS, Palnitkar S, Parfitt AM. Age and distance from the surface but not menopause reduce osteocyte density in human cancellous bone. Bone 2002;31:313 – 8. [20] Qiu SJ, Boyce TM, Schaffler MB. Osteocyte loss and microdamage in aging human compact bone. Trans Orthop Res Soc 1997;22:89. [21] Reilly GC. Observations of microdamage around osteocyte lacunae in bone. J Biomech 2000;33:1131 – 4. [22] Rubin CT, Lanyon LE. Limb mechanics as a function of speed and gait: a study of functional strains in the radius and tibia of horse and dog. J Exp Biol 1982;101:187 – 211.
S. Qiu et al. / Bone 37 (2005) 10 – 15 [23] Schaffler MB, Choi K, Milgrom C. Aging and matrix microdamage accumulation in human compact bone. Bone 1995;17:521 – 5. [24] Vashishth D, Verborgt O, Divine G, Schaffler MB, Fyhrie DP. Decline in osteocyte lacunar density in human cortical bone is associated with accumulation of microcracks with age. Bone 2000;26: 375 – 80.
15
[25] Verborgt O, Gibson GJ, Schaffler MB. Loss of osteocyte integrity in association with microdamage and bone remodeling after fatigue in vivo. J Bone Miner Res 2000;15:60 – 7. [26] Zioupos P. Accumulation of in-vivo fatigue microdamage and its relation to biomechanical properties in ageing human cortical bone. J Microsc 2001;201:270 – 8.
The morphological association between microcracks and osteocyte lacunae in human cortical bone Shijing Qiua,*, D. Sudhaker Raoa, David P. Fyhrieb, Saroj Palnitkara, A. Michael Parfittc a
Bone and Mineral Research Laboratory, E&R Building 7071, Henry Ford Hospital, 2799 W Grand Boulevard, Detroit, MI 48202, USA b Bone and Joint Center, Henry Ford Hospital, Detroit, MI 48202, USA c Division of Endocrinology and Center for Osteoporosis and Metabolic Bone Disease, University of Arkansas for Medical Sciences, Little Rock, AR 72205, USA Received 5 May 2004; revised 14 December 2004; accepted 11 January 2005 Available online 5 May 2005
Abstract We studied the spatial relationship between the osteocyte lacunar – canalicular network and microdamage accumulation in bone matrix. Rib sections from 9 white women aged 50 – 60 were stained with basic fuchsin and examined using bright-field and fluorescence microscopy. The results showed that the numerical and length density of cracks were 5-fold higher in interstitial bone than in osteons (P < 0.001). Osteocyte lacunar density was 17% lower in interstitial bone than in osteonal bone (P < 0.001). In addition, the osteocyte lacunae in interstitial bone were significantly fewer (by 16%) in the area adjacent to microdamage as compared with the area remote from microdamage (P < 0.001). The proportion of fields with lacunar density less than 728/mm2, the cut-off point calculated from ROC analysis, was 30% in osteonal bone, 55% in interstitial bone remote from microcracks and 83% adjacent to microcracks. The mean values of lacunar density in these bones were 10%, 22% and 27% lower than the cut-off point, respectively. The likelihood of microdamage was 3.8 times higher in bone with osteocyte lacunar density <728/mm2. About 73% of the crack profiles were spatially associated, at least partly, with bone fragments in which osteocyte lacunae were absent. We conclude that microdamage and osteocyte deficiency occur in the same bone regions; there is likely a causal relationship between them but we are unable to say which comes first. D 2005 Elsevier Inc. All rights reserved. Keywords: Interstitial bone; Osteocytes; Microdamage; Bone age
Introduction In the skeleton, microdamage can occur as a result of repetitive mechanical loading and contributes to a reduction in skeletal mechanical properties, specifically elastic modulus and fracture toughness [3,9,16]. There is evidence that fatigue-related microcracks increase with aging and are prone to accumulate in interstitial bone [12,23]. Coincidentally, osteocytes decrease with aging, especially in interstitial bone. Many investigators have suggested that microdamage in bone matrix is associated with loss of osteocytes [5 –7,12,19,24]. More recently, Vashishth et al.
* Corresponding author. Fax: +1 313 916 8201. E-mail address: [email protected] (S. Qiu). 8756-3282/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2005.01.023
[24] found that microdamage was profoundly increased in cortical bone in which mean osteocyte lacunar density was below 600/mm2, but they did not examine the spatial relationship between microdamage and osteocyte deficit. Schaffler et al. [23] and Norman et al. [15] separately reported that approximately 80% of microcracks were located in interstitial bone but did not adjust for the larger area of interstitial bone than osteonal bone. Accordingly, there is no direct evidence that accumulation of microdamage coincides with osteocyte deficiency. Bone microdamage is generally defined as linear cracks detectable by light microscopy. Osteocyte lacunae in bone are regarded by some investigators as stress concentrators and as preferential sites for microdamage initiation [10,14,21]. Reilly et al. [21] provided empirical evidence that microcracks frequently arose from osteocyte lacunae
S. Qiu et al. / Bone 37 (2005) 10 – 15
under fatigue loading, but the load they applied can yield a permanent bone deformation in less than 1000 cycles, indicating that the strain magnitude was greater than 5000 microstrain (Ae) [13]. According to McCreadie’s theoretical calculations [10], the maximum strain surrounding the lacunae is about 1.7 times higher than that applied on bone. Nicollella et al. [14] measured perilacunar bone matrix strain in bovine cortical bone using a digital stereoimaging technique. They found that the average strain around osteocyte lacunae was 1.5– 4.5 times greater than the strain applied on bone, whereas the peak strain in local perilacunar matrix may reach 10 times greater [14]. For normal bone under physiological conditions, the peak strain is about 1500 Ae, and the habitual strains produced in daily life range between 400– 800 Ae [8,22], both of which are considerably lower than the strain applied in Reilly’s experiment. Hence, the damage of osteocyte lacunae in Reilly’s study is likely associated with high strains in perilacunar matrix. There are no data, to our knowledge, on the incidence of damaged osteocyte lacunae in the normal skeleton. The purpose of the present study was to determine the morphological association between microdamage and osteocyte lacunae at the microscopic levels. We attempted to find out whether the accumulation of microdamage was spatially associated with osteocyte deficit.
Materials and methods Archived human rib sections were used for this study. The sections were obtained from 9 healthy white women aged 50 – 60 years at autopsy. The method of sample acquisition and processing has been previously described [7]. In brief, the 3-in. rib segments were placed in 1% basic fuchsin and 40% ethyl alcohol for 4 weeks and then rinsed in a large volume of tap water for 48 h. After hydration, cross and longitudinal sections were cut from each rib, hand ground to a thickness of about 50 Am and mounted on the slide. The cross sections were examined using a Nikon microscope equipped with a CCD video camera (Optronics, Goleta, CA). The microscopic image was imported to a Bioquant NOVA image analysis system (R and M
11
Biometrics Inc., Nashville, TN). For microdamage measurements, only linear cracks were included. Bone area (B.Ar) and the number and length of cracks (Cr.N and Cr.L) were measured using bright-field light microscopy (10 objective). Total bone area excluded all bone cavities except osteocyte lacunae, and intact osteon area excluded the haversian canal. Interstitial bone area was calculated from the subtraction of intact osteon area from total bone area. The mean length (mCr.L), numerical density (Cr.N/B.Ar) and length density (Cr.L/B.Ar) of cracks were separately computed for the entire bone, osteonal bone and interstitial bone. Because of unclear attribution, the cracks that involved both osteon and interstitial bone were only counted in total bone but not in osteon or in interstitial bone. Once confirmed by bright-field microscopy, each crack was examined by fluorescence (UV light) microscopy (20 objective) for determining the morphological relationship between microcracks and the osteocyte lacunar– canalicular network; osteocyte lacunae and canaliculi can be identified in basic fuchsin stained sections by blue epifluorescence [12,24]. Microcrack profiles were classified into two types. In type I profiles, cracks were associated, at least partly, with bone that was lacking osteocyte lacunae (Figs. 1A and B); in type II profiles, cracks were entirely localized within bone with intact lacunar – canalicular network (Fig. 1C). The different crack profiles so defined do not necessarily represent different type of crack, but rather the probability that the section plane will intercept a crack in regions with different lacunar density. The numerical density (Cr.N/B.Ar) was calculated for each type of crack profile. The relationship between osteocyte lacunar density and microdamage was determined only in interstitial bone because very few microcracks were present in osteons. Unlike osteons, which can be defined by the cement line, interstitial bone may not show clear cement lines for compartment definition. Therefore, we developed a special method to determine the osteocyte lacunar density adjacent to and remote from microdamage. The bright-field and fluorescence (UV light) microscopy (20 objective) were performed in interstitial compartments with microcrack(s). The microscopic images were saved in TIFF file. Adobe
Fig. 1. The morphological relationship between microcrack profiles and osteocyte lacunar – canalicular network: Type I: the profiles are partly (A) or entirely (B) associated with the bone that is lacking osteocyte lacunae and canaliculi; type II: the profile is entirely associated with the osteocyte lacunar – canalicular network (C).
12
S. Qiu et al. / Bone 37 (2005) 10 – 15
Photoshop was used for image processing. The bright-field and fluorescence microscopic images obtained from the same field were merged. The cracks, osteon boundary and osteocyte lacunae can be clearly shown in the merged image (Fig. 2). Two concentric circles were laid over the merged image. The area containing one and sometimes several linear cracks was covered only by the inner circle (200 Am in diameter in this case), which enclosed bone adjacent to microdamage. The area between the inner and outer circle (300 Am in diameter in this case) enclosed bone remote from microdamage. The number of osteocyte lacunae was counted within each region occupied by interstitial bone as well as in a neighboring osteon. The areas of measurements in each section depended on the number and distribution of microcracks; in this study, 6 – 17 areas were measured in different subjects. The osteocyte lacunar density was calculated by lacunar number/bone area (Lc.N/B.Ar). Differences between variables were assessed by one way ANOVA and Student’s t test for normally distributed data and by Kruskal – Wallis analysis of variance on ranks and Mann – Whitney rank sum test for nonnormally distributed data. These statistical analyses were performed using Sigmastat software (SPSS, Inc., Chicago, IL). The level of statistical significance was accepted at P < 0.05. The ability of the osteocyte lacunar density to discriminate between cracked and uncracked bone was examined by construction of receiver operating characteristic (ROC) curves. Using the cut-off point of osteocyte lacunar density derived from ROC analysis, the likelihood of bone microdamage associated with osteocyte deficiency was assessed through logistic analysis. Both ROC and logistic analyses were performed using the
Table 1 Microdamage in rib cortical bone 2
Size (mm ) Cr.N/B.Ar (/mm2) Cr.L/B.Ar (Am/mm2) mCr.L (Am)
Osteon
Interstitial bone
P
2.79 0.25 14.9 61.9
12.9 1.32 82.9 63.5
<0.001 <0.001 <0.001 NS
(0.62) (0.26) (15.0) (22.3)
(3.42) (0.63) (38.8) (26.9)
MEDCALC program (MedCalc Software, Mariakerke, Belgium).
Results Histomorphometric data of bone area, numerical and surface density and mean length of cracks are listed in Table 1. Osteons accounted for approximately 18% of the entire bone in cross section. About 3% of the cracks appeared in intact osteons, 80% in interstitial bone and 17% involved both. The numerical and length density of cracks in interstitial bone were significantly (more than 5 times) greater than in osteons (P < 0.001) (Table 1). Fluorescence microscopy showed that 73% of the crack profiles were type I. The numerical density of type I profiles was significantly higher than that of type II (P < 0.001, Fig. 3). In interstitial bone, the osteocyte lacunar density was 25% lower in the area adjacent to a crack than the area remote from a crack (P = 0.009). In addition, the lacunar density in interstitial bone, even in the area remote from crack(s), was 17% lower than in osteons (P < 0.001, Fig. 4). ROC analysis was shown in Fig. 5. Values for sensitivity, specificity and area under the curve were 81.9, 45.7 and 0.681 (95% CI: 0.614 –0.744), respectively. The cut-off point for osteocyte lacunar density was 728/mm2 (Fig. 5). In osteonal bone, 30% of microscopic fields had lacunar density below 728/mm2, but in interstitial bone remote from microdamage, this proportion increased to 55% and in bone adjacent to microdamage to 83%. For the low-density fields in osteonal bone and in interstitial bone remote from and adjacent to microdamage, the mean values of lacunar density were 10%, 22% and 27% lower than the cut-off point, respectively (Fig. 6). The value in osteonal bone was significantly higher than that in interstitial bone both remote from and adjacent to microcracks (P < 0.001), but there was no significant difference between the latter two areas. Logistic analysis demonstrated that the likelihood of bone microdamage was 3.8 times higher in the area with osteocyte lacunae <728/mm2 than in the area >728/mm2 (Odds ratio = 3.81. 95% CI: 2.03– 7.14, P < 0.001).
Discussion Fig. 2. Measurement of osteocyte lacunar density in interstitial bone. The inner and outer circles represent the compartments adjacent to and remote from microdamage, respectively. Osteonal bone (shown by *) in the circles is excluded.
In recent years, the association between osteocytes and microdamage has been investigated using basic fuchsin stained sections [4,12,20,24]. A limitation of
S. Qiu et al. / Bone 37 (2005) 10 – 15
13
Fig. 3. The density of different types of crack profile (Cr.N/B.Ar). Type I is significantly more than type II ( P < 0.001).
these studies is that the osteocytes have not been properly preserved. According to the procedures devised by Frost et al. [7], the rib segments were placed in 40% ethanol that contained 1% basic fuchsin for 4 weeks. Although this method could stain osteocyte lacunae and microcracks, it was not appropriate for bone cell fixation. Consequently, these investigations, including the current study, only observed osteocyte lacunae instead of osteocytes. Since the empty lacunae that remain after osteocyte death will eventually become filled by mineralized osteoid tissue [6,18], the loss of osteocytes may be associated with a decrease in total lacunae, leading to an impairment of the integrity of the osteocyte lacunar– canalicular network. Bone is a tissue that undergoes repeated cyclic loading, so fatigue damage, which is in the form of microcracks, is prone to occur in bone matrix. Previous investigations on microdamage have mainly concerned the density of cracks in whole bone [11,15,24,26]. There is little information on the difference in microdamage between osteons and interstitial bone. We demonstrated for the first time that both numerical and length densities
of cracks were significantly higher in interstitial bone than in osteons. O’Brien et al. [17] observed microcrack accumulation in cortical bone during fatigue testing and found that in the early part of bone fatigue life, (approximately 10% of total life) microcracks initiated in interstitial bone. These lines of evidence suggest that interstitial bone is a preferential place for microdamage accumulation. Up to now, only indirect evidence suggested an association between fatigue damage and osteocyte deficiency. As we and others have shown, fatigue damage tends to occur in interstitial bone [17,23] in which the osteocyte density is lower [6,19], and older subjects with lower osteocyte lacunar density also have more microcracks in bone matrix [24]. Vashishth et al. [24] reported
Fig. 4. The osteocyte lacunar density (Lc.N/B.Ar) in osteon (On) and in interstitial bone adjacent to and remote from microdamage (Ad-IrB and ReIr.B). Lacunar density in On is significantly higher than that in Ad-Ir.B (P < 0.001) and Re-Ir.B (P = 0.001); there is also significant difference between the latter two groups (P = 0.009).
Fig. 6. For the low-density fields (<728/mm2), the osteocyte lacunar density in osteonal bone was significantly higher than that in interstitial bone remote from and adjacent to microdamage (P < 0.001), but there was no significant difference between the latter two areas.
Fig. 5. ROC curve for discrimination between cracked and uncracked interstitial bone by osteocyte lacunar density. The area under the curve is 0.681 (95% CI: 0.614 – 0.744).
14
S. Qiu et al. / Bone 37 (2005) 10 – 15
that a significant increase in microdamage would occur in the subjects whose osteocyte lacunar density was <600/mm2, but their data cannot prove that microdamage is spatially increased in areas with osteocyte deficiency. We developed two novel methods for determining the spatial relationship between microdamage and osteocyte lacunar deficiency. First, we measured the densities of osteocyte lacunae in osteonal bone and in interstitial bone that were adjacent to and remote from microcracks. The results indicated that osteocyte lacunae were significantly fewer in the area adjacent to microcracks than in the area remote from the cracks. ROC analysis showed that osteocyte lacunar density was able to discriminate between damaged and undamaged bone (Fig. 5); the cut-point was 728/mm2. The likelihood of microdamage was 3.8 times higher in an area with <728/mm2 of osteocyte lacunae. Second, we observed the spatial relationship between microcracks and the lacunar– canalicular network. Two types of microcrack profiles were found by using fluorescence microscopy. Type I profiles, in which the crack was wholly or partly associated with bone lacking osteocyte lacunae, were much more frequent than type II profiles, in which the crack was associated with an intact lacunar –canalicular network. We can conclude that the overwhelming majority of microcracks are observed in bone in which the osteocyte lacunar– canalicular network is attenuated. To our knowledge, this is the first demonstration of the spatial coincidence between osteocyte deficiency and microdamage accumulation. Because interstitial bone is composed of the fragments of osteons and other lamellar bone that remains from previous remodeling cycles and is accessed relatively less by bone remodeling due to its deep-seated position, the age of interstitial bone is considerably greater than that of osteons and varies between different regions depending on the local remodeling history [1,23]. Since the likelihood of microdamage and the frequency of osteocyte death both increase with the age of the bone, it is to be expected that both phenomena would be more frequent in interstitial bone than in osteonal bone, but the close spatial relationship suggests a causal relationship, although the direction of causality cannot be inferred from a cross sectional study. One possibility is that microdamage occurs first and leads to osteocyte death in the same region. In favor of this possibility is that local osteocyte death is an early response to microdamage [25] and is normally followed by the origination of a new BMU to remove the microdamaged bone [2]. A failure or long delay of the repair process could account for the spatial relationship we observed. A second possibility is that osteocyte death occurs first and leads to perilacunar hypermineralization and increased bone brittleness, thus predisposing to microdamage [7,18]. Further work will be needed to distinguish these possibilities.
Acknowledgment This work was supported by fund from the National Institutes of Health (RO3-AR 48641, DK43858).
References [1] Akkus O, Polyakova-Akkus A, Adar F, Schaffler MB. Aging of microstructural compartments in human compact bone. J Bone Miner Res 2003;18:1012 – 9. [2] Bentolila V, Boyce TM, Fyhrie DP, Drumb R, Skerry TM, Schaffler MB. Intracortical remodeling in adult rat long bones after fatigue loading. Bone 1998;23:275 – 81. [3] Burr D. Microdamage and bone strength. Osteoporos Int 2003; 14(Suppl 5):67 – 72. [4] Frank JD, Ryan M, Kalscheur VL, Ruaux-Mason CP, Hozak RR, Muir P. Aging and accumulation of microdamage in canine bone. Bone 2002;30:201 – 6. [5] Frost HM. In vivo osteocyte death. J Bone Jt Surg 1960;42A:138 – 43. [6] Frost HM. Micropetrosis. J Bone Jt Surg 1960;42A:144 – 50. [7] Frost HM. Presence of microscopic cracks in vivo in bone. Bull Henry Ford Hosp 1960;8:25 – 35. [8] Martin RB, Burr DB. Fatigue of bone. In: Martin RB, Burr DB, editors. Structure, function, and adaptation of compact bone. New York’ Raven Press; 1989. p. 186 – 213. [9] Mashiba T, Turner CH, Hirano T, Forwood MR, Johnston CC, Burr DB. Effects of suppressed bone turnover by bisphosphonates on microdamage accumulation and biomechanical properties in clinically relevant skeletal sites in beagles. Bone 2001;28: 524 – 31. [10] McCreadie BR, Hollister SJ. Strain concentrations surrounding an ellipsoid model of lacunae and osteocytes. Comput Methods Biomech Biomed Engin 1997;1:61 – 8. [11] Mori S, Burr DB. Increased intracortical remodeling following fatigue damage. Bone 1993;14:103 – 9. [12] Mori S, Harruff R, Ambrosius W, Burr DB. Trabecular bone volume and microdamage accumulation in the femoral heads of women with and without femoral neck fractures. Bone 1997;21:521 – 6. [13] Nicolella DP, Lankford J. Strain concentration effects of osteocyte lacunae. Trans Orthop Res Soc 2002 [27 pp. Paper: 0088. Dallas, Texas]. [14] Nicolella DP, Moravits DM, Lankford J, Bonewald L. Bone matrix strain is amplified at osteocyte lacunae in cortical bone. J Bone Miner Res 2004;19:S72. [15] Norman TL, Wang Z. Microdamage of human cortical bone: incidence and morphology in long bones. Bone 1997;20:375 – 9. [16] Norman TL, Yeni YN, Brown CU, Wang Z. Influence of microdamage on fracture toughness of the human femur and tibia. Bone 1998;23: 303 – 6. [17] O’Brien FJ, Taylor D, Lee TC. Microcrack accumulation at different intervals during fatigue testing of compact bone. J Biomech 2003; 36:973 – 80. [18] Parfitt AM. Bone age, mineral density, and fatigue damage. Calcif Tissue Int 1993;53:S82 – 5 [discussion S85-6]. [19] Qiu S, Rao DS, Palnitkar S, Parfitt AM. Age and distance from the surface but not menopause reduce osteocyte density in human cancellous bone. Bone 2002;31:313 – 8. [20] Qiu SJ, Boyce TM, Schaffler MB. Osteocyte loss and microdamage in aging human compact bone. Trans Orthop Res Soc 1997;22:89. [21] Reilly GC. Observations of microdamage around osteocyte lacunae in bone. J Biomech 2000;33:1131 – 4. [22] Rubin CT, Lanyon LE. Limb mechanics as a function of speed and gait: a study of functional strains in the radius and tibia of horse and dog. J Exp Biol 1982;101:187 – 211.
S. Qiu et al. / Bone 37 (2005) 10 – 15 [23] Schaffler MB, Choi K, Milgrom C. Aging and matrix microdamage accumulation in human compact bone. Bone 1995;17:521 – 5. [24] Vashishth D, Verborgt O, Divine G, Schaffler MB, Fyhrie DP. Decline in osteocyte lacunar density in human cortical bone is associated with accumulation of microcracks with age. Bone 2000;26: 375 – 80.
15
[25] Verborgt O, Gibson GJ, Schaffler MB. Loss of osteocyte integrity in association with microdamage and bone remodeling after fatigue in vivo. J Bone Miner Res 2000;15:60 – 7. [26] Zioupos P. Accumulation of in-vivo fatigue microdamage and its relation to biomechanical properties in ageing human cortical bone. J Microsc 2001;201:270 – 8.