Effects of Advanced Age on the Morphometry and Degenerative State of the Cervical Spine in a Rat Model.код для вставкиСкачать
THE ANATOMICAL RECORD 294:1326–1336 (2011) Effects of Advanced Age on the Morphometry and Degenerative State of the Cervical Spine in A Rat Model ANDREW C. LAING,1* RILEY COX,2 WOLFRAM TETZLAFF,3,4 4,5 AND THOMAS OXLAND 1 Injury Biomechanics and Aging Laboratory, Department of Kinesiology, University of Waterloo, Waterloo, Ontario, Canada 2 School of Biomedical Physiology and Kinesiology, Simon Fraser University, Burnaby, British Columbia, Canada 3 Department of Zoology, University of British Columbia, Vancouver, British Columbia, Canada 4 International Collaboration on Repair Discoveries (ICORD), Vancouver, British Columbia, Canada 5 Orthopaedics and Injury Biomechanics Group, Departments of Orthopaedics and Mechanical Engineering, University of British Columbia, Vancouver, British Columbia, Canada ABSTRACT Aging causes changes in the geometry of the human cervical spine that may inﬂuence the tissue response to applied loads. Rat models are often used to study spinal cord injuries (SCI) and have the potential to enhance our understanding of the effect of age on SCI. The goal of this study was to characterize the morphometry and degenerative state of the cervical spine in Fisher 344 rats, and to determine the inﬂuence of age on these variables. Fifteen rats were split into three age groups: young adult (3 months of age), aged (12–18 months) and geriatric (30 months). Following tissue harvest we used a lCT scanner to image the cervical and upper thoracic spine from each specimen. Analysis software was used to measure variables including canal pinch diameter (the most rostral point on the dorsal aspect of a vertebral body to the most caudal aspect of the lamina on the immediately rostral vertebra), vertebral canal depth, width, and area, vertebral body height, depth, width, and area, and intervertebral disc thickness. Orthopaedic surgeons used midsagittal images to rate the degenerative state of the intervertebral discs. For all measures except disc thickness there was a signiﬁcant increase (mean (SD) ¼ 15.0 (9.7)%) for the aged compared to young specimens (P < 0.05). There were signiﬁcant differences between the aged and geriatric specimens for only vertebral body depth (P ¼ 0.016) and area (P ¼ 0.020). Intervertebral disc degeneration was signiﬁcantly greater on the ventral aspect of the spinal column (P < 0.001), with a trend toward increased degeneration in the geriatric specimens (P ¼ 0.069). The results suggest that agerelated morphometric differences may need to be accounted for in experiC 2011 mental aging models of SCI in rats. Anat Rec, 294:1326–1336, 2011. V Wiley-Liss, Inc. Key words: spine; age; rat; morphometry; degeneration; cervical Grant sponsor: The Natural Sciences and Engineering Research Council of Canada (NSERC); Grant sponsor: Christopher and Dana Reeve Foundation; Grant number: LB10801-2. *Correspondence to: Andrew C. Laing, Department of Kinesiology, University of Waterloo, 200 University Ave West, WaterC 2011 WILEY-LISS, INC. V loo, Ontario, Canada N2L 3G1. Fax: 519-746-6776. E-mail: [email protected] Received 7 July 2010; Accepted 26 March 2011 DOI 10.1002/ar.21436 Published online 28 June 2011 in Wiley Online Library (wileyonlinelibrary.com). EFFECTS OF AGE ON SPINE MORPHOMETRY IN RATS INTRODUCTION There are approximately 12,000 new cases of spinal cord injury (SCI) in North America each year (O’Connor, 2006; NSCISC, 2010). Because of the chronic nature of many of these injuries, the lifetime medical costs typically range between $500,000 and $5,000,000 per case, whereas the total economic cost (including medical care and lost wages) is estimated in the tens of billions of dollars per year in North America alone (Priebe et al., 2007). Even more important are the effects of SCI on personal suffering and decreased quality of life. Because of the aging of the population in most developed nations, the incidence of SCI in elderly persons is growing faster than in any other age group (similar to the trends for other fall-related injuries such as hip fractures) (Nobunaga et al., 1999; Kannus et al., 2000; O’Connor, 2006; Kannus et al., 2007). For example in the province of Ontario in Canada, those 65 years of age and older have a higher rate of SCI than the 20–39 year group, leading to a bimodal age proﬁle for SCI (Pickett et al., 2003; Pickett et al., 2006). The prevalence of comorbidities before SCI is signiﬁcantly higher for elderly versus young persons, and secondary complications after SCI are more frequent among individuals with preexisting medical conditions (Krassioukov et al., 2003). In light of the epidemiologic evidence which demonstrates the high incidence of SCI in elderly persons (Nobunaga et al., 1999; Kannus et al., 2000; Pickett et al., 2003; O’Connor, 2006; Pickett et al., 2006; Kannus et al., 2007), one signiﬁcant limitation of our understanding of these injuries is that the vast majority of experimental studies have been conducted with young animal specimens. To date, only ﬁve studies have investigated the potential inﬂuence of advanced age on SCI by investigating rats greater than 12 months of age (rats have a lifespan of over 2 years). Genovese et al. (2006) found that following spinal cord clip compression, 18-month old rats suffered more severe trauma than 3month old rats characterized by increased edema and neutrophil inﬁltration (based on myeloperoxidase staining), increased immunoreactivity for nitrotyrosine, decreased motor test scores Basso, Beattie, Bresnahan (BBB) Locomotor Rating Scale, and a higher mortality rate. Following spinal hemisection, Gwak et al. (2004) reported that 12-month old rats demonstrated greater locomotor deﬁcits (BBB, paw withdrawal latency) than 40 and 60-day old rats. Chaovipoch et al. (2006) inferred a protracted time course of cell death following complete crush atT8–9 in 12-month compared to 2-month old rats, based on the number of TUNEL positive (apoptotic) cells in the gray matter. Finally, Siegenthaler et al. (2008a; 2008b) reported that following thoracic spinal cord contusion injuries, 12 (aged) and 30-month old (geriatric) rats showed less spontaneous recovery of walking, slower bladder function recovery, greater demyelination and less remyelination than 3-month old rats. Each of these studies demonstrated a signiﬁcant inﬂuence of age. However, to provide more insights into the factors underlying these differences we require enhanced knowledge of the age-related changes in uninjured rats. Although age-related changes in vertebral column morphology may have little inﬂuence on spinal cord tissue damage during injury mechanisms such as clip compression, hemisection, or crush, they could potentially 1327 play a substantial role during more clinically relevant injury mechanisms including contusion (simulating a burst fracture) (Scheff et al., 2003) or fracture-dislocation (caused by relative shearing of two adjacent vertebral bodies) (Choo et al., 2007; 2008). For example, if the vertebral canal is signiﬁcantly larger in older rats, a given vertebral dislocation displacement would cause less relative shearing of the spinal cord than in young animals. Consequently, to conﬁdently apportion the cause of age-related tissue damage following SCI to structural versus physiological pathway differences, characterization of these factors in healthy animals across a range of ages is required. As the vast majority of SCI in older adults occur in the cervical as opposed to thoracic region (Nobunaga et al., 1999; Pickett et al., 2006), there is a clear need for experimental animal models that investigate agerelated differences in cervical SCI. However, very little is known about how advanced age inﬂuences the morphometry of the rat cervical spine. Flynn and Bolton (2007) produced some of the only quantitative data on cervical vertebral canal dimensions in rats. Similar to the trend in Caucasian humans (Tatarek, 2005), they reported that the anterior–posterior (AP) depth of the canal tended to decrease from C2 (avg (SD) ¼ 3.62 (0.31) mm) to T1 (avg (SD) ¼ 2.59 (0.13) mm). In contrast, mediolateral dimensions were smallest at the C2 level (avg (SD) ¼ 4.13 (0.27) mm) and progressively increased at more caudal segments (C7 avg (SD) ¼ 5.17 (0.19) mm). Unfortunately, their experimental protocol did not allow for the calculation of several important variables that inﬂuence spinal cord injury and cervical myelopathy. Speciﬁcally, by separating each vertebra and removing all soft tissues during dissection, variables measured across vertebral segments [e.g., canal pinch diameter (CPD) (Ivancic et al., 2006)] and markers of spine degeneration (including osteophyte development and changes to the intervertebral disc) could not be quantiﬁed. An important unanswered question is whether such morphometric measures would differ as a function of age in adult rats. The objectives of this study were twofold. Our ﬁrst aim was to use a custom-developed micro-computed tomography collection and analysis procedure to determine the inﬂuence of age on the morphometry of the cervical vertebral column in rats. The second aim was to determine whether the degree of cervical intervertebral disc degeneration inferred from radiographs, as assessed by orthopaedic surgeons, was inﬂuenced by rat age. These data will aid in determining whether rats are a suitable model to assess age-related changes in the cervical spine, and will assist in interpreting the results from studies that use a rat model to assess the inﬂuence of age on the response of the spinal cord to mechanical assault. MATERIALS AND METHODS Tissue Specimens Tissue harvest involved anaesthetization with isoﬂurane followed by transcardial perfusion with warm saline and then ice-cold 4% paraformaldehyde with 0.25% glutaraldehyde, at which point the cervical/upper thoracic spinal columns were removed from 15 male Fisher 344 rats. We selected this strain of rat for our aging studies 1328 LAING ET AL. as they weigh less than Sprague-Dawley rats, and have successfully been used as an aging animal model in nerve regeneration studies (Apel et al., 2009; Apel et al., 2010). Five animals were 85 days of age at the time of tissue harvest (classiﬁed as the ‘‘young’’ or 3-month old age group), three were 12-month old and one was 18month old (together forming the ‘‘aged’’ group) and six were 30-month old (termed the ‘‘geriatric’’ group). Age group classiﬁcations were made on the basis of longevity studies that report 50 and 90% mortality ages of 23 and 28 months, respectively, for male Fisher 344 rats dying of natural causes (Thurman et al., 1994), and correspond to age-group names used in previous rat models of age and SCI (Siegenthaler et al., 2008a). The average (SD) mass of the animals at the time of tissue harvest was 261 (24) g for the Young group, 475 (63) g for the Aged group, and 566 (65) for the Geriatric group. The soft tissues surrounding the vertebral column (including muscle and fascia) were removed by gross dissection. The articulations between adjacent vertebrae were conserved with the spinal cord inside, and the specimens were trimmed to include only the ﬁrst cervical (C1) to the second thoracic (T2) vertebrae. We used nylon ﬁshing line and plastic zip ties to securely mount the dorsal side of each vertebral column along the longitudinal aspect of a polyvinyl chloride baseplate (15 40 2 mm). Pilot results (collected using the imaging and measurement procedures outlined below) from a specimen mounted in neutral and 20 extended postures demonstrated minimal inﬂuence of extension on our spine geometry measures (e.g., vertebral canal depth was within 2.0% at C3, 2.2% at C4, 0.8% at C5, 0.9% at C6, and 1.8% at C7; CPD was within 2.3% at C2/C3, 2.2% at C3/C4, 1.4% at C4/C5, 0.1% at C5/C6 and C6/C7, 1.9% at C7/T1). Therefore, specimens were mounted in a neutral conﬁguration that removed any natural curvature of the spine. Specimens were inserted into low density polyethylene sample vials (Nalgene 18-mL sample vials with snap caps, Fisher Scientiﬁc, Nepean, Canada) and stored in a 10% buffered formalin solution. This study received approval from the Animal Care Committee at the University of British Columbia (certiﬁcate #A08-0594). Imaging Procedures A microCT scanner (lCT35, SCANCO Medical AG, Switzerland) was used for imaging purposes. Each specimen was scanned in a formalin solution using an in vitro scanning protocol (70 kVp, 114lA, and 100 ms integration time) resulting in an isotropic 30 lm voxel size. A 30-mm section of the vertebral column was scanned resulting in 1,000 axial (transverse plane) images (Fig. 1). The samples were scanned with the long axis of the vertebral column aligned along the vertical axis of the micro-CT image coordinate system. This minimized any beam-hardening effects by minimizing the bone tissue volume that the x-ray beam passed through. In addition, the image reconstruction algorithm within the analysis software (Scanco Evaluation Program v 6.0, SCANCO Medical AG, Switzerland) provided correction for beamhardening artefact. Following scanning, the software’s evaluation script ‘‘reformat to axial cuts’’ was used to transform the data into sagittal plane images of the specimen (Fig. 2). Image Analysis All geometric variables were measured using the Scanco image analysis software noted above. Our specimens were always positioned in the lCT such that the sagittal plane on axial images was horizontal (Fig. 1). To determine the image that represented the sagittal midline, one axial slice at each of C2, C4, and C6 was evaluated. Speciﬁcally, the row of pixels that best approximated the sagittal midline of the vertebra was recorded for the representative C2, C4, and C6 image slices. We averaged these three values to determine the corresponding sagittal image slice that best approximated the midline for the overall vertebral column. From this midsagittal image, several measurements were made for each vertebra from C2 to T1 (Fig. 2). First, ventral vertebral body height (VBHV; distance from the rostral to the caudal point on the ventral side of the vertebral body) and dorsal vertebral body height (VBHD; distance from the rostral to the caudal point on the dorsal side of the vertebral body) were measured. The dorsal midpoint of the vertebral body (calculated from VBHD) was used in measuring vertebral canal depth (VCD; distance to the lamina along a line perpendicular to the dorsal surface of the vertebral body) and vertebral body depth (VBD; distance to the ventral aspect of the vertebral body along a line perpendicular to the dorsal surface of the vertebral body). CPD was calculated as the most rostral point on the dorsal aspect of a vertebral body to the most caudal aspect of the lamina on the immediately rostral vertebra. Intervertebral disc thickness (IDT) was measured as the distance of a line, originating at the ventral-dorsal midpoint on the rostral aspect of a vertebral body and perpendicular to the rostral surface, to the caudal aspect of the immediately rostral vertebral body (Fig. 2). Finally, vertebral canal area (VCAsag) was measured by tracing the outline of the vertebral canal from the rostral aspect of C3 to the caudal aspect of C7. We calculated additional geometric parameters from the axial images (Fig. 1). The dorsal midpoint of the vertebral body (deﬁned above) informed the selection of speciﬁc axial images from the entire image stack. Vertebral canal width (VCW) at levels C3 to T1 was measured as the widest transverse distance across the vertebral canal. Axial vetebral canal area (VCAax) at levels C3 to T1, and vertebral body area (VBA) from C2 to T1, were measured by manually tracing a line around each structure. Image analysis was performed by a single assessor across all specimens. To test inter-rater reliability, a second assessor independently measured each geometric variable for 9 of the 15 specimens. Analysis software (SPSS Version 17.0, SPSS, Chicago) was used to calculate intraclass correlation coefﬁcients (ICC) using an alpha, two-way random effects model evaluating for absolute agreement. The ICC values for each variable (e.g., vertebral canal depth) were averaged across vertebral levels. The percent differences (and ICC values) between the measures from the two observers averaged 2.7% (0.760) for VBHV, 1.5% (0.895) for VBHD, 1.2% (0.853) for VCD, 2.5% (0.969) for VBD, 1.2% (0.896) for CPD, 6.6% (0.856) for IDT, 1.4% (0.959) for VCAsag, 1.4% (0.845) for VCW, 1.9% (0.901) for VCAax, and 6.7% (0.829) for VBA. EFFECTS OF AGE ON SPINE MORPHOMETRY IN RATS 1329 Fig. 1. Axial microCT images of a young specimen labelled from C2 to T1. Measurement approach for VBA, VCW, and VCAax provided in the C3, C6, and T1 images, respectively. Clinical Assessment of Degeneration Three orthopaedic spine surgeons independently rated the degree of intervertebral disc degeneration in each specimen. Each surgeon was board certiﬁed and engaged in a Spine Fellowship run by the Vancouver Spine Program at the University of British Columbia and Vancouver General Hospital. The surgeons were initially provided with a sample midsagittal image that illustrated the bony anatomy of the cervical spine in a healthy rat. They were then provided with the midsagittal lCT image of each specimen. Based on Kettler and Wilke’s (2006) review of existing grading systems and demonstration of inter-rater reliability (Cote et al., 1997), the surgeons used Kellgren et al.’s (1963) 5-point scale (Table 1) to rate the overall degree of intervertebral disc degeneration (degen) on the anterior (ventral) and posterior (dorsal) aspects of the C2 to T1 vertebral column. The ICCs across raters were 0.741 for ventral degen and 0.443 for dorsal degen, which correspond to substantial and moderate inter-rater reliability, respectively (Landis and Koch, 1977). The three ratings for each specimen and aspect (ventral or dorsal) combination were averaged to provide a single ventral and a sin- gle dorsal numeric outcome on a continuous scale for each specimen. Statistical Analysis All statistical analyses were performed with a software package using an a of 0.05 (SPSS Version 17.0, SPSS, Chicago). We used linear mixed models using a restricted maximum likelihood estimation approach to test the effect of age group (young, aged, and geriatric) and cervical level (C2, C3, etc.) on each of the multisegment dependent variables, and of age group and vertebral body aspect (ventral or dorsal) on ratings of intervertebral disc degeneration. This analysis approach had the advantage over a mixed model analysis of variance that specimens with missing data cells (which occurred for one specimen at the C2 level) were not listwise excluded. The models were run with both homogeneous and heterogeneous autoregressive covariance structures—the ones that provided the lowest Akaike’s Information Criteria (an index of model ﬁt in which goodness-of-ﬁt improves as values decrease) are reported. A one factor (between group) analysis of 1330 LAING ET AL. Fig. 2. Midsagittal microCT image of a young specimen from the second cervical (C2) to second thoracic (T2) vertebra. Measurement approach for ventral and dorsal vertebral body height (VBHV and VBHD) illustrated at the C4 level, CPD at the C4/5 level, and vertebral body depth (VBD) and vertebral canal depth (VCD) at the C6 level. Inset focusing on C5/C6 demonstrates method for measuring IDT. TABLE 1. Kellgren et al. (1963) scale for grading intervertebral disc degeneration in the cervical spine Grade Criteria 0 1 2 3 4 Absence of degeneration in the disc Minimal osteophytosis Minimal osteophytosis; possible narrowing of the disc space; some sclerosis of vertebral plates Moderate narrowing of the disc space; deﬁnite sclerosis of the vertebral plates; osteophytosis Severe narrowing of the disc space; sclerosis of the vertebral plates; multiple large osteophytes variance was used to assess the effect of age on VCAsag. When the interaction effects were ordinal (i.e., the magnitude of the differences between age groups differed across some vertebral levels but in all cases the rank order of young, aged, geriatric remained constant), we interpreted them in concert with main effects (Howell, 2002). As our primary interest was in age-related effects, we used independent t-test pairwise comparisons (with Bonferroni corrections) to identify signiﬁcant differences between groups when the results indicated signiﬁcant main effects for age. For cases of signiﬁcant main effects of vertebral level, we described the general trends observed. The statistical powers associated with post hoc tests for each variable across age groups were calculated with the G*Power 3.1 software application (Faul et al., 2007). Finally, to provide some insight into the inﬂuence of animal mass vs. age [which are known to be positively associated (Thurman et al., 1994)] on vertebral canal 1331 EFFECTS OF AGE ON SPINE MORPHOMETRY IN RATS geometry, we used Pearson correlation analysis to assess the association between mass and age, and between mass and VCD, VCW, and VCAax, at the C5 level, and CPD at the C4/C5 level. Furthermore, we performed stepwise multiple regression analyses to determine the additional variance (beyond age) that mass explained for the C5 vertebral canal measures. RESULTS Tables 2–5 summarize the results for all variables across age groups and vertebral levels, whereas Table 6 indicates the test statistic (F) and probability (P) values from the statistical analyses. For geometric variables, we observed signiﬁcant main effects of age on all outcomes except IDT, in addition to signiﬁcant main effects of vertebral level. There were signiﬁcant interaction effects (age X vertebral level) for all geometric variables except VCAax. However, as the majority of the interac- tions were ordinal in nature we interpreted both main and interaction effects (Howell, 2002). In general, the magnitude of the geometric measures increased with advancing age (Tables 2–5). Post hoc tests (Table 6) indicated that there were signiﬁcant differences between the young and aged specimens, and young and geriatric specimens, on all geometric variables except IDT. Speciﬁcally, compared to the young specimens, the values (excluding IDT) were on average 15.0% (SD ¼ 9.7%) larger for the aged group (ranging from a 9.1% increase for VCW to a 31.3% increase for VCAsag), and 21.5% (SD ¼ 13.4%) larger for the geriatric specimens (ranging from an 11.4% increase for VCW to a 37.5% increase for VCAsag). In contrast, the aged and geriatric groups were signiﬁcantly different from each other for only VBD and VBA with the geriatric specimens an average (SD) of 15.4 (10.7)% and 14.0 (7.6)% larger, respectively. There was less consistency in the changes in geometric measures across vertebral levels. For vertebral canal TABLE 2. Average (SD) values for canal pinch diameter (CPD) and intervertebral disc thickness (IDT) across age groups and vertebral levels. % difference values are relative to the Young group with 1 values representing an increase Cervical vertebral level CPD (mm) Young Aged % difference Geriatric % difference IDT (mm)a Young Aged Geriatric C2/C3 C3/C4 C4/C5 C5/C6 C6/C7 C7/T1 Avg 3.85 (0.03) 4.53 (0.09) 17.7 4.62 (0.21) 20.0 3.26 (0.11) 3.85 (0.17) 18.1 3.95 (0.12) 21.2 3.03 (0.13) 3.56 (0.15) 17.5 3.78 (0.14) 24.8 2.97 (0.05) 3.35 (0.09) 12.8 3.50 (0.14) 17.8 3.13 (0.07) 3.33 (0.08) 6.4 3.50 (0.14) 11.8 3.25 (0.07) 3.56 (0.10) 9.5 3.56 (0.11) 9.5 – – 13.7 – 17.5 0.54 (0.07) 0.44 (0.07) 0.46 (0.05) 0.54 (0.03) 0.47 (0.03) 0.56 (0.16) 0.51 (0.08) 0.42 (0.04) 0.44 (0.06) 0.51 (0.08) 0.44 (0.06) 0.49 (0.05) 0.46 (0.07) 0.41 (0.05) 0.45 (0.03) 0.33 (0.05) 0.33 (0.04) 0.41 (0.04) – – – a % difference not included as no signiﬁcant effect of age observed. TABLE 3. Average (SD) values for vertebral canal geometry variables across age groups and vertebral levels. % difference values are relative to the Young group with 1 values representing an increase Cervical vertebral level C2 C3 VCD (mm) Young 3.84 (0.08) 3.44 (0.03) Aged 4.49 (0.12) 3.94 (0.14) % difference 16.9 14.5 Geriatric 4.55 (0.30) 3.98 (0.10) % difference 18.5 15.7 VCW (mm) Young – 4.25 (0.16) Aged – 4.43 (0.15) % difference – 4.2 Geriatric – 4.52 (0.10) % difference – 6.4 VCAax (cm2) Young – 0.119 (0.005) Aged – 0.145 (0.010) % difference – 21.8 Geriatric – 0.152 (0.006) % difference – 27.7 C4 C5 C6 C7 T1 Avg 3.10 (0.05) 3.50 (0.13) 12.9 3.69 (0.09) 19.0 2.99 (0.03) 3.42 (0.14) 14.4 3.65 (0.19) 22.1 2.93 (0.06) 3.29 (0.12) 12.3 3.45 (0.13) 17.7 2.90 (0.05) 3.23 (0.09) 11.4 3.27 (0.08) 12.8 2.87 (0.04) 3.17 (0.10) 10.5 3.14 (0.09) 9.4 – – 13.3 – 16.5 4.75 (0.10) 5.00 (0.12) 5.3 5.08 (0.11) 6.9 4.74 (0.09) 5.18 (0.19) 9.3 5.26 (0.06) 11.0 5.01 (0.13) 5.48 (0.14) 9.4 5.57 (0.08) 11.2 5.06 (0.14) 5.70 (0.24) 12.6 5.80 (0.15) 14.6 5.03 (0.15) 5.62 (0.21) 11.7 5.73 (0.07) 13.9 – – 9.1 – 11.4 0.126 (0.004) 0.126 (0.002) 0.131 (0.004) 0.130 (00.005) 0.121 (0.005) – 0.148 (0.009) 0.152 (0.010) 0.156 (0.010) 0.156 (0.007) 0.144 (0.011) – 17.5 20.6 19.1 20.0 19.0 19.7 0.157 (0.005) 0.162 (0.005) 0.164 (0.004) 0.161 (0.008) 0.151 (0.002) – 24.6 28.6 25.2 23.8 24.8 25.8 VCD, vertebral canal depth; VCW, vertebral canal width; VCAax, axial vertebral canal area. 1332 LAING ET AL. TABLE 4. Average (SD) values for vertebral body geometry variables across age groups and vertebral levels. % difference values are relative to the Young group with 1 values representing an increase Cervical vertebral level C2 VBHV (mm) Young Aged % difference Geriatric % difference VBHD (mm) Young Aged % difference Geriatric % difference VBA (cm2) Young Aged % difference Geriatric % difference VBD (mm) Young Aged % difference Geriatric % difference C3 C4 C5 C6 C7 T1 Avg 3.01 (0.13) 4.04 (0.46) 34.2 4.21 (0.81) 39.9 3.04 (0.15) 3.19 (0.10) 4.9 3.04 (0.54) 0.0 2.71 (0.11) 2.97 (0.12) 9.6 2.88 (0.24) 6.3 2.52 (0.08) 2.84 (0.15) 12.7 2.94 (0.15) 16.7 2.53 (0.09) 2.83 (0.16) 11.9 2.81 (0.05) 11.1 2.57 (0.11) 2.85 (0.15) 10.9 2.86 (0.09) 11.3 2.75 (0.05) 3.20 (0.13) 16.4 3.20 (0.12) 16.4 – – 14.4 – 14.5 2.95 (0.14) 3.48 (0.30) 18.0 3.58 (0.83) 21.4 3.06 (0.10) 3.72 (0.08) 21.6 3.76 (0.12) 22.9 2.85 (0.11) 3.27 (0.14) 14.7 3.41 (0.06) 19.6 2.65 (0.15) 3.10 (0.12) 17.0 3.17 (0.07) 19.6 2.71 (0.14) 3.16 (0.15) 16.6 3.27 (0.10) 20.7 2.65 (0.07) 3.02 (0.14) 14.0 3.06 (0.08) 15.5 2.71 (0.10) 3.24 (0.09) 19.6 3.27 (0.12) 20.7 – – 17.3 – 20.0 0.035 (0.005) 0.033 (0.003) 0.037 (0.003) 0.035 (0.002) 0.038 (0.003) 0.038 (0.003) 0.039 (0.004) – 0.028 (0.001) 0.036 (0.002) 0.042 (0.003) 0.043 (0.001) 0.043 (0.004) 0.050 (0.006) 0.049 (0.004) – 20.0 9.1 13.5 22.9 13.2 31.6 25.6 13.7 0.030 (0.001) 0.042 (0.007) 0.051 (0.005) 0.043 (0.005) 0.050 (0.005) 0.059 (0.009) 0.058 (0.005) – 14.3 27.3 37.8 22.9 31.6 55.3 48.7 29.9 1.61 (0.16) 1.28 (0.36) 20.5 1.66 (0.19) 3.1 1.46 (0.06) 1.63 (0.21) 11.6 1.95 (0.20) 33.6 1.61 (0.07) 1.97 (0.09) 22.4 2.07 (0.16) 28.6 1.58 (0.05) 1.96 (0.03) 24.1 1.93 (0.30) 22.2 1.46 (0.08) 1.78 (0.03) 21.9 2.03 (0.15) 39.0 1.40 (0.06) 1.77 (0.21) 26.4 2.19 (0.16) 56.4 1.47 (0.14) 1.88 (0.40) 27.9 2.20 (0.14) 49.7 – – 16.3 – 33.2 VBHV, ventral vertebral body height; VBHD, dorsal vertebral body height; VBA, vertebral body area; VBD, vertebral body depth. TABLE 5. Average (SD) values for mid-sagittal vertebral canal area (VCAsag; from the rostral aspect of C3 to the caudal aspect of C7) and intervertebral disc degeneration (degen). % difference values are relative to the Young group with positive values representing an increase degen 2 Young Aged % Difference Geriatric % Difference VCAsag (cm ) ventral dorsal 0.48 (0.01) 0.63 (0.04) 31.3 0.66 (0.02) 37.5 1.33 (0.62) 1.33 (0.81) 0 2.61 (1.02) 96.2 0.80 (0.51) 0.67 (0.47) 16.3 1.39 (0.84) 73.8 measures (Fig. 3), CPD decreased from C2/C3 (average (SD) ¼ 4.34 (0.38) mm) to C5/C6 (3.28 (0.26) mm) before plateauing caudally. VCD consistently decreased from C2 (4.28 (0.39) mm) to T1 (3.06 (0.16) mm), whereas VCW increased from C3 (4.41 (0.17) mm) to T1 (5.47 (0.35) mm). VCAax was largest at C6 (0.151 (0.016) cm2), decreasing to similar values at both C3 (0.139 (0.016) cm2) and T1 (0.139 (0.015) cm2). Regarding vertebral body measures across levels of the vertebral column, VBHV decreased from C3 (average (SD) ¼ 3.08 (0.34) mm) to C6 (2.72 (0.17) mm) before increasing (to 3.05 (0.24) mm) at T1 – a similar trend was observed for VBHD. VBD was smallest at C2 (1.53 (0.28) mm), with local peaks at C4 (1.89 (0.24) mm) and T1 (1.87 (0.39) mm). VBA generally increased from C2 0.031 (0.004) cm2) to T1 (0.049 (0.009) cm2). We observed signiﬁcant interaction effects for all geometric variables except VCAax, most of which were ordinal in nature (Tables 2–6; Fig. 3). Age-related differences were greatest at the rostral segments for VCD, CPD, and VBHV. In contrast, age-related differences tended to be more prevalent at caudal regions for VBD, VCW, and VBA. There was a nonsigniﬁcant trend between degen and age (P ¼ 0.069; Tables 5 and 6), with a tendency toward greater radiographic evidence of degeneration in the geriatric specimens [average (SD) ¼ 2.00 (1.10)] compared to the young [1.07 (0.60)] and aged [1.00 (0.71)] groups. Degen was signiﬁcantly more pronounced (P < 0.001) on the ventral [1.84 (1.02)] compared to dorsal [1.00 (0.70)] aspect of the vertebral bodies. No signiﬁcant interaction effects were observed (P ¼ 0.171). The statistical power (Table 6) of our analyses was lowest when assessing potential differences across the aged and geriatric groups. Averaging across vertebral levels or dorsal/ventral aspects, the mean probability of correctly rejecting the null hypothesis (i.e., no difference across age groups) was above 80% for seven of the eleven dependent variables for the young versus aged comparisons, and eight of the eleven variables for the young versus geriatric comparisons. In contrast, the highest power reached for the aged versus geriatric comparisons was 0.509. Animal age was signiﬁcantly and positively associated with mass (r ¼ 0.905, P < 0.001). The results of overall analyses (including all specimens) demonstrated that mass was signiﬁcantly associated with C4/C5 CPD (P < 0.001), C5 VCD (P < 0.001), C5 VCW (P < 0.001), and 1333 EFFECTS OF AGE ON SPINE MORPHOMETRY IN RATS TABLE 6. Summary across all dependent variables of test statistic (F) and probability (P) values associated with main and interaction effects, and probability (P) and powerb values for posthoc tests comparing across age groups P CPD IDT VBHV VBHD VCD VCW VBA VBD VCAax VCAsag degen Pairwise comparisons (P)a F Power Age Vert level Age X level Age Vert level Age X level Youngaged Younggeriatric Agedgeriatric Youngaged Younggeriatric Agedgeriatric <0.001 0.09 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.069 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 N/A <0.001 0.01 0.038 0.03 0.012 0.006 0.005 0.002 <0.001 0.752 N/A 0.171 88 2.8 14.2 27.8 72.9 41.8 26.7 23.4 59.4 57.3 3.4 151.6 15.7 27.4 66.8 82.2 148.1 15.9 10.9 22.4 N/A 26.7 3.9 2.4 2.4 2.8 3.1 2.9 3.9 4.5 0.7 N/A 2.1 <0.001 0.117 0.002 <0.001 <0.001 <0.001 0.009 0.028 <0.001 <0.001 1 <0.001 1.000 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.149 0.07 0.226 1.000 0.928 0.187 0.505 0.020 0.016 0.105 0.373 0.145 0.984 0.353 0.827 0.978 1.000 0.873 0.755 0.765 0.996 1.000 0.057 1.000 0.293 0.751 0.922 1.000 0.983 0.881 0.830 1.000 1.000 0.421 0.286 0.273 0.079 0.191 0.267 0.184 0.509 0.504 0.443 0.255 0.369 CPD, canal pinch diameter; IDT, Intervertebral disc thickness; VBHV, ventral vertebral body height; VBHD, dorsal vertebral body height; VCD, vertebral canal depth; VCW, vertebral canal width; VBA, vertebral body area; VBD, vertebral body depth; VCAax, axial vertebral canal area, VCAsag, mid-sagittal vertebral canal area; degen, intervertebral disc degeneration. a Bonferonni adjustments incorporated. b Power calculated with Gpower 3.1 (Faul et al., 2007) using a t-test family, ’Means: difference between two independent means’ option, two tailed test, and a ¼ 0.05. The mean value across vertebral levels or ventral-dorsal aspects is presented. TABLE 7. Pearson correlation coefﬁcients (r) and signiﬁcance (p) values for correlation analyses assessing the relationship between animal mass and vertebral canal geometry measures at the ﬁfth cervical vertebrae (C5) Specimens Young Aged Geriatric All ages r p r p r p r p CPD @ C4/C5 VCD @ C5 VCW @ C5 VCAax @ C5 0.536 0.351 0.075 0.925 0.088 0.869 0.868 <0.001 0.654 0.232 0.396 0.604 0.289 0.578 0.81 <0.001 0.719 0.171 0.256 0.744 0.076 0.887 0.881 <0.001 0.557 0.329 0.809 0.191 0.323 0.532 0.731 0.002 CPD, canal pinch diameter; VCD, vertebral canal depth; VCW, vertebral canal width; VCAax, axial vertebral canal area. mal mass provided no substantial improvements over age in the percentage of variance explained in our spinal canal geometry measures. Speciﬁcally, when mass was added as a predictive factor to the reference model that included animal age, the adjusted r2 values decreased slightly for C4/C5 CPD (r2 ¼ 0.809 vs. r2 ¼ 0.817) and C5 VCD (r2 ¼ 0.785 vs. r2 ¼ 0.799) and increased slightly for C5 VCW (r2 ¼ 0.757 vs. r2 ¼ 0.717) and C5 VBAax (r2 ¼ 0.836 vs. r2 ¼ 0.827). Fig. 3. Average (SD) values of: A: CPD; B: vertebral canal depth (VCD); and C: VCW across age groups and vertebral levels. C5 VCAax (P ¼ 0.002) (Table 7). However, there were no signiﬁcant associations between mass and these variables when evaluating each age cohort separately (Table 7). In addition, regression analyses indicated that ani- DISCUSSION The goal of this study was to perform a detailed characterization of the morphology and degenerative state of the cervical spine in Fisher 344 rats, and to determine the inﬂuence of age on these variables. These data are essential for interpreting the results of animal models of SCI that assess the inﬂuence of age on tissue damage severity. Regarding our ﬁrst aim, we observed 1334 LAING ET AL. that 9 of 10 geometric variables measured from the cervical spinal columns of Fisher 344 rats were signiﬁcantly larger for the 12–18-month old (by 15.0% on average) and 30-month old (by 21.5% on average) specimens compared to young adult (3 months) specimens. However, signiﬁcant age-related increases were generally not observed between rats in the aged (12–18 months) and geriatric (30 months) groups. Related to our second aim, radiographic signs of intervertebral disc degeneration were more evident on the ventral compared to the dorsal aspect, with a nonsigniﬁcant trend of more advanced degeneration in the geriatric specimens. In addition to our novel ﬁndings of age-related changes, the cervical canal dimensions from our young specimens correspond to previous results from rats. Our values from Fisher 344 rats are similar to those measured by Flynn & Bolton from Sprague-Dawley rats (2007) with vertebral canal depth being largest at C2 and progressively decreasing caudally to T1 [by 25.2% in this study and 28.5% in Flynn and Bolton (2007)]. Our VCW values are also similar, increasing 17% caudally from C3 to C7. Our results do not correspond as well to those measured from sheep (Wilke et al., 1997) in which vertebral canal depth has been reported to increase (by 17%) from C3 to C7. It is unclear whether these relate to functional differences between species, or are the result of different measurement protocols [microCT in the current study vs. a handheld micrometer in Wilke et al. (1997)]. The intervertebral disc degeneration ﬁndings have important implications for rat models of aging and SCI. Intervertebral disc degeneration is common, occurring in up to 50% of humans above the age of 40 years of age, and in 85% of persons over the age of 60 (Lehto et al., 1994; Matsumoto et al., 1998). It also has implications for SCI, with cervical disc generation often associated with anterior compression of the spinal cord, and with accelerated progression in elderly persons (Okada et al., 2009). Slight to moderate disc degeneration can result in intervertebral instability (Dai, 1998; Miyazaki et al., 2008), which may potentially increase the risk of vertebral dislocations and neurological injuries associated with spinal cord shearing. In contrast, severe intervertebral disc degeneration is associated with increased intervertebral stability (Dai, 1998; Miyazaki et al., 2008) in part the result of osteophytosis at the discovertebral junction. Although they may assist in stabilizing the joint, osteophytic materials on posterior discovertebral junctions can result in cervical spondylosis and local pressure points that contribute to an increased risk of spinal cord compression injuries (Shedid and Benzel, 2007). In this study, based on radiographic signs we observed no change in the degenerative state of the discs between young and aged specimens, and only a trend toward increased degeneration in the geriatric group. This corresponds to Zhang et al. (2009) who, through histological analysis of midsagittal tissue sections, observed no differences in intervertebral disc degenerative in the lumbar spine between 6- and 22-month old SpragueDawley rats. Less expected was our observation that evidence of degeneration was signiﬁcantly greater on the ventral aspect of the discs and vertebral bodies compared with the dorsal aspect. Few osteophytes were observed within the vertebral canal. Consequently, while intervertebral disc degeneration in a rat model may alter the mechanical properties of the cervical spinal column (e.g., stability, stiffness), the lack of bone spurs and discontinuities within the vertebral canal suggest little risk of age-related differences in local pressure points on the spinal cord during transverse or shear loading. Similar to previous ﬁndings (Thurman et al. 1994), we observed a signiﬁcant association between increasing animal age and body mass. However, animal mass provided minimal improvements over animal age in predicting vertebral canal geometry at the ﬁfth cervical vertebrae. Furthermore, animal mass was not predictive of our vertebral canal measures when individual age cohorts were evaluated separately (Table 7). These results could suggest that age-related increases in mass may be more related to changes in skeletal geometry, whereas variability in mass at a given age may be indicative of differences in adiposity. However, the small sample size in these subset analyses is a major limitation. Consequently, additional studies are required to provide a more detailed characterization of the inﬂuence of mass on vertebral column morphometry in rodent models of SCI. Our results generally correspond to the trends in vertebral canal dimensions across levels of the cervical spine measured from humans. Similar to humans, vertebral canal depth in our specimens was largest at C2 and progressively decreased caudally to T1 (Panjabi et al., 1991; Tatarek, 2005). We observed VCW to progressively increase caudally from C2 before plateauing in the C7 to T1 region. This corresponds to increases in human values from C3 to C6, but does not reﬂect the decreases reported between C2-C3, and C6-C7 (Panjabi et al., 1991; Tatarek, 2005). Our trends for VCAax are similar to that of Panjabi et al. (1991), with values decreasing from C2 to C3, then increasing to C6 and C7, before decreasing again at C7 or T1. However, unlike common observances in older humans (Okada et al., 2009), we observed no evidence of cervical spinal stenosis or cervical myelopathy in our aged or geriatric rats. This difference could be a result of the relatively small sample in our geriatric group as, similar to humans, they may only afﬂict a small proportion of the population. Alternatively, such between-species differences may in fact exist as a result of differential loading patterns on the cervical spinal column between bipedal and quadripedal mammals (e.g., greater shear forces in the latter). Further studies are required to investigate such questions from both evolutionary and biomechanical perspectives. These data have implications for studies that use rat models to assess age-related changes in the distribution and severity of spinal cord injuries. First, there were no differences between the aged and geriatric groups for nine of our eleven dependent variables. This suggests that, from a morphological perspective, 12–18-month old rats have similar spinal geometric characteristics compared to rats twice their age, and could be considered for use in spinal column aging studies. Second, these data indicate that there are differences on the order of 15% between geometric measures of young and aged Fisher 344 male rats. Consequently, to produce similar mechanical loads to the spinal cord of young and aged rats during induced SCI, different injury system parameters may need to be employed. For example, as the vertebral canal depth was 13.3% (on average) larger in aged rats, EFFECTS OF AGE ON SPINE MORPHOMETRY IN RATS using a similar contusion depth across age groups may result in less mechanical strain of the spinal cord (and less severe tissue damage) in the aged specimens. These are important issues to be considered when developing experimental protocols for studies assessing the inﬂuence of aging on SCI—further studies are required to provide additional insight in this area. Vertebral fracture-dislocation has been cited as the most common mechanism of cervical SCI in humans (Sekhon and Fehlings, 2001), is associated with ‘‘complete’’ SCI in humans (Pickett et al., 2006), and in rats causes more severe histological indices of damage compared to contusion and ﬂexion-distraction injury mechanisms (Choo et al., 2007; 2008; 2009). Despite this, fracture-dislocation is relatively understudied injury mechanism in experimental SCI research. The geometric variable most relevant to AP cervical dislocation is CPD, and changes in this variable have been assessed during simulated frontal and rear impacts (Ito et al., 2004; Ivancic et al., 2006). Using 3-month old Sprague-Dawley rats, Choo et al. (2009) classiﬁed a 2.6-mm dislocation as moderate-to-severe with a mortality rate of 10%. The current results from Fisher 344 rats indicate that a 2.6mm C4/C5 dislocation would be 85.8% of CPD for 3month olds (the age most commonly used for experimental SCI studies), but only 73.0% and 68.8% of CPD for 12–18- and 30-month olds, respectively. This further indicates the potential need to account for age-related changes in spine geometry in studies that investigate whether advanced age inﬂuences SCI outcomes following fracture-dislocation injury mechanisms. The imaging and analysis procedures used in this study demonstrated high inter-rater reliability with measurements across raters averaging within 1.2% for both vertebral canal depth (ICC ¼ 0.853) and CPD (ICC ¼ 0.896). Therefore, this experimental approach appears sufﬁciently accurate to assist in the development of age-speciﬁc/equivalent models of cervical SCI in rats. There were several limitations associated with this study. First, we only included male specimens in our study. However, we feel justiﬁed in this as Flynn and Bolton (2007) reported no sex-related differences in cervical VCW or depth in Sprague-Dawley rats (even though the males’ masses were on average 31% larger). Second, despite our best efforts, we were unable to remove all of the vertebral column extension in our specimens. The importance of this issue is highlighted by ﬁndings that spine ﬂexion/extension induced by dynamic front and rear impacts can result in a narrowing of CPD in the cervical spine (Ito et al., 2004; Ivancic et al., 2006). However, we feel this will have a limited inﬂuence on our ﬁndings as we used a standard mounting technique across all specimens, and pilot results from a single specimen (see ‘‘Tissue specimens’’ section) indicated that vertebral column extension during our static ﬁxation had minimal inﬂuence on our microCT geometry measures. Third, using the current protocol, we were unable to measure the geometry of the spinal cord relative to the vertebral canal. Determination of whether spinal cord morphometry scales with vertebral canal dimensions will form an important research question toward the development of consistent SCI models across age groups. Finally, our study only assessed the degree of intervertebral disc degeneration based on radiographic images. Although we feel justiﬁed in our conclu- 1335 sions based on evidence of signiﬁcant associations between radiographic and histological indices of disc degeneration (Benneker et al., 2005) and the demonstrated reliability of the modiﬁed Kellgren scale (Cote et al., 1997), histological examination of the discs would provide additional insights into potential age-related changes in these structures (Zhang et al., 2009). This study investigated the inﬂuence of advanced age on cervical spine morphometry in Fisher 344 rats. It provides the ﬁrst reported measures of CPD, VCA, VBA, and intervertebral disc degeneration for this animal model, and conﬁrms previous ﬁndings related to vertebral canal depth and width (Flynn and Bolton, 2007). The results indicate that spinal column measures are 15% larger in 12–18-month old compared with 3-month old rats, with few additional changes observed in 30month old specimens. These data provide important information for the development of valid animal models of aging and SCI. ACKNOWLEDGEMENTS The authors would like to thank Dr. Peter J. Apel for providing the aged and geriatric tissue samples used in this study. The authors would also like to acknowledge the contributions of the orthopaedic surgeons (Dr. Stephen Kingwell, Dr. Christian Dipaola, Dr. Brian Lenehan) who performed the assessments of intervertebral disc degeneration. LITERATURE CITED Apel PJ, Alton T, Northam C, Ma J, Callahan M, Sonntag WE, Li Z. 2009. How age impairs the response of the neuromuscular junction to nerve transection and repair: an experimental study in rats. J Orthop Res 27:385–393. Apel PJ, Ma J, Callahan M, Northam CN, Alton TB, Sonntag WE, Li Z. 2010. Effect of locally delivered IGF-1 on nerve regeneration during aging: an experimental study in rats. Muscle Nerve 41:335–341. Benneker LM, Heini PF, Anderson SE, Alini M, Ito K. 2005. Correlation of radiographic and MRI parameters to morphological and biochemical assessment of intervertebral disc degeneration. Eur Spine J 14:27–35. Chaovipoch P, Jelks KA, Gerhold LM, West EJ, Chongthammakun S, Floyd CL. 2006. 17beta-estradiol is protective in spinal cord injury in post- and pre-menopausal rats. J Neurotrauma 23:830– 852. Choo AM, Liu J, Dvorak M, Tetzlaff W, Oxland TR. 2008. Secondary pathology following contusion, dislocation, and distraction spinal cord injuries. Exp Neurol 212:490–506. Choo AM, Liu J, Lam CK, Dvorak M, Tetzlaff W, Oxland TR. 2007. Contusion, dislocation, and distraction: primary hemorrhage and membrane permeability in distinct mechanisms of spinal cord injury. J Neurosurg Spine 6:255–266. Choo AM, Liu J, Liu Z, Dvorak M, Tetzlaff W, Oxland TR. 2009. Modeling spinal cord contusion, dislocation, and distraction: characterization of vertebral clamps, injury severities, and node of Ranvier deformations. J Neurosci Methods 181:6–17. Cote P, Cassidy JD, Yong-Hing K, Sibley J, Loewy J. 1997. Apophysial joint degeneration, disc degeneration, and sagittal curve of the cervical spine. Can they be measured reliably on radiographs? Spine (Phila Pa 1976) 22:859–864. Dai L. 1998. Disc degeneration and cervical instability. Correlation of magnetic resonance imaging with radiography. Spine (Phila Pa 1976) 23:1734–1738. Faul F, Erdfelder E, Lang AG, Buchner A. 2007. G*Power 3: a ﬂexible statistical power analysis program for the social, 1336 LAING ET AL. behavioral, and biomedical sciences. Behav Res Methods 39: 175–191. Flynn JR, Bolton PS. 2007. Measurement of the vertebral canal dimensions of the neck of the rat with a comparison to the human. Anat Rec (Hoboken) 290:893–899. Genovese T, Mazzon E, Di Paola R, Crisafulli C, Muia C, Bramanti P, Cuzzocrea S. 2006. Increased oxidative-related mechanisms in the spinal cord injury in old rats. Neurosci Lett 393: 141–146. Gwak YS, Hains BC, Johnson KM, Hulsebosch CE. 2004. Effect of age at time of spinal cord injury on behavioral outcomes in rat. J Neurotrauma 21:983–993. Howell DC. 2002. Statistical methods for psychology. 5th ed. CA: Duxbury, Paciﬁc Grove, ISBN 053437770X. Ito S, Panjabi MM, Ivancic PC, Pearson AM. 2004. Spinal canal narrowing during simulated whiplash. Spine (Phila Pa 1976) 29:1330–1339. Ivancic PC, Panjabi MM, Tominaga Y, Pearson AM, Elena Gimenez S, Maak TG. 2006. Spinal canal narrowing during simulated frontal impact. Eur Spine J 15:891–901. Kannus P, Niemi S, Palvanen M, Parkkari J. 2000. Continuously increasing number and incidence of fall-induced, fracture-associated, spinal cord injuries in elderly persons. Arch Intern Med 160:2145–2149. Kannus P, Palvanen M, Niemi S, Parkkari J. 2007. Alarming rise in the number and incidence of fall-induced cervical spine injuries among older adults. J Gerontol A Biol Sci Med Sci 62:180–183. Kellgren J, Jeffrey M, Ball J. 1963. Atlas of stand radiographs. Vol. II. The epidemilogy of chronic rheumatism. Oxford: Blackwell Scientiﬁc. p 14–19. Kettler A, Wilke HJ. 2006. Review of existing grading systems for cervical or lumbar disc and facet joint degeneration. Eur Spine J 15:705–718. Krassioukov AV, Furlan JC, Fehlings MG. 2003. Medical co-morbidities, secondary complications, and mortality in elderly with acute spinal cord injury. J Neurotrauma 20:391–399. Landis JR, Koch GG. 1977. The measurement of observer agreement for categorical data. Biometrics 33:159–174. Lehto IJ, Tertti MO, Komu ME, Paajanen HE, Tuominen J, Kormano MJ. 1994. Age-related MRI changes at 0.1 T in cervical discs in asymptomatic subjects. Neuroradiology 36:49–53. Matsumoto M, Fujimura Y, Suzuki N, Nishi Y, Nakamura M, Yabe Y, Shiga H. 1998. MRI of cervical intervertebral discs in asymptomatic subjects. J Bone Joint Surg Br 80:19–24. Miyazaki M, Hong SW, Yoon SH, Zou J, Tow B, Alanay A, Abitbol JJ, Wang JC. 2008. Kinematic analysis of the relationship between the grade of disc degeneration and motion unit of the cervical spine. Spine (Phila Pa 1976) 33:187–193. Nobunaga AI, Go BK, Karunas RB. 1999. Recent demographic and injury trends in people served by the Model Spinal Cord Injury Care Systems. Arch Phys Med Rehabil 80:1372–1382. NSCISC. 2010. National spinal cord injury statistical center: facts and ﬁgures at a glance. Available at www.nscisc.uab.edu. (accessed online on July 6, 2010). O’Connor PJ. 2006. Trends in spinal cord injury. Accid Anal Prev 38:71–77. Okada E, Matsumoto M, Ichihara D, Chiba K, Toyama Y, Fujiwara H, Momoshima S, Nishiwaki Y, Hashimoto T, Ogawa J, Watanabe M, Takahata T. 2009. Aging of the cervical spine in healthy volunteers: a 10-year longitudinal magnetic resonance imaging study. Spine (Phila Pa 1976) 34:706–712. Panjabi MM, Duranceau J, Goel V, Oxland T, Takata K. 1991. Cervical human vertebrae. Quantitative three-dimensional anatomy of the middle and lower regions. Spine (Phila Pa 1976) 16:861–869. Pickett GE, Campos-Benitez M, Keller JL, Duggal N. 2006. Epidemiology of traumatic spinal cord injury in Canada. Spine 31:799–805. Pickett W, Simpson K, Walker J, Brison RJ. 2003. Traumatic spinal cord injury in Ontario, Canada. J Trauma 55:1070–1076. Priebe MM, Chiodo AE, Scelza WM, Kirshblum SC, Wuermser LA, Ho CH. 2007. Spinal cord injury medicine. 6. Economic and societal issues in spinal cord injury. Arch Phys Med Rehabil 88:S84– S88. Scheff SW, Rabchevsky AG, Fugaccia I, Main JA, Lumpp JE, Jr. 2003. Experimental modeling of spinal cord injury: characterization of a force-deﬁned injury device. J Neurotrauma 20:179–193. Sekhon LH, Fehlings MG. 2001. Epidemiology, demographics, and pathophysiology of acute spinal cord injury. Spine 26:S2–12. Shedid D, Benzel EC. 2007. Cervical spondylosis anatomy: pathophysiology and biomechanics. Neurosurgery 60:S7–S13. Siegenthaler MM, Ammon DL, Keirstead HS. 2008a. Myelin pathogenesis and functional deﬁcits following SCI are age-associated. Exp Neurol 213:363–371. Siegenthaler MM, Berchtold NC, Cotman CW, Keirstead HS. 2008b. Voluntary running attenuates age-related deﬁcits following SCI. Exp Neurol 210:207–216. Tatarek NE. 2005. Variation in the human cervical neural canal. Spine J 5:623–631. Thurman JD, Bucci TJ, Hart RW, Turturro A. 1994. Survival, body weight, and spontaneous neoplasms in ad Libitum-fed and foodrestricted Fischer-344 rats. Toxicol Pathol 22:1–9. Wilke HJ, Kettler A, Wenger KH, Claes LE. 1997. Anatomy of the sheep spine and its comparison to the human spine. Anat Rec 247:542–555. Zhang YG, Sun ZM, Liu JT, Wang SJ, Ren FL, Guo X. 2009. Features of intervertebral disc degeneration in rat’s aging process. J Zhejiang Univ Sci B 10:522–527.