- Email: [email protected]
Effects of postmortem storage by freezing on ligament tensile behavior
MEI-9190861300+ .oO Pcrgamon Journals Ltd
EFFECTS
SAVIO
OF POSTMORTEM STORAGE BY FREEZING ON LIGAMENT TENSILE BEHAVIOR
L.-Y. Woo, CARLO
A. ORLANDO,
JONATHANF. CAMP and WAYNE H. AKESOX
Division of Orthopaedics and Rehabilitation, University of California, San Diego, La Jolla, CA and San Diego Veterans Administration Medical Center. La Jolla. CA, U.S.A. Abstract-The purpose of this study is to examine the effect of prolonged postmortem freezing storage (between 1 l/2 and 3 months at -20°C) on the structural properties of the medial collateral ligament (MCL)-bone complex as well as the mechanical properties of the MCL substance from the rabbit knee. Tensile testing of the femur-MCL-tibia specimen was performed and no statistically significant changes were noted between the fresh and stored samples in terms of the cyclic stress relaxation, the load-deformation characteristics, as well as the load, deformation and energy absorbing capability at failure. The area of hysteresis of the stored samples was significantly reduced in the first few cycles, however. The mechanical properties of the MCL substance, as represented by the stress-strain curves, tensile strength and ultimate strain also did not change following storage. We conclude, therefore, proper and careful storage by freezing would have little or no effect on the biomechanical properties of the ligaments.
to have no consistent changes as compared to those after immediately sacrifice. tested samples Histologically, the stored ligaments disclosed progressive blurring of the collagenous bundles combined with decomposition of cells and ground substances. Thus, these authors suggested that the lack of changes in tensile behavior may be due to the high resistance of collagen fibers to autolytic changes. Subsequently, Viidik and Lewin (1966) tested the ACL-bone complex stored at -20°C for 1 week and found no consistent changes in the load-elongation characteristics as well as the load and elongation at failure, when compared to the fresh samples. The only exception was that the failure energy of the stored samples was slightly higher. Matthews and Ellis (1968). compared the cyclic behavior between fresh and stored cat extensor tendons and found that the overall stiffness did not change significantly in the first 3 h after death. However, there was a decrease in the apparent elastic modulus for the tendons preserved by freezing at - 10°C for 2 weeks. More recently, Noyes and Grood (1976) examined the effect of freezing storage for 4 weeks (at - 15>C)on Rhesus monkeys’ ACL. These authors found no storage induced changes in ligament biomechanical properties or cross-sectional area. Dorlot et 01. (1980) studied the human ACL stored at - 18°C from 5 to 60 days and found a slight increase in stiffness and no change in failure load as compared to those for the fresh samples. Barad et al. (1982) also studied the monkeys’ ACL by testing the right side after overnight storage at 4°C and the left side after storage at - 80°C for 3-5 weeks. The structural and mechanical characteristics of the ligament after deep freezing were found to be slightly inferior (not significantly) to the ligaments stored overnight. It appears that the current literature on the effect of freezing storage has been conflicting as well as incon-
INTRODUCTION As the methodology utilized for the biomechanical evaluation of ligaments and tendons becomes more precise as well as more complex, it is simply not feasible in many instances to perform the test on live animals or immediately after sacrifice. In the case of hunian samples, frozen storage is usually necessary even though samples can be taken without freezing. Therefore, postmortem freezing of samples for storage purposes prior to biomechanical testing has become a common and accepted practice. In the case of human specimens, samples are likely obtained from freshfrozen cadavers. Further, freezing of tendon and ligament allografts from cadaver donors for transplantation has been advocated in recent years (Peacock, 1959; Shino er al., 1984). All of the above are rationales for collection of accurate data on the mechanical and structural behavior of soft tissues before and after postmortem freezing storage. The concept of studying the possible changes in properties of soft tissue following freezing or other postmortem storage procedures is not new. Wertheim (1847) tested dog tendons from one limb immediately after death (fresh), and the contralateral five days later, and found no differences in the modulus of elasticity. Subsequent studies include those of Gratz (1931), Cronkite (1936). Stucke (1950) and Hardy (1951). Smith (1954) found that the rabbit anterior cruciate ligaments (ACL) became less extensible within 1 h after death, and suggested that data on ligaments can be valid only if tests are performed within 30 min after death. Viidik et al. (1965) evaluated the effects on ACL stored at room temperature (18-20°C) from 2 to 96 h postmortem. The mechanical characteristics of the bone-ACL complex of the stored samples were found
Received I7 June 1985; in revisedform 24 September 1985. 399
clusive. In addition, most of the previous studies have not differentiated the mechanical properties of the
400
S.
L.-Y. Woo. C. A. ORLANW J. F. CAMP and W. H. AKESON
ligament substance from the structural properties of the bone-ligament-bone complex. With the availability of new experimental apparatus and methodology, it has become possible to accurately document both of the aforementioned properties simultaneously from a single test specimen (Woo et al., 1981, 1983; WOO, 1982). Therefore, the objective of this study is to examine the effect of prolonged postmortem freezing on the structural properties of the medial colhteral ligament (MCL)-bone complex including the changes in cyclic behavior (stress relaxation, area of hysteresis, etc.) as well as the mechanical properties of the ligament substance. MATERIALS AND MFTHODS
Ten, male New Zealand white rabbits, weighing 3.7 +O.I kg were sacrificed and the hindlimbs were immediately disarticulated at the hip joint, radiographed and randomly assigned to the fresh (control) or postmortem freezing storage (experimental) groups. In the experimental group, each hindhmb was double wrapped in saline soaked gauze and then sealed in a plastic bag and stored airtight at - 20°C for 3 months. In the control group, all extraosseous and periarticular connective tissues around the knee were Pissected, leaving only the MCL intact. The proximal femur and the distal tibia were then cut away leaving bones approximately 5 cm in length from the joint line. The femur and the tibia were then drilled sagittally and perpendicular to the longitudinal direction of the ligament so that a threaded stainless steel pin, 3 mm in diameter and 6cm long, could be inserted through
each of the drilled holes. These pins were then used to secure the femur-MCL-tibia complex onto a set of specially designed clamps at 90” knee Bexion (Fig. 1). The bones were further anchored onto the clamps by means of nylon straps. Minor adjustments and aiignments were then made so that the tensile load would be applied directly along the long axis of the MCL substance. Care was taken throughout the preparation to ensure that the bone-ligament complex was well irrigated with saline to minimize tissue dehydration. After the bone-MCL complex was clamped, the width and thickness of the mid-MCL substance was carefully measured using a vernier caliper at the joint line and 5 mm proximal and distal to it. These measurements were repeated three times, and the repeatability was within 0.1 mm for the thickness measurement and 0.2 mm for the width measurement. The cross-sectional area of the ligament substance was thus determined by assuming the ligament to have a rectangular cross-section. The MCL was then stained perpendicularly (to the length of the MCL) with three parallel, thin dark lines using an elastic (Verhoff) stain. The center line was placed along the joint line directly above the tibia1 plateau and the other two lines were placed 0.5 cm proximal (femoral region) and distal (tibia1 region) to it. The outer two lines were used as gauge length marks for ligament strain determination. The experimental apparatus employed in this study included the use of the video dimensional analyzer (VDA) system to measure the tissue surface strain which was presented in detail in a previous investigation (Woo et al., 1983). Here, only a schematic diagram of the apparatus is shown (Fig. 1). The tensile load and
Video Cassette
Recorder
l!!
Load Cell
Video
Camera
Femu
Tibia
Tensile Load
TV
Strip Chart Recorder
Monitor
VDA
System
Fig. 1. Experimental apparatus used to measure the mechanical and structural properties of ligament and bone-ligament complex.
Effects of
postmortem storage
deformation were obtained from the Instron testing machine, while the tensile strain was determined by the VDA system. The experimental procedure used was as follows: each clamped specimen was first submerged in a 37°C saline bath and mounted as a system on an Instron testing machine. A small preload of 0.5 N was applied. After 30 min. the specimen was then subjected to ten cycles of loading and unloading between deformations of 0 and 1.0 mm at an extension rate of 1 cm min- ‘. Following cyclic testing, the specimens were stretched to failure at a rate of 1cm min-’ (corresponded to 0.4 ‘I0s- ’ for MCL substance) and load-deformation curves for the femur-MCL-tibia complex (structural properties) were obtained. With the cross-sectional area of the MCL substance determined. the mechanical properties (stress-strain curve) of the MCL substance were also obtained. For the experimental (stored) specimens, the limbs were first defrosted overnight at 4°C and thawed at room temperature on the day of testing. The dissection, specimen preparation and testing procedures were identical to those described for the fresh specimens. The mode of failure noted for both fresh and stored specimens (except in one case) were all by tibia1 avulsion. This is a common mode for skelet&y immature (open epiphysis) rabbits (Woo et al., 1985). Thus the data obtained for this group of animals did not include the ultimate tensile properties of the MCL substance, and therefore, an additional group of nine male rabbits with closed epiphysis (body weight of 4.3 +0.2 kg and on average a few months older) were studied. It has been shown that skeletally mature rabbits at this age range would have 67% of MCL substance failure even at slow strain rates (Woo et al., 1985).The older group of rabbits was sacrificed and the hindlimbs were again disarticulated and randomly divided into control and experimental samples. For the experimental sample, the period of storage was 1 l/2 months. The biomechanical testing procedures used for the older group of animals were identical to those described for the younger animals.
401
To compare the data obtained from the cyclic loading and unloading test, it was convenient to divide (normalize) the peak load for each cycle by the peak load at the first cycle. This ‘reduced cyclic stress relaxation’ also helped to minimize the animal-toanimal differences. Expressed in this way, one noticed that there were no significant differences between the fresh and the stored samples although the former stress relaxed to slightly lower values than the latter (Fig. 2). In terms of the area of hysteresis, the values obtained for the stored samples were smaller during the first few cycles and statistically significant differences between the stored and fresh samples were found (Fig. 3). This trend continued throughout the remaining cycles, but the difference diminished and became not statistically significant at the tenth cycle. Prolonged freezing storage appeared to have little effect on the structural characteristics of the MCL-bone complex (Fig. 4a). The load-deformation curves between the fresh and the stored samples were quite similar. In addition, the
stress Fig. 2. Cyclic relaxation behavior of the bone-ligament complex from the younger (open epiphysis) animal group. Data are expressed in mean f S.E.
m
Fresh In = 10)
0
smea
fn= 101
RESULTS
Postmortem storage by freezing caused a slight, but not statistically significant reduction in the crosssectional area of the MCL substance. For the younger animals, the cross-sectional area (mean f S.E.) was 3.9 +O.l mm’ for the control, and 3.7 fO.l mm* for the experimental (p > 0.10). while for the older animals, these values were 3.9 kO.1 mm’ and 3.6+0:2 mm* (p > 0.10). respectively. For the younger animals, nineteen out of twenty samples failed by avulsion at the tibia1 junction. There was one case of mid-substance failure. For the older animals, only five paired samples had mid-substance failures, while the remaining four pairs of limbs had either one or both sides fail by tibia1 avulsion.
J&j& 10thCycle paired l.lest
p
p>o.ro
Fig. 3. Histogram showing the differences in the area of hysteresis values (mean &SE.) between fresh and stored samples from the younger (open epiphysis) animal group.
402
S.
L.-Y. Woo, C. A.
ORLANDO, J.
F. CAMP and W. H.
AKESON
0 srorcd(” = 10)
(a) STRUCTURAL PROPERTIES OF BONE-MCL COMPLEX
ibt
240 -
MECHANICAL PROPERTIES OF MCL SUBSTANCE
60-
zoo+I / I+
160-
failursby
tibmlavulslon
,’ ,’ ,’ ,’ .i ,‘I’
zz 2 120-
a a
3 2 ,m z
s
I 1
2 3 4 Deformation (mm)
I
I
5
6
2
4
I
1
I
I
6
8
10
12
Strain (%)
Fig. 4. Comparison of the structural properties of the MCL-bonecomplex and the mechanical properties of the MCL substance from the fresh and stored young (open epiphysis) rabbit knees. Data are expressed as mean f SE.
failure
properties
also exhibited
no statistical
dif-
ferences between the paired samples in terms of the maximum load, maximum deformation, and the maximum energy absorbed at failure (Fig, 5). The mechanical properties of the ligament substance were again similar between the fresh and stored samples, with no discernable differences in their stress-strain curves (Fig. 4b). The data on the effect of storage on the structural parameters of the MCL-bone complex from the five matched pairs (of older animals with mid-substance failure) are shown in Table 1. In general, the results obtained paralleled those for the younger group of animals. The load-deformation and stress-strain curves were similar between the fresh and stored samples. For the properties of the MCL substance at failure (tensile strength and ultimate strain), there also existed no statistical differences between the fresh and stored samples (Fig. 6). @ [3
Maximum Load
Paired
t.teat
p>o.10
DISCUSSION
In this study, the effect of prolonged freezing storage on the biomechanical properties of bone-MCL-bone complex from the rabbit knee has been evaluated. The structural parameters studied include: cyclic stress relaxation, area of hysteresis, and the load-deformation characteristics (together with the failure properties) of the MCL-bone complex. The mechanical parameters include the stress-strain curves along with the tensile strength and ultimate strain properties of the MCL substance. In terms of the structural properties of the MCL-bone complex, there exist no statistically significant changes following 3 months of frozen storage. These findings are in general agreement with the majority of the reported data (Viidik and Lewin, 1966; Noyes and Grood, 1976; Dorlot et al., 1980, Barad et al., 1982). In recent years, the measurements of stress relaxFresh n = 10) Sfofed (n = 10)
Maximum Deformation
Energy Absorbed at Failure
P>O.S
P>O.so
Fig. 5. Comparison of the structural properties at tensile failure (mean + S.E.) for the bone-MCL-bone complex of the younger (open epiphysis) rabbits.
403
Effects of postmortem storage Table 1. Structural properties of the bone-ligament complex from the older (closed epiphysis) animal group. Note that the p values shown represent those obtained by paired t-test Stored 45 days (n = 5)
Fresh (n = 5) (1) Area of hysteresis First cycle (Nmm)
5.86 f 1.60
2.20 * 0.54 p < 0.05
Tenth cycle (Nmm)
1.36 + 0.50
0.58 + 0.30 p > 0.10
(2) Structural properties (at failure) Pm., (N) 368.4 & 15.0
316.2 + 22.3 p > 0.05
Wk
(mm)
6.6 + 0.6
6.6 f 0.5 p > 0.50
Energy absorbed at failure (N mm)
Tensile Strength 12.0
0 75 % X 50 a D 25
9.0 5 2 6.0 UI 3.0
pawed
P>O.40
p > 0.50
Ultimate Strain
100
1170.0f 200.0
1330.0 f 200.0
p>o40
t.test
Fig. 6. Mechanical properties at tensile failure (mean + S.E.) for the fresh and stored MCLsubstance of rabbits with closed epiphyses.
ation as well as area of hysteresis have been included in our laboratory as a part of the routine biomechanical evaluation of soft tissue behavior (Woo et al., 1981; Woo, 1982). It is felt that the mechanical behaviors of the ligaments and tendons in the subfailure region are of significant importance, as the data obtained more closely represent those in the physiological loading range. In this study, no significant change in the cyclic stress relaxation behavior between the fresh and stored samples is found. However, there are significant decreases in the area of hysteresis during the first few cycles of loading and unloading following prolonged freezing storage. In the subsequent cycles, these differences diminish, although the trend continues. The causes for the alteration in the area of hysteresis for the stored samples during the first few cycles are presently unknown. Presumably, it could be attributed to the fact that storage may have caused decomposition of cells and ground substances and changes in the fluid movements (Stouffer and Butler, 1984) in the ligament, or other changes in the bone-ligament junction. We
have noted that the stored ligaments are slightly less glistening in appearance. Histological H&E stained thin sections from quick frozen samples revealed that the stored ligament had more cell degradation (i.e. took up less stain) than that for the fresh samples. Undoubtedly, studies using other freezing methods and storing samples at different temperature, e.g. - 8O”C, to minimize cell disruption wiil be of interest for further investigations. With the aid of the VDA system, the surface strain of the ligament substance alone can be measured, without the nonuniform strain contributions from other areas of the bone-ligament complex. Further, the ligament strains are determined within the gauge length marks located at the mid-substance, and therefore, o priori knowledge of the original (entire) length of the ligament (which is extremely difficult to determine) is not required. Thus, it is believed that the stress and strain values obtained in this study are quite accurate. With such increased accuracy, it becomes possible to compare the mechanical properties of the ligament substance before and after prolonged postmortem storage by freezing and also lends confidence to the data obtained, i.e. the lack of changes in the stress-strain curves, as well as the ultimate failure properties of the MCL substance secondary to freezing. Based on the results obtained in this study, we can conclude that the biomechanical properties of ligaments and tendons after prolonged freezing storage should not be different from those for the fresh samples. We recommend, however, that the ligament and tendon from experimental animals should be stored intact (with muscles etc.) rather than in their dissected state to minimize the potential water loss. Any drying out of the sample would have significantly affected both the structural and mechanical properties of the soft tissues as well as their bone or muscle attachments. Although it is always best to perform biomechanical tests of ligament and tendons in situ and in live animals (Tipton et al., 1974) or within a few
Jo1
S. L.-Y. Woo, C. A. ORLANDO, J. F. CAMP~II~ W. H. AKESON
minutes after sacrifice (Frank
et al., 1983). the results
obtained in this study indicate that proper and careful
storage would not alter the biomechanical properties (with the exception of the area of hysteresis) of ligaments and tendons. frozen
Arknowledgemenrs-The authors gratefully acknowledge financial support from the RR&D of the Veterans Administration Medical Center, NIH Grant AM34264 and the Malcolm and Dororhy Coutts institute for Joint Reconstruction and Resrarch.
REFERENCES Barad, S.. Cabaud. H. E. and Rodrigo. J. J. (1982) Effects of storage at -8o‘C as compared to 4°C on the strength of rhesus monkey anterior cruciate ligaments. Trans. Orthop. Res. Sot. 7, 378 (Abstract). Cronkite, A. E. (1936) The tensile strength of human tendons. Anat. Rec. 64, 173-186. Dorlot, J. M, Ait ba Sidi, M., Gremblay, G. M. and Drouin, G. (1980) Load-elongation behavior of the canine anterior cruciate ligament. J. biomech. Engng 102, 190-193. Frank, C. B., Woo, S. L.-Y.. Amiel, D., Harwood. F.. Gomez, M. A. and Akeson. W. (1983) Medial collateral.ligament healing. a multidisciplinary assessment in rabbits. Am. J. Sports Med. 11, 379-389. Gratz, C. M. (1931) Tensile strength and elasticity tests dn human fascia lata. J. Bone If. Surg. 13, 334. Hardy, R. H. (1951) Observation on the structure and properties of the plantar calcaneonavicular ligament in man. J. Anal., Lond. 85, 135. Matthews, L. S. and Ellis, D. (1968) Viscoelastic properties of cat tendon: Effects of time after death and preservation by freezing. J. Biomechonics 1. 65-71. Noyes. F. R. and Grood. E. S. (1976) The strength of the anterior cruciate ligament in humans and rhesus monkeys. J. Bone Jt. Surg. 8, 1074-1082.
Peacock, E. E. Jr. (1959) Morphology of homologous and heterologous tendon grafts. Surgery Gynec. Obsret. 109, 735-742. Shine. K.. Kawasaki, T.. Hirose. H.. Gotoh. I., Inoue, H. and Ono. K. (1984) Replacement of the anterior cruciate ligament by an allogeneic tendon graft. J. Bone Jr. Surg. 5. 67268 1. Smith. J. W. (1954) The elastic properties of the anterior cruciate ligament of the rabbit. J. Anar. 88, 369-380. Stouffer, D. C. and Butler, D. L. (1984) An analysis of crimp unfolding, fluid expulsion and fiber failure in collagen fiber bundles. Adoonces in Bioengineering (Edited by Spilker. R. L.), pp. 4&47 (Abstract). American Society of Mechanical Engineers, New York. Stucke, K. (1950) Ober das elastische Verhatten der achillessehne im belastungsveruch. Arch. klin. Chir. 265, 579. Tipton, C. M., Matthews, R. D.,and Sandage, D. S. (1974) In situ measurement ofjunction strength and ligament elongation in rats. J. oppl. Phys. 37, 758-761. Viidik, A., Sanquist, L. and Magi, M. (1965) Influence of postmortal storage on tensile strength characteristics and histology of rabbit ligaments. Acfu orthop. scond. Suppl. 79. Viidik A. and Lewin, T. (1966) Changes in tensile strength characteristics and histology of rabbit ligaments induced by different modes of postmortal storage. Acto orrhop. stand. 37, 141-1.55. Wertheim, M. G. (1847) Memoirs sur I’elasticiteet lacohesion des principaux tissus der corps humain. Annls. Chim Phys. 21, 385-414. Woo, S. L-Y., Gomez, M. A., Akeson, W. H. (1981) The time and history dependent viscoelasticproperties of the canine medial collateral ligaments. J. biomech. Engng 103, 293-298. Woo, S. L.-Y (1982) Mechanical properties of tendons and ligaments. Biorheology 19, 385-396. Woo, S. L-Y., Gomez, M. A., Seguchi, Y., Endo. C. M. and Akeson, W. H. (1983) Measurement of mechanical properties of ligament substance from a bone-ligamentbone preparation. J. orthop. Res. 1, 22-29. Woo, S. L-Y., Orlando, C. A., Frank, C. B., Gomez, M. A. and Akeson, W. H. (1985) Tensile properties of medial collateral ligament as a function of age. Trons. Orthop. Res. Sot. lo,38 (Abstract). Full length manuscript to appear in J. orthop. Res.)
EFFECTS
SAVIO
OF POSTMORTEM STORAGE BY FREEZING ON LIGAMENT TENSILE BEHAVIOR
L.-Y. Woo, CARLO
A. ORLANDO,
JONATHANF. CAMP and WAYNE H. AKESOX
Division of Orthopaedics and Rehabilitation, University of California, San Diego, La Jolla, CA and San Diego Veterans Administration Medical Center. La Jolla. CA, U.S.A. Abstract-The purpose of this study is to examine the effect of prolonged postmortem freezing storage (between 1 l/2 and 3 months at -20°C) on the structural properties of the medial collateral ligament (MCL)-bone complex as well as the mechanical properties of the MCL substance from the rabbit knee. Tensile testing of the femur-MCL-tibia specimen was performed and no statistically significant changes were noted between the fresh and stored samples in terms of the cyclic stress relaxation, the load-deformation characteristics, as well as the load, deformation and energy absorbing capability at failure. The area of hysteresis of the stored samples was significantly reduced in the first few cycles, however. The mechanical properties of the MCL substance, as represented by the stress-strain curves, tensile strength and ultimate strain also did not change following storage. We conclude, therefore, proper and careful storage by freezing would have little or no effect on the biomechanical properties of the ligaments.
to have no consistent changes as compared to those after immediately sacrifice. tested samples Histologically, the stored ligaments disclosed progressive blurring of the collagenous bundles combined with decomposition of cells and ground substances. Thus, these authors suggested that the lack of changes in tensile behavior may be due to the high resistance of collagen fibers to autolytic changes. Subsequently, Viidik and Lewin (1966) tested the ACL-bone complex stored at -20°C for 1 week and found no consistent changes in the load-elongation characteristics as well as the load and elongation at failure, when compared to the fresh samples. The only exception was that the failure energy of the stored samples was slightly higher. Matthews and Ellis (1968). compared the cyclic behavior between fresh and stored cat extensor tendons and found that the overall stiffness did not change significantly in the first 3 h after death. However, there was a decrease in the apparent elastic modulus for the tendons preserved by freezing at - 10°C for 2 weeks. More recently, Noyes and Grood (1976) examined the effect of freezing storage for 4 weeks (at - 15>C)on Rhesus monkeys’ ACL. These authors found no storage induced changes in ligament biomechanical properties or cross-sectional area. Dorlot et 01. (1980) studied the human ACL stored at - 18°C from 5 to 60 days and found a slight increase in stiffness and no change in failure load as compared to those for the fresh samples. Barad et al. (1982) also studied the monkeys’ ACL by testing the right side after overnight storage at 4°C and the left side after storage at - 80°C for 3-5 weeks. The structural and mechanical characteristics of the ligament after deep freezing were found to be slightly inferior (not significantly) to the ligaments stored overnight. It appears that the current literature on the effect of freezing storage has been conflicting as well as incon-
INTRODUCTION As the methodology utilized for the biomechanical evaluation of ligaments and tendons becomes more precise as well as more complex, it is simply not feasible in many instances to perform the test on live animals or immediately after sacrifice. In the case of hunian samples, frozen storage is usually necessary even though samples can be taken without freezing. Therefore, postmortem freezing of samples for storage purposes prior to biomechanical testing has become a common and accepted practice. In the case of human specimens, samples are likely obtained from freshfrozen cadavers. Further, freezing of tendon and ligament allografts from cadaver donors for transplantation has been advocated in recent years (Peacock, 1959; Shino er al., 1984). All of the above are rationales for collection of accurate data on the mechanical and structural behavior of soft tissues before and after postmortem freezing storage. The concept of studying the possible changes in properties of soft tissue following freezing or other postmortem storage procedures is not new. Wertheim (1847) tested dog tendons from one limb immediately after death (fresh), and the contralateral five days later, and found no differences in the modulus of elasticity. Subsequent studies include those of Gratz (1931), Cronkite (1936). Stucke (1950) and Hardy (1951). Smith (1954) found that the rabbit anterior cruciate ligaments (ACL) became less extensible within 1 h after death, and suggested that data on ligaments can be valid only if tests are performed within 30 min after death. Viidik et al. (1965) evaluated the effects on ACL stored at room temperature (18-20°C) from 2 to 96 h postmortem. The mechanical characteristics of the bone-ACL complex of the stored samples were found
Received I7 June 1985; in revisedform 24 September 1985. 399
clusive. In addition, most of the previous studies have not differentiated the mechanical properties of the
400
S.
L.-Y. Woo. C. A. ORLANW J. F. CAMP and W. H. AKESON
ligament substance from the structural properties of the bone-ligament-bone complex. With the availability of new experimental apparatus and methodology, it has become possible to accurately document both of the aforementioned properties simultaneously from a single test specimen (Woo et al., 1981, 1983; WOO, 1982). Therefore, the objective of this study is to examine the effect of prolonged postmortem freezing on the structural properties of the medial colhteral ligament (MCL)-bone complex including the changes in cyclic behavior (stress relaxation, area of hysteresis, etc.) as well as the mechanical properties of the ligament substance. MATERIALS AND MFTHODS
Ten, male New Zealand white rabbits, weighing 3.7 +O.I kg were sacrificed and the hindlimbs were immediately disarticulated at the hip joint, radiographed and randomly assigned to the fresh (control) or postmortem freezing storage (experimental) groups. In the experimental group, each hindhmb was double wrapped in saline soaked gauze and then sealed in a plastic bag and stored airtight at - 20°C for 3 months. In the control group, all extraosseous and periarticular connective tissues around the knee were Pissected, leaving only the MCL intact. The proximal femur and the distal tibia were then cut away leaving bones approximately 5 cm in length from the joint line. The femur and the tibia were then drilled sagittally and perpendicular to the longitudinal direction of the ligament so that a threaded stainless steel pin, 3 mm in diameter and 6cm long, could be inserted through
each of the drilled holes. These pins were then used to secure the femur-MCL-tibia complex onto a set of specially designed clamps at 90” knee Bexion (Fig. 1). The bones were further anchored onto the clamps by means of nylon straps. Minor adjustments and aiignments were then made so that the tensile load would be applied directly along the long axis of the MCL substance. Care was taken throughout the preparation to ensure that the bone-ligament complex was well irrigated with saline to minimize tissue dehydration. After the bone-MCL complex was clamped, the width and thickness of the mid-MCL substance was carefully measured using a vernier caliper at the joint line and 5 mm proximal and distal to it. These measurements were repeated three times, and the repeatability was within 0.1 mm for the thickness measurement and 0.2 mm for the width measurement. The cross-sectional area of the ligament substance was thus determined by assuming the ligament to have a rectangular cross-section. The MCL was then stained perpendicularly (to the length of the MCL) with three parallel, thin dark lines using an elastic (Verhoff) stain. The center line was placed along the joint line directly above the tibia1 plateau and the other two lines were placed 0.5 cm proximal (femoral region) and distal (tibia1 region) to it. The outer two lines were used as gauge length marks for ligament strain determination. The experimental apparatus employed in this study included the use of the video dimensional analyzer (VDA) system to measure the tissue surface strain which was presented in detail in a previous investigation (Woo et al., 1983). Here, only a schematic diagram of the apparatus is shown (Fig. 1). The tensile load and
Video Cassette
Recorder
l!!
Load Cell
Video
Camera
Femu
Tibia
Tensile Load
TV
Strip Chart Recorder
Monitor
VDA
System
Fig. 1. Experimental apparatus used to measure the mechanical and structural properties of ligament and bone-ligament complex.
Effects of
postmortem storage
deformation were obtained from the Instron testing machine, while the tensile strain was determined by the VDA system. The experimental procedure used was as follows: each clamped specimen was first submerged in a 37°C saline bath and mounted as a system on an Instron testing machine. A small preload of 0.5 N was applied. After 30 min. the specimen was then subjected to ten cycles of loading and unloading between deformations of 0 and 1.0 mm at an extension rate of 1 cm min- ‘. Following cyclic testing, the specimens were stretched to failure at a rate of 1cm min-’ (corresponded to 0.4 ‘I0s- ’ for MCL substance) and load-deformation curves for the femur-MCL-tibia complex (structural properties) were obtained. With the cross-sectional area of the MCL substance determined. the mechanical properties (stress-strain curve) of the MCL substance were also obtained. For the experimental (stored) specimens, the limbs were first defrosted overnight at 4°C and thawed at room temperature on the day of testing. The dissection, specimen preparation and testing procedures were identical to those described for the fresh specimens. The mode of failure noted for both fresh and stored specimens (except in one case) were all by tibia1 avulsion. This is a common mode for skelet&y immature (open epiphysis) rabbits (Woo et al., 1985). Thus the data obtained for this group of animals did not include the ultimate tensile properties of the MCL substance, and therefore, an additional group of nine male rabbits with closed epiphysis (body weight of 4.3 +0.2 kg and on average a few months older) were studied. It has been shown that skeletally mature rabbits at this age range would have 67% of MCL substance failure even at slow strain rates (Woo et al., 1985).The older group of rabbits was sacrificed and the hindlimbs were again disarticulated and randomly divided into control and experimental samples. For the experimental sample, the period of storage was 1 l/2 months. The biomechanical testing procedures used for the older group of animals were identical to those described for the younger animals.
401
To compare the data obtained from the cyclic loading and unloading test, it was convenient to divide (normalize) the peak load for each cycle by the peak load at the first cycle. This ‘reduced cyclic stress relaxation’ also helped to minimize the animal-toanimal differences. Expressed in this way, one noticed that there were no significant differences between the fresh and the stored samples although the former stress relaxed to slightly lower values than the latter (Fig. 2). In terms of the area of hysteresis, the values obtained for the stored samples were smaller during the first few cycles and statistically significant differences between the stored and fresh samples were found (Fig. 3). This trend continued throughout the remaining cycles, but the difference diminished and became not statistically significant at the tenth cycle. Prolonged freezing storage appeared to have little effect on the structural characteristics of the MCL-bone complex (Fig. 4a). The load-deformation curves between the fresh and the stored samples were quite similar. In addition, the
stress Fig. 2. Cyclic relaxation behavior of the bone-ligament complex from the younger (open epiphysis) animal group. Data are expressed in mean f S.E.
m
Fresh In = 10)
0
smea
fn= 101
RESULTS
Postmortem storage by freezing caused a slight, but not statistically significant reduction in the crosssectional area of the MCL substance. For the younger animals, the cross-sectional area (mean f S.E.) was 3.9 +O.l mm’ for the control, and 3.7 fO.l mm* for the experimental (p > 0.10). while for the older animals, these values were 3.9 kO.1 mm’ and 3.6+0:2 mm* (p > 0.10). respectively. For the younger animals, nineteen out of twenty samples failed by avulsion at the tibia1 junction. There was one case of mid-substance failure. For the older animals, only five paired samples had mid-substance failures, while the remaining four pairs of limbs had either one or both sides fail by tibia1 avulsion.
J&j& 10thCycle paired l.lest
p
p>o.ro
Fig. 3. Histogram showing the differences in the area of hysteresis values (mean &SE.) between fresh and stored samples from the younger (open epiphysis) animal group.
402
S.
L.-Y. Woo, C. A.
ORLANDO, J.
F. CAMP and W. H.
AKESON
0 srorcd(” = 10)
(a) STRUCTURAL PROPERTIES OF BONE-MCL COMPLEX
ibt
240 -
MECHANICAL PROPERTIES OF MCL SUBSTANCE
60-
zoo+I / I+
160-
failursby
tibmlavulslon
,’ ,’ ,’ ,’ .i ,‘I’
zz 2 120-
a a
3 2 ,m z
s
I 1
2 3 4 Deformation (mm)
I
I
5
6
2
4
I
1
I
I
6
8
10
12
Strain (%)
Fig. 4. Comparison of the structural properties of the MCL-bonecomplex and the mechanical properties of the MCL substance from the fresh and stored young (open epiphysis) rabbit knees. Data are expressed as mean f SE.
failure
properties
also exhibited
no statistical
dif-
ferences between the paired samples in terms of the maximum load, maximum deformation, and the maximum energy absorbed at failure (Fig, 5). The mechanical properties of the ligament substance were again similar between the fresh and stored samples, with no discernable differences in their stress-strain curves (Fig. 4b). The data on the effect of storage on the structural parameters of the MCL-bone complex from the five matched pairs (of older animals with mid-substance failure) are shown in Table 1. In general, the results obtained paralleled those for the younger group of animals. The load-deformation and stress-strain curves were similar between the fresh and stored samples. For the properties of the MCL substance at failure (tensile strength and ultimate strain), there also existed no statistical differences between the fresh and stored samples (Fig. 6). @ [3
Maximum Load
Paired
t.teat
p>o.10
DISCUSSION
In this study, the effect of prolonged freezing storage on the biomechanical properties of bone-MCL-bone complex from the rabbit knee has been evaluated. The structural parameters studied include: cyclic stress relaxation, area of hysteresis, and the load-deformation characteristics (together with the failure properties) of the MCL-bone complex. The mechanical parameters include the stress-strain curves along with the tensile strength and ultimate strain properties of the MCL substance. In terms of the structural properties of the MCL-bone complex, there exist no statistically significant changes following 3 months of frozen storage. These findings are in general agreement with the majority of the reported data (Viidik and Lewin, 1966; Noyes and Grood, 1976; Dorlot et al., 1980, Barad et al., 1982). In recent years, the measurements of stress relaxFresh n = 10) Sfofed (n = 10)
Maximum Deformation
Energy Absorbed at Failure
P>O.S
P>O.so
Fig. 5. Comparison of the structural properties at tensile failure (mean + S.E.) for the bone-MCL-bone complex of the younger (open epiphysis) rabbits.
403
Effects of postmortem storage Table 1. Structural properties of the bone-ligament complex from the older (closed epiphysis) animal group. Note that the p values shown represent those obtained by paired t-test Stored 45 days (n = 5)
Fresh (n = 5) (1) Area of hysteresis First cycle (Nmm)
5.86 f 1.60
2.20 * 0.54 p < 0.05
Tenth cycle (Nmm)
1.36 + 0.50
0.58 + 0.30 p > 0.10
(2) Structural properties (at failure) Pm., (N) 368.4 & 15.0
316.2 + 22.3 p > 0.05
Wk
(mm)
6.6 + 0.6
6.6 f 0.5 p > 0.50
Energy absorbed at failure (N mm)
Tensile Strength 12.0
0 75 % X 50 a D 25
9.0 5 2 6.0 UI 3.0
pawed
P>O.40
p > 0.50
Ultimate Strain
100
1170.0f 200.0
1330.0 f 200.0
p>o40
t.test
Fig. 6. Mechanical properties at tensile failure (mean + S.E.) for the fresh and stored MCLsubstance of rabbits with closed epiphyses.
ation as well as area of hysteresis have been included in our laboratory as a part of the routine biomechanical evaluation of soft tissue behavior (Woo et al., 1981; Woo, 1982). It is felt that the mechanical behaviors of the ligaments and tendons in the subfailure region are of significant importance, as the data obtained more closely represent those in the physiological loading range. In this study, no significant change in the cyclic stress relaxation behavior between the fresh and stored samples is found. However, there are significant decreases in the area of hysteresis during the first few cycles of loading and unloading following prolonged freezing storage. In the subsequent cycles, these differences diminish, although the trend continues. The causes for the alteration in the area of hysteresis for the stored samples during the first few cycles are presently unknown. Presumably, it could be attributed to the fact that storage may have caused decomposition of cells and ground substances and changes in the fluid movements (Stouffer and Butler, 1984) in the ligament, or other changes in the bone-ligament junction. We
have noted that the stored ligaments are slightly less glistening in appearance. Histological H&E stained thin sections from quick frozen samples revealed that the stored ligament had more cell degradation (i.e. took up less stain) than that for the fresh samples. Undoubtedly, studies using other freezing methods and storing samples at different temperature, e.g. - 8O”C, to minimize cell disruption wiil be of interest for further investigations. With the aid of the VDA system, the surface strain of the ligament substance alone can be measured, without the nonuniform strain contributions from other areas of the bone-ligament complex. Further, the ligament strains are determined within the gauge length marks located at the mid-substance, and therefore, o priori knowledge of the original (entire) length of the ligament (which is extremely difficult to determine) is not required. Thus, it is believed that the stress and strain values obtained in this study are quite accurate. With such increased accuracy, it becomes possible to compare the mechanical properties of the ligament substance before and after prolonged postmortem storage by freezing and also lends confidence to the data obtained, i.e. the lack of changes in the stress-strain curves, as well as the ultimate failure properties of the MCL substance secondary to freezing. Based on the results obtained in this study, we can conclude that the biomechanical properties of ligaments and tendons after prolonged freezing storage should not be different from those for the fresh samples. We recommend, however, that the ligament and tendon from experimental animals should be stored intact (with muscles etc.) rather than in their dissected state to minimize the potential water loss. Any drying out of the sample would have significantly affected both the structural and mechanical properties of the soft tissues as well as their bone or muscle attachments. Although it is always best to perform biomechanical tests of ligament and tendons in situ and in live animals (Tipton et al., 1974) or within a few
Jo1
S. L.-Y. Woo, C. A. ORLANDO, J. F. CAMP~II~ W. H. AKESON
minutes after sacrifice (Frank
et al., 1983). the results
obtained in this study indicate that proper and careful
storage would not alter the biomechanical properties (with the exception of the area of hysteresis) of ligaments and tendons. frozen
Arknowledgemenrs-The authors gratefully acknowledge financial support from the RR&D of the Veterans Administration Medical Center, NIH Grant AM34264 and the Malcolm and Dororhy Coutts institute for Joint Reconstruction and Resrarch.
REFERENCES Barad, S.. Cabaud. H. E. and Rodrigo. J. J. (1982) Effects of storage at -8o‘C as compared to 4°C on the strength of rhesus monkey anterior cruciate ligaments. Trans. Orthop. Res. Sot. 7, 378 (Abstract). Cronkite, A. E. (1936) The tensile strength of human tendons. Anat. Rec. 64, 173-186. Dorlot, J. M, Ait ba Sidi, M., Gremblay, G. M. and Drouin, G. (1980) Load-elongation behavior of the canine anterior cruciate ligament. J. biomech. Engng 102, 190-193. Frank, C. B., Woo, S. L.-Y.. Amiel, D., Harwood. F.. Gomez, M. A. and Akeson. W. (1983) Medial collateral.ligament healing. a multidisciplinary assessment in rabbits. Am. J. Sports Med. 11, 379-389. Gratz, C. M. (1931) Tensile strength and elasticity tests dn human fascia lata. J. Bone If. Surg. 13, 334. Hardy, R. H. (1951) Observation on the structure and properties of the plantar calcaneonavicular ligament in man. J. Anal., Lond. 85, 135. Matthews, L. S. and Ellis, D. (1968) Viscoelastic properties of cat tendon: Effects of time after death and preservation by freezing. J. Biomechonics 1. 65-71. Noyes. F. R. and Grood. E. S. (1976) The strength of the anterior cruciate ligament in humans and rhesus monkeys. J. Bone Jt. Surg. 8, 1074-1082.
Peacock, E. E. Jr. (1959) Morphology of homologous and heterologous tendon grafts. Surgery Gynec. Obsret. 109, 735-742. Shine. K.. Kawasaki, T.. Hirose. H.. Gotoh. I., Inoue, H. and Ono. K. (1984) Replacement of the anterior cruciate ligament by an allogeneic tendon graft. J. Bone Jr. Surg. 5. 67268 1. Smith. J. W. (1954) The elastic properties of the anterior cruciate ligament of the rabbit. J. Anar. 88, 369-380. Stouffer, D. C. and Butler, D. L. (1984) An analysis of crimp unfolding, fluid expulsion and fiber failure in collagen fiber bundles. Adoonces in Bioengineering (Edited by Spilker. R. L.), pp. 4&47 (Abstract). American Society of Mechanical Engineers, New York. Stucke, K. (1950) Ober das elastische Verhatten der achillessehne im belastungsveruch. Arch. klin. Chir. 265, 579. Tipton, C. M., Matthews, R. D.,and Sandage, D. S. (1974) In situ measurement ofjunction strength and ligament elongation in rats. J. oppl. Phys. 37, 758-761. Viidik, A., Sanquist, L. and Magi, M. (1965) Influence of postmortal storage on tensile strength characteristics and histology of rabbit ligaments. Acfu orthop. scond. Suppl. 79. Viidik A. and Lewin, T. (1966) Changes in tensile strength characteristics and histology of rabbit ligaments induced by different modes of postmortal storage. Acto orrhop. stand. 37, 141-1.55. Wertheim, M. G. (1847) Memoirs sur I’elasticiteet lacohesion des principaux tissus der corps humain. Annls. Chim Phys. 21, 385-414. Woo, S. L-Y., Gomez, M. A., Akeson, W. H. (1981) The time and history dependent viscoelasticproperties of the canine medial collateral ligaments. J. biomech. Engng 103, 293-298. Woo, S. L.-Y (1982) Mechanical properties of tendons and ligaments. Biorheology 19, 385-396. Woo, S. L-Y., Gomez, M. A., Seguchi, Y., Endo. C. M. and Akeson, W. H. (1983) Measurement of mechanical properties of ligament substance from a bone-ligamentbone preparation. J. orthop. Res. 1, 22-29. Woo, S. L-Y., Orlando, C. A., Frank, C. B., Gomez, M. A. and Akeson, W. H. (1985) Tensile properties of medial collateral ligament as a function of age. Trons. Orthop. Res. Sot. lo,38 (Abstract). Full length manuscript to appear in J. orthop. Res.)