Anterior cruciate
ligament reconstruction using bone-patellar tendon-bone allografts
A
and biomechanical evaluation in goats
biological
DAVID J. DREZ, Jr,* MD, JESSE DeLEE,† MD, JOHN P. HOLDEN,‡ MS, STEVEN ARNOCZKY,§ DVM, FRANK R. NOYES,‡ MD, AND THOMAS S. ROBERTS,*∥ MD *
University Knee and Sportsmedicine Center, From theTheLouisiana State of Texas Health Sciences Center at San
Louisiana, t
Lake Charles, Antonio, San Antonio,
University
Texas, &Dag er; The Noyes-Giannestras Biomechanics Laboratory, University of Cincinnati, Cincinnati, Ohio, and § the Hospital for Special Surgery, New York, New York ABSTRACT
Recently, allografts have been proposed for use in ACL reconstruction. The use of an allograft would reduce the operative time and the morbidity associated with the use of autogenous tissues for ACL reconstructions. In 1985, Curtis et al.8 evaluated ACL reconstructions in dogs using freezedried fascia lata allografts. The results were promising. Because of the results of this previous study and the fact that many ACL reconstructions are done with a bonepatellar tendon-bone complex, we felt that an evaluation of bone-patellar tendon-bone allografts in an animal model was indicated. The purpose of this study was to determine the biological and biomechanical fate of freeze-dried, ethyl-
Twenty-eight goats underwent ACL reconstruction with freeze-dried bone-patellar tendon-bone allografts in one knee, the opposite knee serving as a control. One group of 16 knees was evaluated, in groups of four, at 6, 12, 26, and 52 weeks by histologic and vascular injection techniques. The other group of 12 knees was evaluated in two groups of six at 26 and 52 weeks by morphological and biomechanical techniques of analysis. Within the first 12 weeks these allografts were revascularized ; in the first 26 weeks they had matured to resemble normal connective tissue. Graft stiffness was 29% of the control value and maximum force to failure was 43% of the control value. The results of this study indicated that freeze-dried bone-patellar tendon-bone allografts are biomechani-
cally and biologically grafts.
similar to
patellar
ene
oxide-sterilized, bone-patellar tendon-bone allografts a goat model.
used for ACL reconstruction in
MATERIALS AND METHODS
tendon auto-
We used 28 Angora goats between 2 and 3 years of age, with a weight of approximately 22 to 32 kg each. Goats were chosen because the biomechanical properties of the goat ACL and the human ACL are similar.&dquo; We divided the animals into two groups, one for histological and vascular analysis (16 animals) and one for biomechanical and morphological analysis (12 animals). All of the goats had the ACL of one hindlimb reconstructed with a freeze-dried bone-patellar tendon-bone allograft. The opposite hindlimb served as a control. We sacrificed groups of four animals at 6, 12, 26, and 52 weeks for histologic analysis. For morphological and biomechanical analysis, we sacrificed two groups of six goats at 26 and 52 weeks.
Using a portion of the patellar tendon for ACL reconstruction has been advocated by many authors. 2,3,9.11,15,16 Independent studies by Alm et all and Arnoczky et al.~ established that free autogenous patellar tendon grafts were rapidly revascularized, underwent cellular repopulation and neo-organization of the collagen fibrils, and at 6 months postimplantation, the grafts histologically resembled normal ligamentous tissue. 11 Address correspondence and repnnt requests to Thomas S. Roberts, MD, 2500 North State Street, Jackson, MS 39216-4505. 256
257
Surgical procedure After anesthetizing the animals with intramuscular ketamine hydrochloride 125 to 150 mg and xylazine 25 mg, we checked the knees for range of motion and stability. One knee of each animal was then shaved, prepared, and draped in a sterile fashion. Our surgical approach was a lateral parapatellar incision extending from the distal one-third of the thigh to just below the tibial tubercle. The fascia lata and lateral capsule were opened in line with the skin incision, and the patella was subluxated medially, exposing the knee joint. At this point, we completely resected the ACL. On manual examination, all of the knees demonstrated gross anterior subluxation of the tibia after ACL excision. A drill hole was made from the medial aspect of the proximal tibia into the knee just anterior and medial to the anatomical insertion of the ACL on the proximal tibia. This was then overdrilled with a 6 mm drill bit. We also prepared a groove in the lateral femoral condyle so that placement of the graft would be in as near an anatomical position as possible. We obtained freeze-dried bone-patellar tendon-bone allografts from animals used in a previous study. The allografts, prepared by the Bone Bank Foundation of San Antonio, Texas, averaged 6 mm in width and were prepared from both the medial and lateral portions of the patellar tendon complex. The allografts were frozen at -70°C, freezedried, and then sterilized with ethylene oxide. Following this, the allografts were purged for 6 hours. Before implantation, the grafts were reconstituted with 250 mm of normal saline for 30 to 60 minutes and cooled to 5°C until ready for use.
Using a rasp, we smoothed the tibial tunnel and femoral margins and then pulled the graft into the tunnel and positioned it in the groove. The graft was secured with a small barbed Richard’s staple placed over the bone plug in a groove prepared at the distal end of the tibial tunnel. We reduced the anteriorly subluxated knee and pulled the graft taut over the lateral femoral condyle in order to eliminate the abnormal laxity. The graft was oriented in the femoral groove so that the cancellous bone of the graft was in direct contact with the cancellous bone of the &dquo;groove&dquo; in the lateral femoral condyle. Another staple secured the bone plug to the lateral surface of the femur. We rechecked the knee for stability to be certain that the anterior subluxation had been obliterated. After irrigation with normal saline, we closed the synovium and capsule with running 3-0 nylon sutures, the subcutaneous tissue with running 2-0 vicryl sutures, and the skin with interrupted 30 nylon sutures. The goats were given 1 g of cefazolin sodium perioperatively. Postoperatively, they recovered from anesgroove
thesia in a controlled area without immobilization for the limb. Afterward, we allowed them to roam freely in a fenced area
measuring approximately
30 X 30 meters.
Laboratory techniques At the conclusion of each experimental period, we sacrificed four animals via barbiturate overdose. From two of the
animals, the allograft was removed en bloc, fixed in 10% buffered formalin, and embedded in paraffin. Five micrometer-thick sections were cut, stained with hematoxylin and eosin, and examined by light and polarized light microscopy. Using a modified Spalteholz technique,’ we evaluated the vascularity of the graft in the remaining two animals.
Biomechanical
testing
Anteroposterior force displacement test of the whole knee joint. Anteroposterior (AP) force displacement measurements were made at 45° of knee flexion before the joints were dissected. Limbs were thawed overnight at room temperature. We removed all soft tissue from the tibia and femur to within 7 cm of the joint line and mounted the limb in an instrumented measurement device. The femur was held in grips that allowed adjustment of femoral position with 6 df while the tibia hung vertically. The load application mechanism restricted external tibial rotations during the test, while allowing vertical pistoning of the tibia. A load cell monitored the applied AP forces, and a linear variable differential transformer (LVDT) measured AP displacement at the level of the tibial tubercle. We determined total AP translation between AP forces of 30 N at the joint line. The translation in the low stiffness region (primary laxity) was determined by using the method of Markoff et al. 17 The anterior translation in the high stiffness region (secondary anterior laxity) was the additional translation, from the end of the primary laxity region, required to reach 30 N of anterior force. We calculated anterior joint stiffness from a tangent to the loading curve at 30 N of anterior force. Rating of articular cartilage. Specimens were dissected to prepare femur-graft-tibia units for tensile failure testing. During this process, we photographed all joints and recorded gross joint degeneration, including the size, depth, and location of any cartilage lesions, meniscal tears, and osteophyte formations. We graded cartilage changes as follows: 0, no observable gross changes; 1, intact surface color changes and/or surface irregularities; 2, surface fibrillation (disruption) alone or with loss of cartilage but with no bone exposed; 3, exposed bone loss less than 10% of the surface in the given region; and 4, greater than 10% of the surface area having exposed bone and fragmentation of the cartilage around the lesion. We rated eight regions: each tibial condyle, the anterior and posterior halves of each femoral condyle, the trochlear groove, and the articular surface of the patella. Because of the nature of the data (rating scores) and the relatively small sample size, we used the Wilcoxon signedrank test, a nonparametric method, to test for differences between control and operated knees. Paired difference scores (operated versus nonoperated) were first compared between the two groups of animals sacrificed at 6 and 12 months. Because we found no statistically significant differences in the ratings, we used the data from all 12 animals collectively
258
significant differences between operated and nonoperated knees at each of the eight locations.
to test for
Specimen geometry. We removed the medial femoral condyle to provide better visualization of, and access to, the ligament. Using vernier calipers, we measured graft and control ACL lengths with control lengths computed as the mean length of fibers on the anterior, posterior, medial, and lateral sides of the ligament. Tissue cross-sectional areas were measured using an area micrometer with a blade pressure of 0.12 mPa applied for 2 minutes along the midportion of the graft. The blade pressure was reproducible. Tensile failure tests. The tibia and femur were then mounted in grips at a flexion angle of 30° and the intraarticular portion of the graft was visually aligned in a vertical orientation under the load cell axis. We conducted failure tests at room temperature by elongating the specimen at a rate of 100% per second of the average measured length of the ligament of graft. An LVDT on the actuator monitored actuator travel and a clip gauge attached directly to the tibia
ripheral portion of the graft (Fig. 1), whereas the central portion of the graft remained avascular and acellular (Fig. 2). The intraarticular portion of the graft was surrounded by a vascular synovial envelope and demonstrated evidence of intrinsic revascularization. The intraosseous portion of the graft was relatively avascular at this time (Fig. 3). Although there was no histologic evidence of an inflammatory or rejection response in the intraarticular portion of the allograft, the intraosseous portion of one graft demonstrated patches of lymphocytic infiltration suggestive of an immune response (Fig. 4). Interestingly, this was the only specimen to exhibit this response. At 3 months, the allograft appeared grossly the same. However, as in the 6 week specimens, the degree of arthro-
fibrosis and degenerative joint disease varied. The findings similar to those in the 6 week specimens. The revas-
were
and femur provided the relative displacement of the bone ends. A lab minicomputer was used to initiate the test and to measure and store the force, actuator motion, and clip gauge data. We determined failure mechanisms from both direct posttest examination of the failed tissues and review of 16 mm cinefilms taken at 200 frames per second. The clip gauge and force measurements were used to determine the following structural properties: ligament stiffness in the linear loading region, maximum force before failure, and energy to maximum force. The clip gauge was required primarily for control specimen tests in which the large forces caused bone deflection despite the use of support brackets intended to prevent deflection of the tibia and femur.
RESULTS Clinical analysis of the goats showed that the animals walked without a limp by the 6th postoperative week. Five animals developed an early effusion, but none showed any other evidence of an infection. No positive cultures were obtained and the effusions were felt to be secondary to inflammation. The effusions all resolved before sacrifice. One goat died from pulmonary complications subsequent to the ACL reconstruction and was not analyzed. That goat was
Figure 1. Photomicrograph of the peripheral intraarticular portion of an allograft at 6 weeks postimplantation. Notice the vascular channels (arrow) that have already formed, as well as the early cellular repopulation.
replaced.
ACL grafts
at the time of sacrifice. felt to have a positive anterior drawer sign or Lachman on physical examination. No other form of ligamentous laxity was demonstrated. No roentgenwere
None of the knees
grossly intact were
ographic analysis was performed. Histology and vascularity Six weeks after transplantation, the knees demonstrated various degrees of arthrofibrosis and degenerative joint disease. Small cartilaginous lesions were present, but no osteophytes were noted. Histologic evaluation of the allograft revealed the presence of vessels and fibroblasts in the pe-
Figure 2. The central portion of this allograft at 6 weeks demonstrates the necrosis and lack of vascularity in this portion of the graft.
259
Figure 3. This 6 week specimen prepared by the Spalteholz technique demonstrates a lack of vascularity in the tibial tunnel (arrow), but the periphery of the intraarticular portion of the graft is well vascularized. The PCL and fat been removed to facilitate observation of the graft.
pad
have
cularization process had spread throughout the allograft with the majority of the graft being revascularized (Fig. 5). Histologic sections revealed an increased cell population and a more organized appearance to the tissue (Fig. 6). By 6 months, the graft was completely revascularized and the tissue had the appearance of normal, dense, regularly oriented connective tissue (Fig. 7). The 12 month specimens were similar in both gross and histologic appearance to the 6 month specimens.
Rating of articular cartilage and
biomechanical tests
Rating of articular cartilage. Table 1 shows the results of the analysis of articular cartilage changes in the eight regions of the knee. The mean values and the distribution of ratings are shown for operated and nonoperated knees separately. Statistical differences between operated and nonoperated knees are based on paired values within animals. Operated knees had significantly higher grades (i.e., more degeneration) than nonoperated knees at seven of the eight locations. We noted a large number of Grade 4 ratings (more than
4. This 6 week allograft demonstrates patches of a cellular infiltrate that is indicative of a reactive response. The cellular infiltrate was present only in the tibial tunnel area of the allograft. This is the only portion of any of the specimens that demonstrated any type of reactive pattern.
Figure
50% of cartilage eroded to bone) in the patellofemoral joint of the operated knees. In addition to these ratings of the articular cartilage, meniscal irregularities were noted in eight knees and osteophyte formation in seven knees. AP force displacement tests. Table 2 shows results of measurements from the AP force displacement tests. The control knees demonstrated a total AP translation of 1.1 mm, with only 0.2 mm of translation in the region of primary laxity. These results show the near absence of slackness in the anterior cruciate and posterior cruciate ligaments. There were no significant differences in total AP translation, primary and secondary anterior translation, and anterior stiffness between the 6 and 12 month allograft groups. Reconstructed knees of both groups had statistically greater total AP translation than controls. As a reference, tests conducted by Holden et al. 12 of three goat knees with the ACL resected showed total AP translation of 8.9 ± 0.4 mm. The increase in total translation in the reconstructed knees was due mostly to a significant increase in primary anterior translation, which averaged 2.8 mm. In contrast, the secondary anterior translation averaged 0.6 mm. Although this
260
z
I1
...,#,&dquo;,..~Y&dquo;pil,
1’ t i
,
s ,
I
I I
44
I
1
k
I B
.,
°:w4
G
Y‘~.
I
I
ligament reconstruction using bone-patellar tendon-bone allografts
A
and biomechanical evaluation in goats
biological
DAVID J. DREZ, Jr,* MD, JESSE DeLEE,† MD, JOHN P. HOLDEN,‡ MS, STEVEN ARNOCZKY,§ DVM, FRANK R. NOYES,‡ MD, AND THOMAS S. ROBERTS,*∥ MD *
University Knee and Sportsmedicine Center, From theTheLouisiana State of Texas Health Sciences Center at San
Louisiana, t
Lake Charles, Antonio, San Antonio,
University
Texas, &Dag er; The Noyes-Giannestras Biomechanics Laboratory, University of Cincinnati, Cincinnati, Ohio, and § the Hospital for Special Surgery, New York, New York ABSTRACT
Recently, allografts have been proposed for use in ACL reconstruction. The use of an allograft would reduce the operative time and the morbidity associated with the use of autogenous tissues for ACL reconstructions. In 1985, Curtis et al.8 evaluated ACL reconstructions in dogs using freezedried fascia lata allografts. The results were promising. Because of the results of this previous study and the fact that many ACL reconstructions are done with a bonepatellar tendon-bone complex, we felt that an evaluation of bone-patellar tendon-bone allografts in an animal model was indicated. The purpose of this study was to determine the biological and biomechanical fate of freeze-dried, ethyl-
Twenty-eight goats underwent ACL reconstruction with freeze-dried bone-patellar tendon-bone allografts in one knee, the opposite knee serving as a control. One group of 16 knees was evaluated, in groups of four, at 6, 12, 26, and 52 weeks by histologic and vascular injection techniques. The other group of 12 knees was evaluated in two groups of six at 26 and 52 weeks by morphological and biomechanical techniques of analysis. Within the first 12 weeks these allografts were revascularized ; in the first 26 weeks they had matured to resemble normal connective tissue. Graft stiffness was 29% of the control value and maximum force to failure was 43% of the control value. The results of this study indicated that freeze-dried bone-patellar tendon-bone allografts are biomechani-
cally and biologically grafts.
similar to
patellar
ene
oxide-sterilized, bone-patellar tendon-bone allografts a goat model.
used for ACL reconstruction in
MATERIALS AND METHODS
tendon auto-
We used 28 Angora goats between 2 and 3 years of age, with a weight of approximately 22 to 32 kg each. Goats were chosen because the biomechanical properties of the goat ACL and the human ACL are similar.&dquo; We divided the animals into two groups, one for histological and vascular analysis (16 animals) and one for biomechanical and morphological analysis (12 animals). All of the goats had the ACL of one hindlimb reconstructed with a freeze-dried bone-patellar tendon-bone allograft. The opposite hindlimb served as a control. We sacrificed groups of four animals at 6, 12, 26, and 52 weeks for histologic analysis. For morphological and biomechanical analysis, we sacrificed two groups of six goats at 26 and 52 weeks.
Using a portion of the patellar tendon for ACL reconstruction has been advocated by many authors. 2,3,9.11,15,16 Independent studies by Alm et all and Arnoczky et al.~ established that free autogenous patellar tendon grafts were rapidly revascularized, underwent cellular repopulation and neo-organization of the collagen fibrils, and at 6 months postimplantation, the grafts histologically resembled normal ligamentous tissue. 11 Address correspondence and repnnt requests to Thomas S. Roberts, MD, 2500 North State Street, Jackson, MS 39216-4505. 256
257
Surgical procedure After anesthetizing the animals with intramuscular ketamine hydrochloride 125 to 150 mg and xylazine 25 mg, we checked the knees for range of motion and stability. One knee of each animal was then shaved, prepared, and draped in a sterile fashion. Our surgical approach was a lateral parapatellar incision extending from the distal one-third of the thigh to just below the tibial tubercle. The fascia lata and lateral capsule were opened in line with the skin incision, and the patella was subluxated medially, exposing the knee joint. At this point, we completely resected the ACL. On manual examination, all of the knees demonstrated gross anterior subluxation of the tibia after ACL excision. A drill hole was made from the medial aspect of the proximal tibia into the knee just anterior and medial to the anatomical insertion of the ACL on the proximal tibia. This was then overdrilled with a 6 mm drill bit. We also prepared a groove in the lateral femoral condyle so that placement of the graft would be in as near an anatomical position as possible. We obtained freeze-dried bone-patellar tendon-bone allografts from animals used in a previous study. The allografts, prepared by the Bone Bank Foundation of San Antonio, Texas, averaged 6 mm in width and were prepared from both the medial and lateral portions of the patellar tendon complex. The allografts were frozen at -70°C, freezedried, and then sterilized with ethylene oxide. Following this, the allografts were purged for 6 hours. Before implantation, the grafts were reconstituted with 250 mm of normal saline for 30 to 60 minutes and cooled to 5°C until ready for use.
Using a rasp, we smoothed the tibial tunnel and femoral margins and then pulled the graft into the tunnel and positioned it in the groove. The graft was secured with a small barbed Richard’s staple placed over the bone plug in a groove prepared at the distal end of the tibial tunnel. We reduced the anteriorly subluxated knee and pulled the graft taut over the lateral femoral condyle in order to eliminate the abnormal laxity. The graft was oriented in the femoral groove so that the cancellous bone of the graft was in direct contact with the cancellous bone of the &dquo;groove&dquo; in the lateral femoral condyle. Another staple secured the bone plug to the lateral surface of the femur. We rechecked the knee for stability to be certain that the anterior subluxation had been obliterated. After irrigation with normal saline, we closed the synovium and capsule with running 3-0 nylon sutures, the subcutaneous tissue with running 2-0 vicryl sutures, and the skin with interrupted 30 nylon sutures. The goats were given 1 g of cefazolin sodium perioperatively. Postoperatively, they recovered from anesgroove
thesia in a controlled area without immobilization for the limb. Afterward, we allowed them to roam freely in a fenced area
measuring approximately
30 X 30 meters.
Laboratory techniques At the conclusion of each experimental period, we sacrificed four animals via barbiturate overdose. From two of the
animals, the allograft was removed en bloc, fixed in 10% buffered formalin, and embedded in paraffin. Five micrometer-thick sections were cut, stained with hematoxylin and eosin, and examined by light and polarized light microscopy. Using a modified Spalteholz technique,’ we evaluated the vascularity of the graft in the remaining two animals.
Biomechanical
testing
Anteroposterior force displacement test of the whole knee joint. Anteroposterior (AP) force displacement measurements were made at 45° of knee flexion before the joints were dissected. Limbs were thawed overnight at room temperature. We removed all soft tissue from the tibia and femur to within 7 cm of the joint line and mounted the limb in an instrumented measurement device. The femur was held in grips that allowed adjustment of femoral position with 6 df while the tibia hung vertically. The load application mechanism restricted external tibial rotations during the test, while allowing vertical pistoning of the tibia. A load cell monitored the applied AP forces, and a linear variable differential transformer (LVDT) measured AP displacement at the level of the tibial tubercle. We determined total AP translation between AP forces of 30 N at the joint line. The translation in the low stiffness region (primary laxity) was determined by using the method of Markoff et al. 17 The anterior translation in the high stiffness region (secondary anterior laxity) was the additional translation, from the end of the primary laxity region, required to reach 30 N of anterior force. We calculated anterior joint stiffness from a tangent to the loading curve at 30 N of anterior force. Rating of articular cartilage. Specimens were dissected to prepare femur-graft-tibia units for tensile failure testing. During this process, we photographed all joints and recorded gross joint degeneration, including the size, depth, and location of any cartilage lesions, meniscal tears, and osteophyte formations. We graded cartilage changes as follows: 0, no observable gross changes; 1, intact surface color changes and/or surface irregularities; 2, surface fibrillation (disruption) alone or with loss of cartilage but with no bone exposed; 3, exposed bone loss less than 10% of the surface in the given region; and 4, greater than 10% of the surface area having exposed bone and fragmentation of the cartilage around the lesion. We rated eight regions: each tibial condyle, the anterior and posterior halves of each femoral condyle, the trochlear groove, and the articular surface of the patella. Because of the nature of the data (rating scores) and the relatively small sample size, we used the Wilcoxon signedrank test, a nonparametric method, to test for differences between control and operated knees. Paired difference scores (operated versus nonoperated) were first compared between the two groups of animals sacrificed at 6 and 12 months. Because we found no statistically significant differences in the ratings, we used the data from all 12 animals collectively
258
significant differences between operated and nonoperated knees at each of the eight locations.
to test for
Specimen geometry. We removed the medial femoral condyle to provide better visualization of, and access to, the ligament. Using vernier calipers, we measured graft and control ACL lengths with control lengths computed as the mean length of fibers on the anterior, posterior, medial, and lateral sides of the ligament. Tissue cross-sectional areas were measured using an area micrometer with a blade pressure of 0.12 mPa applied for 2 minutes along the midportion of the graft. The blade pressure was reproducible. Tensile failure tests. The tibia and femur were then mounted in grips at a flexion angle of 30° and the intraarticular portion of the graft was visually aligned in a vertical orientation under the load cell axis. We conducted failure tests at room temperature by elongating the specimen at a rate of 100% per second of the average measured length of the ligament of graft. An LVDT on the actuator monitored actuator travel and a clip gauge attached directly to the tibia
ripheral portion of the graft (Fig. 1), whereas the central portion of the graft remained avascular and acellular (Fig. 2). The intraarticular portion of the graft was surrounded by a vascular synovial envelope and demonstrated evidence of intrinsic revascularization. The intraosseous portion of the graft was relatively avascular at this time (Fig. 3). Although there was no histologic evidence of an inflammatory or rejection response in the intraarticular portion of the allograft, the intraosseous portion of one graft demonstrated patches of lymphocytic infiltration suggestive of an immune response (Fig. 4). Interestingly, this was the only specimen to exhibit this response. At 3 months, the allograft appeared grossly the same. However, as in the 6 week specimens, the degree of arthro-
fibrosis and degenerative joint disease varied. The findings similar to those in the 6 week specimens. The revas-
were
and femur provided the relative displacement of the bone ends. A lab minicomputer was used to initiate the test and to measure and store the force, actuator motion, and clip gauge data. We determined failure mechanisms from both direct posttest examination of the failed tissues and review of 16 mm cinefilms taken at 200 frames per second. The clip gauge and force measurements were used to determine the following structural properties: ligament stiffness in the linear loading region, maximum force before failure, and energy to maximum force. The clip gauge was required primarily for control specimen tests in which the large forces caused bone deflection despite the use of support brackets intended to prevent deflection of the tibia and femur.
RESULTS Clinical analysis of the goats showed that the animals walked without a limp by the 6th postoperative week. Five animals developed an early effusion, but none showed any other evidence of an infection. No positive cultures were obtained and the effusions were felt to be secondary to inflammation. The effusions all resolved before sacrifice. One goat died from pulmonary complications subsequent to the ACL reconstruction and was not analyzed. That goat was
Figure 1. Photomicrograph of the peripheral intraarticular portion of an allograft at 6 weeks postimplantation. Notice the vascular channels (arrow) that have already formed, as well as the early cellular repopulation.
replaced.
ACL grafts
at the time of sacrifice. felt to have a positive anterior drawer sign or Lachman on physical examination. No other form of ligamentous laxity was demonstrated. No roentgenwere
None of the knees
grossly intact were
ographic analysis was performed. Histology and vascularity Six weeks after transplantation, the knees demonstrated various degrees of arthrofibrosis and degenerative joint disease. Small cartilaginous lesions were present, but no osteophytes were noted. Histologic evaluation of the allograft revealed the presence of vessels and fibroblasts in the pe-
Figure 2. The central portion of this allograft at 6 weeks demonstrates the necrosis and lack of vascularity in this portion of the graft.
259
Figure 3. This 6 week specimen prepared by the Spalteholz technique demonstrates a lack of vascularity in the tibial tunnel (arrow), but the periphery of the intraarticular portion of the graft is well vascularized. The PCL and fat been removed to facilitate observation of the graft.
pad
have
cularization process had spread throughout the allograft with the majority of the graft being revascularized (Fig. 5). Histologic sections revealed an increased cell population and a more organized appearance to the tissue (Fig. 6). By 6 months, the graft was completely revascularized and the tissue had the appearance of normal, dense, regularly oriented connective tissue (Fig. 7). The 12 month specimens were similar in both gross and histologic appearance to the 6 month specimens.
Rating of articular cartilage and
biomechanical tests
Rating of articular cartilage. Table 1 shows the results of the analysis of articular cartilage changes in the eight regions of the knee. The mean values and the distribution of ratings are shown for operated and nonoperated knees separately. Statistical differences between operated and nonoperated knees are based on paired values within animals. Operated knees had significantly higher grades (i.e., more degeneration) than nonoperated knees at seven of the eight locations. We noted a large number of Grade 4 ratings (more than
4. This 6 week allograft demonstrates patches of a cellular infiltrate that is indicative of a reactive response. The cellular infiltrate was present only in the tibial tunnel area of the allograft. This is the only portion of any of the specimens that demonstrated any type of reactive pattern.
Figure
50% of cartilage eroded to bone) in the patellofemoral joint of the operated knees. In addition to these ratings of the articular cartilage, meniscal irregularities were noted in eight knees and osteophyte formation in seven knees. AP force displacement tests. Table 2 shows results of measurements from the AP force displacement tests. The control knees demonstrated a total AP translation of 1.1 mm, with only 0.2 mm of translation in the region of primary laxity. These results show the near absence of slackness in the anterior cruciate and posterior cruciate ligaments. There were no significant differences in total AP translation, primary and secondary anterior translation, and anterior stiffness between the 6 and 12 month allograft groups. Reconstructed knees of both groups had statistically greater total AP translation than controls. As a reference, tests conducted by Holden et al. 12 of three goat knees with the ACL resected showed total AP translation of 8.9 ± 0.4 mm. The increase in total translation in the reconstructed knees was due mostly to a significant increase in primary anterior translation, which averaged 2.8 mm. In contrast, the secondary anterior translation averaged 0.6 mm. Although this
260
z
I1
...,#,&dquo;,..~Y&dquo;pil,
1’ t i
,
s ,
I
I I
44
I
1
k
I B
.,
°:w4
G
Y‘~.
I
I
