HM Medical Clinic



The American Journal of Sports Graft Fixation in Cruciate Ligament Reconstruction
Jeff Brand, Jr., Andreas Weiler, David N. M. Caborn, Charles H. Brown, Jr. and Darren L. Johnson Am J Sports Med The online version of this article can be found at: can be found at:
The American Journal of Sports Medicine
Additional services and information for
0363-5465/100/2828-0761$02.00/0THE AMERICAN JOURNAL OF SPORTS MEDICINE, Vol. 28, No. 5 2000 American Orthopaedic Society for Sports Medicine Graft Fixation in Cruciate Ligament

Jeff Brand, Jr.,* MD, Andreas Weiler,† MD, David N. M. Caborn,* MD, Charles H. Brown, Jr.,‡ MD, and Darren L. Johnson,*§ MD From the *University of Kentucky School of Medicine, Lexington, Kentucky, †Sports Traumatology & Arthroscopy Service, Humboldt-University, Berlin, Germany, and ‡Brigham and Women's Hospital, Department of Orthopaedic Surgery, Boston, Massachusetts s.Current rehabilitation protocols after knee ligament sur-gery stress immediate full range of motion, return of neu- Cruciate ligament reconstruction has progressed dra- romuscular function, proprioception, and early weight- matically in the last 20 years. Anatomic placement of bearing forces up the kinetic chain. In the early ligament substitutes has fostered rehabilitation efforts postoperative period, graft fixation is the weak link within that stress immediate and full range of motion, imme- the entire system. No commonly used graft fixation has diate weightbearing, neuromuscular strength and co- ultimate failure strength or stiffness comparable with the ordination, and early return to athletic competition (3 native cruciate ligament (Table 1). Fixation methods must months). This has placed extreme importance on se-cure graft fixation at the time of ligament reconstruc- be rigid and stiff to allow current rehabilitation principles.
tion. Current ligament substitutes require a bony or soft Current fixation techniques involve soft tissue and bone tissue component to be fixed within a bone tunnel or on within a bone tunnel or periosteal fixation away from joint the periosteum at a distance from the normal ligament attachment site. Fixation devices have progressed Bone-patellar tendon-bone, quadrupled hamstring ten- from metal to biodegradable and from far to near- don, or quadriceps tendon-bone are the most commonly normal native ligament attachment sites. Ideally, the used ligamentous substitutes in cruciate ligament recon- biomechanical properties of the entire graft construct struction. Using these ligament substitutes with current would approach those of the native ligament and facil- fixation devices, we have been unable to reproduce the itate biologic incorporation of the graft. Fixation should normal transition zones of insertion of the ACL and PCL.
be done at the normal anatomic attachment site of the Given that anatomic structure dictates function, the me- native ligament (aperture fixation) and, over time, allow chanical profile of the ligament substitute has not been the biologic return of the histologic transition zone from reproduced. Variables that we are able to measure in the ligament to fibrocartilage, to calcified fibrocartilage, to basic science laboratory at time zero of ligament recon- bone. The purpose of this article is to review current struction include data on ultimate failure load, yield point, fixation devices and techniques in cruciate ligament stiffness, displacement to failure, and mode of failure.
Correlation of these results with clinical outcome has notbeen reported.
Our purpose is to review all current information with regard to ligament substitute fixation of bone and soft The importance of secure graft fixation in ligament recon- tissue grafts. It is important for the surgeon to be aware of struction has changed dramatically in the last 20 year- the difference in fixation techniques with the associatedbiologic consequences. Different graft substitutes may re-quire different fixation techniques that have direct bio- § Address correspondence and reprint requests to Darren L. Johnson, MD, logic implications. Knowledge of these fixation techniques Chairman of Orthopaedic Surgery, University of Kentucky School of Medicine, will allow the clinician to make necessary intraoperative and Kentucky Clinic, K-439, 740 Limestone, Lexington, KY 40536.
One author has commercial affiliation with a product named in this study.
postoperative decisions in cruciate ligament reconstruction.

Brand et al. American Journal of Sports Medicine There is evidence that less than 454 N is sufficient for Ultimate Load to Failure and Stiffness of Current Graft activities of daily living. In a clinical study, Shelbourne Selections in Cruciate Ligament Surgery and Gray79 reported use of a button for both the tibial and Ultimate Strength femoral fixation of a patellar tendon reconstruction, which Graft Selection (ref.) has a failure strength of 248 N.43 Excellent clinical and objective knee stability was maintained with an acceler- ated rehabilitation program in their series of patients.
Patellar tendon25 Quadrupled hamstring tendon (semitendinosisand gracilis)35 Biomechanical Properties Quadriceps tendon85 Stiffness is the slope of the linear region of the load-elongation curve and is usually reported in units such asnewtons per millimeter (N/mm). As a graft and its fixation IDEAL GRAFT FIXATION device are loaded with a tensile force, displacement in thegraft and fixation device occurs equal to an amount de- scribed by its stiffness. Present graft fixation alternativesare less stiff than the native ACL and graft choices. This Because there is only one means of graft fixation that can be compared with a chain secured to posts by bungee approaches the strength of the native ACL, the question cords at either end of the chain. As force is applied to the is, "How much strength is required of a cruciate ligament chain, the bungee cords, not the chain, will displace under reconstruction for activities of daily living and a progres- tensile load. Mechanically, the majority of tendon fixation sive rehabilitation program?" constructs are less stiff than the interference screw Noyes et al.69 have estimated the strength required for against a bone plug, which has been considered the stan- activities of daily living to be 454 N based on the failure dard for fixation (see Tables 4, 7, 8, and 9). Thus, given strength of the ACL. They state that, "It seems reasonable that ultimate failure strength is comparable between the to assume that under normal conditions biological tissues two given fixation choices, tendon constructs may displace are subjected to forces ranging from one-tenth to not more or slip more before they fail, creating laxity in the graft than one-fifth of their breaking loads." The same group concludes "For the posterior cruciate ligament (PCL) these Many tendon fixation devices are "indirect." They rely force levels would be increased." on linkage material to connect the tendon to the fixation Morrison,64–66 a bioengineer, writing in the late 1960s device. A biomechanical study compared strain that was and early 1970s in a series of three articles relating force induced by cyclic loading in a patellar tendon graft and a plate and gait analysis data, made calculations and con- quadrupled hamstring tendon graft and found that the clusions regarding the forces in the ACL and PCL; theseare shown in Table 2. Markolf et al.58 used a cadavericmodel that ignored muscle forces and examined the forceson the ACL, a patellar tendon graft, and an overtensioned(45 N) patellar tendon graft. The patellar tendon graftexperienced higher forces than the native ACL (peakforce, 297 N) and overtensioning the graft increased theforces experienced by the graft (up to 497 N). In a similarstudy by Markolf et al.,59 the PCL forces were examined inthe intact PCL and a patellar tendon graft was used toreconstruct the PCL. The forces in the PCL study weremuch lower, generally less than 100 N. The higher forcesdeveloped in some grafts in hyperextension and hyperflex-ion, leading the authors to recommend avoiding thesemotions after reconstruction.
Estimations of Forces Present in the Cruciate Ligaments in Activities of Daily Living64–66 Descending stairs Figure 1. A schematic diagram demonstrating the bungee
effect or longitudinal graft-tunnel motion.

Vol. 28, No. 5, 2000 Graft Fixation in Cruciate Ligament Reconstruction tape-tissue interface (5.4%) had notably more strain than tendon graft incorporation has been shown to occur the tape (2.9%) or the tissue (1.1%) alone.52 If strain or sooner. A rabbit model with a free semitendinosus graft laxity is in line with the linkage, it is referred to as the intraarticularly placed through bone tunnels and fixed bungee cord effect32 (Fig. 1). These shearing forces may be with suture suggested that the graft healed in the tunnel responsible for tunnel expansion, also known as the wind- within 3 weeks.34 In a similar ACL reconstruction study with a sheep model fixed with biodegradable interference In the native cruciate ligament, the point of fixation is screws directly against the free autologous Achilles ten- at the joint surface. However, most tendon fixation con- don graft, intraligamentous failure was demonstrated by 6 structs are placed at a distance from the joint surface with weeks.92 Evidence of bone plug incorporation before soft a staple, screw and suture, or soft tissue washer. When tissue healing in a bone tunnel is not definite based on interference fixation is placed closer to the joint surface, animal studies. Weightbearing and rehabilitative exer- there is increased knee stability at a variety of flexion cises increase stress that the new, reconstructed ligament angles and also improved graft isometry (Fig. 2) (Ref. 41; will have to respond to and react. These activities occur at C. Morgan, unpublished data, 1994).
the time when the weak link of the reconstruction is thefixation of the graft. In our laboratory experience, evenlow cyclic loads, up to 110 N, cause shear forces in the bone Biologic Properties tunnel on the graft.12 Strength and stiffness in the fixa-tion is the key to diminishing this graft-bone tunnel mo- It has been stated that bone plug incorporation occurs tion as healing progresses.
before tendon incorporation in a bone tunnel,76 but basicscience on this matter is not definite. In a study by Clancyet al.,23 bone plug-patellar tendon-bone plug in a bone FEMORAL AND TIBIAL FIXATION tunnel was histologically incorporated at 8 weeks aftersurgery in a rhesus monkey, when it was first histologi- There are two key differences that need to be considered cally examined. After 3 months, all biomechanical testing between femoral and tibial fixation, that of bone density resulted in interstitial failure of the reconstructed grafts, and the angle at which force is applied to the graft attach- with no bony avulsions occurring, thus implying bone plug ment. The bone quality and geometry of the tibia is dif- incorporation in the bone tunnel.
ferent from that of the femur.11 The Dual Photon Absorp- In a dog extraarticular tendon model, the tendon graft tometry (DEXA) of the tibial metaphysis has been pulled out of the bone tunnel until 12 weeks postopera- determined to be less than the femoral metaphysis in the tively, indicating that the tendon was not healed in the same knee of elderly cadavers11 and in young women.89 bone tunnel. The graft was an extraarticular, long digital The line of force on the graft is directly in line with the extensor tendon and was left attached distally under ten- tibial tunnel. The line of force on the graft is obliquely sion; it was not an intraarticular free graft.73 However, orientated to the femoral tunnel in the weightbearingposition, which is extension. Based on radiographic stud-ies, the femoral tunnel does not become colinear with theligament graft until approximately 100° of knee flexion.57 Kohn and Rose46 have found a lower ultimate load of tibial failure when using interference fixation for boneplug fixation.
Study methods of present biomechanical studies vary ex-tensively from institution to institution, making compar-ative statements of fixation methods and devices difficult(Table 3). Variables that we are able to measure in thebasic science laboratory at time zero of ligament recon-struction include data on ultimate failure load, yield point,stiffness, displacement to failure, and mode of failure.
Stiffness, an important descriptive variable that predictsthe displacement or slippage of a device before it fails, hasnot been reported in all biomechanical studies. Anothervariable, bone mineral density, with direct clinical appli-cations is varied throughout present studies. Bone min-eral density is correlated with the results of tendon inter- Figure 2. A schematic representation of a transtibial ACL
ference fixation and may be important in other forms of reconstruction using a quadrupled hamstring tendon graft fixation as well.11 The results of animal studies, which and direct biodegradable interference screw fixation placed have a higher and more consistent bone mineral density, at joint surfaces.
have yielded higher failure values using interference fix- Brand et al. American Journal of Sports Medicine Biomechanical Study Methods for Fixation Devices Comparison of "Pull-out" Studies using Human Tissue Brown et al., older cadavers Semifix, bone mulch, Anterior drawer to interference fixation, semitendinosis and endobutton, press-fit gracilis tendonpatellar tendon Brown et al.,14 older cadaver Interference screws, Femur only, in line compared rear entry Caborn et al.,19 older cadavers Femur only, in line interference screw Caborn et al.,18 older cadavers BioScrew, titanium Femur only, in line interference screw Gerich et al.,33 cadavers (ages Interference screws, Tibia only, "axial to Patellar tendon bone Johnson and vanDyk,44 Interference screw Femoral preparation cadavers (ages 47–70) only, in line with biodegradable screw Kohn and Rose,46 cadavers, Interference screw, Tibia preparation median age, 30 (22–60) influence of screw Femoral preparation, diameter, compared Kurosaka et al.,48 cadavers, Anterior drawer to interference fixation, Magen et al.,56 cadavers (ages Tibia preparation, in Quadrupled hamstring line with the tunnel Matthews et al.,62 cadavers Interference screw, Tibia preparation, suture and post with femoral preparation, graft tensionedperpendicularly tobone preparation Pena et al.,70 cadavers (ages Interference screw, Femoral preparation, BioScrew and metal Rowden et al.,74 young Interference screw Anterior drawer to cadavers (mean age, 26) Steiner et al.,87 cadavers Suture and post, post Anterior drawer to interference screw quadrupled hamstring Weiler et al.,97 cadaver (mean Button, screw and Anterior drawer to Hamstring, hamstring washer, RCI screw, ation than comparable human studies.56 The bone min- BONE PLUG GRAFT FIXATION—TIBIAL FIXATION eral density of elderly cadavers may be as little as halfthat of a young healthy person who sustains cruciate ligament damage as a teenager. Because of the scarcity ofspecimens, the same specimen is often tested multiple Although there are alternative means of fixation in graft times. Techniques vary from the clinical situation to the tunnel-length mismatch, this mismatch is considered the laboratory. For instance, the interference screw is often primary indication for staple fixation of a bone plug. An- placed under direct visualization, minimizing the possibil- other method of fixation for graft tunnel-length mismatch ity of divergence, which certainly occurs in vivo.51 If a include a longer femoral tunnel with a proportionately device is tested in line with the tunnel, the worst-case longer interference screw to create aperture fixation. Var- scenario, the failure load may be less than if the device is ious means of shortening the graft to match tunnel length loaded at an angle to the tunnel that will increase the have also been described. A set of doubled staples in a shear forces. Because of viscoelastic properties of the shallow trough (with an ultimate load at failure of 588 N) graft-bone construct, the rate the graft is loaded will affect compared favorably with interference fixation (506 to 758 the stiffness. Rehabilitation and ambulation stresses are N) in failure, and the staples were significantly stiffer examples of cyclic loading and are not accounted for with (86.3 N/mm) than interference fixation (49.2 to 54.9 static testing at time zero fixation of the graft N/mm) in a young (mean age, 44) human cadaveric model.
Vol. 28, No. 5, 2000 Graft Fixation in Cruciate Ligament Reconstruction Unfortunately, the incidence of bone block breakage (27%) fixation may be combined with other types of fixation such was significantly greater than that of the interference as a suture and post, EndoButton (Acufex, Inc., Mansfield, screw fixation (1%).33 Massachusetts), or screw and washer.
Currently, a screw 9 mm in diameter and at least 20 Screws Used as a Post mm in length is the standard used for fixation. The dif-ference between the outside diameter of the screw and the Steiner et al.87 did a study that reported a screw used as core diameter is the most important consideration.97 Kohn a post, linked with suture, and combined with an inter- and Rose46 showed that a 9-mm tibial interference screw ference screw against a bone plug. They found that this disengaged from the bone tunnel at significantly more had a failure strength (674 N), which approximated that of maximum tensile strength and linear load to failure com- the intact ACL (560 N) (Table 4). The angle at which the pared with a 7-mm screw (Table 4). Screw length beyond screw is placed determines whether the graft is tensioned 20 mm in conjunction with a bone plug does not appear to as the screw is tightened or whether the graft is relaxed as be necessary.14,39 the screw is tightened. Although a low-profile screw with The gap between the bone plug and bone tunnel and the a flatter head is available from many of the orthopaedic interaction with screw diameter influences the fixation manufacturers, conventional screws are often removed strength. Brown et al.15 suggested that interference because of pain. The post and suture can serve as a backup (screw outer diameter minus tunnel bone block gap) was to tibial interference fixation that is compromised by poor correlated with failure, but gap size alone is not associated bone quality or bone plug fracture.
with failure. Similarly, a separate porcine biomechanicalstudy showed that a 1- or 2-mm gap with a 7-mm screw Interference Fixation yielded equal failure strength to a 3- or 4-mm gap with a Whatever fixation strength is required for activities of 9-mm screw.17 Alternatively, when faced with a gap or daily living and a progressive rehabilitation program ap- bone of poor quality, a bone shim may improve the fixation pears to be met by the strength and stiffness of interfer- ence fixation, which, for this reason, has been described as Despite the clinical success of interference fixation, the standard of graft fixation.87 Interference fixation was complications, usually preventable, have been reported.
first described by Lambert50 in a study using a 6.5-mm Counter tension through the bone plug sutures can reduce cancellous screw. In 1987, Kurosaka et al.48 demonstrated graft advancement as the interference screw is placed.61 superior strength with a larger diameter screw (9 mm) for Screw laceration of either the bone plug suture or of the interference fixation. When poor bone stock exists—which graft itself are clinical concerns. If the sutures that are may be due to revision, tunnel widening, or graft tunnel- attached to the bone plug are lacerated with the threads length mismatch— or additional fixation strength is from the screw, poor graft fixation cannot be salvaged needed for large or noncompliant patients, interference with a suture-and-post construct. Suture laceration can be Tibial Fixation Options for Bone-Patellar Tendon-Bone Plug in a Bone Tunnela Suture (#5) to button48 Anterior drawer to knee Button failed, suture pulled through the bone plug Staple1 patella tendon48 Anterior drawer to knee Graft slipped under the staple Doubled staples on patella Tibia only, "axial to Graft slipped under staple, 27% tendon in a trough33 bone block breakage Suture and post87 Anterior drawer to knee Bone-tendon rupture, bone plug fracture, tibial post pull-out 6.5 mm AO interference screw48 Anterior drawer to knee Grafts pulled out of the tunnel 9 mm interference screw48 Anterior drawer to knee Grafts pulled out of the tunnel Interference screw and suture Anterior drawer to knee Bone plug fractured, pull-out around tibial screw and suturerupture 7 mm interference screw46 Tibia only, parallel to Tendon tearing, slipping of the 9 mm interference screw46 Tibia only, parallel to Tendon tearing, slipping of the 9 ⫻ 30 mm interference screw33 Tibia only, "axial to Tendon tearing or bone plug 9 ⫻ 25 mm biodegradable Tibia only, parallel to Bone plug slipped, tendon tearing a The standard deviations or ranges of variability are reported in parentheses following the mean.
Brand et al. American Journal of Sports Medicine avoided with the use of 20-gauge wire through the holes in ing to their degradation. Group one consists of slow de- the bone plug. Graft laceration may require another graft grading and highly crystalline poly-(L-lactide) and poly- option.61 Two cases of bone plug comminution have been (L-co-D,L-lactide) stereocopolymers with a low D,L amount.
reported: one was salvaged by reversing the graft and These materials are considered to have high mechanical placing the fractured bone plug on the tibial side and properties among the poly-alpha-hydroxy acids, but their fixing it with a suture and post, the other had to be revised degradation can last up to several years and is incomplete to another graft choice.5 Pain in the area of the tibial because of a possible accumulation of insoluble crystalline screw that was caused by hardware has been reported by implant remnants.6,22,27,71 Group two is represented by 3% of patients, and screw removal was very successful in amorphous poly-(L-co-D,L-lactide) stereocopolymers with a relieving this pain.49 high D,L amount and the purous poly-(D,L-lactide). Thesematerials degrade completely within 1 to 2 years, but their Biodegradable Interference Screws mechanical properties are lower compared with the poly-(L-lactide).84 The third group consists of fast-degrading The terms "biodegradable" or "bioabsorbable" are used copolymers such as poly-(D,L-lactide-co-glycolide) or polyg- interchangeably to characterize materials that disinte- lycolide-co-trimethylencarbonate, whose strength reten- grate after implantation and are subsequently excreted.
tion lasts for only several weeks.
Materials that disintegrate in the body have been used by For many years, biodegradable implants have been orthopaedic surgeons over the past 3 decades and these thought to offer advantages over metal analogs. Metal materials allow for better available implants. In cruciate implants can distort magnetic resonance imaging (Fig. 3) ligament surgery, several different biodegradable inter-ference screws consisting of different polymeric raw ma- and release metal ions into the surrounding tissue.40, 80 terials are currently available (Table 5). A large number of Further disadvantages include the need for a second sur- studies have investigated their biomechanical and clinical gical procedure for implant removal and a revision sur- gery complicated by the presence of a metal implant. In Biodegradable implants consist mainly of the poly-al- cruciate ligament surgery, the major advantages of biode- pha-hydroxy acids, polylactide and polyglycolide, includ- gradable interference screws is an uncompromised revi- ing their copolymers, poly-(D,L-lactide-co-glycolide) and sion surgery. This is especially important because the number of revisions has risen dramatically within the last as poly-(L-lactide), poly-(L-co-D,L-lactide) and poly-(D,L- few years.68,88 The difficulties encountered with retained lactide) are also used (Table 5). These raw materials repre- metal screws in revision surgery has been described.77 In sent substantially different material characteristics, such case of revision after using biodegradable interference as degradation kinetics, mechanical properties, and bio- screws, surgery may be performed like a primary proce- compatibility. Generally, it is considered reasonable to dure if the material has degraded and osseous replace- divide these materials into three different groups accord- ment has taken place with an appropriate amount of Advantages and Disadvantages of Different Biodegradable Interference Screws Implant (Manufacturer) Raw material (Abbreviation) Biologically Quiet Interference Amorphous material, osseous Low initial fixation strength, fast Screw (Instrument Makar glycolide) 85/15% replacement within an degradation,71 only one size Inc., Okemos, MI) appropriate time62 Bio-Interference Screw (Arthrex poly-(L-lactide) (PLLA) High initial fixation strength,71 Semicrystalline PLLA with Corp., Naples, FL) different sizes available recrystallization and possibleincomplete degradation BioScrew (Linvatec Corp., poly-(L-lactide) (PLLA) High initial fixation strength, high Highly crystalline PLLA with torsional strength,71 different incomplete degradation62 Endo-Fix (Acufex Inc., High initial fixation strength71 Low torsional strength, crystalline copolymer, fast 67.5/32.5% (PGA-co- degradation with possible adverse tissue response,20,30,71only one size available Phantom Absorbable Screw poly-(L-lactide) (PLLA) High initial fixation strength Highly crystalline PLLA with (DePuy Orthopaedic incomplete degradation Technology Inc., Tracy, CA) Phusiline Interference Screw High initial fixation strength Low torsional strength, (Phusis mate´riaux different sizes available71 semicrystalline polymer with biore´sorbables, Le Versoud, re-crystallization and incomplete degradation71 Sysorb (Sulzer Orthopedics High initial fixation strength, high Possible viscoplastic deformation, torsional strength, amorphous only one size available material, osseous replacementwithin an appropriate time62,71

Vol. 28, No. 5, 2000 Graft Fixation in Cruciate Ligament Reconstruction incorporation of a bone-tendon-bone graft, and little isknown about the fixation properties of the biodegradablescrews within this period while the material is degrading.
Only a few in vivo studies have investigated changes infixation strength of biodegradable interference screwsover time. Walton and Cameron90 used polyglycolide-co-trimethylencarbonate screws (Endofix, Acufex Inc.) in asheep model and reported that the fixation strength ofthese screws remained comparable with that of metalscrews for 12 weeks. Therin et al. (unpublished data,1996) also investigated the in vivo biocompatibility anddegradation of a poly-(L-co-D,L-lactide) screw (Phusiline,Phusls mate´riaux biore´sorbables, St. Ismier, France) in asheep model and reported proper bone healing measuredby polychrome sequential labeling. Champion et al.21 in-vestigated the pushout loads of a poly-(L-lactide) interfer-ence screw (Phantom, DePuy Inc., Tracy, California) in acanine model over 24 weeks, and suggested that thesescrews withstand ACL forces during the healing stage ofreconstruction. The clinical use of biodegradable interfer-ence screws for bone-tendon-bone graft fixation was firstdescribed in the middle 1990s (Refs. 3, 44; Therin et al.,unpublished data, 1996). To date, several midterm studiescomparing metal and biodegradable interference screws inclinical studies have reported no significant difference inclinical outcome.3,31,60 The major disadvantage of biodegradable screws is screw breakage or drive failure during insertion (Refs. 3,43, 82, 97; C. Morgan, unpublished data, 1994). A screw'sresistance to breakage may depend on several factors,including core diameter, drive diameter, and drive shape.
The drive designs of some biodegradable interference Figure 3. A, a coronal section MRI of the femoral tunnel at 2
screws are direct copies of their metallic counterparts.
months postoperatively of a biodegradable screw (arrows).
Others have specially designed drive systems that may There is no artifact from the screw and it appears to be provide a better force transmission to the screw core, opposed to the quadrupled hamstring tendon graft. B, coro- thereby increasing implant resistance to breakage (Table nal section MRI of another patient with a biodegradable 5). A recent report demonstrated that implant design may screw interference fixation at 1 year. The screws have nearly be more important than the mechanical properties of the completely degraded (arrows) leaving a bright signal, but polymeric raw material to improve torsional strength.97 again the femoral graft is well opposed to the interference To avoid screw breakage, care should be taken to insert the screw convergent to the tunnel-bone block gap. Toreduce peak screw insertion torque, especially in thedense femoral bone, the manufacturer's recommendations newly formed bone at the former implant site. In addition, to use a notching device or a tap should be followed.
functional loads can be assumed earlier by the healing There are still concerns about an appropriate biocom- bone while the material is degrading.22 Another advan- patibility of other biodegradable materials because of re- tage in cruciate ligament surgery is a decreased potentialof graft laceration during screw insertion, which has been ports on severe foreign-body reactions associated with the described to occur when using metal screws.30,61 use of self-reinforced and highly crystalline polyglycolide Several recent biomechanical studies compared the ini- implants.8,20,36,91 Today, other materials such as polylac- tial fixation strength of biodegradable and conventional tide and its copolymers and stereocopolymers are consid- titanium interference screws in human and animal cadav- ered to have better biocompatibility,9,16,36,91 and clini- eric models for bone-tendon-bone graft fixation. These cally relevant foreign-body reactions have not yet been studies showed that most biodegradable screws provide described in the clinical reports on biodegradable interfer- similar fixation strength and concluded that the use of ence screws. However, further studies should take into these screws may allow for an accelerated postoperative consideration that foreign-body reactions may principally rehabilitation program.1, 19, 70, 75, 97 While these investiga- accompany the use of each biodegradable implant and, to tions studied only the initial fixation strength, it is known finally judge the appropriateness of such an implant, long- that approximately 6 weeks are required for the bony term studies are necessary.37 Brand et al. American Journal of Sports Medicine BONE PLUG GRAFT FIXATION— Interference Fixation Two studies with human tissue compared a metal 7-mm diameter screw placed intraarticularly, as in endoscopicACL reconstruction, with an outside-in technique using a The EndoButton is used primarily with bone plug fixation 9-mm screw and found similar strength and stiffness (Ta- in femoral tunnel blow-out. Interference fixation is pref- ble 7) (Ref. 87; Brown et al., unpublished data, 1996).
erable in routine femoral bone plug fixation. The En- Although screw divergence from the bone plug is com- doButton, a modification of the button, was designed to be mon when postoperative radiographs are evaluated criti- used in the endoscopic ACL reconstruction for femoral cally,51 it is not considered a clinical concern. Dworsky et fixation and now has been described for use in PCL recon- al.28 described the endoscopically placed interference struction as well.4,81 Doubling the linkage materials has screw acting as a "wedge," effectively blocking the femoral significantly increased their mechanical properties (Table bone plug from being displaced into the joint. Further- 6) (C. H. Brown et al., unpublished data, 1996).
more, if the angle of screw divergence from the femoralbone plug is greater than 20°, there is a significant reduc-tion of the pullout strength in biomechanical testing.45 However, in the clinical situation, Fanelli et al.29 showed The Mitek Anchor (Mitek, Westwood, Massachusetts) is a that there was no increase in fixation failure with diver- four-pronged device that is linked to a graft by suture or gent interference screws placed endoscopically at angles tape in a fashion similar to that of the EndoButton. When greater than 20°.
comparing the Mitek device with the EndoButton in apatellar tendon-bone plug model, there was no significantdifference in failure or stiffness (Table 7) (Brown et al., SOFT TISSUE FIXATION—TIBIAL FIXATION unpublished data, 1996). This device can be used similarlyto the EndoButton in cases of femoral fixation salvage for femoral tunnel blow-out.
A single staple used with the semitendinosus tendon isneither strong nor stiff.48 The tendon graft looped over a Press-Fit Femoral Bone Plug second staple, now called the "belt-buckle" technique,markedly improved fixation in a porcine model.56 The Malek et al.57 have reported press-fitting the femoral bone failure load was 705 N with a stiffness of 174 N/mm (Table plug in an effort to avoid the complications of interference 8). Staples can frequently cause pain at the site of implan- screw fixation. Brown et al. (unpublished data, 1996) com- tation and must be removed. Although the belt-buckle pared the press-fit of the bone plug (ultimate load at technique has been used successfully, fixation is perios- failure, 350 N) with the patellar tendon bone plug with teal and is at a distance from joint surfaces.
interference fixation (398 N), EndoButton (554 N), andMitek Anchor (511 N). No statistical difference was notedin failure or stiffness (Table 7). A clinical study with Screws Used as a Post press-fit fixation on the femoral side and interference A screw can be used with a standard metal washer as a screw fixation on the tibial side noted one case of femoral post to tie suture around or it can be used with a soft bone plug fracture.10 Two cases of revision to an interfer- tissue washer against tendon. A screw with a soft tissue ence screw were required because of "insufficient femoral washer placed directly against a quadrupled tendon graft is slightly stronger and stiffer than the screw used as apost with suture (821 ⫾ 219 N compared with 573 ⫾ 109N, respectively) (Table 8).87 A screw with a soft tissue washer is the preferred method of tibial soft tissue fixa- Linkage Material Propertiesa tion, compared with a screw linked with suture, because of its superior stiffness and avoidance of relatively elastic Mersilene tape (Ethicon, Inc., Sommerville, NJ) Doubled Mersilene Meadox (Meadox Medical Inc., Oakland, NJ) The washerplate, WasherLoc (Arthrotek, Biomet, Inc., Endotape (Smith and Warsaw, Indiana), is a multiple-pronged washer and Nephew Endoscopy, screw used to fix the tibial end of the quadrupled ham- Inc., Andover, MA) string tendon graft. It is placed at the distal end of the Three #5 Ethibond tibial tunnel and can be recessed to diminish the promi- sutures, (Ethicon, Inc.) nence of the screw head. The ultimate failure load was 905 N (SD, 291 N) and the stiffness was 273 N (SD, 56 N), From Brown et al., unpublished data, 1996. The standard deviations are reported in parentheses following the mean.
which is similar to that of the native ACL (Table 8).56 Vol. 28, No. 5, 2000 Graft Fixation in Cruciate Ligament Reconstruction Femoral Fixation Options for Bone-Patellar Tendon-Bone Plug in a Bone Tunnela Anterior drawer to Tibial bone block fracture or suture breakage, tibial sidefixation failure Mitek deviceb Anterior drawer to Patellar tendon failure, fracture tibial bone block, sutures torethrough bone block Anterior drawer to Tibial bone plug pulled out, fracture tibial bone block,patellar tendon failed Interference screw from Anterior drawer to Pull-out around the screw Endoscopic interference screw87 Anterior drawer to Bone plug fractured, femoral screw pull-out, bone tendonrupture Interference screw outside-in14 Anterior drawer to Bone block pull-out, bone block Endoscopic interference screw14 Anterior drawer to Bone block pull-out, bone block Metal endoscopic interference Femur only, parallel Femoral fixation failure, fracture of bone plug, tearingof graft BioScrew endoscopic Femur only, parallel Femoral fixation failure, interference screw19 fracture of bone plug, tearingof graft Metal interference screw70 Femur only, parallel Pullout and bone block fracture BioScrew interference screw70 Femur only, parallel Bone block pull-out Metal interference screw44 Femur only, parallel Failure between the cortical and cancellous bone of thebone plug Femur only, parallel Failure between the cortical and cancellous bone of thebone plug a The standards deviations or ranges of variability are reported in parentheses following the mean.
b From Brown et al., unpublished data, 1996.
Tibial Fixation Options for a Soft Tissue Graft in a Bone Tunnela,b Stapled semitendinosis48 Anterior drawer to knee Tendon pulled out of staple QHT with suture and post87 Anterior drawer to knee Suture tendon stretches, QHT with screw and a soft Anterior drawer to knee Tendon stretches or tibial QHT with a washerplate56 Tibia only, parallel to No failure mode given QHT with the RCI titanium Anterior drawer to knee Tendons pulled out or QHT with the RCI titanium Tibia only, parallel to No failure mode given QHT with the RCI titanium Anterior drawer to knee Failed at the tibial socket QHT with biodegradable Tibia only, parallel to Graft slipped around tibial interference screw 1 mm QHT with biodegradable Tibia only, parallel to Graft slipped around tibial interference screw 1⁄2 mm a QHT, quadrupled hamstring graft.
b The standard deviations are reported in parentheses following the mean.
c Brown et al., unpublished data, 1996.
Brand et al. American Journal of Sports Medicine Biomechanically, this is the only tibial soft tissue fixation creased failure strength of the quadrupled hamstring ten- that approximates the ACL in failure and stiffness.
don graft and an 89% increase in stiffness.35 Patients who have had ACL reconstruction with trans- SOFT TISSUE FIXATION—FEMORAL fixion devices have had outcomes similar to those reported in the literature. Two patients had the pin repositionedafter migration.24 One of those patients and later another Transfixion Fixation patient had the pin removed because of iliotibial bandirritation. This device has since been modified to address The Trans-Fix (Arthrex, Naples, Florida) and the Bone the prominence of the pin head.24 Mulch Screw (Arthrotek) are examples of transfixion fix- The cross-pin offers stiffness superior to the EndoBut- ation. There was no significant difference in failure load or ton linked with a continuous loop. In fact, stiffness of the stiffness, Trans-Fix (523 N) versus the EndoButton with cross-pin approaches that of the ACL. The device does Endotape (520 N) (Smith & Nephew Endoscopy, Inc.). In require a second counter incision to deploy the cross-pin.
paired knees, there was no difference in failure between Fixation by this device is deeper in the tunnel, allowing for the Bone Mulch screw (583 N) and EndoButton (628 N).
the graft to move in the tunnel, which has been associated The Bone Mulch Screw was slightly stiffer; 24.4 N/mm with tunnel expansion.
compared with 21.2 N/mm for the EndoButton (Brown etal., unpublished data, 1996). In cyclic biomechanical test- ing, both the Trans-Fix (238 N/mm) and the Bone MulchScrew (257 N/mm) possessed stiffness superior to the En- A biomechanical study in young human cadavers found doButton linked with either the Endotape (183 N/mm) or that a hamstring tendon construct fixed with an EndoBut- the continuous loop (179 N/mm). The Trans-Fix (1042 N) ton and a tibial post failed at 612 N ⫾ 73 N compared with and the Bone Mulch Screw (978 N) were stronger to fail- 416 N ⫾ 66 N in the patellar tendon group with interfer- ure than the Endobutton linked with Endotape (644 N), ence fixation.74 The stiffness did not significantly vary but the highest level of failure was reported with the between groups. It was commented in the study that ei- EndoButton linked with a continuous loop (1342 N) ther construct was only 20% to 30% of the failure strength (Brown et al., unpublished data, 1999) (Table 9). In addi- of the native ACL (2195 N ⫾ 427 N) (Table 9).74 Direct tion to a favorable failure strength and stiffness, transfix- biomechanical comparison between EndoButton linked ion devices may allow independent tensioning of the four with a continuous loop and linked with Endotape revealed strands of the quadrupled hamstring tendon. In a labora- similar stiffness data, but a much higher failure with the tory study, this resulted in a statistically significant in- continuous loop, 1345 N versus 644 N for the Endotape Femoral Fixation Options for a Soft Tissue Graft in a Bone Tunnela,b QHT with Trans-Fixc Anterior drawer to Cross-pin toggled graft slipped off, tibial fixation failure QHT with Bone Mulchc Anterior drawer to Tibial fixation failure, implant QHT with an EndoButton, Anterior drawer to mersilene tapec QHT with EndoButton Anterior drawer to Tape broke, tibial fixation failure, and Endotapec tendon failure, implant pulled QHT with EndoButton Anterior drawer to Implant pulled through bone, and three #5 suturec tibial fixation failure, suturefailure, tendon failure QHT with EndoButton Anterior drawer to Tibial fixation failure, implant pulled through the bone, tape Semitendinosus fixed with Anterior drawer to the EndoButton and QHT with Mitekc Anterior drawer to Implant pulled through bone Femur only, parallel to No stiffness reported Failed by graft slipping QHT with BioScrew18 Femur only, parallel to No stiffness reported Failed by graft slipping QHT BioScrew, 0.5 mm Femur only, parallel to No stiffness reported Failed by graft slipping a QHT, quadrupled hamstring graft.
b The standards deviations are reported in parentheses following the mean.
c From Brown et al., unpublished data, 1996.
Vol. 28, No. 5, 2000 Graft Fixation in Cruciate Ligament Reconstruction linked EndoButton (Brown et al., unpublished data, ured from the time of operative fixation to later follow-up in some patients.53 The authors are considering back-up Biomechanically, the EndoButton linked with tape has fixation to the interference screw for patients with sus- motion of the graft in the tunnel of up to 3 mm under pected lower bone mineral density or with poor screw physiologic cyclic loads.38 This longitudinal motion or bun- purchase. Despite low loads found with biomechanical gee effect has been associated with tunnel expansion in testing, a recent clinical report comparing transtibial clinical trials.54,67 The natural history of tunnel expan- hamstring and patellar tendon graft interference screw sion is undetermined at present, but it is of obvious con- fixation found no significant difference in outcome.26 cern to surgeons using hamstring tendon fixation with Several factors exist that influence the initial fixation linked devices. Extensive tunnel expansion complicates strength of hamstring tendon grafts fixed with interfer- revision surgery because of bone loss and may jeopardize ence screws. These factors are especially important to fixation of the graft. Despite this reservation, the Endo- increasing fixation strength on the tibial site, which has Button has been a popular and clinically successful form of been considered to be the weak link of such a reconstruc- femoral hamstring tendon fixation.
tion. Initially, Morgan (unpublished data, 1994) intro-duced a bone-hamstring tendon-bone composite graft for an all-inside ACL reconstruction. In a biomechanicalstudy of this technique, Liu et al.55 found substantially Brown et al. (unpublished data, 1996) compared the Mitek lower loads and a high slippage for this bone-hamstring Anchor directly with the EndoButton in paired elderly tendon-bone composite graft compared with a bone-patel- human specimens with a quadrupled hamstring tendon lar tendon-bone graft in a porcine knees model. Shin et al.
graft. The EndoButton was significantly stronger (618 N (unpublished data, 1996) introduced the harvest of a ham- compared with 412 N, P ⫽ 0.03), but stiffness was compa- string tendon graft with a distally attached tibial bone rable (Table 9). The Mitek Anchor failed by pulling plug, a method that has been used by Sta¨helin and through the bone.
Weiler83 for the tibial fixation of a hamstring tendon graftin an all-inside technique. In another recent biomechani- Interference Fixation cal study it was demonstrated that the harvest of a semi-tendinosus tendon graft with a distally attached bone plug The use of interference screw fixation of a multiple-looped provides similar fixation strength when compared with hamstring tendon graft has recently raised strong interest the conventional bone-tendon-bone graft fixation, given in soft tissue fixation in cruciate ligament reconstruc- that both grafts were fixed with biodegradable interfer- tion.26,32,78,82,83 The direct tendon-to-bone interference screw fixation allows an anatomic fixation close to the To enhance the direct tendon-to-bone interference fit joint line, which has been demonstrated to increase knee fixation without bone blocks, a precise match of tunnel stability and graft isometry.41,42,63,96 Additionally, an an- size to graft diameter is necessary; a recent biomechanical atomic interference fit fixation may overcome biomechani- study compared 1- and 0.5-mm tunnel sizing and found cal disadvantages of conventional extraarticular ham- that sizing the tunnels in increments of 0.5 mm increases string tendon graft fixation techniques, such as suture fixation strength significantly.86 In a separate biome- stretch-out, graft tunnel motion, and the so-called wind- chanical study investigating the effect of screw geometry shield-wiper effect.37,38,54,95,96 It has been hypothesized on hamstring tendon interference fit fixation, it was dem- that these biomechanical disadvantages may contribute to onstrated that by increasing both screw length and screw the creation of high shearing forces at the tunnel wall, diameter, fixation strength was significantly improved. In which may also delay an osseous graft incorporation and this study the influence of screw length (23 versus 28 mm) lead to tunnel enlargement.54,67,95 was greater than that of thread diameter (screw diameter ⫽ Recent biomechanical studies compared biodegradable graft size versus screw diameter ⫽ graft size ⫹ 1 mm).93 and blunt-threaded titanium interference screws (RCI, To further ascertain the appropriateness of this new Smith & Nephew Donjoy, Carlsbad, California) for ham- technique for hamstring tendon graft fixation in cruciate string tendon interference fit fixation and found that bio- ligament surgery, it is essential to understand tendon-to- degradable and titanium screws provide similar or supe- bone healing progression with interference screw fixation.
rior fixation strength over conventional hamstring tendon In a recent animal study there was evidence that the fixation.18,86,94 In these reports, the mean failure load of healing under interference screw compression follows dif- a transtibial ACL reconstruction with hamstring tendons ferent patterns than what has been described in animal and interference screw fixation exhibited substantially models using noncompressing extraarticular fixation tech- lower loads than the estimated forces in the native ACL or niques.7,34,73,92 In this animal model, Weiler et al.92 the graft during daily activities (Refs. 2, 58, 69, 96; Brown found that the healing progresses only partially via the et al., unpublished data, 1996). Therefore, it has been development of a so-called fibrous interface (Fig. 4) This advocated by some that the initial strength of transtibial usually develops between the tendon graft and the bone hamstring tendon interference fit fixation may not allow surface. Their findings indicate that direct contact healing for an accelerated postoperative rehabilitation (Refs. 56, between the graft and the bone surface may exist if com- 96; Brown et al., unpublished data, 1996). A clinical study pression fixation is used and may also overcome the de- showed an increase in anterior tibial translation meas- layed tendon-bone healing if extraarticular fixation is used.
Brand et al. American Journal of Sports Medicine Graft fixation remains the weak link in the early postop-erative period of ligament reconstruction. Technologicaladvancements in surgical techniques have allowed for animmediate return of neuromuscular function within theextremity. Fixation must not only withstand these earlyphysiologic forces but must also facilitate biologic incorpo-ration of the graft construct in its entirety. The specificanatomic location of the attachment site will have pro-found effects on fiber recruitment patterns within theligament substitute. Fixation of a bone plug in a bonetunnel with a metal or bioabsorbable interference screwappears to meet our current demands. Present soft tissuefixation within a bone tunnel or extratunnel may notpossess the same biomechanical or biologic properties as abone in a bone tunnel fixed with an interference screw.
Devices that are linked to the graft or placed nonanatomi-cally have been associated with motion through the graftconstruct and have spurred the search for direct fixationat the joint surface.
Controversy remains as to the suitability of soft tissue fixation for progressive rehabilitation. Other fixation de-vices are used and tested—such as transfixion femoralfixation, hybrid fixation, and tibial washerplate fixa-tion—to more closely achieve the normal mechanical char-acteristics of the native ligament graft.
FUTURE DIRECTIONS Ideally, the biomechanical properties of the entire graftconstruct would approach those of the native ligament andfacilitate biologic incorporation of the graft. Fixation Figure 4. A, transversal cut of the tibial tunnel at 6 weeks
should be done at the normal anatomic attachment site of after ACL reconstruction in a sheep model. An Achilles ten- the native ligament (aperture fixation) and, over time, don split graft was directly fixed with a biodegradable poly- allow the biologic return of the histologic transition zone (D,L-lactide) interference screw. The graft is directly in contact from ligament to fibrocartilage to calcified fibrocartilage with the surrounding bone tissue. A fibrous interface between to bone. The transition from ligament to bone may occur the graft and the bone is only partially developed, suggesting without a fibrous interzone with compressive interference a different mechanism of graft incorporation as compared fixation. Manipulating the biologic environment with gene with conventional extracortical fixation techniques. At the top therapy or tissue engineering may speed graft incorpora- of the figure is the indentation from the biodegradable screw tion. Biodegradable screws can serve as a carrier for these in the tendon tissue (Masson Goldner's trichrome stain). B, substances or other growth factors to aid in graft incorpo- fluorescence microscopy of a specimen at 9 weeks. There is ration in the bone tunnel and fill the bone defect that may an intensive fluorochrome activity at the interface between be left by biodegradable screw absorption.
the graft and the bone tissue, indicating an early closure of Diminishing individual fiber movement within the ten- the fibrous interface. The green stain (calcein green) was don graft and the elimination of linkage materials will administered at 1 week, the yellow stain (tetracycline) was improve future soft tissue fixation. Combination of fixa- administered at 5 weeks.
tion devices, for example, use of a screw and washer and abiodegradable interference screw directly against a ten-don graft, or "hybrid fixation," may be useful in the inter- When using biodegradable interference screw fixation mediate future. Biodegradable bone cement that allows for a soft tissue graft, there are concerns about a possible for immediate fixation of the graft and eventual replace- compromise of the graft incorporation when the screw ment with normal osseous tissue may be developed. This degrades. In the model of Weiler et al.,92 an intermediate biodegradable bone cement may need to be combined with degrading poly-(D,L-lactide) interference screw was used; current graft fixation choices until it cures and achieves it disintegrated macroscopically at 24 weeks. At this time, maximum strength. Fixation that allows immediate and no graft pull-out from the tunnel was observed. This indi- secure fixation will aid rehabilitation, hasten return of cates that screw degradation may not compromise graft muscle tone and force, and benefit patient outlook.
incorporation after all.
Critical evaluation of patient satisfaction through out- Vol. 28, No. 5, 2000 Graft Fixation in Cruciate Ligament Reconstruction come-based research received recent emphasis at the 66th 17. Butler JC, Branch TP, Hutton WC: Optimal graft fixation. The effect of gap size and screw size on bone plug fixation in ACL reconstruction. Arthros- annual meeting of the Academy of Orthopaedic Surgeons copy 10: 524 –529, 1994 (D. W. Jackson, unpublished data, 1999). The relationship 18. Caborn DNM, Coen M, Neef R, et al: Quadrupled semitendinosus-gracilis between less stiff and less strong graft fixation that is autograft fixation in the femoral tunnel. A comparison between a metal anda bioabsorbable interference screw. Arthroscopy 14: 241–245, 1998 currently available, the interplay with rehabilitative ef- 19. Caborn DNM, Urban WP Jr, Johnson DL, et al: Biomechanical compari- forts, and laxity of the reconstructed knee has not been son between BioScrew and titanium alloy interference screws for bone- established. At present, there is not a strong clinical as- patellar tendon-bone graft fixation in anterior cruciate ligament reconstruc-tion. Arthroscopy 13: 229 –232, 1997 sociation between fixation that performs well in labora- 20. Casteleyn PP, Handelberg F, Haentjens P: Biodegradable rods versus tory testing and objective knee stability. If this association Kirschner wire fixation of wrist fractures. A randomised trial. J Bone Joint is proven, the clinical association of laxity to clinical out- Surg 74B: 858 – 861, 1992 21. Champion AR, Cutshall TA, Van Sickle DC: In vitro and vivo evaluation of come and patient satisfaction can be investigated. Pres- a bioresorbable interference screw. Trans Orthop Res Soc 20: 638, 1995 ently, there are few clinical studies directed to these is- 22. Claes LE: Mechanical characterization of biodegradable implants. Clin sues. Comparative studies of different modes of graft Mater 10: 41– 46, 1992 23. Clancy WG Jr, Narechania RG, Rosenberg TD, et al: Anterior and pos- fixation will be important in this effort. Longitudinal stud- terior cruciate ligament reconstruction in rhesus monkeys. A histological ies, although difficult and fraught with methods problems, microangiographic and biomechanical analysis. J Bone Joint Surg 63A: will establish the relationship between patient satisfac- 1270 –1284, 1981 24. Clark R, Olsen RE, Larson BJ, et al: Cross-pin femoral fixation: A new tion, residual laxity, and degenerative arthrosis.
technique for hamstring anterior cruciate ligament reconstruction of theknee. Arthroscopy 14: 258 –267, 1998 25. Cooper DE, Deng XH, Burstein AL, et al: The strength of the central third patellar tendon graft: A biomechanical study. Am J Sports Med 21: 818 –824, 1993 Thank you to Paula Hurley, MS Ed, and Bruce Pohlig for 26. Corry IS, Webb JM, Clingeleffer AJ, et al: Arthroscopic reconstruction of the anterior cruciate ligament—A comparison of patellar tendon autograft editorial assistance. A large number of researchers have and four-strand hamstring tendon autograft. Am J Sports Med 27: 444 – contributed to the study of graft fixation who are not recognized in this article because of space constraints. We 27. Daniels AU, Chang MK, Andriano KP: Mechanical properties of biode- gradable polymers and composites proposed for internal fixation of bone.
apologize for these omissions.
J Appl Biomater 1: 57–78, 1990 28. Dworsky BD, Jewell BF, Bach BR Jr: Interference screw divergence in endoscopic anterior cruciate ligament reconstruction. Arthroscopy 12: 29. Fanelli GC, Desai BM, Cummings PD, et al: Divergent alignment of the 1. Abate JA, Fadale PD, Hulstyn MJ, et al: Initial fixation strength of polylactic femoral interference screw in single incision endoscopic reconstruction of acid interference screws in anterior cruciate ligament reconstruction. Ar- the anterior cruciate ligament. Contemp Orthop 28(1): 21–25, 1994 throscopy 14: 278 –284, 1998 30. Fellinger M, Passler J, Wilburger R, et al: Komplikationen und mogliche 2. Aune AK, Ekeland A, Cawley PW: Interference screw fixation of hamstring Fehlerquellen bei der Anwendung von interferenzschrauben in der arthro- vs patellar tendon grafts for anterior cruciate ligament reconstruction.
skopischen Kreuzbandchirurgie. Arthroskopie: 33–38, 1993 Knee Surg Sports Traumatol Arthrosc 6: 99 –102, 1998 31. Frank A: Arthroscopic ACL reconstruction with patellar tendon. Absorb- 3. Barber FA, Elrod BF, McGuire DA, et al: Preliminary results of an absorb- able interference screw versus metal screws. Proc Europ Soc Sports able interference screw. Arthroscopy 11: 537–548, 1995 Traumatol Knee Surg Arthrosc, 8th Congress, Nice, France, 1998 4. Barrett GR, Papendick L, Miller C: EndoButton button endoscopic fixation 32. Fu FH, Bennett CH, Ma B, et al: Current trends in anterior cruciate technique in anterior cruciate ligament reconstruction [Technical Note].
ligament reconstruction. Part II: Operative procedures and clinical corre- Arthroscopy 11: 340 –343, 1995 lations. Am J Sports Med 28: 124 –130, 2000 5. Berg EE: Autograft bone-patella tendon-bone plug comminution with loss 33. Gerich TG, Cassim A, Lattermann C, et al: Pullout strength of tibial graft of ligament fixation and stability [abstract]. Arthroscopy 12: 232–235, 1996 fixation in anterior cruciate ligament replacement with a patellar tendon 6. Bergsma JE, de Bruijn WC, Rozema FK, et al: Late degradation tissue graft: Interference screw versus staple fixation in human knees. Knee response to poly(L-lactide) bone plates and screws. Biomaterials 16: Surg Sports Traumatol Arthrosc 5: 84 – 89, 1997 34. Grana WA, Egle DM, Mahnken R, et al: An analysis of autograft fixation 7. Blickenstaff KR, Grana WA, Egle D: Analysis of a semitendinosus au- after anterior cruciate ligament reconstruction in a rabbit model.
tograft in a rabbit model. Am J Sports Med 25: 554 –559, 1997 Am J Sports Med 22: 344 –351, 1994 8. Bo¨stman OM: Intense granulomatous inflammatory lesions associated 35. Hamner DL, Brown CH Jr, Steiner ME, et al: Hamstring tendon grafts for with absorbable internal fixation devices made of polyglycolide in anklefractures. Clin Orthop 278: 193–199, 1992 reconstruction of the anterior cruciate ligament: Biomechanical evaluation 9. Bo¨stman OM, Pihlajama¨ki HK, Partio EK, et al: Clinical biocompatibility of the use of multiple strands and tensioning techniques. J Bone Joint and degradation of polyevolactide screws in the ankle. Clin Orthop 320: Surg 81A: 549 –557, 1999 36. Hoffmann R, Krettek C, Hetkamper A, et al: Osteosynthesis of distal 10. Boszotta H: Arthroscopic anterior cruciate ligament reconstruction using a radius fractures with biodegradable fracture rods. Results of two years patellar tendon graft in press-fit technique: Surgical technique and follow- follow-up [in German]. Unfallchirurg 95: 99 –105, 1992 up. Arthroscopy 13: 332–339, 1997 37. Hoffmann R, Weiler A, Helling HJ, et al: Local foreign-body reactions to 11. Brand JC, Caborn P, Steenlage E, et al: Interference screw fixation biodegradable implants. A classification [in German]. Unfallchirurg 100: strength of a quadrupled hamstring tendon is directly correlated to bone 658 – 666, 1997 mineral density measured by dual photon absorptometry (DEXA). Arthros- 38. Hoher J, Livesay GA, Ma CB, et al: Hamstring graft motion in the femoral copy 15: 58, 1999 bone tunnel when using titanium button/polyester tape fixation. Knee Surg 12. Brand JC, Danaceau S, Hamilton D, et al: Comparison of interference Sports Traumatol Arthrosc 7: 215–219, 1999 fixation of tendon and bone plug for the quadriceps tendon in cruciate 39. Hulstyn M, Fadale PD, Abate J, et al: Biomechanical evaluation of inter- ligament reconstruction. Arthroscopy 15: 58, 1999 ference screw fixation in a bovine patellar bone-tendon-bone autograft 14. Brown CH Jr, Hecker AT, Hipp JA, et al: The biomechanics of interference complex for anterior cruciate ligament reconstruction. Arthroscopy 9: 417– screw fixation of patellar tendon anterior cruciate ligament grafts.
Am J Sports Med 21: 880 – 886, 1993 40. Imhoff A, Burkart A, Roscher E: Foreign-body reaction to the bioabsorb- 15. Brown G, Pen˜a F, Grøntvedt T, et al: Fixation strength of interference able suretac device. Arthroscopy 16: 91–95, 2000 screw fixation in bovine, young human, and elderly human cadaver knees: 41. Ishibashi Y, Rudy TW, Livesay GA, et al: The effect of the anterior cruciate Influence of insertion torque, tunnel-bone block gap, and interference.
ligament graft fixation site at the knee on knee stability: Evaluation using Knee Surg Sports Traumatol Arthrosc 3: 238 –244, 1996 a robotic testing system. Arthroscopy 13: 177–182, 1997 16. Bucholz RW, Henry S, Henley MB: Fixation with bioabsorbable screws for 42. Johnson DH, Houle JB, Almazan A: Comparison of intraoperative AP the treatment of fractures of the ankle. J Bone Joint Surg 76A: 319 –324, translation of two different modes of fixation of the grafts used in ACL reconstruction [abstract]. Arthroscopy 14: 425, 1998 Brand et al. American Journal of Sports Medicine 43. Johnson DP: Operative complications from the use of biodegradable 72. Race A, Amis AA: The mechanical properties of the two bundles of the Kurosawa screws. J Bone Joint Surg 80B (Suppl 1): 103, 1998 human posterior cruciate ligaments. J Biomech 27: 13–24, 1994 44. Johnson LL, vanDyk GE: Metal and biodegradable interference screws: 73. Rodeo SA, Arnoczky SP, Torzilli PA, et al: Tendon-healing in a bone Comparison of failure strength. Arthroscopy 12: 452– 456, 1996 tunnel. A biomechanical and histological study in the dog. J Bone Joint 45. Jomha NM, Raso VJ, Leung P: Effect of varying angles on the pullout Surg 75A: 1795–1803, 1993 strength of interference screw fixation. Arthroscopy 9: 580 –583, 1993 74. Rowden NJ, Sher D, Rogers GJ, et al: Anterior cruciate ligament graft 46. Kohn D, Rose C: Primary stability of interference screw fixation: Influence fixation: Initial comparison of patellar tendon and semitendinosus au- of screw diameter and insertion torque. Am J Sports Med 22: 334 –338, tografts in young fresh cadavers. Am J Sports Med 25: 472– 478, 1997 75. Rupp S, Krauss PW, Fritsch EW: Fixation strength of a biodegradable 47. Kousa P, Ja¨rvinen TLN, Pohjonen T, et al: Fixation strength of a biode- interference screw and a press-fit technique in anterior cruciate ligament gradable screw in anterior cruciate ligament reconstruction. J Bone Joint reconstruction with a BPTB graft. Arthroscopy 13: 61– 65, 1997 Surg 77B: 901–905, 1995 76. Safran MR: Graft selection in knee surgery [current concepts]. Am J Knee 48. Kurosaka M, Yoshiya S, Andrish JT: A biomechanical comparison of Surg 8: 168 –180, 1995 different surgical techniques of graft fixation anterior cruciate ligament 77. Safran MR, Harner CD: Technical considerations of revision anterior reconstruction. Am J Sports Med 15: 225–229, 1987 cruciate surgery. Clin Orthop 325: 50 – 65, 1996 49. Kurzweil PR, Frogameni AD, Jackson DW: Tibial interference screw re- 78. Scranton PE Jr, Pinczewski L, Auld MK, et al: Outpatient endoscopic moval following anterior cruciate ligament reconstruction. Arthroscopy 11: quadruple hamstring anterior cruciate ligament reconstruction. Oper Tech Orthop 6: 177–180, 1996 50. Lambert KL: Vascularized patellar tendon graft with rigid internal fixation 79. Shelbourne KD, Gray T: Anterior cruciate ligament reconstruction with for anterior cruciate ligament insufficiency. Clin Orthop 172: 85– 89, 1983 autogenous patellar tendon graft followed by accelerated rehabilitation. A 51. Lemos MJ, Albert J, Simon T, et al: Radiographic analysis of femoral two- to nine-year followup. Am J Sports Med 25: 786 –795, 1997 interference screw placement during ACL reconstruction: Endoscopic ver- 80. Shellock FG, Mink JH, Curtin S, et al: MR imaging and metallic implants sus open technique. Arthroscopy 9: 154 –158, 1993 for anterior cruciate ligament reconstruction: Assessment of ferromag- 52. Levine RE, Simonian PT, Wright TM, et al: Cyclic creep response of netism and artifact. J Magn Reson Imaging 2: 225–228, 1992 hamstring and patellar tendon ACL grafts. Arthroscopy 14: 436, 1998 53. Liew A, Johnson D: Bioabsorbable interference screw fixation of ham- 81. Shino K, Nakagawa S, Nakamura N, et al: Arthroscopic posterior cruciate string graft in ACL reconstruction. Arthroscopy 15: 547–548, 1999 ligament reconstruction using hamstring tendons: One-incision technique 54. L'Insalata JC, Klatt BF, Fu FH, et al: Tunnel expansion following anterior with EndoButton [technical note]. Arthroscopy 12: 638 – 642, 1996 cruciate ligament reconstruction: A comparison of hamstring and patellar 82. Stahelin AC, Feinstein R, Friederich NF: Clinical experience using bioab- tendon autografts. Knee Surg Sports Traumatol Arthrosc 5: 234 –238, sorbable interference screw for ACL reconstruction. Orthop Trans 19: 55. Liu SH, Kabo JM, Osti L: Biomechanics of two types of bone-tendon-bone 83. Sta¨helin AC, Weiler A: All-inside anterior cruciate ligament reconstruction graft for ACL reconstruction. J Bone Joint Surg 77B: 232–235, 1995 using semitendinosus tendon and soft threaded biodegradable interfer- 56. Magen HE, Howell SM, Hull ML: Structural properties of six tibial fixation ence screw fixation [technical note]. Arthroscopy 13: 773–779, 1997 methods for anterior cruciate ligament soft tissue grafts. Am J Sports Med 84. Sta¨helin AC, Weiler A, Ru¨fenacht H, et al: Clinical degradation and bio- 27: 35– 43, 1999 compatibility of different bioabsorbable interference screws: A report of six 57. Malek MM, DeLuca JV, Verch DL, et al: Arthroscopically assisted ACL cases. Arthroscopy 13: 238 –244, 1997 reconstruction using central third patellar tendon autograft with press fit 85. Sta¨ubli HU, Schatzmann L, Brunner P, et al: Quadriceps tendon and femoral fixation. Instr Course Lect 45: 287–295, 1996 patellar ligament: Cryosectional anatomy and structural properties in 58. Markolf KL, Burchfield DM, Shapiro MM, et al: Biomechanical conse- young adults. Knee Surg Sports Traumatol Arthrosc 4: 100 –110, 1996 quences of replacement of the anterior cruciate ligament with a patellar 86. Steenlage E, Brand JC, Caborn D, et al: Interference screw fixation of a ligament allograft. Part II: Forces in the graft compared with forces in the quadrupled hamstring graft is improved with precise match of tunnel to intact ligament. J Bone Joint Surg 78A: 1728 –1734, 1996 graft diameter. Arthroscopy 15: 59, 1999 59. Markolf KL, Slauterbeck JR, Armstrong KL, et al: A biomechanical study of 87. Steiner ME, Hecker AT, Brown CH Jr, et al: Anterior cruciate ligament replacement of the posterior cruciate ligament with a graft. Part II: Forces graft fixation: Comparison of hamstring and patellar tendon grafts.
in the graft compared with forces in the intact ligament. J Bone Joint Surg Am J Sports Med 22: 240 –247, 1994 79A: 381–386, 1997 88. Uribe JW, Hechtman KS, Zvijac JE, et al: Revision anterior cruciate 60. Marti C, Imhoff AB, Bahrs C, et al: Metallic versus bioabsorbable interfer- ligament surgery: Experiences from Miami. Clin Orthop 325: 91–99, 1996 ence screw for fixation of bone-patellar tendon-bone autograft in arthro- 89. Vuori I, Heinonen A, Sievanen H, et al: Effects of unilateral strength scopic anterior cruciate ligament reconstruction: A preliminary report.
training and detraining on bone mineral density and content in young Knee Surg Sports Traumatol Arthrosc 5: 217–221, 1997 women. A study of mechanical loading and deloading on human bones.
61. Matthews LS, Soffer SR: Pitfalls in the use of interference screws for Calcif Tissue Int 55: 59 – 67, 1994 anterior cruciate ligament reconstruction: Brief report. Arthroscopy 5: 225– 90. Walton M, Cameron M: Efficacy of an absorbable interference screw for graft fixation anterior cruciate ligament reconstruction: A study using a 62. Matthews LS, Yahiro MA, Lawrence SJ, et al: Fixation strengths of bone- sheep model. J Bone Joint Surg 78B (Suppl 2 & 3): 126, 1996 patellar tendon-bone grafts [abstract]. Am J Sports Med 18: 556 –557, 91. Weiler A, Helling H-J, Kirch U, et al: Foreign-body reaction and the course of osteolysis after polyglycolide implants for fracture fixation. Experimental 63. Morgan CD, Kalmam VR, Grawl DM: Isometry testing for anterior cruciate study in sheep. J Bone Joint Surg 78B: 369 –376, 1996 ligament reconstruction revisited. Arthroscopy 11: 647– 659, 1995 92. Weiler A, Hoffman RFG, Bail HJ, et al: Tendon healing in a bone tunnel.
64. Morrison JB: The mechanics of the knee joint in relation to normal walking.
Histological analysis after biodegradable interference fit fixation. Arthros- J Biomech 3: 51– 61, 1970 copy 15: 548 –549, 1999 65. Morrison JB: Function of the knee joint in various activities. Biomed Eng 4: 93. Weiler A, Hoffmann RFG, Siepe CJ, et al: The influence of screw geom- etry on hamstring tendon interference fit fixation. Am J Sports Med 28: 66. Morrison JB: Bioengineering analysis of force actions transmitted by the knee joint. Biomed Eng (April): 164, 1968 67. Nebelung W, Becker R, Merkel M, et al: Bone tunnel enlargement after 94. Weiler A, Hoffmann RFG, Sta¨helin AC, et al: Hamstring tendon fixation anterior cruciate ligament reconstruction with semitendinosus tendon us- using interference screws. A biomechanical study in calf tibial bone.
ing EndoButton fixation on the femoral side. Arthroscopy 14: 810 – 815, Arthroscopy 14: 29 –37, 1998 95. Weiler A, Hoffmann RFG, Sudkamp NP, et al: Replacement of the anterior 68. Noyes FR, Barber-Westin SD: Revision anterior cruciate ligament sur- cruciate ligament. Biomechanical study for patellar and semitendinosus gery: Experiences from Cincinnati. Clin Orthop 325: 116 –129, 1996 tendon fixation with poly(D,L-lactide) interference screw [in German].
69. Noyes FR, Butler DL, Grood ES, et al: Biomechanical analysis of human Unfallchirurg 102: 115–123, 1999 ligament grafts used in knee-ligament repairs and reconstructions. J Bone 96. Weiler A, Scheffler S, Go¨ckenjan A, et al: Different hamstring tendon graft Joint Surg 66A: 344 –352, 1984 fixation techniques under incremental cyclic loading conditions [abstract].
70. Pena F, Gro¨ntvedt T, Brown GA, et al: Comparison of failure strength Arthroscopy 14: 425– 426, 1998 between metallic and absorbable interference screws. Influence of inser- 97. Weiler A, Windhagen HJ, Raschke MJ, et al: Biodegradable interference tion torque, tunnel-bone block gap, bone mineral density, and interfer- screw fixation exhibits pull-out force and stiffness similar to titanium ence. Am J Sports Med 24: 329 –334, 1996 screws. Am J Sports Med 26: 119 –128, 1998 71. Pistner H, Bendix RR, Muhling J, et al: Poly(L-Lactide): A long term study 98. Woo SL-Y, Hollis JM, Adams DJ, et al: Tensile properties of the human in vivo. Part III. Analytical characterization. Biomaterials 14: 291–298, femur-anterior cruciate ligament-tibia complex: The effects of specimen age and orientation. Am J Sports Med 19: 217–225, 1991


Modeling the Invasion of Community-AcquiredMethicillin-Resistant Staphylococcus aureus intoHospitals Erica M. C. D'Agata,1 Glenn F. Webb,2 Mary Ann Horn,2,3 Robert C. Moellering, Jr.,1 and Shigui Ruan4 1Division of Infectious Diseases, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts; 2Department ofMathematics, Vanderbilt University, Nashville, Tennessee; 3Division of Mathematical Sciences, National Science Foundation, Arlington, Virginia;and 4Department of Mathematics, University of Miami, Coral Gables, Florida