Biomaterials 21 (2000) 2623}2629

Bioscrew "xation of patellar tendon autografts

F. Alan Barber *, Burton F. Elrod
, David A. McGuire, Lonnie E. Paulos
FACS, Plano Orthopedic and Sports Medicine Center, 5228 West Plano Parkway, Plano, TX 75093, USA

Southern Sports Medicine & Orthopaedic Center, Nashville, TN, USA
Knee and Arthroscopic Surgery, Anchorage, Al, USA
Orthopedic Specialty Hospital, Salt Lake City, UT, USA

Abstract

Biodegradable interference screw "xation was studied using patellar tendon autografts in a randomized, prospective multicenter
comparison of the PLLA Bioscrew and a metal interference screw. sixty-eight of 114 (60%) patients (34 Bioscrew and 34 metal screw)
enrolled were available for follow up an average of 35 months after surgery (range 24}65). There were 42 males and 26 females with an
average age of 29 years (16}50). Tourniquet times and associated surgical "ndings were similar for the two groups. Postoperative KT,
Tegner, and Lysholm scores were not statistically di!erent between the two groups. Average follow up KT maximum manual
side-to-side di!erence was 0.9 mm for Bioscrews and 1.4 mm for metal screws. Postoperative Lachman and pivot shift testing were
equivalent for the two groups. Other than six of the 7 mm Bioscrews breaking during insertion in the femoral tunnel, there were no
problems related to the PLLA implants. No statistical di!erence was found between the Bioscrew and the metal screw groups, and
there were no osteolytic or other reactive bone changes observed associated with the PLLA Bioscrew. These data indicate that the
PLLA Bioscrew functions reliably as an interference "xation screw for patellar tendon autografts.  2000 Elsevier Science Ltd.
All rights reserved.

Keywords: Interference "xation screw; Bioscrew; Polylactic acid; Patellar tendon autograft

1. Introduction Interference screws are commonly used with patellar

Arthroscopically assisted anterior cruciate ligament
reconstruction is commonly performed using a patellar
tendon autograft [1}3] selected because of its inherent
strength, availability, and the secure "xation achieved by
the associated bone plugs. Good initial "xation of the
properly positioned and tensioned graft is critical, but the
importance of this initial "xation decreases as the graft is
incorporated into the adjacent bone tunnels. Rodeo [4]
demonstrated that this process is substantially complete
by 6 weeks for those grafts with bone plugs (patellar
tendon and quadriceps) and 12 weeks for tendons. Recent
evidence from [5] suggests that good outlet "xation of
soft tissue grafts using interference screws can be as
short as 6 weeks. Regardless of the "xation method, once
graft incorporation is complete, the "xation device is
unnecessary.
* Corresponding author. Tel.:#1-972-250-5700; fax:#1-972-250-5747.
tendon autografts but have some technical problems.
Noncannulated interference screws can diverge or over
penetrate into the joint [6], and any metal screw can
lacerate the graft [7]. A retained interference screw
can complicate revision surgery, degrade magnetic
images, and make radiographic assessment of graft
placement di$cult. Graft "xation devices should
not injure the graft, diverge, loosen, and become promin-
ent. Ideally, the device should &disappear' when it is
no longer needed. A bioabsorbable screw avoids con-
cerns associated with metal implants including problems
with postoperative imaging, graft laceration, cold in-
tolerance, complication with any revision surgery, and
problems associated with retained metal in a subsequent
infection.
The Bioscrew is a bioabsorbable interference screw
(Fig. 1) made from poly L-lactic acid (PLA). This report
reviews the results of a prospective randomized study of
the e$cacy of the Bioscrew compared to a standard
metal interference screw and discusses the biology of
PLLA and clinical experience with PLLA interference
"xation screws for patellar tendon autografts.

0142-9612/00/$ - see front matter  2000 Elsevier Science Ltd. All rights reserved.
PII: S 0 1 4 2 - 9 6 1 2 ( 0 0 ) 0 0 1 3 0 - 7
2624 F.A. Barber et al. / Biomaterials 21 (2000) 2623}2629
Fig. 1. The biodegradable Bioscrew is made of poly L-lactic acid.
2. Methods
A randomized, prospective study compared patella
tendon autografts interference "xed using either the Bio-
screw or a titanium interference screw (Linvatec, Largo,
FL). The inclusion criteria were unilateral knee instabil-
ity, positive Lachman's and positive pivot shift tests, KT
maximum manual side-to-side di!erences greater than
3 mm, a minimum age of 16 years with nearly closed knee
growth plates, adequate bone density, compliance with
the study protocol, and a commitment for at least two
years follow up. The exclusion criteria were active infec-
tion, a history of blood supply limitations and/or pre-
vious infections that might retard healing, PCL tears,
and prior knee ligament surgery. Randomization was
accomplished by using sealed envelope opened after ac-
ceptance into the study that revealed the patient's assign-
ment to either the Bioscrew group or the metal screw
group.
Preoperative assessment included history, physical ex-
amination, radiographs, and preoperative Lysholm knee
scoring scale [8] and Tegner [9] activity level scales. KT
measurements were obtained. The surgical "ndings and
procedures were recorded. All patients underwent an
arthroscopic patellar tendon autograft ACL reconstruc-
tion "xed by interference screws.
Postoperative assessments included Lysholm and
Tegner scales, meniscal tests, ligament tests, KT tests,
and knee radiographs. Statistical comparisons of cat-
egorical variables were performed using Fisher's exact
test while continuous variables were compared by the
Student's t-test. All data summaries and statistical com-
parisons were performed by statistical analysis system
(SAS) software. Statistical signi"cance was assigned at
p(0.05.
The Bioscrew is made from puri"ed poly L-lactide. The
screws used in this study were 7 or 9 mm in diameter and
either 20 or 25 mm long. Currently available Bioscrews
also come in 8 mm diameters. These cannulated screws
slide over a Nitinol guide wire and are inserted by a tri-
angular #uted screwdriver. The screw pitch is 10 threads
per inch and a tapering screw tip. The "rst entrance angle
is 303 and the second angle is 153. Both metal and
biodegradable thread thickness are comparable. Screw
lengths and diameters were also the same. The threads
are buttressed on the leading edge and tapered on the
trailing edge. The current version of the 7 mm Bioscrew
has an increased core thickness (minor diameter) that
reduces the risk of screw breakage. The minor core dia-
meters of the 7, 8 and 9 mm screws are now 0.192, 0.202,
and 0.222 in, respectively. The major diameters for the 7,
8, and 9 mm Bioscrews are 0.282, 0.315, and 0.358 in,
respectively.
2.1. Surgical technique
A central third patellar tendon autograft was har-
vested through an anterior incision and prepared for
insertion. A notchplasty cleared any impinging bone
from the superior and lateral intercondylar notch. After
a tibial drill guide located the tibial tunnel site, a tibial
guide wire was advanced into position and then the
tunnel was created using a cannulated drill (usually
10 mm) over the guide wire. A transtibial aiming guide
was passed into position at the posterior superior inter-
condylar notch and the femoral guide wire was passed
through the femur. The transtibial guide was removed
and a cannulated reamer advanced over the guidewire to
create the femoral tunnel. The slotted femoral guide wire
was threaded with the heavy sutures attached to the
patellar tendon autograft and pulled up through the
femur delivering the sutures out the side of the thigh.
These sutures then advanced the autograft into position.
The autograft was "xed into position with the selected
interference screw "rst in the femur, and then, after
proper tensioning of the graft, in the tibia. Seven-mil-
limeter diameter interference screws were usually selected
for femoral "xation and 9 mm screws for the tibia. In
a few cases, an outside}in method through a second
incision was used to create the femoral tunnel and facili-
tate graft passage. This technique was phased out as
more experience was gained with the single incision tech-
nique.
Postoperative management emphasized achieving
and maintaining full extension, progressive weight bear-
ing with the elimination of crutches within the "rst
two weeks postoperatively, and a rapid return to normal
daily activities. CPM and ice therapy and knee exten-
sion bracing was used as well. Open-chain exercises
were avoided and the patients were instructed to
avoid knee extension against resistance in the ter-
minal thirty degrees for six months. Closed-chain
exercises including stationary biking, stair climbers, and
ski machines were allowed. Jogging began at two
months, noncontact pivoting at three months, and, with
a derotational brace, unlimited contact sports at four
months.
 F.A. Barber et al. / Biomaterials 21 (2000) 2623}2629
3. Results
For this review, 68 of a total of 114 (60%) patients who
underwent a patellar tendon autograft reconstruction
had complete follow-up data at a minimum of 24 months
(34 Bioscrew, 34 metal). Average follow-up was 35
months (range, 24}65 months) with 42 males and 26
females. The average Bioscrew group age was 29 years
(range, 16}50 years) and the average metal screw group
age was 30 years (range, 16}46 years). Tourniquet times
for both groups averaged 67 min. Additional surgical
"ndings (such as chondromalacia and meniscal tears)
were similar for both groups. No demographic di!er-
ences existed between these two groups including age,
height, weight, race, sex, side of dominance, number of
subluxation episodes, or side of injury. No statistically
signi"cant di!erences were present between the two
groups, and no di!erences were found related to com-
plications, motion loss, or results at follow-up based on
the time from injury to surgery.
Only one interference "xation screw (Bioscrew or
metal screw) was placed in the femoral tunnel to secure
the femoral side of the graft. In the Bioscrew group,
a single tibial tunnel Bioscrew was used except in one case
that required two &nested' Bioscrews. In the metal group
a single metal interference screw was used as well except for
three cases in which staples were used for tibial "xation.
Tegner scores for the Bioscrew group (3.97) and metal
screw group (3.88) at "nal follow up were equivalent.
Average Lysholm scores for Bioscrews (94) and metal
screws (95) also were not statistically di!erent. The fol-
low-up Lysholm scores showed excellent (95}100) or
good (84}94) results in 94% of Bioscrews and 91% of
metal screws. Follow-up KT tests showed an average KT
maximal manual di!erence at follow-up of 0.9 mm for
Bioscrews and 1.4 mm for metal screws. Again, no statis-
tical di!erences were found.
Motion was measured with the patient supine by
a goniometer. There was no loss of extension greater than
63 in any Bioscrew or metal screw cases. One Bioscrew
and one metal screw failed to achieve at least 1203
#exion. A Lachman test was performed at follow-up and
for the Bioscrew group was absent in 19, 1# in 14, and
2#in none. The Lachman was absent for the metal screw
group in 22, 1# in 10, and 2# in 1. The pivot shift test
was negative in 27 and 1#in 6 Bioscrews and negative in
29 and 1#in 4 metal screws at follow-up. A joint e!usion
was found in 1 Bioscrew and 1 metal screw knee at follow
up. Patellofemoral crepitus was present in 9 of 34 Bio-
screws and in 11 of 34 metal screws. Patellofemoral
tenderness was present in only 3 of 34 Bioscrews and 3 of
34 metal screws at follow-up.
Radiographs at intervals up to 65 months failed to
demonstrate any lytic changes associated with the Bio-
screws. Normal osseous incorporation and maturation
were consistently observed. The Bioscrews also permitted
2625
unimpeded radiographic assessment of the graft and tun-
nel location immediately postoperative (Fig. 2). Radio-
graphs were consequently able to clearly demonstrate
bone plug incorporation into the tunnels. Over the sub-
sequent follow up period, radiographs continued to dem-
onstrate complete osseous incorporation without lytic
changes or reactive bone (Fig. 3). Magnetic resonance
imaging was possible with the Bioscrew and gave a clear
assessment of the graft status (Fig. 4). No lytic or in#am-
matory response was ever demonstrated at the Bioscrew
locations. The PLLA Bioscrew gradually reabsorbed
over the postoperative period as observed on various
incidental MRIs obtained at di!erent intervals extending
to four years.
4. Discussion
The clinical use of the absorbable Bioscrew began in
1992 and since then there have been no reports of adverse
events related to this biodegradable material used as an
interference "xation device. Reports of this clinical ex-
perience have been previously published [10,11]. Paral-
leling this clinical e!ort is an in vivo sheep study on the
Bioscrew that examined the tissue response over a four-
year period [12]. Both the human clinical experience and
the sheep histology data show that the PLLA interfer-
ence "xation screw works well with no material related
problems. PLLA has several clinical advantages: absorb-
ability, degradation to products that are nontoxic and
naturally occurring metabolites, and an initial high
mechanical strength with good strength retention over
time [13]. The gradual strength decline associated with
PLLA allows for a similar slow gradually decreased
stress shielding.
PLLA interference screws should be di!erentiated
from other bioabsorbable fracture "xation devices that
have been associated with local sterile in#ammatory
reactions in the early post implantation period [14].
Adverse reactions to absorbable implants have also oc-
curred in the shoulder [15]. These cases show intense
in#ammatory reactions and foreign-body-type giant cells
in a granulomatous pattern. The implants in questions
were made of polyglycolide (PGA). Osteolytic areas in
the ankle up to a centimeter in diameter accompanied
implant decomposition and developed 6 and 12 weeks
after implantation [16]. These "ndings are consistent
with animal studies in which in#ammatory reactions
occurred associated with polyglycolide, lactide}glycolide
copolymers, and polydioxanone implants [16]. Recently,
subcutaneous cyst formation and tibial bone tunnel en-
largement were attributed to an in#ammatory response
caused by a biodegradable screw composed of poly-D,
L-lactic acid. This material biodegrades much more
rapidly that the PLLA used for the Bioscrew, which
may contribute to this "nding [17].
2626 F.A. Barber et al. / Biomaterials 21 (2000) 2623}2629
Fig. 2. The biodegradable interference screw permits clear radiographic assessment of graft and tunnel placement.
The comparison of postoperative subjective perfor-
mance as evaluated by the Tegner activity scores showed
no di!erence between those reconstructions using metal
or Bioscrew interference "xation. Similarly, a review of
the objective data provided by physical examinations,
Lysholm scores, and KT testing showed no di!erences
between the cases in which metal screws and Bioscrews
were used. The radiographic examinations demonstrated
a progressive incorporation of the autografts into the
femoral and tibial bone tunnels without adverse occur-
rences. Careful attention was given to the Bioscrew
radiographs and no evidence of osteolysis or reaction
was found. There was no evidence of Bioscrew divergence
from the bone tunnels as was occasionally observed in
the metal screws. This absence of divergence is believed
to be the result of the less rigid and more malleable
nature of the PLA material that will conform to and
follow the drilled tunnel rather than &"nd' a new course.
No cases of graft laceration occurred in either Bioscrew
or metal group. It is believed that the &softer' nature of the
Bioscrew would make such an occurrence less likely than
with the metal screw. The Bioscrew actually compresses
slightly as it is inserted into the interference position
between the tunnel wall and the autograft bone, conform-
ing to the walls of the tunnel.
PLLA can be produced in amorphous, crystalline, and
self-reinforced forms that can be controlled by polymer
preparation and manufacture. The form or combination
clearly a!ects implant behavior. The Bioscrew is princi-
pally amorphous. Less than half of the material is crystal-
line. Amorphous implants are produced by injection
molding or extruded with the melted material rapidly
shaped after melting. The polymer in such a manufactur-
ing process is not kept at a high temperature for pro-
longed times. Predominately crystalline implants are
produced when the amorphous PLLA polymer is an-
nealed at a high temperature (above the glass transition
temperature) to obtain a higher yield strength and higher
modulus (than the amorphous PLLA). Self-reinforced
material is created when the PLLA polymer is injected or
extruded (&orientruded') into a rod that is then drawn
along a single axis imparting a molecular orientation to
the polymer. This results in very high tensile strength and
tensile modulus, and creates a highly oriented crystal-
line/semicrystalline composite structure [18]. The cry-
stalline content of implants ranges from 15 to 74% for
 F.A. Barber et al. / Biomaterials 21 (2000) 2623}2629
Fig. 3. The AP (a) and lateral (b) radiographs 50 months after surgery show no lytic change or reactive bone formation. Complete osseous
incorporation and Bioscrew absorption occurred.
PLLA [19}21] and can be 0% or completely amorphous
when there is a combination of levo and dextro stereo-
isomers as in PDLLA [20].
While the initial material con"guration is signi"cant,
in the biologic environment the amorphous crystalline
ratio is in constant #ux and the crystallinity percentage
changes [22]. Even implants with very low crystallinity
demonstrate an increase over time to become a very high
crystalline structure prior to absorption. Degradation
induced crystallization and increased mobility of the
polymeric fragments allow the fragments to rearrange
themselves spatially and enhance the crystallinity [22].
At the same time, there is a marked drop in the molecular
weight. Based upon this information, it is presumed that
the Bioscrew undergoes such a process with declining
molecular weight at the same time as the amorphous
crystalline ratio changes and degradation proceeds.
Poly L-lactic acid implants degrade principally by hy-
drolysis. The lactic acid generated is incorporated in the
tricarboxylic acid cycle and excreted by the lungs as
carbon dioxide and water [23]. The six-month half-life of
this polymer contrasts sharply to the seven to fourteen
2627
day half-life of some copolymers of polylactic acid and
polyglycolic acid. The slow hydrolysis of the Bioscrew
was not associated with any localized osteolysis or other
appreciable reaction. The osteolytic phenomena asso-
ciated with other polymers, when present, appear be-
tween 6 and 12 weeks after implantation [24].
The short-term changes of the PLLA Bioscrew up to
12 weeks are minimal. During this period the bone plugs
of the patellar tendon autograft become completely in-
corporated into the femoral and tibial tunnels. This is
supported by the clinical experience that demonstrated
no di!erence between the Bioscrew and metal screw
implant sites. The in vivo data showed that no lytic
change occurred and the PLLA screw functioned as if it
were histologically inert [12]. There were no adverse
e!ects on the tendon or surrounding tissue.
Final Bioscrew degradation occurs by four years.
Fragments of PLLA are found encapsulated by multi-
nucleated giant cells and macrophages in dense "brous
tissue surrounded by a sclerotic rim. The associated
lymph nodes contain multinucleated giant cells and clus-
ters of macrophages with an amorphous eosinophilic
2628 F.A. Barber et al. / Biomaterials 21 (2000) 2623}2629
Fig. 4. MRI at 14 months post surgery demonstrates no localized
reaction and good bone healing associated with the Bioscrew "xation.
cytoplasm containing fragments of the monomer. This
material is similar to that associated with other PLLA
degradation studies [25].
MRIs and radiographs during the follow up period
demonstrated little obvious degradation in the "rst few
months. Despite the expected ongoing hydrolysis over
time, it is felt that the intramedullary location of these
implants and the slow release of the breakdown products
eliminate the development of a clinically signi"cant in-
#ammatory process, although one probably does occur
at the cellular level. Because of the slow nature of the
degradation, the polymer load being released is limited
and the response is more restrained than that associated
with such rapidly degraded polymers such as PGA.
There were no cases of localized osteolytic activity or
other radiographically apparent reaction in our series.
The absence of such a phenomena over a follow up
period as long as 65 months, suggests that this process is
unlikely to appear in association with an implant of pure
poly-L-lactic acid.
The only adverse occurrence attributed to the Bio-
screw was insertion breakage of the 7 mm screws in the
femoral tunnel. This was in an earlier version of the screw
and later design changes increasing the core diameter,
helped resolve this situation. No 9 mm screws broke.
There was no statistical di!erence between the two clini-
cal groups studied. The PLLA Bioscrew performance
was equivalent to the metal interference "xation screw in
the patellar tendon autograft, and provides a reasonable
alternative to metal interference screws.
Acknowledgements
The authors gratefully appreciate the technical assis-
tance provided by James N. Click PA-C, Victor Gonzales
ORT, Stephen Hendricks, Suzanne Zastrow RN, Sandra
K. Starkey RN, Laura Sene!, and Ann E. Greenwald MS
as well as the statistical analysis provided by Carol A.
Weideman Ph.D.
References
[1] Brostrom L, Gillquist J, Liljedahl S-O, Lindvall N. Behandling av
invetererad ruptur av framre korsbandet. Lakrtidn 1968;64:4477.
[2] Gillquist J, Liljedahl S-O, Lindvall N. Reconstruction for old
rupture of the anterior cruciate ligament. Injury 1971;4:271}8.
[3] Alm A, Gillquist J. Reconstruction of the anterior cruciate liga-
ment by using the medial third of the patellar ligament. Acta Chir
Scand 1974;140:289}96.
[4] Rodeo SA, Arnoczky SP, Torzilli PA, Hidaka C, Warren RF.
Tendon-healing in a bone tunnel: a biomechanical and histologi-
cal study in the dog. J Bone Jt Surg 1993;75-A:1795}803.
[5] Weiler A, Ho!mann RFG, Bail HJ, Rehm O, SuK dkamp NP.
Tendon healing in a bone tunnel: histological analysis after biode-
gradable interference "t "xation. Arthroscopy 1999;15:548}9.
[6] Franco MG, Bach Jr BR, Bush-Joseph CA. Intraarticular place-
ment of Kurosaka interference screws. Arthroscopy 1994;
10:412}7.
[7] Bach Jr BR. Potential pitfalls of Kurosaka screw interference
"xation for ACL surgery. Am J Knee Surg 1989;2:76}82.
[8] Lysholm J, Gillquist J. Evaluation of knee ligament surgery re-
sults with special emphasis on use of a scoring scale. Am J Sports
Med 1982;10:150}4.
[9] Tegner Y, Lysholm J. Rating systems in the evaluation of knee
ligament injuries. Clin Orthop 1985;198:43}9.
[10] Barber FA, Elrod BF, McGuire DA, Paulos LE. Preliminary
results of an absorbable interference screw. Arthroscopy
1995;11:537}48.
[11] McGuire DA, Barber FA, Elrod BF, Paulos LE. Bioabsorbable
interference screws for graft "xation in anterior cruciate ligament
reconstruction. Arthroscopy 1999;15:463}73.
[12] Barber FA. Biology and clinical experience of absorbable mate-
rials in ACL "xation. Tech Orthop 1999;14:34}42.
[13] Jadhav BS, Tunc DC. Four year absorption study of three mor-
phologically di!erent forms of poly-L-lactide in rabbits. In:
Gebelein C, Carrahon C, editors. Biotechnology and bioactive
polymers. New York: Plenum Press, 1994. p. 169}76.
[14] Bostman OM. Intense granulomatous in#ammatory lesions asso-
ciated with absorbable internal "xation devices made of poly-
glycolide in ankle fractures. Clin Orthop 1992;278:193}9.
[15] Edwards DJ, Hoy G, Saies AD, Hayes MG. Adverse reactions to
an absorbable shoulder "xation device. J Shoulder Elbow Surg
1994;3:230}3.
 F.A. Barber et al. / Biomaterials 21 (2000) 2623}2629
[16] Tunc DC, Jadhav BS. In: Gebelein CG, Dunn RL, editors. Pro-
gress in biomedical polymers. New York: Plenum Press, 1990.
p. 239.
[17] Martinek V, Friederich NF. Tibial and pretibial cyst formation
after anterior cruciate ligament reconstruction with bioabsorb-
able interference screw "xation. Arthroscopy 1999;15:317}20.
[18] Bostman OM. Osteolytic changes accompanying degradation of
absorbable fracture "xation implants. J Bone Jt Surg 1991;73-B:
679}82.
[19] Engelberg I, Kohn J. Physio-mechanical properties of degradable
polymers used in medical applications: a comparative study. Bio-
materials 1991;12:292}304.
[20] Gilding DK, Reed AM. Biodegradable polymers for use in sur-
gery*polyglycolic/poly(lactic acid) homo- and copolymers: 1.
Polymer 1979;20:1459}64.
2629
[21] Suuronen R, Pohjonen T, Taurio R, et al. Strength retention of
self-reinforced poly-L-lactide screws and plates: an in vivo and in
vitro study. J Biomater Sci Polym Ed 1992;3:426}31.
[22] Migliaresi C, Fambri L, Cohn D. A study on the in vitro de-
gradation of poly(lactic acid). J Biomater Sci Polym Ed 1994;
5:591}606.
[23] Hollinger JO, Battistone GC. Biodegradable bone repair mate-
rials synthetic polymers and ceramics. Clin Orthop 1986;207:
290}305.
[24] Bostman OM. Current concepts review absorbable implants for
the "xation of fractures. J Bone Jt Surg 1991;73-A:148}53.
[25] Verheyen CCPM, de Wijn JR, van Blitterswijk CA, Rozing PM,
de Groot K. Examination of e!erent lymph nodes after 2 years of
transcortical implantation of poly(L-lactide) containing plugs:
a case report. J Biomed Mater Res 1993;27:1115}8.