Abstract
Background: A prospective study was done to
determine the functional results, patient satisfaction, and graft
failure rate after fifty-seven consecutive revision replacements
of the anterior cruciate ligament with use of a bone-patellar
tendon-bone autogenous graft.
Methods: Fifty-four patients (fifty-five
operations) were followed in this study. Concurrent operative procedures
were performed during the revision procedure in thirty-seven knees
(67%). These procedures included repair of a meniscal tear
in twenty knees (36%) and reconstruction of deficient posterolateral
or medial ligament structures in seventeen knees (31%).
Nine knees (16%) had a high tibial osteotomy to correct
varus malalignment before the revision operation. The results were evaluated
with the Cincinnati Knee Rating System.
Results: There were significant improvements in
the scores for pain (p < 0.0001), activities of daily living
(p < 0.01), sports participation (p < 0.001),
patient satisfaction (p < 0.0001), and overall rating of
the knee (p < 0.0001). Thirty-three (60%)
of the replaced ligaments were functional, nine (16%) were
partially functional, and thirteen (24%) had failed.
Conclusions: Many knees (93%) had compounding
problems, including articular cartilage damage, prior meniscectomy,
loss of secondary ligament restraints, varus malalignment, and concomitant
ligament replacement or meniscal repair. Therefore, the results
were generally less favorable than those following primary operations.
The rate of graft failure was three times higher than our previously
reported failure rate after primary replacements of the anterior
cruciate ligament with a bone-patellar tendon-bone
autogenous graft. Even so, symptoms and functional limitations with
regard to daily and sports activities were found to have decreased
and patient satisfaction improved. We advocate correction of varus
malalignment prior to anterior cruciate procedures. Associated posterolateral
ligament deficiencies should be surgically corrected during anterior
cruciate procedures to prevent excessive loading on the graft from
abnormal lateral tibiofemoral joint opening. Meniscal tears, including
complex tears that extend into the avascular zone, can be concurrently
repaired successfully during the revision.
Replacement of the anterior cruciate ligament is performed
in 102,000 patients yearly in the United States1.
Failure of this operation to provide joint stability has been reported
to occur in 3% to 10% of knees2-6.
Therefore, each year between 3000 and 10,000 patients either require
a revision replacement of the anterior cruciate ligament or follow
a conservative treatment program. The reasons for failure include
improper graft placement7, graft
impingement due to inadequate notchplasty8,
improper graft tensioning9, inadequate
graft fixation due to the fixation device or deficient bone stock10, use of a graft of diminished tensile
strength or size11, failure to
correct associated ligament instabilities12,
and reinjury. There is no single standard revision procedure; graft
choices include bone-patellar tendon-bone or Achilles tendon
allografts13-15 as well as bone-patellar
tendon-bone15-18, semitendinosus-gracilis17, and quadriceps tendon-patellar
bone autografts18.
We previously reported the results of revision replacement of the
anterior cruciate ligament with use of a bone-patellar tendon-bone
allograft in a prospective study of seventy-five consecutive
knees14. There was a significant
improvement in the scores for symptoms and functional limitations
(p < 0.01); however, the rate of graft failure was 33% (twenty-five
knees). For this reason, we recommended that allografts not be considered
as the first choice for revision procedures. We then initiated the
current prospective study of bone-patellar tendon-bone
autografts for revision anterior cruciate replacements. In a previous
prospective investigation of chronic ruptures of the anterior cruciate ligament,
we reported that this autogenous graft replacement had a low failure
rate of just 7% (six of ninety knees)6.
We are aware of only three prior studies of revision anterior cruciate
surgery in which a bone-patellar tendon-bone autogenous
graft had been used; however, none provided objective measurements
to establish a failure rate16-18.
The first purpose of the current study was to determine the functional
outcomes and failure rate after fifty-seven consecutive
revision procedures involving replacement of the anterior cruciate
ligament with a bone-patellar tendon-bone autogenous
graft. The second purpose was to report the indications for and
results of concomitant surgical procedures to treat associated ligament
instabilities and varus malalignment of the lower limb. All patients
had the operation performed by the same surgeon, followed a predefined
rehabilitation program, and were followed for at least two years
postoperatively. The results were evaluated by a senior clinical
research associate instead of the surgeon.
Study Group
This prospective study, which was approved by our institutional
review board, involved fifty-seven consecutive anterior cruciate
revision procedures performed in fifty-six patients from
August 1990 to August 1996. Two knees were lost to follow-up.
Therefore, fifty-five operations in fifty-four patients
(96% of the original cohort) form the basis of this report.
Thirty-six operations were done in men, and nineteen were
done in women. The mean age at surgery was twenty-seven
years (range, fourteen to forty-eight years). Forty-four
patients (80%) had injured the knee during a sports activity.
The mean time from the first knee injury to the index procedure
was eighty months (range, six to 218 months). In three patients
(three knees), the revision procedure failed before the two-year
examination, and the data for these patients are included in the
failure category. The remaining fifty-one patients (fifty-two
knees) were examined at a mean of thirty-three months (range,
twenty-four to seventy-four months) postoperatively.
Two hundred and forty-two operative procedures (Table I) had been done
before the revision operations. A high tibial osteotomy had been
done in one patient prior to referral and in eight other patients
at our center19, as a staged treatment
approach. These patients had a mean of 6.5° of varus (range, 4°
to 12° of varus) of the mechanical axis preoperatively. The revision
anterior cruciate procedure was performed at a mean of eight months
(range, one to thirty-one months) after the high tibial osteotomy.
Evaluation
The objective evaluation was performed with a KT-2000 arthrometer
(MedMetric, San Diego, California) at 134 N of total anterior-posterior
force preoperatively and postoperatively. All tests were performed
by the same examiner, for whom the 90% confidence limit
had previously been reported to be 1.7 mm (right knee compared with
left knee) for measurement of total anterior-posterior
translation at 134 N20. Eight
patients had bilateral rupture of the anterior cruciate ligament,
and their data were excluded from the comparisons of the arthrometric
values. A comprehensive evaluation of the knee included assessment
of tibiofemoral, patellofemoral, and knee stability as well as alignment-related
factors21,22. The results of the
arthrometric and pivot-shift tests were used to classify
the anterior cruciate grafts as functional, partially functional,
or nonfunctional14. Grafts were
designated as partially functional when the arthrometer showed an
increase of 3 to 5.5 mm, the Lachman test was only slightly positive
with a hard stop, and the pivot-shift test was negative. Grafts
were classified as nonfunctional when 6 mm of increased anterior-posterior
displacement was detected on arthrometric testing or the pivot-shift
test was positive.
The Cincinnati Knee Rating System, for which the reliability, validity,
and responsiveness had been established, was used to determine the
results23. All fifty-one patients
without failure before the two-year examination completed a questionnaire
and were then interviewed by a research associate. The overall rating
score (on a scale of 0 to 100 points) was based on twenty factors,
and the results were determined by the change in the number of points from
the preoperative evaluation to the latest follow-up evaluation.
A modification in the rating of symptoms was required because the
majority of patients did not return to strenuous sports activities.
With this modification14, the
patient is asked if symptoms occurred with any type of sports activity.
The occupational levels of the patients were rated, and patients
provided their own rating of the overall knee conditon by circling
one number on a scale of 1 to 1023.
The preoperative and most recent radiographs (45° posteroanterior
flexion weight-bearing24,
lateral, and patellofemoral axial views) were evaluated for the
placement of the femoral and tibial graft tunnels and narrowing
of the patellofemoral and tibiofemoral joints. The appearance of
the articular cartilage was classified at the revision operation25. The cartilage was rated as abnormal
if there was a lesion of 15 mm in diameter and fissuring and fragmentation
of more than one-half of the depth of the involved articular
surface, or if any subchondral bone was exposed.
Operative Procedure
All knees had an absent or nonfunctional anterior cruciate ligament
at the revision operation, and all were treated with a bone-patellar
tendon-bone autograft. In thirty-nine knees, the ipsilateral
patellar tendon had not been previously used and was harvested.
In sixteen knees, the patellar tendon had been previously used;
in five of these knees the contralateral patellar tendon was harvested,
and in eleven knees the ipsilateral patellar tendon was reharvested.
The mean time between the first patellar tendon graft procedure
and the revision operation in these eleven knees was forty-two
months (range, seven to 168 months). In all eleven, the patellar
tendon had a grossly normal appearance and the prior bone defects
had healed.
First, diagnostic arthroscopy was performed to inspect the articular
surfaces and the menisci. Sixteen meniscal tears were partially
debrided, and twenty-two other meniscal tears were repaired.
Fourteen of the twenty-two tears were in the periphery, and eight
extended into the central, avascular region. Seventeen of the tears
were new, and five were tears that had failed to heal and were being
subjected to a second attempt at repair. An inside-out
arthroscopic-assisted technique with limited posteromedial
or posterolateral incisions and multiple superior and inferior vertically
divergent nonabsorbable 2-0 sutures (TiCron; Davis and
Geck, Wayne, New Jersey) was used to restore the meniscus to its
normal anatomic location26,27.
The endoscopic anterior cruciate procedure has been previously
described6. Any remaining anterior
cruciate ligament or graft fibers were removed, and a limited notchplasty
was performed as required to allow normal knee hyperextension without
graft impingement. A 4-cm incision was made medial to the
tibial tubercle, and the central one-third of the patellar
tendon, 9 to 10 mm in width, was harvested. The bone-patellar tendon-bone
graft was placed at the anatomic femoral and tibial insertion sites
of the anterior cruciate ligament. In thirty-five knees
(64%), a double-incision arthroscopic-assisted
technique was used in order to change the orientation of the femoral
graft tunnel so that it would be in the correct anatomic site. In
twenty knees (36%), a single-incision endoscopic
technique was used to place and fix the graft. No knee required
a bone graft for abnormally large tibial or femoral osseous defects.
The interference femoral screw from the prior procedure was retained
in knees in which the original, misplaced femoral tunnel was adjacent
to the new tunnel; the screw was retained to prevent collapse of
the bone bridge between the two tunnels. In the other knees, the
screw was removed and the graft was fixed with a new interference
screw. In fifteen knees that had had a prior high tibial osteotomy
or had osteopenic tibial cancellous bone, an interference screw
and a tibial post (a staple or a cortical screw) with sutures in
the bone portion of the graft were added to provide secure fixation.
The defect in the patellar tendon was closed. A core tibial reamer
was used to remove cancellous bone, which was placed into the patellar and
tibial graft sites.
Sixteen knees (29%) required reconstruction of the posterolateral
ligament structures because of abnormal increases in lateral joint
opening and external tibial rotation. Thirteen knees had a proximal
advancement of the posterolateral structures28,
and three knees had replacement of the deficient posterolateral
structures with autogenous29 or
allogeneic30 graft. The indication
for replacement was a prior traumatic rupture of these structures.
One knee required replacement of a deficient medial collateral ligament
with a double-bundle semitendinosus autogenous graft.
Postoperative Rehabilitation
All patients began immediate knee motion and muscle-strengthening
exercises on the first postoperative day. A Bledsoe brace (Medical
Technology, Grand Prairie, Texas) was worn for the first six weeks.
Partial weight-bearing was begun during the second week
and was gradually advanced to full weight-bearing by the
sixth week for all patients except those who had had a concomitant
posterolateral reconstructive procedure. Those patients were allowed
toe-touch weight-bearing for four weeks and then
slowly advanced to full weight-bearing by the twelfth week.
A Bledsoe Thruster brace, designed to decrease loads on the lateral
ligament reconstruction and to prevent abnormal lateral tibiofemoral
joint opening during walking, was used for six to nine months by
these patients. All patients were allowed a full range of knee motion
immediately postoperatively and had to have achieved a range of
flexion of at least 0° to 90° by the second postoperative week.
In patients who had had a concomitant posterolateral procedure, a
range of 0° to 90° of flexion was allowed for the first four weeks
and was increased to 135° by the eighth week28.
Any patient who had difficulty regaining a normal range of motion
of the knee was enrolled in a specific treatment program31.
Muscle-strengthening and flexibility exercises were
begun on the first postoperative day and included mobilization of
the patella, straight-leg raises, isometric exercises,
and electrical muscle stimulation. Closed-chain exercises
(toe raises, wall-sitting isometrics for quadriceps control to muscle
exhaustion [knee flexion angle between 30° and 45°],
and mini-squatting) were begun as soon as the patients had achieved weight-bearing
equal to at least one-half of their body weight. By the
fifth to sixth week, patients began open-chain exercises on
weight machines, proprioception training, and general cardiovascular
conditioning32. A running program
was begun by the sixth month by patients who had no effusion or
cartilage damage in the knee, quadriceps strength of at least 70% of
that of the contralateral limb, and no more than a 3-mm
increase in anterior-posterior displacement. A return to
full activities was permitted between the ninth and twelfth months
provided that the patient did not have symptoms. Patients with articular
cartilage damage avoided strenuous exercises and were advised to
return to light recreational sports33.
Statistical Analysis
Paired two-tailed Student t tests, contingency table
analyses, single linear regression analyses, and chi-square
tests were used to determine significant differences between preoperative
and follow-up data. Stepwise linear regression analyses were
performed between the twenty individual factors on which the overall
rating was based and the total point score at follow-up.
The effects of articular cartilage damage, reharvest of the patellar
tendon, and use of either the double-incision or the endoscopic
technique were assessed separately. The results were compared among
three subgroups: patients treated with revision of the anterior
cruciate ligament alone, those in whom the revision had been done
after a high tibial osteotomy, and those with other, concurrent
ligament procedures.
In order to evaluate two of the primary study outcomes (the pain
score and the overall rating score), sample-size calculations and
the power to detect a difference between preoperative and postoperative
mean scores were determined. With fifty-five knees in the
study, it was found that the investigation had sufficient power
(90%) to detect those differences at a significance level
of 0.05.
Symptoms and Patient Satisfaction
The mean score for pain improved from 3.7 1.8 points preoperatively
to 5.5 1.1 points at the time of follow-up (p < 0.0001).
Preoperatively, twenty-two (42%) of the fifty-two knees
that were evaluated were moderately or severely painful with activities
of daily living; postoperatively, only three knees (6%)
were painful with daily activities (Fig. 1). The mean score for giving-way
improved from 4.4 1.6 points preoperatively to 5.8 0.6 points
postoperatively (p < 0.0001). The mean score for patient
satisfaction improved from 3.0 1.7 points preoperatively to 6.1
2.1 points postoperatively (p < 0.0001). The overall condition
of forty-six knees (88%) was rated by the patient
as improved, the condition of three (6%) was rated as the
same, and the condition of three was rated as worse (Fig. 2).
Activities of Daily Living and Sports
The mean scores for walking, stair-climbing, and squatting improved
postoperatively (p < 0.01). Preoperatively, thirteen knees
(25%) caused severe or moderate difficulty with normal walking;
postoperatively, only two knees (4%) caused problems with
walking (Fig. 3).
The mean scores for running, jumping, hard twisting, cutting, and
pivoting also improved (p < 0.001). Preoperatively, forty
knees (77%) caused severe problems with running; postoperatively,
twenty-one knees (40%) caused such problems (Fig. 4). Preoperatively,
twenty-two patients (twenty-two knees; 42%) participated
in sports activities, all with symptoms or noteworthy limitations.
Postoperatively, thirty-two patients (thirty-two knees;
62%) had returned to sports without symptoms, eight patients
(eight knees; 15%) were participating with symptoms and
against advice, and eleven patients (twelve knees; 23%)
had not returned to sports because of the knee condition.
Occupation
Preoperatively, seven patients were disabled by the knee condition,
seventeen were working without symptoms, and nine were working with
severe symptoms and limitations. Postoperatively, two patients were
disabled by the knee condition, twenty-three patients had
returned to the same occupation without symptoms, eight had increased
their occupational level and were working without symptoms, and
four were working in a strenuous occupation with symptoms. The remainder
of the patients were not in the workforce; they were students or
homemakers.
Knee Stability and Examination
The mean value for anterior-posterior displacement decreased from
11.2 3.9 mm (range, 5.0 to 21.0 mm) preoperatively to 2.2 4.9
mm (range, -6.0 to 15.5 mm) postoperatively (p < 0.0001)
(Fig. 5).
Preoperatively, all fifty-five knees had a grade-2 or 3
pivot shift. At the time of follow-up, forty-three
knees (78%) had a grade-0 or 1 shift; eleven (20%),
a grade-2 shift; and one (2%), a grade-3 shift. On the
basis of both the pivot-shift and the arthrometric test results,
thirty-three (60%) of the replaced ligaments were
functional, nine (16%) were partially functional, and thirteen
(24%) had failed.
Preoperatively, comparison with the contralateral knee revealed
a mean of 8 mm (range, 5 to 15 mm) of increase in lateral joint
opening in the sixteen knees that subsequently had a concomitant
posterolateral ligament procedure. A mean of 9° of increase in external
tibial rotation was also found in these knees. Postoperatively,
no increase in lateral joint opening or external tibial rotation
was found in twelve knees, and 3 to 6 mm of increase in lateral
joint opening was found in four knees. The one knee that had a graft
replacement of the medial collateral ligament had 12 mm of increase
in medial joint opening preoperatively and no increase postoperatively.
There were no joint effusions, and all knees had at least 135° of
flexion. All knees had at least 0° of extension, except for two
that lacked 5° of full extension.
Overall Rating
The mean overall rating improved from 61 10 points (range, 43
to 79 points) preoperatively to 87 11 points (range, 62 to 100
points) at the time of follow-up (p < 0.0001).
All knees had an improvement (mean, 27 points; range, 2 to 46 points). The
stepwise linear regression analysis showed that four factors (knee
arthrometer values, score for twisting, score for walking, and pivot-shift
values) were significant in the determination of the overall rating
(Table II).
Effects of Articular Cartilage Damage, Reharvest
of Bone-Patellar Tendon-Bone Graft, and Other Operative Procedures
There was a significant difference between the thirty-one knees
(56%) that had abnormal articular cartilage surfaces and
the knees that had normal surfaces with regard to the patient’s
ability to perform twisting and turning activities and to return
to strenuous sports activities (p < 0.05). At the time of
follow-up, four of the eleven knees that had had reharvest of
the patellar tendon had a functional anterior cruciate graft; one,
a partially functional graft; and six, a nonfunctional graft.
Patients who had had a high tibial osteotomy before the revision
procedure or a concomitant ligament stabilization procedure had
poorer results. Compared with the patients who had had the revision
procedure only, those who had had a prior high tibial osteotomy
had significantly lower scores for mean patient satisfaction and
the overall rating (p < 0.05) and those who had had a concomitant
ligament stabilization procedure had a significantly lower overall
rating (p < 0.05) (Table III). The nine knees that had had a
high tibial osteotomy were corrected to a mean of 4° of valgus (range,
1° to 6° of valgus) postoperatively. The overall condition of seven
of the knees was rated by the patient as improved, and the condition
of two was rated as the same as the preoperative condition. Preoperatively,
seven knees were painful with daily activities; at the time of follow-up,
only one was painful with daily activities.
There was no difference in the knee arthrometer test results, graft
function, or pain or satisfaction scores between the knees that
had had the double-incision arthroscopic-assisted
technique and those that had had the endoscopic technique.
Complications and Reoperations
Two knees required additional treatment to regain normal motion.
One knee was gently manipulated with the patient under anesthesia
seven weeks postoperatively. In another knee, an arthroscopic lysis
of adhesions and release of scar tissue was performed ten weeks
postoperatively. Both knees regained a full range of motion.
Subsequent operative procedures were done in seven knees. Arthroscopy
and removal of a painful tibial screw was performed in two knees.
One knee had a twisting reinjury and a successful repair of a new
meniscal tear, and another knee had a partial meniscectomy after
a lateral repair of a longitudinal tear in the outer one-third
of the meniscus had failed to heal. Three knees showed an increase
in anterior-posterior displacement on arthrometric testing
four weeks postoperatively. Arthroscopy performed six weeks postoperatively
showed that the collagenous portion of the bone-patellar
tendon-bone graft was lax but intact. It was presumed that
the graft had slipped at the tibial osseous fixation site and, therefore,
refixation was performed.
The majority of knees requiring revision procedures have compounding
problems, including articular cartilage damage, prior meniscectomy,
loss of other ligament restraints, varus malalignment, and the need
for concomitant ligament procedures or meniscal repair. In our study,
only four knees (7%) did not have one or more of these
compounding problems. Accordingly, revision anterior cruciate procedures13,14,16-18 have less favorable results
than do primary replacements. Still, we were encouraged that 88% (forty-six)
of the knees were rated by the patients as improved overall compared
with the preoperative status, 81% (forty-two)
were less painful, and 62% (thirty-two) were in
patients who had returned to (mostly light) athletics without symptoms.
Effect of Articular Cartilage Damage on Functional
Outcome
The condition of the articular cartilage had a significant effect on
the type of sports activities to which the patients returned postoperatively,
a finding that agreed with those of prior studies14,33.
When we counsel patients with articular cartilage damage, we try
to provide realistic expectations of the surgery, which are reduction
of pain and instability with activities of daily living and perhaps
a return to light recreational activities. Eighteen (58%)
of the thirty-one patients in this subgroup returned to
light recreational activities postoperatively without problems.
Treatment of Associated Varus Malalignment
and Ligament Deficiency
Nine knees required a high tibial osteotomy to correct varus malalignment
before the revision anterior cruciate operation. We have previously
discussed the indications for correcting lower-limb malalignment,
with use of either opening-wedge or closing-wedge
techniques, prior to ligament procedures (Fig. 6)19,34. When a varus-aligned
knee requires an osteotomy before a ligament procedure but has no
abnormal lateral joint opening or external tibial rotation, the
osteotomy and the anterior cruciate revision can be done at the
same setting19. However, when
a varus-aligned knee has abnormal lateral joint opening
and a varus thrust on gait, the valgus-producing high tibial osteotomy
is done first and is followed later by the anterior cruciate revision.
Seventeen knees (31%) in this investigation had unrecognized or
untreated lateral, posterolateral, or medial ligament deficiency,
which was probably a factor in the failure of the first anterior
cruciate procedure. An anterior cruciate graft is subjected to excessive
tensile loading when there is associated lateral or medial collateral
ligament insufficiency12 because
of the abnormal medial or lateral tibiofemoral joint opening that
occurs with activity. The preoperative clinical examination is used
to detect associated instabilities and, particularly, varus recurvatum,
lateral joint opening, and excessive external tibial rotation35. At arthroscopy, we use the gap
test with varus or valgus loading to detect abnormal medial or lateral
joint opening (Figs. 7-A, 7-B, and 7-C). There will be 12 mm of joint
opening at the periphery of the lateral tibiofemoral joint at 30°
of flexion in knees with deficient posterolateral structures. We
have classified varus-aligned knees into three categories that have
implications with regard to the recommended treatment plan19. The first category (primary varus)
comprises knees that have varus alignment due to the underlying
tibiofemoral alignment with no associated posterolateral ligament
deficiency or abnormal lateral joint opening. The second category
(double varus) comprises knees that have associated deficiency of
the lateral collateral ligament. In these knees, varus alignment
is increased as a result of both tibiofemoral osseous alignment and
abnormal lateral joint opening. The third category (triple varus)
includes knees that have deficiency of all of the posterolateral
structures. Varus alignment increases on standing, and a varus recurvatum
position is present. Importantly, in double-varus knees, the lateral
ligamentous tissues may shorten after osteotomy and eliminate any
abnormal lateral joint opening36.
Therefore, a posterolateral ligament reconstruction is not required
at the time of the revision anterior cruciate replacement. In triple-varus
knees, however, reconstruction of the posterolateral ligament structures
is required and is done at the time of the revision replacement19.
The operative procedures indicated for knees with deficient posterolateral
tissues have been described28-30.
The procedure is chosen on the basis of the quality and integrity
of these tissues as determined at surgery. A proximal advancement
of the posterolateral structures is indicated in knees with chronic
interstitial failure of these tissues with no prior traumatic disruption
and with intact attachment sites28.
On surgical inspection, the lateral collateral ligament, the popliteus
muscle-tendon-ligament unit, and other posterolateral
tissues appear normal but lax. Rather than selective tightening
or recession, these structures are advanced proximally at the femoral
attachment site in the line of the normal attachment of the lateral
collateral ligament with the knee flexed to 30° and in neutral tibial
rotation. Fixation is achieved by placing a four-prong
staple at the normal femoral attachment site of these structures
to maintain length-tension behavior. The second option
is to use allograft30 or autograft
augmentation between the femoral epicondyle and the fibular attachment
anterior and posterior to the lateral collateral ligament. The third
option, in knees with traumatic rupture, is to use a bone-patellar
tendon-bone autograft or allograft to replace the lateral
collateral ligament and an Achilles tendon-bone allograft
or a semitendinosus double-loop tendon graft to replace
the popliteus muscle-ligament-tendon complex29. Sutures are placed between the
allograft and the fibular head to restore the popliteofibular ligament.
We routinely correct all associated ligament ruptures during an
anterior cruciate procedure, and, in this investigation, all but
one of the associated ligament instabilities were successfully corrected.
Overall Graft Failure Rate
In the current study, the overall rate of graft failure was 24% (thirteen
of fifty-five knees). This is a threefold increase compared
with the rate of 7% (six of ninety knees) that we previously
reported after primary anterior cruciate replacement6. In that study, a subgroup of sixty
knees with chronic ligament rupture had an 8% failure rate.
We are not aware of any other reports, in the English-language literature,
of graft failure rates after revision anterior cruciate procedures
with a bone-patellar tendon-bone autogenous graft. In the present study,
the rate of graft failure was higher in knees that had had a concomitant
ligament procedure than it was in those that had not had such a
procedure (35% compared with 16%; p < 0.05).
However, we do not know if the increased failure rate in knees with
combined ligament instabilities was due to a bias from the small
number of knees (seventeen) with that factor.
Prior studies of revision anterior cruciate replacements with allografts
showed even higher failure rates (33% [twenty-five of
seventy-five knees]14 and
36% [nine of twenty-five knees13]).
We previously reported that the rate of failure of allografts in
knees with chronic anterior cruciate deficiency decreased when the
replacement was combined with a lateral extra-articular
iliotibial band procedure37. In
a series of 104 patients, the rate of failure was 16% (ten
of sixty-four knees) when the allograft had been used alone
compared with 3% (one of forty knees) when the allograft
had been combined with the extra-articular procedure (p < 0.05). This
finding suggests that, when certain knees with excessive laxity
of the secondary restraints undergo revision, there may be a benefit
from a combined intra-articular graft replacement and extra-articular
iliotibial band procedure.
Effect of Reharvest of the Patellar Tendon
In nine of the eleven knees in which the patellar tendon was reharvested,
the original harvest and our revision procedure were performed at
least two years apart. The patellar tendon had a normal appearance
and the prior bone defects had healed. Even so, six of these grafts
subsequently failed. Although authors of early studies38,39 postulated that reharvest of
a patellar tendon graft was feasible, other investigators demonstrated
persistent changes on magnetic resonance imaging and morphologically,
up to seven years postoperatively, that could adversely affect the biomechanical
properties of the healed donor site16,40-42.
Kartus et al. reported that knees that had received a reharvested
graft at revision had lower functional scores and a higher rate
of postoperative complications, including patellar fracture and
patellar tendon rupture16. We
do not believe that the patellar tendon should be reharvested for
revision anterior cruciate procedures, even when those procedures
are performed many years after the original graft harvest.
Meniscal Repair During the Revision Procedure
One important finding from this study was the number of meniscal
repairs (twenty-two in twenty knees) performed during the
revision procedures. Seventeen repairs were done for a new meniscal
tear, and five were second repairs of a tear that had failed to
heal. In eight knees, the tear extended into the central, avascular
region, in which a partial meniscectomy would have resulted in a
near-total meniscectomy. We previously reported the repair
technique for complex tears that extend into the avascular region27. Vertically divergent nonabsorbable
sutures are placed in the superior and inferior surfaces of the
meniscus every 4 mm to reduce and repair meniscal tears, with meticulous
closure of any meniscal gap. Abrasion of the perisynovial tissues
and meniscus is used to stimulate a vascular response. In the prior study,
we reported that 159 (80%) of 198 knees that had a single
or complex meniscal tear extending into the avascular zone had no
tibiofemoral joint symptoms27.
We do not believe that meniscal fixators provide the same meticulous
approximation of damaged meniscal tissues. At the time of follow-up
in the current study, only one of the meniscal repairs had required
partial resection and none of the remainder were associated with
tibiofemoral symptoms.
Comparison of Study Results with Those in
the Literature
There is discrepancy in the literature regarding symptoms and functional
limitations following revision anterior cruciate operations. Wirth
and Kohn noted that, at a mean of eight years following revision
procedures with use of autogenous bone patellar tendon-bone
grafts, 75% (sixty-five) of eighty-seven
knees continued to have symptoms and only 60% (fifty-two)
had results that were satisfactory to the patient18.
Uribe et al. reported that, although only 54% (twenty-nine) of
fifty-four patients returned to the same activity level
that they had had before the original knee injury, all patients thought
that they had benefited from the anterior cruciate revision when
they were evaluated at a mean of thirty-two months postoperatively17. Those authors used a variety of
grafts for the revision procedure, including bone-patellar
tendon-bone autografts, bone-patellar tendon-bone
allografts, and semitendinosus-gracilis autografts. As shown in
our study, different types of knees require anterior cruciate revision,
and strict comparisons must account for the effects of compounding
variables such as articular cartilage damage, meniscectomy, associated ligament
ruptures, and varus alignment.
Preoperative Planning for Revision Surgery
It is necessary to carefully assess the factors that may have been
related to the prior failure so that these problems may be addressed
before or during the revision procedure (Fig. 8). An extensive preoperative evaluation
includes a meticulous physical examination for associated ligament
deficiencies; observation of gait for abnormal hyperextension, varus
thrusting, or other abnormalities requiring preoperative correction43; and analysis of radiographs and
magnetic resonance imaging studies to determine graft tunnel location,
size, and abnormal widening. Full double-standing radiographs and
posteroanterior radiographs with the knee in 45° of flexion are
usually required to assess osseous alignment and tibiofemoral joint narrowing.
Anterior cruciate revisions can be associated with a wide spectrum
of problems, ranging from unrealistic expectations regarding return
to strenuous sports by patients with articular cartilage damage
or a prior meniscectomy to associated anatomic abnormalities that
require surgical correction. These abnormalities include ligament
deficiencies, varus osseous alignment with medial arthrosis or lateral
tibiofemoral gapping, muscular weakness, painful neuroma, complex
regional pain syndrome, and anterior knee pain secondary to damage to
the patellofemoral cartilage. Nearly one-half of the knees
in this investigation had a major concomitant procedure in addition
to the anterior cruciate revision. Furthermore, a surgeon-rehabilitation
team is required to provide instruction about rehabilitation to
ensure that the postoperative exercise program will be successful.
A lack of complete knee extension prior to revision surgery poses
special problems, and it is important to obtain full knee motion
before the operative procedure. The prior graft may have been placed
too far anteriorly or a cyclops lesion may be producing a mechanical
block to full extension. If the lack of knee extension is >5°,
we routinely stage the procedure, first performing an arthroscopic
débridement. Anterior intercondylar blockage is corrected
by débridement of the prior graft and notchplasty. There
may be associated tightness of the posterior capsular structures,
which may have shortened, limiting knee extension.
Preparation of Osseous Tunnels for Revision
Graft Placement
Several methods can be used to address a widely misplaced femoral
or tibial graft tunnel at the time of the revision surgery14. The graft must be placed at the
anatomic tibial and femoral sites, and no prior osseous tunnel can
allow the graft to displace from this site or can adversely affect
adequate fixation. Stacking two interference cancellous screws (Fig. 9) may secure
the graft at the attachment site and fill the bone tunnel. This
technique is indicated when the bone tunnel is 3 to 5 mm larger
than the graft. Otherwise, the more anteriorly placed bone tunnel
requires a staged cancellous graft procedure. We were able to avoid
bone-grafting of the prior femoral or tibial tunnel in all knees
in our study.
In other knees, a second incision may be used to orient the femoral
tunnel along another pathway, maintaining an adequate bone bridge
between the prior and the newly created tunnel. In cases where the
original tunnel was misplaced but is adjacent to the new femoral
tunnel site, care must be taken to drill the tunnel in a stepwise
manner, beginning with a 6-mm tunnel and then enlarging
it to the correct diameter and anatomic location. We do not remove
the previously placed interference screw, which provides support
to buttress the bone and assists in the fixation of the newly located
graft.
In some knees, the previously placed femoral tunnel is located adjacent
to the anatomic site and it is not possible to locate the new graft
tunnel correctly without overlapping and breaking into the old tunnel.
With use of a posterolateral second incision, posterior notching
of the lateral femoral condyle into the normal position of the anterior
cruciate insertion may be performed. This allows the graft to be
inserted within the anatomic attachment site of the anterior cruciate
ligament and still maintains an osseous bridge from the prior osseous
tunnel. Internal graft fixation with two small-fragment cancellous screws
and a backup suture post provides secure osseous fixation.
An enlarged tibial graft tunnel poses unique problems. Sufficient
graft-fixation strength may be obtained by placing the distal interference
screw and the graft adjacent to the tibial tubercle, thereby avoiding
weaker cancellous bone fixation. Graft osseous sutures are tied
to a tibial post. In knees with excessive widening at the tibial
insertion site, fixation may be achieved distally; however, the
graft undergoes a windshield-wiper effect with knee flexion.
When the tibial tunnel extends posteriorly past the normal insertion
site of the anterior cruciate ligament, the graft assumes a vertically
oriented position and fails to resist anterior tibial translation
effectively. Enlargement of the tibial graft tunnel can also affect graft
incorporation. This is an indication for staged autogenous bone-grafting
of the tibial tunnel in order to subsequently obtain an anatomic
graft replacement. At the subsequent revision procedure, a bone-patellar
tendon-bone or Achilles tendon-bone allograft
should be available as a backup. Some knees still have inadequate
tibial bone stock, and a larger bone-patellar tendon-bone
graft may be customized to achieve anatomic placement of the graft.
Permission for the possible use of an allograft should be obtained
at the preoperative counseling session.
Current Graft Choices for Revision Anterior
Cruciate Surgery
On the basis of the results of our investigation, a bone-patellar tendon-bone
autograft is currently our first choice for revision anterior cruciate
surgery. In knees in which the patellar tendon was previously harvested,
we use the contralateral patellar tendon. If this graft is not available,
we still perform an autograft replacement, with a quadriceps tendon-bone
graft from the ipsilateral knee and with complete grafting of any residual
patellar bone defect. When autograft tissues are not available,
a bone-patellar tendon-bone allograft or Achilles tendon-bone
allograft may be considered. We routinely use a lateral extra-articular
iliotibial band procedure with the allograft procedure37, to reduce the high failure rate
previously reported after anterior cruciate allograft revisions14.
In conclusion, anterior cruciate revisions are performed in different
patient subsets. The subgroup that is simplest to treat does not
have associated articular cartilage damage, a previous meniscectomy,
varus alignment, or other major ligament instabilities. The revision
procedure can be expected to provide results and patient satisfaction
that are close to those of initial replacement procedures. The revision
operation should be performed early, after the patient has undergone
adequate rehabilitation following the first procedure and is prepared from
a time and motivational standpoint. When revision procedures are
delayed in athletically active individuals, deterioration can be
expected, with subsequent meniscal tears and articular cartilage
damage. This represents a second subgroup of patients, in whom anterior
cruciate revision may still be indicated but who may not be able
to return to athletics because of the damage to the articular cartilage
and the meniscal loss44. The third
group of knees in which anterior cruciate revision may be performed
has associated ligament instabilities, usually involving the lateral
collateral ligament and the posterolateral structures. Combined
ligament procedures are performed in this group, followed by immediate
motion and a comprehensive rehabilitation program postoperatively28,30. The fourth subgroup of patients
requires a high tibial osteotomy. The varus alignment and associated
varus thrust with activity increase tensile forces on the lateral
and posterolateral soft tissues and frequently produce a chronic
interstitial failure, increased lateral joint opening and posterolateral
tibial subluxation, and the need for concomitant posterolateral
ligament reconstruction. This group is the most difficult to treat. To
decrease the risk of postoperative complications and morbidity,
we advocate that corrective tibial osteotomy be performed first,
as a staged procedure, with the anterior cruciate and posterolateral
ligament reconstructive procedures done months later19. The results of the treatment of
these more complex cases are heterogeneous, depending on the response
of the patients to the combined operative procedures.
Owings MF,Kozak LJ. Ambulatory and inpatient procedures in the United States, 1996. Vital Health Stat 13,1998;139: 1-119. 1391
1998
[PubMed]
Bach BR Jr, Levy ME, Bojchuk JTradonsky
SBush-Joseph CA,Khan NH. Single-incision endoscopic anterior cruciate
ligament reconstruction using patellar tendon autograft. Minimum
two-year follow-up evaluation. Am J Sports Med,1998;26: 30-40. 2630
1998
[PubMed]
Corry IS, Webb JM, Clingeleffer
AJ,Pinczewski LA. Arthroscopic reconstruction of the anterior cruciate ligament. A
comparison of patellar tendon autograft and four-strand hamstring
tendon autograft. Am J Sports Med,1999;27: 444-54.. 27444
1999
[PubMed]
Marcacci M, Zaffagnini S, Iacono
F, Neri MP, Loreti I,Petitto A. Arthroscopic intra- and extra-articular
anterior cruciate ligament reconstruction with gracilis and semitendinosus
tendons. Knee Surg Sports Traumatol Arthrosc,1998;6: 68-75. 668
1998
[PubMed]
Meystre JL, Vallotton J,Benvenuti
JF. Double semitendinosus anterior cruciate ligament reconstruction:
10-year results. Knee Surg Sports Traumatol Arthrosc,1998;6: 76-81. 676
1998
[PubMed]
Noyes FR,Barber-Westin
SD. A comparison of results in acute and chronic anterior
cruciate ligament ruptures of arthroscopically assisted autogenous patellar
tendon reconstruction. Am J Sports Med,1997;25: 460-71.. 25460
1997
[PubMed]
Hefzy MS, Grood ES,Noyes FR. Factors affecting the region of most isometric femoral
attachments. Part II: The anterior cruciate ligament. Am J Sports Med,1989;17: 208-16. 17208
1989
[PubMed]
Tanzer M,Lenczner E. The relationship of intercondylar notch size and content
to notchplasty requirement in anterior cruciate ligament surgery. Arthroscopy,1990;6: 89-93. 689
1990
[PubMed]
Noyes FR, Butler DL, Paulos LE,Grood
ES. Intra-articular cruciate reconstruction. I: Perspectives
on graft strength, vascularization, and immediate motion after replacement. Clin Orthop,1983;172: 71-7. 17271
1983
[PubMed]
Matthews LS,Soffer SR. Pitfalls in the use of interference screws for anterior
cruciate ligament reconstruction: brief report. Arthroscopy,1989;5: 225-6. 5225
1989
[PubMed]
Noyes FR, Butler DL, Grood ES, Zernicke
RF,Hefzy MS. Biomechanical analysis of human ligament grafts used in knee-ligament
repairs and reconstructions. J Bone Joint Surg Am,1984;66: 344-52. 66344
1984
[PubMed]
Wascher DC, Markolf KL, Shapiro
MS,Finerman GA. Direct in vitro measurement of forces in the cruciate
ligaments. Part I: The effect of multiplane loading in the intact knee. J Bone Joint Surg Am,1993;75: 377-86.. 75377
1993
[PubMed]
Johnson DL, Swenson TM, Irrgang
JJ, Fu FH,Harner CD. Revision anterior cruciate ligament surgery: experience
from Pittsburgh. Clin Orthop,1996;325: 100-9. 325100
1996
[PubMed]
Noyes FR, Barber-Westin
SD,Roberts CS. Use of allografts after failed treatment of rupture of
the anterior cruciate ligament. J Bone Joint Surg Am,1994;76: 1019-31. 761019
1994
[PubMed]
Noyes FR,Barber-Westin
SD. Revision anterior cruciate ligament surgery: experience
from Cincinnati. Clin Orthop,1996;325: 116-29. 325116
1996
[PubMed]
Kartus J, Stener S, Lindahl S, Eriksson
BI,Karlsson J. Ipsi- or contralateral patellar tendon graft
in anterior cruciate ligament revision surgery. A comparison of
two methods. Am J Sports Med,1998;26: 499-504. 26499
1998
[PubMed]
Uribe JW, Hechtman KS, Zvifac JE,Tjin-A-Tsoi
EW. Revision anterior cruciate ligament surgery: experience
from Miami. Clin Orthop,1996;325: 91-9. 32591
1996
[PubMed]
Wirth CJ,Kohn D. Revision anterior cruciate ligament surgery: experience
from Germany. Clin Orthop,1996;325: 110-5. 325110
1996
[PubMed]
Noyes FR, Barber-Westin
SD,Hewett TE. High tibial osteotomy and ligament reconstruction for
varus angulated anterior cruciate ligament-deficient knees. Am J Sports Med,2000;28: 282-96. 28282
2000
[PubMed]
Wroble RR, Van Ginkel LA, Grood
ES, Noyes FR,Shaffer BL. Repeatability of the KT-1000 arthrometer
in a normal population. Am J Sports Med,1990;18: 396-9. 18396
1990
[PubMed]
Noyes FR, Grood ES,Suntay WJ. Three-dimensional motion analysis of clinical
stress tests for anterior knee subluxations. Acta Orthop Scand,1989;60: 308-18. 60308
1989
[PubMed]
Noyes FR, Barber SD,Mangine RE. Bone-patellar ligament-bone and fascia
lata allografts for reconstruction of the anterior cruciate ligament. J Bone Joint Surg Am,1990;72: 1125-36. 721125
1990
[PubMed]
Barber-Westin SD, Noyes
FR,McCloskey JW. Rigorous statistical reliability, validity, and responsiveness testing
of the Cincinnati Knee Rating System in 350 subjects with
uninjured, injured, or anterior cruciate ligament-reconstructed
knees. Am J Sports Med,1999;27: 402-16. 27402
1999
[PubMed]
Rosenberg TD, Paulos LE, Parker
RD, Coward DB,Scott SM. The forty-five-degree posteroanterior flexion
weight-bearing radiograph of the knee. J Bone Joint Surg Am,1988;70: 1479-83. 701479
1988
[PubMed]
Noyes FR,Stabler CL. A system for grading articular cartilage lesions at arthroscopy. Am J Sports Med,1989;17: 505-13. 17505
1989
[PubMed]
Buseck MS,Noyes FR. Arthroscopic evaluation of meniscal repairs after anterior
cruciate ligament reconstruction and immediate motion. Am J Sports Med,1991;19: 489-94. 19489
1991
[PubMed]
Rubman MH, Noyes FR,Barber-Westin SD. Arthroscopic repair of meniscal tears that extend
into the avascular zone. A review of 198 single and complex
tears. Am J Sports Med,1998;26: 87-95. 2687
1998
[PubMed]
Noyes FR,Barber-Westin
SD. Surgical restoration to treat chronic deficiency of the
posterolateral complex and cruciate ligaments of the knee joint. Am J Sports Med,1996;24: 415-26. 24415
1996
[PubMed]
Noyes FR,Barber-Westin
SD. Treatment of complex injuries involving the posterior
cruciate and posterolateral ligaments of the knee. Am J Knee Surg,1996;9: 200-14. 9200
1996
[PubMed]
Noyes FR,Barber-Westin
SD. Surgical reconstruction of severe chronic posterolateral
complex injuries of the knee using allograft tissues. Am J Sports Med,1995;23: 2-12. 232
1995
[PubMed]
Noyes FR, Berrios-Torres
S, Barber-Westin SD,Heckmann TP. Prevention of permanent arthrofibrosis after anterior
cruciate ligament reconstruction alone or combined with associated procedures:
a prospective study in 443 knees. Knee Surg Sports Traumatol Arthrosc,2000;8: 196-206. 8196
2000
[PubMed]
Heckmann TP, Noyes FR, Barber-Westin
SD. Autogeneic and allogeneic anterior cruciate ligament
rehabilitation. In: Ellenbecker TS, editor. Knee ligament
rehabilitation. Philadelphia: Churchill Livingstone; 2000.
p 132-50.
Noyes FR,Barber-Westin
SD. Anterior cruciate ligament reconstruction with autogenous patellar
tendon graft in patients with articular cartilage damage. Am J Sports Med,1997;25: 626-34. 25626
1997
[PubMed]
Noyes FR, Simon R. The role
of high tibial osteotomy in the anterior cruciate ligament-deficient
knee with varus alignment. In: DeLee JC, Drez D Jr, editors. Orthopaedic
sports medicine: principles and practice. Volume 3. Philadelphia:
WB Saunders; 1994. p 1401-43.
Noyes FR, Stowers SF, Grood ES, Cummings
J,VanGinkel LA. Posterior subluxations of the medial and lateral tibiofemoral compartments.
An in vitro ligament sectioning study in cadaveric knees. Am J Sports Med,1993;21: 407-14. 21407
1993
[PubMed]
Noyes FR, Barber SD,Simon R. High tibial osteotomy and ligament reconstruction in varus angulated,
anterior cruciate ligament-deficient knees. A two- to
seven-year follow-up study. Am J Sports Med,1993;21: 2-12. 212
1993
[PubMed]
Noyes FR,Barber SD. The effect of an extra-articular procedure on
allograft reconstructions for chronic ruptures of the anterior cruciate
ligament. J Bone Joint Surg Am,1991;73: 882-92. 73882
1991
[PubMed]
Benedetto KP, Sperner G, Gloetzer
WFritschy D,Gautard R. Ultrasonographic followup of patellar tendon following
graft dissection for ACL-replacement. Am J Sports Med,1989;17: 709. 17709
1989
Coupens SD, Yates CK, Sheldon C,Ward
C. Magnetic resonance imaging evaluation of the patellar
tendon after use of its central one-third for anterior
cruciate ligament reconstruction. Am J Sports Med,1992;20: 332-45. 20332
1992
[PubMed]
Bernicker JP, Haddad JL,Lintner
DMDiLiberti TCBocell JR. Patellar tendon defect during the first year after anterior
cruciate ligament reconstruction: appearance on serial magnetic resonance
imaging. Arthroscopy,1998;14: 804-9. 14804
1998
[PubMed]
LaPrade RF, Hamilton CD, Montgomery
RD, Wentorf F,Hawkins HD. The reharvested central third of the patellar tendon.
A histologic and biomechanical analysis. Am J Sports Med,1997;25: 779-85. 25779
1997
[PubMed]
Liu SH, Hang DW, Gentili A,Finerman
GA. MRI and morphology of the insertion of the patellar tendon after
graft harvesting. J Bone Joint Surg Br,1996;78: 823-6. 78823
1996
[PubMed]
Noyes FR, Dunworth LA, Andriacchi
TP, Andrews M,Hewett TE. Knee hyperextension gait abnormalities in unstable knees. Recognition
and preoperative gait retraining. Am J Sports Med,1996;24: 35-45. 2435
1996
[PubMed]
Shelbourne KD,Wilckens JH. Intraarticular anterior cruciate ligament reconstruction
in the symptomatic arthritic knee. Am J Sports Med,1993;21: 685-9. 21685
1993
[PubMed]