Thermal and laser heat shrinkage of collagen recently has
been proposed as a means of treating selected cases of shoulder
instability12. While such treatment
is controversial, clinical reports of capsular shrinkage procedures
involving the shoulder have shown limited but promising results1,2,9,10. Thermal technology also has
been applied to the treatment of anterior cruciate ligament laxity
in the knee11. Basic-science investigators
have elucidated the macroscopic and molecular changes that occur
in the collagen fibrils following thermal shrinkage3-7. However, long-term clinical follow-up
to determine the integrity of these structures following thermal
shrinkage is still lacking.
We present the case of a patient in whom an endoscopic reconstruction
of the anterior cruciate ligament was performed with a hamstring
autograft, which, following reinjury, underwent thermal shrinkage.
The findings at arthroscopy, both during the initial thermal shrinkage
and at fifteen weeks postoperatively, are discussed.
A sixteen-year-old female athlete sustained an anterior cruciate
ligament tear while twisting her left knee during a basketball game.
She subsequently underwent an arthroscopically assisted reconstruction
of the anterior cruciate ligament with a semitendinosus-gracilis
graft and a partial lateral meniscectomy. The graft, a four-bundle
construct, was secured to the femur with endobutton fixation (Smith
and Nephew, Memphis, Tennessee), and it was secured to the tibia
with a soft-tissue interference screw and suture fixation. Following
the reconstruction, the intraoperative Lachman test was negative.
Postoperatively, the patient followed a regimented, physical-therapist-monitored
protocol that included progressive range-of-motion and strengthening
exercises. Four months postoperatively, physical examination revealed
a negative pivot-shift test and a grade of 1+ (less than five millimeters
of side-to-side difference), with a firm end point, on the Lachman
test. Evaluation of laxity with a KT-2000 arthrometer (MEDmetric,
San Diego, California) revealed that the value on the affected side
was within three and one-half millimeters of that on the normal
side on maximum manual testing. Testing with a dynamometer (Biodex,
Shirley, New York) revealed quadriceps and hamstring torques of
83 and 41 percent of body weight, respectively, on the involved
side compared with 95 and 47 percent of body weight, respectively,
on the contralateral side. On the basis of these parameters, the
patient was allowed to progress to full activity.
Five months following the initial reconstruction, the patient
sustained a twisting reinjury to her left knee and felt a pop, with
immediate pain and swelling. Clinical examination suggested an injury to
the anterior cruciate ligament graft as well as a medial meniscal
tear. Magnetic resonance imaging revealed a medial meniscal tear
and showed the anterior cruciate ligament graft to be in continuity.
An examination with the patient under anesthesia revealed a grade
of 2+ on the pivot-shift test and a grade of 2+ (five to ten millimeters
of increased anterior-posterior translation compared with the normal
side) on the Lachman test. Arthroscopy demonstrated an irreparable
bucket-handle tear of the medial meniscus and showed that the anterior
cruciate ligament graft was in continuity but had severe laxity
(Fig. 1).
The medial meniscal tear was debrided, and, since the anterior cruciate
ligament graft appeared to be intact with no evidence of bleeding,
a thermal heat probe (ORATEC; ORATEC Interventions, Menlo Park,
California) with radiofrequency current to induce collagen shrinkage
was used to tighten the graft. This procedure was performed at 65
degrees Celsius and forty watts of energy, as recommended by the
manufacturer. The intra-articular portion of the anterior cruciate
ligament graft shrank (Fig. 2); it then was probed to verify
that appropriate tension had been achieved. Following the shrinkage procedure,
an intraoperative Lachman test revealed no increased anterior tibial
translation compared with the uninjured side.
Postoperatively, the patient participated in an active rehabilitation
program with protection of the thermally treated graft for twelve
weeks. Physical examination of the knee at twelve weeks revealed
a negative pivot-shift test and a grade of 1+ on the Lachman test,
and the patient was allowed to progress to full activity. One week
later, she sustained a noncontact reinjury to her left knee, which
produced functional instability. Physical examination revealed a
grade of 3+ (greater than fifteen millimeters of anterior translation
of the tibia on the femur), with no appreciable end point, on the
Lachman test.
Approximately two weeks following the reinjury, a revision anterior
cruciate ligament reconstruction with use of a bone-patellar tendon-bone
autograft and a medial meniscal repair were performed arthroscopically.
During the procedure, no remnant of the previous hamstring autograft was
identifiable in the notch between the tibial and femoral tunnels
(Fig. 3-A and Fig. 3-B). Following
the reconstruction, an intraoperative Lachman test was negative.
Nine months following the revision, the patient had no subjective
instability and the Lachman test was negative.
Thabit reported using a monopolar device to treat native and
reconstructed anterior cruciate ligaments that were lax but in continuity11. He recommended leaving the posterior
aspect of the ligament untreated to allow for revascularization
and suggested a three to six-month period of activity modification
and rehabilitation after radiofrequency treatment of the anterior
cruciate ligament, prior to return to full activity. In an anecdotal
report on twenty-five patients who were thus treated and were followed
for a maximum of 1.5 years, Thabit reported gross ligamentous failure
in one patient and increased laxity (as measured with the KT-1000
arthrometer) in another. The other twenty-three patients had KT-1000
arthrometer values that were within two millimeters of those for
the contralateral, normal knee11.
Several basic-science studies have elucidated the macroscopic
and molecular changes that occur in the collagen fibrils following
thermal shrinkage3-7,12. On the
basis of both cadaveric and animal models, these reports suggest
that thermal energy results in denaturation of the collagen microstructure3,4,6. The long-term effects of these
structural changes on the biomechanical properties of collagen are unknown.
Schaefer et al. demonstrated, in a rabbit model, that laser-induced
patellar tendon shrinkage (an average of 6.6 percent) resulted in
the tendons stretching out beyond pretreatment lengths over an eight-week
period8. Biomechanical testing
revealed that the tendons were significantly less stiff compared
with the pretreatment values (p = 0.03) and that marked histological
changes were present. However, the rabbit limbs were not immobilized
during the postoperative period, so activity was essentially unrestricted.
In our patient, laxity developed after injury to a hamstring
anterior cruciate ligament autograft. After heat shrinkage of the
graft, two mechanisms for failure seem plausible, and both may have
been contributory. First, biomechanical alteration of the graft
may have predisposed the graft to failure when the patient returned
to sports activity. Wall et al., in a study of bovine tendons, showed
that shrinkage of greater than 15 to 20 percent resulted in substantial
material property changes and denaturation of the fibrillar microstructure12. The amount of shrinkage required
to decrease anterior-posterior translation from greater than ten millimeters
to less than three millimeters (assuming that the anterior cruciate
ligament in our patient was thirty millimeters in length) would
be on the order of 40 percent. Therefore, the estimated degree of
shrinkage required in this case was markedly greater than what is
recommended in the literature.
Second, complete resorption of the graft occurred, suggesting
that alteration in the collagen structure of the graft and heat-related
necrosis may have led to immune-mediated autodigestion. While other mechanisms
may have been responsible for this phenomenon, none are apparent
to us.
We acknowledge the limitations of a single case report; however,
on the basis of our findings, we caution against the use of thermal
shrinkage in the treatment of a lax anterior cruciate ligament graft. We
believe that additional work is necessary to characterize the response
of host and graft collagen tissue to laser and thermal energy before
this modality can be safely applied to such ligament grafts.