Brace-Free Rehabilitation, with Early Return to...

[Reprinted from THE JOURNAL OF BONE AND JOINT SURGERYVol. 78.A, no. 63 pp.814-825, June 1996]

Copyrighted 1996 by The Journal of Bone and Joint Surgery, Inc.Printed in U.S.A

Brace-Free Rehabilitation, with Early Return to Activity,for Knees Reconstructed with a Double-Looped

Semitendinosus and Gracilis Graft*†BY LIEUTENANT COLONEL STEPHEN M. HOWELL‡

AND CAPTAIN MICHAEL A. TAYLOR§, MEDICAL CORPS, UNITED STATES AIR FORCE RESERVE

Investigation performed at the Clinical Investigation Facility, David Grant Medical Center, Travis Air Force Base

Abstract: Forty-one patients in whom operative reconstruction of a torn anterior cruciate ligament had been performed by one surgeon with use of a double-looped semitendinosus and gracilis hamstring graft were studied to determine (1) if a brace-free rehabilitation program compromised the early & ability of the knee; (2) if the stability of the knee deteriorated between four months, when the patient returned to unrestricted activi-ties, and two years; and (3) if the function of the treated knee was completely restored by four months after the operation. The graft was placed arthroscopically, with-out impingement by the intercondylar roof, and was fixed within the tibial tunnel to conserve the length of the graft. The stability and function of thirty-seven of the knees were assessed at four months as part of a larger pro-spec-tive study. Four patients chose not to return for the four-month evaluation. The patients returned to unrestricted sports and work activities after the four-month evalua-tion. At two years, all forty-one patients were evaluated.

At four months, after completion of the brace-free rehabilitation program, thirty-three (82 per cent) of the thirty-seven patients had an absent pivot shift and a normal Lachman test. Twenty-eight (88 per cent) of thirty-four knees had less than three millimeters of dif-ference in laxity compared with the contralateral knee, as determined by testing at the maximum manual force with use of a KT-1000 arthrometer. Stability remained unchanged at two years, justifying the early return to vigorous activities at four months. The girth of the thigh, the extension of the knee, and the Lysholm and Gillquist score were the same at four months as at two years, veri-

814 THE JOURNAL OF BONE AND JOINT SURGERY

fying the success of the brace-free intensive rehabilita-tion program in the restoration of early function to the treated knee. However, some continued improvement was observed in the performance of the one-leg-hop for dis-tance test between four months and two years.

The two structures that are most commonly used for autogenous grafts in the reconstruction of a torn anterior cruciate ligament are the patellar ligament and the ham-string tendons. Most surgeons prefer the patellar ligament because it is readily procured, can be fixed firmly, and tolerates the loads produced by an intensive rehabilitation program. Patients who have a patellar ligament autog-enous graft can return to vigorous activities two to six months after the operation without affecting the stability of the knee at two years after the operation.8,23,24 However, there are concerns that the same rapid rehabilitation may not be successful with the use of other graft materials.17 To our knowledge, the outcome of an intensive rehabilita-tion program and an early return to sports and work activi-ties for patients who have had a reconstruction of the knee with the use of hamstring tendons has not been reported.

This observational study was designed to measure the function and stability of knees that had been recon-structed with a double-looped semitendinosus and gracilis hamstring graft. A functional assessment and arthrometric measurements of stability were made at four months, after completion of a brace-free, intensive rehabilitation pro-gram, and at two years after the operation. The objectives of the study were to determine (1) if the rehabilitation program compromised the early stability of the knee, (2) if the stability of the knee deteriorated after the patients returned to unrestricted activities at four months, and (3) if the function of the knee was completely restored by four months after the operation.

Materials and Methods

PatientsForty-nine consecutive operations were performed by

the senior one of us (S. M. H.), from February 1991 to May 1992, with the use of a four-bundle graft consisting

*One or more of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, educational institution, or other non-profit organization with which one or more of the authors is associ-ated.

†The views expressed in this article are those of the authors and do not reflect the official policy or position of the United States Department of Defense or the United States Government.

‡8100 Timberlake Way, Suite F. Sacramento, California 95823. Please address requests for reprints to Dr. Howell.

§Clinical Investigation Facility, David Grant Medical Center, Travis Air Force Base, California 94535-5300.

BRACE-FREE REHABILITATION, WITH EARLY RETURN TO ACTIVITY, FOR KNEES

TABLE ISTABILITY AND FUNCTIONAL DATA ON THE EIGHT PATIENTS WHO WERE NOT INCLUDED IN THE STUDY*

Variable Case 1 Case 2 Case 3 Case 4 Case 5 Case 6 Case 7 Case 8†

Reason patient was not Could not Moved out Moved out Moved out Died 14 mos. Moved out Moved out Moved out evaluated at 2 yrs. be located of state of area of state postop. of state of state of areaLast objective evaluation 1 2 2 2 4 7 13 13 (mos. postop.)Girth of thigh‡ (cm) 5 cm proximal to superior NA 0 -0.5 -1 0 1 0 1 pole of patella 15 cm proximal to superior NA 0.5 -3.5 -3 -2 0 1 2 pole of patellaExtension of knee (degrees) NA 3 loss§ 0 0 0 0 0 4 loss§Manual maximum test¶ (mm) NA 0 4 0 0 0 0.5 -7End point on Lachman testing Firm Firm Soft Firm Firm Firm Firm FirmPivot shift 0 0 1 0 0 0 0 0Lysholm and Gillquist score NA NA NA NA 97 100 9.5 l00 (points)Hop index# (per cent) NA NA NA NA 100 102 85 100Re-evaluated by telephone NA 49 24 45 14 NA 43 40 (mos. postop.)Level of athletic activity7 NA Strenuous Moderate Strenuous Strenuous Strenuous Strenuous ModerateReconstructed knee stable NA Yes No Yes Yes Yes Yes Yes at last contactReoperation NA Hardware None None None NA None None

removed

815

*NA = not available.†The patient had a bilateral tear of the anterior cruciate ligament.‡The values are given as the difference between the treated and the contralateral limb.§The treated knee had hyperextension but it did not equal that of the contralateral knee.¶The values are given as the difference in anterior laxity between the treated and the contralateral knee.#The hop index was the mean distance hopped by the treated limb divided by that hopped by the contralateral limb, with the result multiplied by 100. A

hop index of 85 per cent or more was considered normal.

VOL. 78-A, NO. 6, JUNE 1996

of a loop of the semitendinosus and a loop of the gracilis tendon to replace a torn anterior cruciate ligament. Only knees that were available for follow-up after two years were included in the study. Eight patients were excluded because they did not return for the two-year evaluation (Table I). The most recent stability measurements for four of these patients were recorded four to thirteen months after the operation; one of the four died fourteen months postoperatively, two were re-evaluated by telephone three years after the operation, and one was lost to follow-up. The most recent stability measurements for three other patients were recorded at two months; one of these patients had an unstable knee, and two were re-evaluated by tele-phone approximately four years after the operation. The remaining patient was lost to follow-up after one month.

Therefore, the present study consisted of forty-one patients; twenty-eight were male and thirteen were female. The mean age at the time of the index operation was thirty-three years (range, fifteen to forty-eight years). All patients were athletically active before the injury, and thirty-eight (93 per cent) injured the knee during a sports activity. The mode of injury was a non-contact, decelera-tion maneuver on a planted foot for twenty-six (63 per cent) of the forty-one patients and a fall from a height or

a collision for fifteen (37 per cent). Thirty-four patients had had no operation on the knee

before the index procedure. Six patients had had one pre-vious operation. Four of them had had an arthroscopic par-tial medial meniscectomy; one, an arthroscopic medial and lateral meniscectomy; and one, an arthroscopic debride-ment. One patient had had two previous operations: an open repair of the medial collateral ligament and an arthroscopic debridement Twenty-five index procedures were performed within six months after the injury, and sixteen were performed after six months. Two patients had a tear of the medial collateral ligament.

All of the patients had an attenuated anterior cru-ciate ligament diagnosed preoperatively and confirmed intraoperatively. Preoperatively, stability was determined objectively with a KT-1000 arthrometer (Med-Metric, San Diego, California) at an applied anterior load of eighty-nine newtons and with a maximum manual anterior force. The difference in anterior displacement between the injured and contralateral knees was calculated. All of the injured knees had at least three millimeters more ante-rior laxity than the contralateral knee, which is diagnostic of an attenuated anterior cruciate ligament.5,6 Subjectively, the end point on Lachman testing was soft in all injured

knees. Intraoperatively, thirty-nine knees had a fully pos-itive pivot shift and, during arthroscopy, all of the ante-rior cruciate ligaments were observed to be attenuated or absent.

The forty-one patients returned for evaluation at a mean of twenty-six months (range, twenty-four to thirty-two months) after the index operation. To control for examiner bias, the one of us who did not perform the operations (M. A. T.) independently examined thirty-seven patients at the most recent evaluation. Because of schedul-ing conflicts, the senior one of us evaluated the remaining four patients. Stability and functional data were also avail-able for thirty-seven patients at four months postopera-tively. These data were obtained from a larger prospective study in which information was collected preoperatively and at one, two, and four months after the procedure by the senior one of us.

Functional AssessmentThe patients categorized their level of activity before

the injury, after the injury but before the operation, and two years after the operation. They chose one of four cat-egories. Strenuous activities included sports that required jumping, pivoting, and hard cutting maneuvers, such as football, soccer, and basketball; moderate activities included those that required strenuous manual work, such as skiing, tennis, baseball, and volleyball light activities included those that required light manual work, such as jogging, running, and cycling; and sedentary activities included housework and desk jobs with no participation in sports.

The girth of the thigh five and fifteen centimeters proximal to the superior pole of the patella as well as the extension of the knee were measured on the treated and contralateral sides. The difference between the girth of the thigh of the treated limb and that of the contralateral limb was calculated, and the girth of the thigh of the treated limb was assigned to one of four categories: (1) at least two centimeters greater than that of the contralateral limb, (2) within one centimeter of that of the contralateral limb, (3) two centimeters less than that of the contralateral limb, or (4) more than two centimeters less than that of the con-tralateral limb. The difference in extension of the knee was also calculated and the extension of the treated knee was assigned to one of three categories: (1) hyperexten-sion equal to that of the contralateral knee, (2) hyperex-tension not equal to that of the contralateral knee, and (3) extension to 0 degrees.

The patients performed a one-leg-hop for distance test by standing on one limb, hopping as far as possible, and landing on the same limb. The distance was measured and recorded. Alternating between the treated and the contra-lateral limb, each limb was tested three times. The hop index was the mean distance hopped by the treated limb divided by the mean distance hopped by the contralateral limb, with the result multiplied by 100. A hop index of 8.5 per cent or more was considered normal.4

The Lysholm and Gillquist score was used to assess the subjective function of the knee, with the help of the patient. Points were assigned for the level of function within several categories: degree of limp; level of weight-bearing; ability to climb stairs and to squat; degree of atrophy of the thigh; and sensation of instability, pain, or swelling during walking, running, or jumping. The score for a normal knee is 100 points.

The International Knee Documentation Committee form was used to evaluate the function of the knee at the most recent follow-up examination.7 Knees were graded as normal (A), nearly normal (B), abnormal (C), or severely abnormal (D) in seven categories: the patient’s assess-ment of the function of the knee, symptoms (such as pain, swelling, and giving-way), motion, stability, crepitus in each knee compartment, morbidity at the donor site, and the one-leg-hop for distance test. The lowest grade in any category was used as the final result for that knee. The grade for a normal knee is A.

Assessment of StabilityThree tests were used to evaluate the stability of the

reconstructed knee. The result of the Lachman test for the treated knee was graded as having either a firm or a soft end point. The result of the pivot-shift test was assessed by comparison of the degree of rotatory subluxation in the treated knee with that in the contralateral knee. The dif-ference in the KT-1000 displacement values between the treated and the contralateral knee was calculated. Ante-rior displacement was recorded to the nearest 0.5 milli-meter with the knee in 20 to 30 degrees of flexion at two different loads. An eighty-nine-newton load was applied with use of the force handle on the arthrometer, and a maximum manual anterior translation was measured by manual application of a high anterior force to the proxi-mal aspect of the calf just distal to the knee joint line.

Stability was determined with the combined results of the Lachman test, the pivot-shift test, and the arthro-metric laxity measurements. A stable knee had three find-ings: a firm end point, no pivot shift or a subtle pivot glide equal to that of the contralateral knee, and a difference of less than three millimeters between the anterior displace-ment of the treated knee and that of the contralateral knee during a maximum manual force. An unstable knee had a soft end point, an increased pivot shift compared with that of the contralateral knee, or an increase in anterior displacement of three millimeters or more.

Roentgenographic AssessmentWe previously described a measurement technique to

determine the percentage of impingement by the intercon-dylar roof, the location of the central axis of the tibial tunnel, and the slope of the intercondylar roof.11-13 These measurements were performed on a lateral roentgenogram of the fully extended knee, made at the most recent fol-low-up examination (Fig. 1).

Impingement by the roof was assessed by study of the relationship of the tibial tunnel to the point of intersec-

S.M. HOWELL AND M.A. TAYLOR816

THE JOURNAL OF BONE AND JOINT SURGERY

tion of the line of the slope of the intercondylar roof with the plane of the articular surface of the tibial plateau. The plane of the tibial plateau was defined by a line between the most superior points of the anterior and posterior mar-gins of the proximal end of the tibia. The percentage of impingement by the roof was calculated by measurement of the distance on the line of the tibial plateau from the point where the line of the anterior edge of the tibial tunnel intersected the plateau to the point where the line of the slope of the intercondylar roof intersected the pla-teau. This distance was then divided by the width of the tibial tunnel, and the result was expressed as a percent-age.13

The location of the central axis of the tibial tunnel was calculated by extension of the line of the central axis of the

tibial tunnel to its intersection with the line of the tibial plateau; the distance from this intersection to the anterior end of the line of the tibial plateau was then measured. This distance was divided by the length of the line of the tibial plateau, and the result was expressed as a percent-age.11,13,16

The slope of the intercondylar roof was measured as the angle subtended by the line of the slope of the inter-condylar roof and the long axis of the femur.11,13,16

ComplicationsA review of the chart and consultation with the patient

at the most recent follow-up examination were used to determine the occurrence of complications. Complications were divided into those associated with procurement of the graft and those related to the intraarticular portion of the operation. The potential morbidity that is associated with procurement of the graft includes wound infection, superficial phlebitis, deep vein thrombosis, loss of sensa-tion in the skin overlying the proximal-lateral aspect of the tibia due to injury to the infrapatellar branch of the saphenous nerve, and weakness during flexion of the knee. The potential morbidity associated with the intraarticular portion of the operation includes infection, arthrofibrosis, prominence of the hardware, and the need for an addi-tional operation.

Operative TechniqueThe senior one of us previously described the intraar-

ticular, arthroscopically assisted reconstruction of the anterior cruciate ligament with use of a double-looped semitendinosus and gracilis autogenous graft by means of a two-incision technique.9

Briefly, the semitendinosus and gracilis tendons were procured with a tendon-stripper (Acufex Micro-surgical, Norwood, Massachusetts). Retained muscle was removed: and sutures were sewn to each end of each tendon. The midpoint of each tendon was looped over a single suture. The suture was used to pull the four-bundle graft through a series of calibrated cylinders (Arthrotek, Ontario, Cal-ifornia). The diameter of the snuggest-fitting cylinder defined the diameter of the four-bundle graft and was used to select the diameter of the cannulated reamer to drill the tibial and femoral tunnels. The four-bundle graft was seven, eight, or nine millimeters in diameter. Four knees (10 per cent) were treated with a seven-millimeter graft; twenty-eight knees (68 per cent), an eight-millime-ter graft; and nine knees (22 per cent), a nine-millimeter graft.

Of the thirty-six knees that had not had a previous operation on the medial meniscus, nine had a partial exci-sion of that structure and five had an arthroscopic suture repair; three knees had an incomplete tear that was left alone. Eight of the forty knees that had not had a previous operation on the lateral meniscus had a partial excision of that structure; an incomplete tear was left alone in three other knees.

BRACE-FREE REHABILITATION, WITH EARLY RETURN TO ACTIVITY, FOR KNEES 817

VOL. 78-A, NO. 6, JUNE 1996

FIG. 1The percentage of impingement by the intercondylar roof, the loca-

tion of the center of the tibial tunnel, and the slope of the intercondylar roof were determined from several landmarks on the lateral roentgenogram of the fully extended knee. The plane of the tibial plateau was defined by line AB. The center of the tibial tunnel (CTT, arrow) bisected the distance from the anterior edge of the tibial tunnel (AETT) to the posterior edge of the tibial tunnel (PETT). The slope of the intercondylar roof was defined as the angle subtended by the line of the slope of the intercondylar roof (IR, dotted line) and the long axis of the femur.

A previously described technique9,10,13,14 was used to customize the placement of the tibial tunnel to account for variability in extension of the knee and the slope of the intercondylar roof among the knee11 and to avoid impinge-ment by the roof. With use of a guide system (Impinge-ment-Free Tibial Guide System; Arthrotek), the center of the tibial tunnel was positioned four to five millimeters posterior and parallel to the slope of the intercondylar roof with the knee in maximum extension within the posterior half of the insertion of the anterior cruciate ligament. The tibial tunnel was drilled, and bone was removed from the intercondylar roof and wall. Elimination of impingement by the roof was confirmed when a metal rod (Impinge-ment-Free Tibial Guide System), the same diameter as the graft, could be advanced freely through the tibial tunnel into the intercondylar notch with the knee in full extension (Fig. 2). The center of the femoral tunnel was positioned five to seven millimeters distal to the proximal edge of the intercondylar roof, at the eleven o’clock orientation for the right knee or the one o’clock orientation for the left knee, with use of a rear-entry or front-entry femoral guide system (Acufex Microsurgical). The femoral tunnel was drilled through the lateral incision.

In our experience, the fixation method that we used in previous studies13-15 did not achieve consistent fixation of the graft directly to bone in approximately 20 per cent of the knees because the gracilis tendon was too short. In the patients in whom this was the case, a suture bridge had to be used to link the short tendon to a fixation post. There were voids between the tapered ends of the tendon and the wall of the bone tunnel. We were concerned that the con-version from mechanical to biological fixation would be protracted if the tunnel was not filled with tendon, and we were reluctant to use this method of fixation in conjunc-tion with intensive rehabilitation.

To satisfy our concerns regarding fixation, we devised a method to anchor short grafts as well as longer grafts consistently without a suture bridge. The length of the graft was conserved by looping the midpoint of each tendon around a fixation post (a 4.0-millimeter- diameter small-fragment cancellous-bone screw; Synthes, Paoli, Penn-

sylvania) countersunk inside the tibial tunnel. A drilling device (Tibial Fixation Device; Arthrotek) positioned the fixation post twenty millimeters inside the tibial tunnel, as measured from the distal end of the tunnel9 (Fig. 3). The smooth section of the screw spanned the bore of the tibial tunnel while the threaded section gained purchase in the cancellous metaphysis and the posterior cortex of the tibia (Fig. 4). The graft was inserted so that the free ends of each tendon extended outside the femoral tunnel while the midpoint of each tendon was looped around the recessed screw. With the knee in full extension, an unmeasured tension was applied manually to the free ends of the ten-dons exiting the femoral tunnel. The graft was secured to the lateral femoral cortex in thirty-nine knees with one or



FIG. 2Illustration demonstrating how bone was removed from the wall and roof of the intercondylar notch until a metal rod of the same diameter as the craft

could be freely advanced into the notch with the knee in maximum extension. The elimination of impingement by the roof was confirmed before the tendon graft was inserted into the knee.

FIG. 3Illustration demonstrating how a drilling device was inserted into the

tibial tunnel to drill a 2.7-millimeter-diameter hole perpendicular to the long axis of the tibia. The drill-hole bisected the cross section of the tibial tunnel and was located twenty millimeters inside the tunnel, as measured from the distal end of the tunnel. This allowed the graft to be fixed within the tibial tunnel, which conserved the length of the graft.

two 6.5-millimeter canceIlous-bonescrews and ligament washers (Synthes). In two knees, two soft-tissue staples (Stryker, Kalamazoo, Michigan) were used because the ligament washers were unavail-able. Fragments produced by the reaming of the bone were impacted into the tibial tunnel with an impinge-ment rod to fill any voids between the tendon, screw,and tunnel wall.

Rehabilitation ProgramThe postoperative regimen included application of a

soft dressing which was kept on for forty-eight hours; continuous passive motion for the first twenty-four to forty-eight hours after the operation; toe-touch weight-bearing for three weeks followed by walking without crutches after three weeks; unrestricted closed and open-chain knee-extension exercises beginning at four weeks resumption of running in a straight line at eight to ten weeks; and an unrestricted return to sports and work activ-ities at four months. No braces were used.

Analysis of the DataA paired Student t test was used to compare continu-

ous data at the four-month and two-year follow-up exami-nations. The Wilcoxon signed-rank test was used when repeated comparisons of ordinal data were required. The Fisher exact test was used when 2 x 2 comparison of nom-inal data we’re appropriate. To determine the percentage of knees that had improved or worsened between four months and two years, paired differences were calculated for the girth of the thigh, the extension of the knee, and the instrumented laxity measurements

The interobserver analysis of the instrumented laxity measurements was performed with use of the contralateral knee because its anterior laxity was assumed to remain (constant over time. This is in contrast to the anterior laxity of the treated knee, which could have increased over time as a result of remodeling of the graft. A paired Student t test was used to compare the maximum manual anterior translation measured in the contralateral knee by the senior one of us at four months with that measured by the other one of us at two years. The four knees examined by the senior one of us at two years were excluded from this analysis.

Results

Functional AssessmentThe level of sports activity was significantly improved

by the operation (p = 0.009) but was not restored to the pre-injury level (p = 0.01) (Table II). Thirty-eight (93 percent) of the forty-one patients had returned to either strenuous (twenty-five patients; 61 per cent) or moderate (thirteen patients; 32 per cent) activities by two years after the operation.

With the numbers available, we could detect no sig-nificant improvement in the girth of the thigh, as measured either five centimeters (p = 0.20) or fifteen centimeters (p = 0 .1 2 proximal to the superior pole of the patella

between the four-month evaluation (per-formed by the senior one of us) and the two-year evaluation (performed by the other one of us, except for four knees) (Table III). The two-year measurement of the girth of the thigh at five centimeters was equal to the four-month measurement for sixteen (43 per cent) of the thirty-seven treated limbs that were examined at both four months and two years. The measurements were within one centimeter of each other for twenty-nine limbs (78 per cent), and they were within two centimeters of each other for thirty-six (97 per cent) (Fig. 5). The two-year measurement at fifteen centime-ters was equal to the four-month measurement for nine (24 per cent) of the thirty-seven treated limbs, was within one centimeter of it for twenty-seven (73 per cent): was within two centimeters of it for thirty (81 per cent), and was within three centimeters of it for thirty-five (94 per cent).

By four months, all patients had regained extension of the knee to at least 0 degrees, with thirty-five (95 per cent) of the thirty-seven knees that were examined at both four months and two years having some measure of hyperex-tension. With the numbers available, there was no sig-nificant improvement in extension between four months and two years (p = 0.79) (Table III). The measurement of extension at two years was equal to that at four months for twenty-one knees (57 per cent), was within 3 degrees of it for thirty-three

(89 per cent), and was within 6 degrees of it for thirty-six (97 per cent) (Fig. 6). Only twenty-nine of the thirty-seven patients who were seen at both four months and two years completed the one-leg-hop for distance test


VOL. 78-A, NO. 6, JUNE 1996

TABLE II

COMPARISON OF LEVELS OF ATHLETIC PARTICIPATION BEFORE THE INJURY OF THE KNEE, PREOPERATIVELY

AND TWO YEARS AFTER THE OPERATION.

Two YearsLevel of Activity? Pre-Injury Preop. Postop‡

Strenous (activities involving 32 (78%) 2 (5%) 25 (61%)jumping, pivoting, and hardcutting, such as football. soccer,and basketball)

Moderate (activities involving 9 (22%) 9 (22%) 13 (32%)strenuous manual work, suchas skiing, tennis, baseballand volleyball)

Light (activities involving light — 11 (27%) 1 (2%)manual work, such as jogging,running, and cycling)

Sedentary (activities such as — 19 (46%) 2 (5%)housework and desk j ob;no sports)

*The values are given as the number of patients.†International Knee Documentation Committee’ categories for level

of athletic participation.‡The level of sports activity was significantly improved by the opera-

tion (p = 0.009) but was not restored to the pre-injury level (p =0.01) (Wil-coxon signed-rank test).

at four months. Eight patients had not regained enough confidence in the treated knee to complete the test. Of the twenty-nine who completed the test, eighteen (62 per cent) achieved a hop index of 85 per cent or more for the treated limb (Table III). The hop index continued to improve significantly between four months and two years (p = 0.0007). At two years, thirty-three (85 per cent) of the thirty-nine patients who completed the test had a hop index of 85 per cent or more. The Lysholm and Gillquist score for the treated knee was virtually the same at four months as it was at two years (p = 0.71). Thirty-seven (90 per cent) of the forty-one patients gave the treated knee a score of 90 points or more.

With the use of the International Knee Documentation Committee form, twenty-six (63 percent) of the treated knees were rated as normal (A); eleven (27 per cent), as nearly normal (B); and four (10 per cent), as abnormal (C) at the time of the most recent follow-up.



Assessment of StabilityThe interobserver analysis of the instrumented laxity

measurements was performed with use of the results of the maximum manual test on the contralateral, uninjured knee. With the numbers available, we could detect no significant difference between the measurements made at four months (by the senior one of us) (average [and stan-dard deviation], 10.4 ± 1.9 millimeters) and those made-at two years (by the other one of us) (average [and standard deviation], 10.2 ± 2.1 millimeters) (p = 0.50). The anterior laxity measured at two years was equal to that measured at four months for sixteen (43 per cent) of the thirty-seven knees examined at both four months and two years, was within one millimeter for thirty (81 per cent), and was within two millimeters for thirty-four (92 per cent)

FIG. 4Illustration demonstrating how the smooth section of the screw spanned

the tibial tunnel and the threaded section had purchase in the posterior cortex of the tibia. The mid-point of both tendons was looped around the tibial fixa-tion post, and the free ends were secured to the lateral femoral cortex with one or two cancellous-bone screws and a soft-tissue washer. Pieces of bone, saved from the reaming of the tunnels, were packed into the distal end of the tibial tunnel between the tendons and the wall of the tibial tunnel.

TABLE III

COMPARISON OF FUNCTION OF THE KNEE AT FOUR MONTHS (BEFORE RETURN TO UNRESTRICTED ACTIVITIES)

WITH THAT AT TWO YEARS AFTER THE OPERATION

4 Mos.* 2 Yrs.* Variable (N = 37+) (N = 41) P Value‡

Girth of thigh§ 0.205 cm proximal to superiorpole of patella

2 cm greater 4 (10%)Within 1 cm 29 (78%) 29 (70%)2 cm less 6 (16%) 5 (12%) >2 cm less

15 cm proximal to superior 0.12pole of patella§22 cm greater 0 4 (10%)Within 1 cm 19 (51%) 23 (56%)2 cm less 11 (30%) 7 (17%)>2 cm less 7 (19%) 7 (17%)

Extension of treated knee 0.79Hyperextends same as 26 (70%) 28 (68%) contralateral kneeHyperextends but not same 9 (24%) 11 (27%) as contralateral kneeExtends to 0 degrees 2 (5%) 2 (5%)

Completed one-leg-hop for 29 39distance test

Hop index 0.0007>l05% 0 3 (8%)95-105% 9 (31%) 24 (62%)85-94% 9 (31%) 6 (15%)75-84% 4 (14%) 4 (10%)6.5-74% 2 (7%) 1 (3%)55-64% 2 (7%) 1 (3%)<55% 3 (10%) 0

Lysholm and Gillquist score 0.71100 points 12 (32%) 16 (39%)95 to 99 points 15 (41%) 12 (29%)90 to 94 points 8 (22%) 9 (22%)79 to 89 points 2 (5%) 4 (10%)

*The values are given as the number of patients.†Four patients did not return for the four-month evaluation.‡According to the paired Student t test.§Cornpared with the contralateral limb.¶The hop index was the mean distance hopped by the treated limb

divided by that hopped by the contralateral limb, with the result multiplied by 100. A hop index of 85 per cent or more was considered normal.

FIG. 5Graph of the percentage of thirty-seven knees with a change in the girth of the thigh between four months and two years. A positive difference indicates

that the measurement of girth was larger at two years than at four months. A negative difference indicates that the measurement was larger at four months than at two years. The measurements were made at five and fifteen centimeters proximal to the superior pole of the patella. The four-month and two-year measure-ments made five centimeters were within one centimeter of each other for 78 per cent (twenty-nine) of the knees and those made at fifteen centimeters were within one centimeter of each other for 73 per cent (twenty-seven) of the knees.

FIG. 6Graph of the percentage of thirty-seven knees with a change in extension between four months and two years. A positive difference indicates that more

extension was measured at two years than at four months. A negative difference indicates that more extension was measured at four months than at two years. The four-month and two-year measurements were equal for 57 per cent (twenty-one) of the knees, were within 3 degrees of each other for 89 per cent (thirty-three), and were within 6 degrees of each other for 97 per cent (thirty-six).

FIG. 7Graph of the results of the interobserver analysis of the instrumented laxity measurements. The anterior translation measured, with the maximum manual

test, in the contralateral knee at four months (by the senior one of us) was compared with that measured at two years (by the other one of us). Measurements were made in the contralateral knee because its anterior laxity was assumed to remain constant over time. A positive difference indicates that more laxity was measured at two years than at four months. A negative difference indicates that more laxity was measured at four months than at two years. The four-month and two-year measurements were identical for 43 per cent (sixteen) of the knees, were within one millimeter of each other for 81 per cent (thirty), and were within two millimeters for 92 per cent (thirty-four).


VOL. 78-A, NO. 6, JUNE 1996



(Fig. 7). The anterior laxity measured at two years with the eighty-nine-newton test was equal to that measured at four months for ten (27 per cent) of the treated knees, was within one millimeter of it for thirty-one (84 per cent), and was within two millimeters of it for thirty-five (95 per cent). The anterior laxity measured at two years with the maximum manual test was equal to that measured at four months for twelve (32 per cent) of the treated knees, was within one millimeter of it for twenty-seven (73 per cent), and was within two millimeters of it for thirty-four (92 per cent) (Fig. 8).

At four months, thirty-three (89 per cent) of the thirty-seven treated knees had an absent pivot shift and a firm end point on Lachman testing (Table IV). Three patients who had a bilateral tear of the anterior cruciate ligament were excluded from the analysis of the instrumented laxity measurements. Twenty-eight (82 per cent) of the thirty-four patients in whom the contralateral knee was unin-jured had either no increase (two patients) or less than a three-millimeter increase in anterior laxity in the treated knee, as determined by the maximum manual test, and the knees were considered stable. The treated knee in two patients who had a three-millimeter increase in anterior laxity was classified as unstable on the basis of the arthro-metric data, even though they had an absent pivot shift and a firm end point on Lachman testing. The remaining four patients who had an increase in laxity of at least three millimeters, as determined by the maximum manual test, also had a positive pivot shift, and the knees were classi-fied as unstable.

With the numbers available, there was no significant deterioration in the stability of the knee between four months postoperatively, when the patients returned to unrestricted sports and work activities, and two years (p =

TABLE IVCOMPARISON OF STABILITY OF THE KNEE AT FOUR MONTHS

(BEFORE RETURN TO UNRESTRICTED ACTIVITIES)WITH THAT AT TWO YEARS AFTER THE OPERATION

Variable 4 Mos.* 2 Yrs.* P ValueNo. of patients 37† 41

No. evaluated by 37 4 senior author (S.M.H.)No. evaluated by 0 37 coauthor (M. A. T.)

Pivot-shift test 0.88‡Absent 33 (89%) 37 (90%)Positive 4 (11%) 4 (10%)

Lachman test 0.88‡Firm end point 33 (89%) 37 (90%)Soft end point 4 (11%) 4 (10%)

89-N test§¶ 0.23#-1 to <O mm 2 (6%) 7 (18%)0 to 2.5 mm 29 (85%) 28 (74%)3 to 4.5 mm 3 (9%) 2 (5%)≥5 mm 0 1(3%)

Manual maximum test§¶ O.56#-1 to <O mm 2 (6%) 4 (11%)0 to 2.5 mm 26 (76%) 30 (79%)3 to 4.5 mm 5 (15%) 1(3%)≥5 mm 1(3%) 3 (8%)

*The values are given as the number of patients.†Four patients did not return for the four-month evaluation.#The Fisher exact test was used for 2 x 2 comparisons.§The values are given as the difference in displacement between the treated knee

and the contralateral knee.¶Three patients had a bilateral tear of the anterior cruciate ligament and were

excluded from the comparison.#According to the paired Student t test.

FIG. 8Graph of the percentage of thirty-seven knees with a change in anterior laxity between four months and two years. A

positive difference indicates that more anterior laxity was measured, with the eighty-nine-newton or maximum manual test, at two years than at four months. A negative difference indicates that more anterior laxity was measured at four months than at two years. The four-month and two-year measurements were identical for 32 per cent (twelve) of the knees, were within one millimeter of each other for 73 per cent (twenty-seven), and were within two millimeters of each other for 92 per cent (thirty-four) when the maximum manual test was used.

0.23 to 0.88) (Table IV). At two years, thirty-seven (90 per cent) of the forty-one treated knees had an absent pivot shift and a firm end point on Lachman testing. Thirty-


VOL. 78-A, NO. 6, JUNE 1996

four (89 per cent) of the thirty-eight patients who had an uninjured, contralateral knee had either no increase (four knees) or less than a three-millimeter increase in anterior laxity in the treated knee, as determined with the maxi-mum manual test. Therefore, at two years, 11 per cent (four) of the thirty-eight knees were classified as unsta-ble.

Two knees that were originally classified as unstable by the senior one of us at four months were reclassified by the other one of us as stable at the most recent fol-low-up evaluation. This reclassification can be explained by the interobserver variability in arthrometric measure-ments of laxity. For example, a patient who h a two-milli-meter increase in anterior laxity has a 38 per cent chance of having either a one or a three-millimeter increase when reexamined by another observer. There-fore, it is to be expected that some knees may be reclassified when mea-surements are made by different observers.

At four months, the senior one of us classified the two knees as unstable because of a three-millimeter increase in anterior laxity and despite a firm end point on Lachman testing and an absent pivot shift. At two-years, the other one of us found only a 2.0 and 2.5 millimeter increase in anterior translation, indicating that these knees were stable. As both authors agreed that the two knees had a firm end point on Lachman testing and an absent pivot shift, and because there is variability in measurements of laxity between observers, it seemed justified to reclassify these two knees as stable. Instability thus occurred in four (10 per cent) of the forty-one knees.

With the numbers available, we could detect no sig-nificant difference in the over-all rate of instability be-tween the twenty-five patients who had been treated within six months after the injury of the anterior cruciate liga-ment and the sixteen patients who had been treated more than six months after the injury (p = 0.63).

Roentgenographic AssessmentImpingement by the roof was eliminated in thirty-

nine (95 per cent) of the forty-one knees, as the anterior border of the tibial tunnel was in line with or posterior to the intersection of the line of the slope of the inter-condy-lar roof with the plane of the articular surface of the tibial plateau. Two knees, both stable, were classified as having mild impingement13, with 10 and 20 percent of the width of the tibial tunnel lying anterior to this intersection. The slopes of the intercondylar roof aver-aged 37 ± 5 degrees but ranged from 26 to 44 degrees. The locations of the central axis of the tibial tunnel averaged 38 ± 5 per cent of the sagittal depth of the tibia, with a wide range (30 to 54 per cent).

ComplicationsTwo patients had a superficial wound infection in the

tibial incision, which was treated in the office with inci-sion and drainage of the wound and at home with changes of the dressing and oral administration of an antibiotic. Four patients had superficial phlebitis involving the saphe-

nous vein that presented within two weeks after the index operation. Treatment consisted of local application of heat and oral administration of a non-steroidal anti-inflam-matory agent. Almost all patients had anesthesia in a several-square-centimeter area of the skin overlying the proximal-lateral aspect of the tibia due to injury of the infrapatellar branch of the saphenous nerve. There were no symptomatic neuromas. One patient had weakness during flexion of the knee. Seven of the forty-one patients had an additional operation: five had removal of all screws and the washer, one had partial excision of the medial meniscus after a failed repair, and one had a lateral release because of pain in the anterior aspect of the knee. No patient needed a manipulation for stiffness. There were no cases of arthrofibrosis or intra-articular infection.

DiscussionThis study was designed to determine if knees recon-

structed with a four-bundle graft, composed of a loop of semitendinosus and gracilis tendon could be safely and effectively rehabilitated without a brace and with the patient returning to vigorous activities four months after the operation. A high rate of instability and poor clinical function at four months would have indicated that the operative technique and rehabilitation program had been poorly designed. An increase in instability between four months and two years would have implied that the com-posite hamstring graft was not mature enough to tolerate the early return to sports and work activities. An improve-ment in the functional assessment of the knee between four months and two years would have shown that the rehabilitation ineffectively restored early function to the knee. A high rate of failure at two years would have sug-gested that a four-bundle hamstring graft was ineffective for replacement of a tom anterior cruciate ligament.

Thirty-seven (90 per cent) of the forty-one treated knees in this study were stable and functional at two years. The knees that were unstable were detected at the four-month follow-up examination.

An analysis of the possible causes for the early fail-ure of the graft in the four patients was not revealing. The rate of failure was higher in the female patients (two of thirteen) than in the male patients (two of twenty-eight); however, because of the low rate of instability, this dif-ference was not significant. The timing of the operation and the condition of the menisci were not implicated; two operations were performed early and the menisci were intact, and two operations were performed later with one meniscus in each knee being partially excised. The opera-tive procedure was consistent; impingement by the roof was avoided in each knee, and there were no intraoperative difficulties. Each patient complied with the postoperative regimen, and none had a reinjury during the rehabilitation phase. Other possible causes for early failure of the graft that were not measured include variability in the preten-sioning of the graft; geometric differences between knees, which may have affected kinematics; and subtle variabil-ity in placement of the femoral tunnel, which may have



affected the tension of the graft. We were unable to deter-mine a specific cause for the early instability.

The stability of the treated knees did not deteriorate between four months and two years, implying that the grafts were mature enough at four months to stabilize the knees effectively. Studies in which an autogenous patel-lar ligament graft was used as a replacement for the ante-rior cruciate ligament have also shown that stability is not affected by a return to vigorous activities as early as two to six month9 or four to six month24 after the recon-struction. Collectively, these studies provide substantial clinical evidence that autogenous graft material used to reconstruct the anterior cruciate ligament in humans is strong and mature enough at four months for the patient to return to sports and work activities safely.

The rate of stability of the knees that had been recon-structed with a double-looped hamstring graft in the pres-ent study matched the rate of stability reported in two studies3,24 and exceeded that reported in five studies1,2,8,20,22 in which an autogenous patellar ligament graft was used to replace the anterior cruciate ligament. Therefore, the double-looped hamstring graft can be used instead of a patellar ligament graft when the goal is an early return to vigorous activities.

The operation involved two principles that may have been important to its success. First, impingement of the graft by the intercondylar roof was avoided, which allowed the knees to regain extension easily without the roof abrad-ing and injuring the graft12-14,16

One explanation for the success of the double-looped-hamstring graft is that it has superior mechanical proper-ties compared with a ten-millimeter-wide patellar ligament graft. The average diameter of the double-looped ham-string graft in our study was eight millimeters, which pro-vides a circular graft with a cross-sectional area of fifty square millimeters. The patellar ligament graft is 3.5 to 4.0 millimeters thick and rectangular19. A ten-millimeter-wide patellar ligament graft has a cross-sectional area of only thirty-five to forty square millimeters. Normal ante-rior cruciate ligaments are an average of five millimeters thick and ten millime-ters wide and have an average cross-sectional area of fifty square millimeters2l, which is iden-tical to that of the double-looped hamstring graft. The

failure strength of the double-looped graft, as calculated from data reported by Noyes et al., is 238 per cent that of the normal anterior cruciate ligament. The failure strength of a ten-millimeter-wide patellar ligament graft is only 138 per cent that of the normal anterior cruciate ligament. The cross-sectional area of the double-looped hamstring graft more closely approximates that of a normal anterior cruciate ligament, and the hamstring graft has a greater margin of strength than the patellar ligament graft. Thus, it is an excellent autogenous graft for reconstruction.

The technique for prevention of impingement by the roof required that the position of the tibial tunnel be cus-tomized to account for the variability, among the knees, in the slope of the intercondylar roof and the extension of the knee11. The slope of the intercondylar roof in the patients in the present study ranged from 26 to 44 degrees, with the center of the tibial tunnel ranging accordingly from 30 to 54 per cent of the sagittal depth of the tibia. The adequacy of the so-called roofplasty and wallplasty was determined before the graft was implanted by insertion of a metal rod, the same diameter as the graft, through the tibial tunnel and into the intercondylar notch with the knee in maximum extension.

The second principle was that firm fixation should be achieved in every knee by conservation of the length of the graft through fixation of the graft within the tibial tunnel. This technique of countersinking the graft was developed because, in our experience13-15, approximately 20 per cent of the gracilis tendons were too short to be directly fixed to bone outside both tunnels. In the present study, the graft was looped around a fixation post within the tibial tunnel, shifting the point of fixation on the tibia four to five centimeters proximally. In every knee, a long, thick piece of the graft exited the femoral tunnel, allow-ing firm fixation of the graft directly to the lateral femoral cortex with a cancellous-bone screw and washer.

In summary: the results of this study support the use of a double-looped semitendinosus and gracilis hamstring graft as an alternative to an autogenous patellar ligament graft when intensive rehabilitation of the knee without a brace and a return to sports or work activities four months after the operation are desired.

References 1. Aglietti, P.; Buzzi, R.; D’Andria, S.; and Zaccherotti, G.: Long-term study of anterior cruciate ligament reconstruc-

tion for chronic instability using the central one-third patellar tendon and a lateral extraarticular tenodesis. Am. J. Sports Med., 20: 38-45,1992.

2. Aglietti, P,; Buzzi, R.; Zaccherotti, G.; and De Biase, P.: Patellar tendon versus doubled semitendinosus and graci-lis tendons for anterior cruciate ligament reconstruction.Am. J. Sports Med., 22: 211-218,1994.

3. Bach, B. R., Jr,; Jones, G. T.; Sweet, F. A.; and Hager, C. A.: Arthroscopy-assisted anterior cruciate ligament reconstruction using patellar tendon substitution. Two- to four-year follow-up results. Am. J. Sports Med., 22: 758-767,1994.

4. Barber, S. D.; Noyes, F. R.; Mangine, R. E.; McCloskey, J. W.; and Hartman, W.: Quantitative assessment of functional limitations in normal and anterior cruciate ligament-deficient knees. Clin. Orthop., 255: 204-214,1990.

5. Daniel, D. M.: Assessing the limits of knee motion. Am. J Sports Med., 19: 139-147,199l.

6. Daniel, D. M.; Stone, M. L.; Sachs, R.; and Malcom, L.: Instrumented measurement of anterior knee laxity in patients with acute anterior cruciate ligament disruption. Am. J Sports Med., 13: 401-407,1985.

7. Engstrom, B.; Wredmark, T.; and Westblad, P.: Patellar tendon or Leeds-Keio graft in the surgical treatment of anterior cruciate ligament ruptures. Intermediate results. Clin. Orthop., 295: 190-197,1993.

8. Glasgow, S. G.; Gabriel, J. P.; Sapega, A. A.; Glasgow, M. T.; and Torg, J. S.: The effect of early versus late return to vigorous activities on the outcome of anterior cruciate ligament reconstruction. Am. J. Sports Med., 21: 243-248,1993.

9. Howell, S. M.: Arthroscopically assisted technique for preventing roof impingement of an anterior cruciate ligament graft illustrated by the use of an autogenous double-looped semitendinosus and gracilis graft. Op. Tech. Sports Med., 1: 58-65,1993.

10. Howell, S. M.: Roof impingement of ACL grafts: diagnosis, cause, prevention, and late surgical correction. In The Crucial Ligaments: Diagnosis and Treatment of Ligamentous Injuries about the Knee, edited by J, A. Feagin, Jr. Ed. 2, pp. 637-648. New York, Churchill Livingstone, 1994.

11. Howell, S. M., and Barad, S. J.: Knee extension and its relationship to the slope of the intercondylar roof. Implications for positioning the tibial tunnel in anterior cruciate ligament reconstructions. Am. J, Sports Med., 23: 288-294,1995.

12. Howell, S. M., and Clark, J. A.: Tibial tunnel placement in anterior cruciate ligament reconstructions and graft impingement. Clin. Orthop., 283: 187-195,1992.

13. Howell, S. M., and Taylor, M. A.: Failure of reconstruction of the anterior cruciate ligament due to impingement by the intercondylar roof. J Bone and Joint Surg, 75-A: 1044-1055, July 1993.

14. Howell, S. M.; Berns G. S.; and Farley, T. E.: Unimpinged and impinged anterior cruciate ligament grafts: MR signal intensity measurements. Radiology, 179: 639-643,199l.

15. Howell, S. M.; Clark, J. A.; and Blasier, R. D.: Serial magnetic resonance imaging of hamstring anterior cruciate ligament autografts during the first year of implantation. A preliminary study. Am. J. Sports Med., 19: 42-47,199l.

16. Howell, S. M.; Clark, J. A.; and Farley, T. E.: Serial magnetic resonance study assessing the effects of impingement on the MR image of the patellar tendon graft. Arthroscopy, 8: 350-358,1992.

17. Johnson, R. J.; Beynnon, B. D.; Nichols, C. E.; and Renstrom, P. A, E H.: Current concepts review. The treatment of injuries of the anterior cruciate ligament. J Bone and Joint Surg., 74-A: 140-151, Jan. 1992.

18. Lysholm, J., and Gillquist, J.: Evaluation of knee ligament surgery results with special emphasis on use of a scoring scale. Am. J Sports Med., 10: 150-154,1982.

1 9. Noyes, F. R.; Butler, D. L.; Grood, E. S.; Zernicke, R. F.; and Hefzy, M. S.: Biomechanical analysis of human ligament grafts used in knee-ligament repairs and reconstructions. J Bone and Joint Surg., 66-A: 344-352, March 1984.

20. O’Brien, S. J.; Warren, R. F.; Pavlov, H.; Panariello, R.; and Wickiewicz, T. L.: Reconstruction of the chronically insufficient anterior cruciate ligament with the central third of the patellar ligament. J Bone and Joint Surg,, 73-A: 278-286, Feb. 1991.

21. Odensten, M., and Gillquist, J.: Functional anatomy of the anterior cruciate ligament and a rationale for reconstruc-tion. J Bone and Joint Surg., 67-A: 257-262, Feb. 1985.

22. Paschal, S. 0.; Stone, K. R.; and Steadman, J. R.: Changes in knee stability after bone-patellar tendon-bone graft anterior cruciate ligament reconstruction and iliotibial band tenodesis. Am. J Knee Surg., 4: 173-185,199l.

23. Shelbourne, K. D., and Nitz P.: Accelerated rehabilitation after anterior cruciate ligament reconstruction. Am. J Sports Med., 18: 292-299,199O.

24. Shelbourne, K. D., and Porter, D. A.: Anterior cruciate ligament-medial collateral ligament injury: nonoperative management of medialcollateral ligament tears with anterior cruciate ligament reconstruction. A preliminary report. Am. J Sports Med., 20: 283-286,1992.


VOL. 78-A, NO. 6, JUNE 1996

Brace-Free Rehabilitation, with Early Return to...

Documents

Transcript of Brace-Free Rehabilitation, with Early Return to...