The American Journal of Sports Medicine

We retrospectively matched 42 patients with unilateral chondral damage in the weightbearing zone of one knee compartment according to sex, age, location of chondral damage, and follow-up time. Two groups of 21 patients were formed. One group had chondral damage only. The other group had chondral damage and a meniscal tear treated with partial meniscectomy. After 12 to 15 years, all patients were reexamined. Twenty-nine percent (N = 6) of the patients who had a partial meniscectomy needed repeat meniscal surgery during followup. No patients with isolated chondral damage developed meniscal symptoms, and only three patients underwent minor surgeries (P < 0.02). At the follow-up evaluation, both groups had similar knee function with a mean Lysholm score of 87 points. However, most patients had reduced their sports ac tivities from competitive individual sports before injury to noncompetitive physical fitness exercise at followup. At the roentgenologic examination, patients who had partial meniscectomies had more severe roentgeno logic signs of osteoarthritis than patients who had chondral damage only (P < 0.03). Meniscectomy, va rus knee alignment at the follow-up evaluation (P < 0.04), and age older than 30 years (P < 0.04) at the time of the operation were associated with a higher incidence of osteoarthritis.

The causes and incidence of rotator cuff injuries in patients under the age of 40 has not been clearly established. The present study focuses on a group of 10 male contact athletes with rotator cuff injuries re lated to trauma sustained during football (ages from 24 to 36 years). Symptoms included pain and dysfunction in all 10 patients and a positive shrug sign in 8 of 10. The diagnoses for these patients were two isolated contusions, five partial-thickness tears, and three full- thickness tears. Surgery was performed on all patients after nonoperative treatment failed. Three partial-thick ness tears were arthroscopically debrided. One full- thickness and two partial-thickness tears were repaired using the arthroscopically assisted miniarthrotomy technique. An open repair was performed in two pa tients. Two isolated rotator cuff contusions were arthro scopically debrided. The average followup was 21 months. Nine of 10 athletes returned to active partici pation in football, 7 of these at their preinjury levels. The diagnosis of rotator cuff injury should be consid ered in a contact athlete who has persistent shoulder pain, impingement signs, weakness, and a positive shrug sign. Arthroscopic debridement of the subacro mial space followed by debridement or repair of rotator cuff tears, as clinically indicated, resulted in a marked improvement in function and rapid return to sport for these patients.

We report the clinical outcome of arthroscopic labral reconstruction using a transglenoid suture technique in a young, active-duty military population. Forty-eight patients (49 shoulders) with varying degrees of gleno humeral instability underwent arthroscopic labral re construction using a transglenoid suture technique. All patients had traumatic injuries to their shoulders and all patients had magnetic resonance imaging scans dem onstrating anterior labral tears. Postoperatively, the patients' shoulders were immobilized for up to 6 weeks. At a mean followup of 30 months (range, 12 to 49), 17 of the 41 patients (41%) with preoperative dislocation or subluxation had recurrent instability. Nine of these patients subsequently underwent open reconstruction procedures for recurrent instability. On the basis of the Rowe rating system, 53% had excel lent or good results and 47% had fair or poor results. The overall perioperative complication rate was 14%. Suprascapular nerve palsy occurred in three cases (6%). Using the Fisher exact test, we determined that immobilization for 6 weeks postoperatively correlated with a lower recurrence rate in the patients with a history of glenohumeral dislocation (P = 0.007). The results of arthroscopic labral reconstruction using transglenoid sutures in the military patient are inferior to the reported 3% to 5% recurrence rate with open Bankart procedures, and the transglenoid pin tech nique jeopardizes the suprascapular nerve.

Seventeen patients underwent posterior capsulotendi nous tensioning procedures to eliminate recurrent pos terior glenohumeral instability. Fourteen patients were evaluated an average of 44 months (range, 18 to 98) after surgery. The average patient age was 27 years. Before surgery, all patients were unable to perform their activities of daily living, occupational activities, and athletic activities. Preoperatively, the average pain rating score on a visual analog scale was 5 of 10 at rest and 9 of 10 with activities. Six patients had previous anterior reconstructions. After surgery, the average range of motion was 174° of forward elevation and 69° of external rotation; internal rotation was to the thumb level of T-8. No patient had a recurrence of posterior instability. After surgery, the average pain rating score was 2 of 10 at rest and 4 of 10 with activities. All patients improved after their operations, but four pa tients were minimally disabled from activities of daily living; six patients experienced shoulder fatigue at work; and four patients had difficulty with sports activ ities. Overall, 13 of the 14 patients were satisfied with their surgical procedures and their outcomes.

We studied glenohumeral rotational range of motion in 39 members of the United States Tennis Association National Tennis Team and touring professional pro gram. We took goniometric measurements of internal and external rotation of dominant and nondominant shoulders at the glenohumeral joint with the humerus at 90° of abduction. We categorized the tennis players by age and by years of tournament play. Results were analyzed by total rotation, internal rotation, external rotation, and dominant-to-nondominant shoulder differ ences. In our results, dominant internal rotation of the shoulder declined and the difference between domi nant and nondominant internal rotation increased with both age and years of tournament play. Men and women tennis players showed the same degree of deficits in range of motion. Significant analysis of vari ance statistics were calculated for dominant internal rotation with years of total play, dominant total rotation with years of total play, and nondominant total rotation with age. Moderate negative correlations were found between dominant internal rotation and years of total play and dominant total rotation and years of total play. These results indicate a loss of internal rotation that seems progressive with longer periods of play. This loss of internal rotation of the shoulder is an absolute loss of motion because total rotation also decreases. Early detection and a corrective training program should be considered because adaptations may result in deleterious biomechanics affecting both perfor mance and risk of injury.

We conducted this cadaveric study to define a biome chanical rationale for rotator cuff function in several deficiency states. A dynamic shoulder testing appara tus was used to examine change in middle deltoid muscle force and humeral translation associated with simulated rotator cuff tendon paralyses and various sizes of rotator cuff tears. Supraspinatus paralysis re sulted in a significant increase (101 %) in the middle deltoid force required to initiate abduction. This in crease diminished to only 12% for full glenohumeral abduction. The glenohumeral joint maintained ball- and-socket kinematics during glenohumeral abduction in the scapular plane with an intact rotator cuff. No significant alterations in humeral translation occurred with a simulated supraspinatus paralysis, nor with 1-, 3-, and 5-cm rotator cuff tears, provided the infraspi natus tendon was functional. Global tears resulted in an inability to elevate beyond 25° of glenohumeral abduction despite a threefold increase in middle deltoid force. These results validated the importance of the supraspinatus tendon during the initiation of abduction. Glenohumeral joint motion was not affected when the "transverse force couple" (subscapularis, infraspina tus, and teres minor tendons) remained intact. Signifi cant changes in glenohumeral joint motion occurred only if paralysis or anatomic deficiency violated this force couple. Finally, this model confirmed that rotator cuff disease treatment must address function in addi tion to anatomy.

To determine the role of the acromioclavicular liga ments in controlling scapular rotation about the distal clavicle and the effects of distal clavicle resection, we used 13 fresh shoulders consisting of the clavicle, acromioclavicular ligaments, coracoclavicular liga ments, and scapula. The range of motion was mea sured using a specially designed goniometer for each of the three orthogonal axes of rotation of the scapula with reference to the clavicle: anterior-posterior axial rotation, protraction-retraction, and abduction-adduc tion. We did two experiments involving sequential sec tioning. Range of motion was measured in the intact shoulder and after each sectioning cut. The order of sectioning in Experiment 1 (six shoulders) was 1) the inferior acromioclavicular ligament, 2) removal of 5 mm of the distal clavicle, and 3) the superior acromiocla vicular ligament. In Experiment 2 (seven shoulders) the order was 1) the superior acromioclavicular ligament, 2) removal of 5 mm of the distal clavicle, and 3) the inferior acromioclavicular ligament. The most important results were 1) only 5 mm of the distal clavicle needs to be resected to ensure that no bone-to-bone contact occurs in rotation postoperatively and 2) there was no difference in the end result (for range of motion in any of the three axes) whether the inferior acromioclavicu lar ligament or the superior acromioclavicular ligament was cut before removal of 5 mm of the distal clavicle.

We reviewed 53 of 58 patients who had primary repairs of posterior cruciate ligament injuries between 1981 and 1988. Sixteen patients had isolated posterior cru ciate ligament ruptures, 16 had complex injuries with capsular and collateral ligament involvement, and 21 had additional anterior cruciate ligament ruptures. For ty-six patients were treated by transosseous multiple- loop sutures and seven with bony avulsions by screw osteosynthesis. The mean follow-up time was 7.5 years (range, 3 to 12). All patients were examined subjectively (questionnaire) and objectively (clinical ex amination, KT-1000 arthrometer, functional testing, ra diographs, and Cybex II isokinetic strength analysis). The results were graded according to the International Knee Documentation Committee evaluation form and the Lysholm score. The average Lysholm score was 82.4 (range, 40 to 100). Thirty-eight patients returned to their preinjury activities at the same intensity level. The patients' subjective assessments were normal or nearly normal in 35 patients. The posterior drawer test was negative or 1 + in 46 patients. Cybex isokinetic strength analysis revealed a decrease in quadriceps muscle strength of the involved limb by 10.5% (P < 0.01). Our data suggest that primary repair of posterior cruciate ligament ruptures provides good results after 8 years in approximately two thirds of the patients. Distal ligamentous ruptures, lack of athletic activity, and tem porary olecranization correlated with poor results. Bony avulsions, midsubstance or proximal ruptures, and athletic activity correlated with good results.

We evaluated 38 subjects with isolated posterior cru ciate ligament-deficient knees at a mean of 13.4 years (range, 5 to 38) after injury to study the occurrence of symptoms, disabilities, and articular degeneration. Each subject completed a standardized questionnaire, physical examination, and had radiographs taken of both knees. Eight (21%) patients had surgeries for meniscal injuries after their posterior cruciate ligament injuries. The mean questionnaire score for function (50-point maximum) was 34.4 ± 6.5 (SD) for the pa tients who did have meniscal surgeries versus 40.0 ± 8.7 for the 30 patients who did not (P = 0.05). Among the 30 patients with isolated posterior cruciate liga ment-deficient knees with normal menisci, 24 (81%) had at least occasional pain and 17 (56%) had at least occasional swelling. As time from injury increased, increased articular degeneration on radiographs was seen (P = 0.037). Our study suggests that the prog nosis for the isolated posterior cruciate ligament-defi cient knee varies. Some patients experience significant symptoms and articular deterioration, while others are essentially asymptomatic and maintain their usual knee function.

We have recently become aware of a strong direct attachment of the popliteal tendon to the fibula. To investigate the importance of this attachment, we ex amined 20 cadaveric knees. The popliteofibular liga ment was identified in all 20 knees. The cross-sectional area of the popliteofibular ligament was 6.9 ± 2.1 mm ², compared with 7.2 ± 2.7 mm² for the lateral collateral ligament. Biomechanical testing of these structures, simulating a purely varus stress on the knee, revealed that the lateral collateral ligament always failed first, followed by the popliteofibular ligament, and then the muscle belly of the popliteus. The mean maximal force to failure of the popliteofibular ligament approached 425 N (range, 204 to 778), compared with 750 N (range, 317 to 1203) for the lateral collateral ligament. Our results indicate that the popliteofibular ligament contributes to posterolateral stability.

We conducted a prospective study of soccer injuries during the Soccer America Dawn to Dark Indoor Soc cer Tournament, which was organized by the Lake Placid Soccer Center, Lake Placid, New York, 1993. Eight hundred twenty-four players competed in open men's, open women's, over-30 men's, and mixed divi sions. The overall rate of injury per 100 player hours was 4.44, with a rate of 5.79 in the open men's, 4.74 in the open women's, 2.73 in the over-30 men's, and 1.54 for the mixed divisions. The differences in injury rates for men versus women and men versus older men were not statistically significant. Twenty-five of the 38 injuries (65.8%) were mild, with 27 injuries (71.4%) occurring in the lower extremities. Ankle sprains were the most common injuries and combined ligamentous injuries to the knee were the most common severe injuries. As the injuries increased in severity, they were more likely to be noncontact injuries. The data demon strate the low incidence of injury in male and female indoor soccer participants. The data also show the similarity in the types of injuries sustained by indoor and outdoor soccer players.

A prospective study of hurling injuries was conducted over the 8 months of one season on 74 players. These athletes averaged 4.30 ± 2.58 hours per week of training and 1.15 ± 0.21 hours per week of matches. Mean time of injury was 1.20 ± 2.53 days in the hospital, 20.34 ± 19.25 days off sport, and 13.34 ± 17.25 days of restricted activity. Together this injury time amounts to 14.3% of the season. There were 92 match- and 43 training-related injuries, giving 342.47 injuries per 10,000 hours of matches and 43.83 injuries per 10,000 hours of training. Overall, there were 369.9 days of injury per 1000 hours of participation. The most common type of injury was muscle strain (24.4% of the 135 total injuries). The hamstrings was the most com mon site of strain, accounting for 41 % of these injuries. Contusions comprised 16.3% of the injuries and sprains comprised 15.6%. The most frequently injured sites were the finger (13%), hamstrings (12%), back (11 %), head (9%), and knee and ankle (9%). Forty-one percent of the injuries were attributed to foul play. The results of the study suggest that the incidence of inju ries in hurling is high and may be attributed to poor conditioning, poor protection, and lack of enforcement of the rules.

Because of the good initial fixation strength of interfer ence screws used in anterior cruciate ligament recon struction, metal interference screws have become the standard method for fixation of bone-patellar tendon- bone grafts. To avoid some of the complications with metal screws, a bioabsorbable interference screw was developed. Data on fixation strength in older human cadavers indicate a similar failure strength between bioabsorbable and metal screws. We studied the fail ure mechanisms, insertion torques, and fixation strengths of absorbable and metal interference screws in cadaveric knees from young and middle-aged do nors. With identical gap and screw size, the mean insertion torque for the metal screws (mean, 1.5 N-m; SD, 0.8) was significantly higher than for the absorb able screws (mean, 0.3 N-m; SD, 0.19). The mean failure load for the metal screws (mean 640 N; SD, 201) was also significantly higher than for the absorb able screws (mean, 418 N; SD, 118).

To correlate clinical results after anterior cruciate liga ment reconstruction with tunnel placement measured radiographically, we prospectively studied 128 patients who had arthroscopically assisted bone-patellar ten don-bone reconstructions. Patients with bilateral ante rior cruciate ligament reconstructions, other significant knee ligament injuries, or those undergoing chondro plasty or meniscal repairs were excluded, leaving 42 patients. The relationship between radiographic tunnel position and clinical results was determined using the Lysholm score, KT-1000 arthrometer testing, the Teg ner activity level, and the pivot shift and Lachman tests. Clinical results correlated positively with posterior fem oral tunnel placement on lateral radiographs and neg atively with excessive anterior tibial tunnel placement. Specifically, when femoral tunnels were placed at least 60% posterior along Blumensaat's line and tibial tun nels were at least 20% posterior along the tibial pla teau, 69% of patients had good or excellent Lysholm scores and 79% had KT-1000 arthrometer maximum manual side-to-side differences of 3 mm or less. When the above criteria were not met, 50% of patients had good or excellent Lysholm scores and 22% had KT- 1000 arthrometer maximum manual side-to-side differ ences of 3 mm or less. This close correlation indicates that satisfactory radiographic tunnel position influences outcome after anterior cruciate ligament reconstruc tion.

We modeled an intraarticular anterior cruciate ligament graft and investigated the effects of attachment orien tation and twist of the graft on its isometry during quadriceps muscle loading. Physiologic levels of quad riceps muscle loads were applied to 15 intact cadaveric knees. We measured the changes in distance between points on the tibia and femur for knee flexion angles between 0° and 120° using a three-dimensional digi tizer. Selected points on the tibia and femur, represent ing graft attachment sites, allowed us to model the graft as a broad band. Distance was used to approximate graft fiber length. A 180° twist in the graft significantly reduced the maximal range of changes in distance when the graft was attached in the anteroposterior direction. Range is defined as the difference between the largest and smallest changes in distance among the fibers of the graft for a given angle of flexion. This reduction enhanced isometry among the fibers of the graft. Enhanced isometry would be expected to en hance load sharing among these fibers, thereby in creasing the overall strength of the graft. For a graft 10 mm wide and 4 mm thick, the dimensions of a typical patellar tendon graft, the best overall isometry was found when the breadth of the graft was attached to the tibia in the mediolateral direction, to the femur along the most isometric line, and with a 180° twist in the graft.

A review of 119 consecutive anterior cruciate ligament reconstructions showed that the time from injury to surgery (early versus delayed) did not make a differ ence in obtaining full range of motion. Only patients with late surgery had a slight decrease in range of motion. Followup data were obtained for 111 recon structions. Twenty-one were early surgeries (1 to 14 days), 22 were delayed surgeries (15 to 28 days), and 68 were late surgeries (more than 28 days). The pa tients involved in the 21 early surgeries obtained 0° of knee extension or better and 135° of knee flexion or better. The patients involved in the 22 delayed recon structions reached 0° of knee extension or better and 135° of flexion or better. Among the patients with the 68 late surgeries, 93% of the knees reached 0° of extension or better and all reached at least 135° of flexion. The five patients who did not achieve full knee extension had extension loses less than 4°. All 111 reconstructions were determined stable when full range of motion was achieved based on clinical exam ination, which included the Lachman test, anterior drawer test, pivot shift, and KT-1000 arthrometer when appropriate.

We measured the bare ankle and the braced angle- torque relationships in 12 uninjured volunteers under static and dynamic conditions within the full range of inversion motion. These relationships were measured with a specially designed mechanical device that al lowed inversion movements with angular velocities up to 850 deg/sec. In testing the bare ankle under static conditions, the torque showed a 10-fold increase within the full range of motion (average, from 0.9 N-m at 7° to about 8 N-m at 48° of inversion). The slope of the angle-torque relationship increased under dynamic conditions giving higher torque values (up to 18 N-m on average). Both orthoses induced similar additional torques that increased linearly, up to about 6 N-m at 45°, with higher angles of inversion. These additional torques are small compared with the amount of stress applied to the foot during a typical ankle sprain situa tion, such as recovering from a jump. Therefore, we propose that orthotic devices increase the ankle torque, counteracting the inversion movement, and also prevent the start of the inversion movement by preloading and maintaining the ankle in a proper ana tomic position with optimal contact between the artic ular surfaces.

Ten uninjured subjects (ages 18 to 30 years) had electromyographic testing of the peroneus longus, per oneus brevis, and tibialis anterior muscles in response to inversion moments at two speeds (50 and 200 deg/ sec) and two joint angles (neutral and 20° of plantar flexion) using a hydraulically controlled tilt platform. Subjects underwent 10 trials of each type of inversion moment on Day 1 testing, which included both legs. On Day 2, subjects again underwent 10 trials of each type of inversion moment, but only on one leg. Reliability was assessed by comparing left and right leg data within muscle groups for Day 1 testing. Repeatability was assessed by comparing Day 1 with Day 2 data. The latency measurements (the time between the be ginning of the inversion moment and the onset of first motor response) for the peroneus brevis and tibialis anterior muscles were found to be reliable and repeat able with no significant differences between the same muscle groups. The peroneus longus muscle had a significant difference between legs but was found to be highly repeatable. Speed of inversion moment and plantar flexion angle both caused significant changes in latency response of the peroneus muscles, with increased speed producing a shorter latency response and increased angle causing a longer latency re sponse. Our results indicate a loss of protective re flexes with increasing plantar flexion.

We studied prospectively the influence of ankle sprains on proprioception as measured by recording the pos tural sway of classical ballet dancers. Excellent bal ance and coordination are important for classical ballet dancers, and postural stability requires adequate pro prioception from the ankle joint. Fifty-three professional dancers from the Royal Swedish Ballet, Stockholm, and 23 nonathletes, the control group, participated in the investigation. Postural sway was recorded and an alyzed with a stabilimeter using a specially designed, portable, computer-assisted force plate. Six dancers sustained ankle sprains during followup. The record ings were obtained of these dancers before and after the injuries. The stabilometry results differed among the male and female dancers and the control group as follows: 1) the male dancers demonstrated a smaller total area of sway, and 2) both the male and female dancers had a smaller mean sway on the left foot than on the right (no mean difference in sway was found between the left and right foot in the control group). In comparison with the condition before injury and with the uninjured foot, the postural stability of the dancer was impaired for several weeks after the ankle sprain. Postural stability gradually improved during rehabilita tion and improvement still occured several weeks after professional dancing had resumed.

We propose a biomechanical model to explain the pathogenesis of iliotibial band friction syndrome in dis tance runners. The model is based on a kinematic study of nine runners with iliotibial band friction syn drome, a cadaveric study of 11 normal knees, and a literature review. Friction (or impingement) occurs near footstrike, predominantly in the foot contact phase, between the posterior edge of the iliotibial band and the underlying lateral femoral epicondyle. The study subjects had an average knee flexion angle of 21.4° ± 4.3° at footstrike, with friction occurring at, or slightly below, the 30° of flexion traditionally described in the literature. In the cadavers we examined, there was substantial variation in the width of the iliotibial bands. This variation may affect individual predisposition to iliotibial band friction syndrome. Downhill running pre disposes the runner to iliotibial band friction syndrome because the knee flexion angle at footstrike is reduced. Sprinting and faster running on level ground are less likely to cause or aggravate iliotibial band friction syn drome because, at footstrike, the knee is flexed be yond the angles at which friction occurs.

The aim of the present study was to characterize the performance ability of the leg extensor apparatus in a group of athletes with jumper's knee and to compare the results with those of a matched control group with out knee symptoms. Patient and control groups (12 players in each) were selected from a population of 141 well-trained male Norwegian volleyball players, of which 55 (39%) satisfied the diagnostic criteria for jumper's knee. The testing program consisted of a standing jump, a countermovement jump, a 15-second rebound jump test, a standing jump with a 20-kg load, and a standing jump with a load corresponding to one-half of the subject's body weight. Jump height and power were measured using a contact mat connected to an electronic timer. The test results of the patient group were significantly higher than those of the con trol group for the countermovement jump (15% in crease), power during rebound jump (41 %), work done in standing jump (12%) and countermovement jump (22%), and the difference between countermovement jump and standing jump (effect of adding eccentric component). Athletes with jumper's knee demon strated better performance in jump tests than uninjured athletes, particularly in ballistic jumps involving eccen tric force generation.

We used dynamic electromyography and a motion analysis system to describe the muscle firing patterns in 10 shoulder muscles and the basic kinematics of a two-handed overhead medicine ball throw. Ten healthy male subjects with no history of shoulder injury were evaluated. The two-handed medicine ball throw was divided into three phases for analysis: cocking, accel eration, and deceleration. The average duration of the throw was 1.92 seconds; the cocking phase repre sented 56%, the acceleration phase 15.5%, and the deceleration phase 28.5% of the throw. In the cocking phase, the upper trapezius, pectoralis major, and an terior deltoid muscles showed high activity (>40% to 60% maximum manual test), and the rotator cuff mus cles had moderate activity (>20% to 40%). In the acceleration phase, five of the muscles demonstrated high levels of activity (>40% to 60%) and the upper trapezius and lower subscapularis muscles had very high levels of activity (>60%). Analysis of the deceler ation phase revealed high activity in the upper trape zius muscle and moderate activity in all other muscles except the pectoralis major. Our findings support the use of medicine ball training as a bridge between static resistive training and dynamic throwing in the rehabil itation of the overhead athlete. This training technique provides a protective method of strengthening that closely simulates portions of the throwing motion.