James S. Keene

The Effect of a Balance Training Program on the Risk of Ankle Sprains in High School Athletes

Background Ankle sprains are the most common musculoskeletal injuries that occur in athletes, and they have a profound impact on health care costs and resources. Hypothesis A balance training program can reduce the risk of ankle sprains in high school athletes. Study Design Randomized controlled clinical trial; Level of evidence, 1. Methods Seven hundred and sixty-five high school soccer and basketball players (523 girls and 242 boys) were randomly assigned to either an intervention group (27 teams, 373 subjects) that participated in a balance training program or to a control group (28 teams, 392 subjects) that performed only standard conditioning exercises. On-site athletic trainers recorded athlete exposures and sprains. Results The rate of ankle sprains was significantly lower for subjects in the intervention group (6.1%, 1.13 of 1000 exposures vs 9.9%, 1.87 of 1000 exposures; P = .04). Athletes with a history of an ankle sprain had a 2-fold increased risk of sustaining a sprain (risk ratio, 2.14), whereas athletes who performed the intervention program decreased their risk of a sprain by one half (risk ratio, 0.56). The ankle sprain rate for athletes without previous sprains was 4.3% in the intervention group and 7.7% in the control group, but this difference was not significant (P = .059). Conclusion A balance training program will significantly reduce the risk of ankle sprains in high school soccer and basketball players.

Stress fractures of the great toe sesamoids

Stress fractures of the great toe sesamoids have not been previously confirmed as a clinical entity. In this report, we present four such cases, and document the diagnosis by histological sections. All of the patients had experienced the insidious onset of activity-related forefoot pain, and had radiographs that demonstrated a sesamoid fracture. None of the fractures healed with 6 weeks of immobilization and prolonged (4-6 months) inactivity. In all cases, the fractured sesamoid was excised, and the hematoxylin and eosin sections of the fracture segments documented the prefracture stress and incomplete repair characteristic of a stress fracture.

Imaging Appearance of the Normal and Partially Torn Ligamentum Teres on Hip MR Arthrography

OBJECTIVEnThe purpose of this study was to evaluate the MR arthrographic appearance of the normal and partially torn ligament teres and to determine if there are imaging criteria for diagnosing partial tears of the ligamentum teres.nnnMATERIALS AND METHODSnOne hundred sixteen patients underwent preoperative MR arthrography and hip arthroscopy. Each MR examination was evaluated independently by two musculoskeletal radiologists for the following: size and width of the ligamentum teres in the proximal, mid, and distal thirds of the ligamentum teres; overall length of the ligamentum; number of bundles (1-3); signal intensity (SI) within the ligamentum teres (low, intermediate, high); ligamentum teres fibers (normal, attenuated, thickened, wavy); ligamentum teres integrity (not torn, degenerated, frayed, partial tear); and femoral head edema at the ligamentum teres origin. Statistical analysis was performed using the Kruskal-Wallis rank sum test and Fisher exact test.nnnRESULTSnTwelve of 116 (10%) subjects had partial ligamentum teres tears. One hundred four subjects had an intact ligamentum teres. The average size and width of the intact ligamentum teres was 12.6 × 4.38, 14.9 × 3.5, and 14.3 × 2.7 mm for proximal, mid, and distal, respectively, with an overall length 27.7 mm. It was most common to visualize two bundles in the proximal portion of the normal ligamentum teres (61 and 64/116). Low, intermediate, and high SI was common on all pulse sequences in normal and partially torn ligamentum teres for both readers (p = 0.33-0.84). For reader 1, there was no statistical difference between ligamentum teres fiber appearance in partial tears (p = 0.20). In contrast, reader 2 found partial tears associated with attenuated and wavy appearance (p = 0.003). Reader 1 diagnosed five of 12 (42%), and reader 2 diagnosed eight of 12 (67%) of the partial ligamentum teres tears (p = 0.47 and p = 0.0004). Edema of the femoral ligamentum teres origin was not associated with partial tears (p = 0.33-0.86). Retrospective review revealed that six partial tears had intra substance linear high SI on T2 images and peripheral irregularity, whereas four other tears had high SI within the ligamentum teres fibers without peripheral irregularity.nnnCONCLUSIONnThe intact and partially torn ligamentum teres can have similar imaging findings on MR arthrography, making the diagnosis of partial ligamentum teres tears difficult. High SI within the substance of the fibers and irregularity suggest partial tearing; however, further research is warranted.

Labral Injuries Due to Iliopsoas Impingement: Can They Be Diagnosed on MR Arthrography?

OBJECTIVEnIliopsoas impingement is a new arthroscopic diagnosis that refers to an anterior labral injury caused by the iliopsoas tendon. Currently, there are no preoperative criteria to establish the diagnosis of iliopsoas impingement. The goal of this study was to determine whether there are imaging criteria that would identify iliopsoas impingement on preoperative MR arthrography.nnnMATERIALS AND METHODSnThis study compared the preoperative MR arthrograms of 23 patients who had iliopsoas impingement diagnosed at hip arthroscopy with the arthrograms of 24 patients who did not have iliopsoas impingement found at hip arthroscopy. All of the arthroscopies were performed by a single orthopedic hip surgeon. In all cases of impingement, there was an isolated injury to the labrum at the 3-oclock position. All were treated by arthroscopic iliopsoas tenotomy performed at the labral level. The MR examinations of the 47 patients were evaluated independently by two musculoskeletal radiologists who were blinded to the diagnosis. The following characteristics of the iliopsoas tendon at the level of the anterior labrum were evaluated: lateral dip, increased signal intensity (SI) between the iliopsoas tendon and labrum, irregularity of the deep margin of the iliopsoas tendon, edema within the iliopsoas tendon or capsule at the 3-oclock position, presence of a labral tear at the 3-oclock position, dimensions of the iliopsoas tendon, and location of iliopsoas tendon as it passed the labrum. Statistical analysis was performed using the Kruskal Wallis test, Fisher exact test, and Cohen kappa. Values for p less than 0.05 were considered significant.nnnRESULTSnNineteen women (mean age, 35 years) and four men (mean age, 36 years) had central iliopsoas impingement. Sixteen women (mean age, 38 years) and eight men (mean age, 35 years) did not have central iliopsoas impingement (p=0.318). For the impingement and nonimpingement groups, lateral dip of the iliopsoas tendon was seen in 15 of 23 (65%) and 17 of 24 (71%) for reader 1 and 18 of 23 (78%) and 11 of 24 (46%) for reader 2, respectively (p=0.76 and 0.036, respectively). There was no difference between the groups for increased SI between the iliopsoas tendon and labrum (p=0.38 and 0.82, respectively), irregular deep margin of the iliopsoas tendon (p=0.61 and 0.35, respectively), thickness of the iliopsoas tendon (p=0.33), or tendon or capsule edema (p=0.37 and 0.77, respectively). Reader 1 found 20 of 23 and reader 2 18 of 23 labral tears at the 3-oclock position in the iliopsoas impingement group, with 13 of 24 and 10 of 24 in the non-iliopsoas impingement group respectively (p=0.024 and 0.017, respectively). The combined iliopsoas tendon width for both readers was 10.2 mm (range, 8.1-14.3 mm) in women and 11.9 mm (range, 11.1-13.4 mm) in men in the iliopsoas impingement group (p=0.0285), and 11.0 mm (range, 9.0-12.6 mm) for women and 11.8 mm (range, 8.7-15.1 mm) for men in the non-iliopsoas impingement group (p=0.159). The iliopsoas tendon most commonly crossed the labrum at the 3-oclock position in both groups (p=0.83-0.17).nnnCONCLUSIONnAn acetabular labral tear at the 3-oclock position should suggest the diagnosis of iliopsoas impingement.

Magnetic resonance imaging of Achilles tendon ruptures

Preoperative magnetic resonance images of three acute Achilles tendon ruptures were correlated with the findings observed during surgical repair of the tendon. Specific comparisons were made regarding the condi tion (shredded, uniform, etc.) and orientation (ante- grade, retrograde, etc.) of the torn fibers, and the width of the diastasis (with and without ankle flexion) between the ends of the tendon. Magnetic resonance imaging (MRI) accurately assessed all of these parameters.

Ultrasound-Guided Corticosteroid Injections for Treatment of Greater Trochanteric Pain Syndrome: Greater Trochanter Bursa Versus Subgluteus Medius Bursa

OBJECTIVEnThe purpose of this study was to evaluate the effectiveness of corticosteroid injections into the greater trochanteric bursa as opposed to the subgluteus medius bursa in patients with greater trochanteric pain syndrome.nnnMATERIALS AND METHODSnWe retrospectively reviewed 183 injections (149 performed in women, 34 performed in men; age range 23-90 years; median, 53 years) performed for treatment of greater trochanteric pain syndrome. A 10-cm visual analog scale survey was used to assess pain level before the procedure and 14 days after the procedure. A 3-mL corticosteroid solution was injected into either the greater trochanteric bursa or the subgluteus medius bursa under direct ultrasound guidance. Procedure images were retrospectively reviewed to determine the site of injection. Diagnostic images obtained at the time of the procedure were also reviewed for findings of tendinopathy, bursitis, and enthesopathy. Statistical analysis of differences in pain reduction was performed, as was analysis for association between pain relief and demographic variables of age, sex, previous injections, and ultrasound findings.nnnRESULTSnSixty-five injections met the inclusion criteria; 56 performed in women and nine performed in men (age range, 30-82 years; median, 53 years). Forty-one injections were into the greater trochanteric bursa and 24 into the subgluteus medius bursa. There was a statistically significant difference in pain reduction between greater trochanteric bursa and subgluteus medius bursa injections with a median pain reduction of 3 as opposed to 0 (p < 0.01). There was no statistically significant association between pain relief and demographic variables or ultrasound findings.nnnCONCLUSIONnCorticosteroid injections into the greater trochanteric bursa may be more effective than injections into the subgluteus medius bursa for treatment of greater trochanteric pain syndrome.

RISK FACTORS FOR RESTRICTED MOTION AFTER ANTERIOR CRUCIATE RECONSTRUCTION

A retrospective review of 373 patients who had undergone anterior cruciate ligament (ACL) reconstruction utilizing the central third of the patellar tendon was undertaken to identify those factors that placed a patient at risk for restricted postoperative motion (flexion < or = 125 degrees or flexion contracture > or = 10 degrees). Stepwise logistic regression analysis determined that the variables most strongly correlated with restricted final range of motion (ROM) were open surgery (P = .0008) and reconstruction performed < or = 7 days after the initial injury (P = .004). Age, associated meniscal repair, or associated collateral ligament injuries did not significantly affect the ROM. A subgroup of 204 patients arthroscopically reconstructed more than 7 days post-injury were significantly less likely to have limited motion when ROM exercises were begun within 2 days of surgery (P = .008). These data support delayed, arthroscopic ACL reconstruction followed by early ROM exercises as useful techniques for avoiding postoperative motion problems.

Correlation of Ultrasound-Guided Corticosteroid Injection of the Quadratus Femoris With MRI Findings of Ischiofemoral Impingement

OBJECTIVEnMRI findings of ischiofemoral impingement (IFI) have been described, but there is little evidence for treatment with ultrasound-guided corticosteroid injection. The purpose of this study was to evaluate the effectiveness of ultrasound-guided corticosteroid injection of the quadratus femoris muscle as a treatment of IFI syndrome and to correlate the MRI findings with injection outcome.nnnMATERIALS AND METHODSnThe medical records of 61 consecutively registered subjects who underwent bony pelvis MRI in which either IFI or quadratus femoris edema was described in the radiology report were retrospectively reviewed. Subjects with MRI findings of IFI and clinical confirmation of pain that could be attributed to IFI were included and divided into injection and control groups based on clinical management. Control subjects had MRI findings and clinical symptoms suggestive of IFI but underwent conservative therapy rather than injection. The control patients had adequate follow-up and clinical documentation to determine their response to treatment. Quadratus femoris muscle edema, fat atrophy, and hamstring tendinopathy were graded from none to severe (grades 0-3). The ischiofemoral and quadratus femoris spaces were also measured. Clinical presentation was classified as typical, somewhat typical, or not typical of IFI. Injection effectiveness was determined by reported pain reduction assessed before, immediately after, and 2 weeks after the procedure with a standard 10-cm visual analog scale. Response to treatment was classified as good (reduction in pain level > 2), mild or partial (reduced by 1 or 2), or no improvement. For patients who did not return their 2-week postinjection pain surveys, injection effectiveness was determined by qualitative assessments found in their clinical notes. A Kruskal-Wallis rank sum test was used to compare effectiveness of injection between groups (p < 0.05). The Fisher exact test was used to evaluate for associations between each MRI finding and injection outcome.nnnRESULTSnOf the 61 patients, 20 patients had both MRI findings and clinical confirmation of pain related to IFI. These 20 patients were included in the study. Fifteen ultrasound-guided injections were performed in seven patients, and these seven patients were included in the injection group (mean age, 47 years; range, 15-66 years); 13 patients were included in the control group (mean age, 42 years; range, 16-62 years). All seven patients in the injection group and 12 of the 13 patients in the control group were women. In the injection group, the mean width of the ischiofemoral space was 12 mm (range, 7-22 mm), and the mean width of the quadratus femoris space was 9 mm (range, 5-16 mm). The mean edema grade was 1.4 (range, 0-3); mean atrophy grade, 1.4 (range, 0-3); and mean hamstring tendinopathy grade, 1 (range, 0-2). In the control group, the mean width of the ischiofemoral space was 9 mm (range, 6-17 mm); mean quadratus femoris space width, 7 mm (range, 3-15 mm); mean edema grade, 1.9 (range, 1-3); mean atrophy grade, 1.2 (range, 0-3); and mean hamstring tendinopathy grade, 1.2 (range, 0-3). No statistical difference was seen between the two groups before treatment. Pain reduction after injection over the 2-week period was statistically significant with a mean reduction of 1.7 (range, 1-2) for the injection group and 0.8 (range, 0-2) for the control group (p < 0.01). Eleven of 15 (73%) of the injections provided good relief, and four of 15 (27%) provided mild relief. None of the 15 injections provided no relief. In the control group, four of 14 (29%) subjects had good relief, three of 14 (21%) had mild relief, and seven of 14 (50%) had no relief (p < 0.01).nnnCONCLUSIONnUltrasound-guided corticosteroid injection of the quadratus femoris muscle shows promise as an effective treatment of IFI syndrome. However, larger longitudinal studies are needed to help establish the role of ultrasound-guided injection in the workup and care of patients presenting with both MRI findings and clinical findings of IFI.

MRI Appearance of the Pectinofoveal Fold

OBJECTIVEnThe pectinofoveal fold is an intraarticular structure of the hip that has had only limited study in the clinical and anatomic literature. This fold may resemble a hip plica; however, symptomatic hip plicae are now being recognized and treated at hip arthroscopy. We wished to determine the frequency and appearance of the pectinofoveal fold on hip MR arthrography. By defining the variations in its appearance, the normal pectinofoveal fold can be distinguished from pathologic hip plicae.nnnMATERIALS AND METHODSnOne hundred fifty-two hip MR arthrography examinations of patients who subsequently underwent hip arthroscopy were retrospectively reviewed. Each MR examination was reviewed for the presence of a pectinofoveal fold. If present, the fold was measured in the anteroposterior, mediolateral, and superior-inferior dimensions; evaluated for smooth or irregular contour; and evaluated for a femoral or capsular site of insertion.nnnRESULTSnThe pectinofoveal fold was visualized on hip MR arthrograms in 144 of the 152 (95%) patients and visualized at hip arthroscopy in 150 of the 152 (99%) patients. The average thickness of the fold was 2.6 mm (range, 1-13 mm) in the mediolateral dimension and 17 mm (range, 1-32 mm) in the anteroposterior dimension. The average length of the fold in the superior-inferior dimension was 23.3 mm (range, 7-44 mm). The pectinofoveal fold had a smooth contour in 75 of the 144 (52%) patients with examinations that showed the fold and an irregular contour in 69 of 144 (48%) patients. The fold was found to insert onto the capsule in 108 of 144 (75%) patients and onto the femur in the remaining 36.nnnCONCLUSIONnThe pectinofoveal fold should almost always be visualized at MR arthrography. The fold can have various appearances and attachment sites, and these normal variations should not be mistaken for fold abnormalities. These findings should be useful in distinguishing this normal structure from normal and pathologic plicae.

Signs of patellar chondromalacia on sagittal T2-weighted magnetic resonance imaging

We incidentally noted distinctive high signal defects or fissures in the patellar articular cartilage on sagittal T2-weighted magnetic resonance (MR) images in 4 patients. At subsequent arthroscopy all 4 patients were found to have patellar chondromalacia. To determine the reliability of these signs, we retrospectively evaluated, in a blinded manner, sagittal T2-weighted MR images of the knee in 75 patients who were undergoing arthroscopic assessment of their patellar articular cartilage. We identified high signal defects or fissures in the patellar cartilage of 5 patients. Patellar chondromalacia was noted at arthroscopy in all 5 patients. Arthroscopy demonstrated patellar chondromalacia in an additional 21 patients with normal MR images. We conclude that high signal defects or fissures on sagittal T2-weighted images are useful signs of patellar chondromalacia. This single imaging sequence will, however, detect only a small number of the cartilage lesions that may be present.