Citations

A cohort of 282 elite amateur ice hockey players were analyzed to 1) record the number, type, location, and severity of head, neck, and facial injuries sustained during games; 2) examine the relationship between injuries and the type of facial protection (none, partial, or full) according to individual playing time; and 3) determine whether full or partial facial protection is associated with an increased incidence of concussions, eye injuries, and neck injuries. Fifty-two injuries (158.9 per 1000 player-game hours) occurred in players wearing no facial protection, 45 (73.5 per 1000 player-game hours) in players wearing partial facial protection (half shield), and 16 (23.2 per 1000 player-game hours) in players wearing full facial protection (full cage or shield). Players wearing no protection were injured at a rate more than twice that of players wearing partial protection and almost seven times higher than those wearing full protection. Concussions occurred in four players wearing no protection, five players wearing partial protection, and two players wearing full protection; these differences were not significant. The risk of eye injury was 4.7 times greater for players wearing no protection compared with those wearing partial protection. No eye or neck injuries occurred in players wearing full protection. This study demonstrates that both full and partial facial protection significantly reduce injuries to the eye and face without increasing neck injuries and concussions.

Abstract The meniscal roots are critical in maintaining the normal shock absorbing function of the meniscus. If a meniscal root tear is left untreated, meniscal extrusion can occur rendering the meniscus nonfunctional resulting in degenerative arthritis. Two main repair techniques are described: (1) suture anchors (direct fixation) and (2) sutures pulled through a tibial tunnel (indirect fixation). Meniscal root repair using a suture anchor technique is technically challenging requiring a posterior portal and a curved suture passing device that can be difficult to manipulate within the knee. We present a technique for posterior medial meniscus root repair using 3 sutures (1 leader, 2 cinch), standard arthroscopy portals, and transtibial fixation. Overall, this technique simplifies a challenging procedure and allows for familiarity and efficiency. Technique Video video See video under supplementary data.

Objective To describe the occurrence and trend of concussions during the international ice hockey tournaments. Design Register study. Setting The International Ice Hockey Federation (IIHF) World Championships and Olympic Winter Games. Participants A total of 3293 games were played in the 169 tournaments by 1212 teams (26130 players) comprising 142244 athletic game exposures. Outcome measures One hundred and sixty concussions were observed. Main Results The average injury rate (IR) for concussion was 1.1 per 1000 ice hockey player-games for all IIHF World Championship tournaments (WC). The injury rate was highest in men's WC A-pool tournaments and Olympic Winter Games (OWG) (IR 1.6), and lowest in women's WC and OWG (IR 0.9). However, the annual injury rate for concussion in men's tournament has been lower than that in World Junior tournaments since 2012. When concussion occurred in a contact with a flexible board injury rate was 0.2 per 1,000 player games and 1.1 if a board and glass were traditional [RR 6.4, (95% CI 1.50–27.61]. In men's tournaments the trend of concussions caused by illegal hits decreased over a study period. After the 2012 Zurich Consensus Guidelines none of the concussed players returned to a play during a same day in men's WC tournaments. Conclusions The risk of concussion was significantly lower if games were played on rinks with flexible board and glass. Between 2006 and 2015, the annual risk of concussion in men's WC has been decreasing most likely due to the decreasing number of concussions caused by illegal hits.

Background: Osteochondritis dissecans (OCD) is a disorder of subchondral bone that causes adverse effects on the overlying cartilage and commonly affects the knee. The incidence of OCD and its long-term effect on the knee joint are controversial. Hypothesis/Purpose: The purpose of this study was to (1) evaluate the rate of osteoarthritis and knee arthroplasty in a population-based cohort of patients with OCD lesions treated operatively and (2) evaluate factors that may predispose patients to knee osteoarthritis and arthroplasty. The investigators hypothesized that OCD lesions diagnosed after skeletal maturity and treatment with fragment excision would be predictive of a diagnosis of osteoarthritis. Study Design: Cohort study; Level of evidence, 3. Methods: The investigators identified 221 patients (mean ± SD age, 26.1 ± 13.6 years) with OCD lesions treated operatively between 1976 and 2010 and followed for 16.3 ± 11.4 years (mean ± SD) from diagnosis. Information related to the diagnosis, laterality of lesion, details of treatment, and progression to osteoarthritis was obtained from the medical record. Surgical treatment was classified as fragment excision, fragment preservation (lesion drilling and/or fragment fixation), or chondral defect grafting (osteochondral allograft or autograft). Factors predictive of osteoarthritis and arthroplasty were examined. Results: There were 134 patients in the fragment excision group, 78 patients in the fragment preservation group, and 9 patients in the chondral defect grafting group. In the fragment excision group, the cumulative incidence of osteoarthritis was 12.0% at 5 years, 17.0% at 10 years, 26.0% at 15 years, 39.0% at 20 years, and 70% at 30 years. The cumulative incidence of arthroplasty was 2.0% at 5 years, 4.0% at 10 years, 4.0% at 15 years, 10.0% at 20 years, and 32.0% at 30 years. In the fragment preservation group, the cumulative incidence of osteoarthritis was 3.0% at 5 years, 7.0% at 10 years, 16.0% at 15 years, 25.0% at 20 years, and 51% at 30 years. The cumulative incidence of arthroplasty was 0.0% at 5 years, 0.0% at 10 years, 3.0% at 15 years, 6.0% at 20 years, and 11.0% at 30 years. No patients in the chondral defect grafting group developed osteoarthritis or underwent arthroplasty. Body mass index (BMI) greater than 25 kg/m 2 (hazard ratio [HR] 3.3; 95% CI, 1.6-7.0), older age at diagnosis (HR 4.9; 95% CI, 1.8-17.3), and fragment excision (HR 2.3; 95% CI, 1.2-4.6) were predictive of osteoarthritis. Conclusion: OCD patients treated with fragment excision have a high rate of osteoarthritis and knee arthroplasty at long-term follow-up. In contrast, patients treated with fragment preservation or chondral defect grafting have lower rates of osteoarthritis and arthroplasty. BMI greater than 25 kg/m 2 , older age at diagnosis, and fragment excision were predictive of osteoarthritis.

The purpose of this study was to assess the correlations of acceleration measurements between an instrumented hockey helmet and the Hybrid III headform and to determine whether collision with a broad surface affects the correlations between these data. A CCM Vector hockey helmet, instrumented with the Head Impact Telemetry System, was fitted onto an instrumented Hybrid III headform and dropped onto an inclined plate to simulate impacts in hockey. The helmet and headform were dropped from 0.15, 0.30, and 0.60 m at different orientations to create impacts to the vertex, occipital, right, left, and forehead surfaces. Peak linear and rotational acceleration magnitudes and head injury criteria were assessed for each system. Correlations were assessed with the Pearson correlation coefficient, and linear regression models were generated. Relative error was assessed, and results were compared with the Wilcoxon signed rank test. Peak acceleration magnitudes generally had strong mathematical correlations, although a 1:1 linear relationship for rotational acceleration data was not observed. Differences between helmet and head peak acceleration magnitudes were significant. The linear relationship varied according to the impact direction. Instrumented helmet and head data demonstrated a strong mathematical correlation, but measures of linear acceleration generally had better 1:1 agreement than rotational acceleration measures. There is some dependence in the relationship between the helmet and head data on the impact direction, with vertex strikes showing the weakest correlation. Although data generated from the helmet may be useful for interpreting head accelerations experienced by on-ice players, additional analysis of the time-series data should be performed to understand its full utility.

Background: The rate of patellofemoral arthritis after lateral patellar dislocation is unknown. Purpose/Hypothesis: The purpose of this study was to compare the risk of patellofemoral arthritis and knee arthroplasty between patients who experienced a lateral patellar dislocation and matched individuals without a patellar dislocation. Additionally, factors predictive of arthritis after patellar dislocation were examined. The hypothesis was that the rate of arthritis is likely higher among patients who experience a patellar dislocation compared with those who do not. Study Design: Cohort study; Level of evidence, 3. Methods: In this study, 609 patients who had a first-time lateral patellar dislocation between 1990 and 2010 were compared with an age- and sex-matched cohort of patients who did not have a patellar dislocation. Medical records were reviewed to collect information related to the initial injury, recurrent dislocation, treatment, and progression to clinically significant patellofemoral arthritis (defined as symptoms with degenerative changes on patellar sunrise radiographs). Factors associated with arthritis (age, sex, recurrence, osteochondral injury, trochlear dysplasia) were examined. Results: At a mean follow-up of 12.3 ± 6.5 years from initial dislocation, 58 patients (9.5%) in the dislocation cohort were diagnosed with patellofemoral arthritis, corresponding to a cumulative incidence of arthritis of 1.2% at 5 years, 2.7% at 10 years, 8.1% at 15 years, 14.8% at 20 years, and 48.9% at 25 years. In the control cohort, 8 patients (1.3%) were diagnosed with arthritis, corresponding to a cumulative incidence of arthritis of 0% at 5 years, 0% at 10 years, 1.3% at 15 years, 2.9% at 20 years, and 8.3% at 25 years. Therefore, patients who experienced a lateral patellar dislocation had a significantly higher risk of developing arthritis (hazard ratio [HR], 7.8; 95% CI, 3.9-17.6; P < .001) than individuals without a patellar dislocation. However, the risk of knee arthroplasty was similar between groups (HR, 2.8; 95% CI, 0.6-19.7; P = .2). Recurrent patellar dislocations (HR, 4.5; 95% CI, 1.6-12.6), osteochondral injury (HR, 11.3; 95% CI, 5.0-26.6), and trochlear dysplasia (HR, 3.6; 95% CI, 1.3-10.0) were associated with arthritis after patellar dislocation. Conclusion: Patellar dislocation is a significant risk factor for patellofemoral arthritis, as nearly half of patients have symptoms and radiographic changes consistent with arthritis at 25 years after lateral patellar dislocation. Osteochondral injury, recurrent patellar instability, and trochlear dysplasia are associated with the development of arthritis.

Background: Second anterior cruciate ligament (ACL) tears after reconstruction occur at a reported rate of 20% to 30%. This high frequency indicates that there may be factors that predispose an athlete to graft failure and ACL tears of the contralateral knee. Purpose: To determine the incidence of second ACL injuries in a geographic population-based cohort over a 10-year observation period. Study Design: Descriptive epidemiological study. Methods: International Classification of Diseases, 9th Revision (ICD-9) codes relevant to the diagnosis of an ACL tear and the procedure code for ACL reconstruction were searched across the Rochester Epidemiology Project, a multidisciplinary county database, between the years of 1990 and 2000. This cohort of patients was tracked for subsequent ACL injuries through December 31, 2015. The authors identified 1041 patients with acute, isolated ACL tears. These patients were stratified by primary and secondary tears, sex, age, activity level, side of injury, sex by side of injury, and graft type. Results: Of the 1041 unique patients with a diagnosed ACL tear in Olmsted County, Minnesota, from 1990 to 2000, there were 66 (6.0%) second ACL tears; 66.7% of these tears occurred on the contralateral side. A second ACL injury was influenced by graft type ( P < .0001), election of ACL reconstruction ( P = .0060), and sex by side of injury ( P = .0072). Nonparametric analysis of graft disruption by graft type demonstrated a higher prevalence of second ACL tears with allografts compared with hamstring ( P = .0499) or patellar tendon autografts ( P = .0012). Conclusion: The incidence of second ACL tears in this population-based cohort was 6.0%, with 66.7% of these tears occurring on the contralateral side from the original injury. There was a high population incidence of second ACL injuries in female patients younger than age 20 years. The utilization of patellar tendon autografts significantly reduced the risk of second ACL injuries compared with allografts or hamstring autografts in this cohort.

Background: Quantitative magnetic resonance (qMR) can be used to measure macromolecules in tissues and is a potential method of observing early cartilage changes in the development of posttraumatic osteoarthritis. Hypothesis/Purpose: We hypothesized that specific patient and surgical factors affecting cartilage matrix composition after anterior cruciate ligament (ACL) reconstruction (ACLR) can be detected using T1ρ and T2 relaxation times. Our purpose was to demonstrate this ability in a multicenter feasibility study. Study Design: Case series; Level of evidence, 4. Methods: A total of 54 patients who underwent ACLR underwent bilateral MRI at baseline before surgery and 6 months postoperatively. Operative findings were recorded. T1ρ and T2 relaxation times were calculated for 6 cartilage regions: the medial femur, lateral femur, medial tibia, lateral tibia, patella, and trochlea. A paired t test compared relaxation times at baseline and 6 months, univariate regression identified regions that influenced patient-reported outcome measures, and analysis of covariance was used to determine the surgical factors that resulted in elevated relaxation times at 6 months. Results: The injured knee had significantly prolonged T1ρ and T2 relaxation times in the tibiofemoral compartment at baseline and 6 months but had shorter values in the patellofemoral compartment compared with the uninjured knee. Prolonged T1ρ and T2 times at 6 months were noted for both the injured and uninjured knees. At 6 months, prolongation of T1ρ and T2 times in the tibial region was associated with lower patient-reported outcome measures. ACLR performed within 30 days of injury had significantly shorter T1ρ times in the tibial regions, and lateral meniscal tears treated with repair had significantly shorter T1ρ times than those treated with excision. Conclusion: Prolonged relaxation times in multiple regions demonstrate how the injury affects the entire joint after an ACL tear. Changes observed in the uninjured knee may be caused by increased loading during rehabilitation, especially in the patellofemoral articular cartilage and distal femur. Relaxation times in the tibial regions may be predictive of patient symptoms at 6 months. These same regions are affected by surgical timing as early as 30 days after injury, but this may partially be reflective of the severity of the preoperative injury and the choice of treatment of meniscal tears.

Background: The cases of patients with medial compartment osteoarthritis who were ≤55 years old and had a proximal tibial osteotomy (PTO) or medial unicompartmental knee arthroplasty (UKA) were compared. Outcomes included postoperative activity level, function, and survivorship free of revision to total knee arthroplasty. Methods: Between 1998 and 2013, data were available for 240 patients between 18 and 55 years old with medial compartment arthritis and varus malalignment who were treated either with PTO (57 patients) or with UKA (183 patients). The mean age was 42.7 years for the 57 patients (41 men and 16 women) in the PTO group versus 49.2 years for the 183 patients (82 men and 101 women) in the UKA group. The Tegner activity level and Lysholm knee scores were evaluated at 3 months and at 1, 2, and 5 years postoperatively as well as at the time of the final follow-up. The end point for survival was defined as revision to total knee arthroplasty. A Wilcoxon rank-sum test was used to evaluate the difference between the groups with respect to the Tegner and Lysholm scores at the respective follow-up intervals. Multivariate regression was used to assess potential confounders. Results: Preoperatively, the PTO and UKA groups had similar Tegner (3.0 ± 1.3 and 2.6 ± 0.09, respectively) and Lysholm scores (69.5 ± 7.3 and 71.6 ± 5.4). Postoperatively, the UKA group had significantly superior mean Tegner scores compared with the PTO group at 3 months (3.82 and 2.02, respectively), at 2 years (4.33 and 3.75), and at the time of the final follow-up (4.48 and 3.08), while the Lysholm scores were higher at 3 months (88.0 and 76.3) and at the final follow-up (90.0 and 80.2) (p < 0.01 for all). Multivariate analysis showed UKA to be an independent predictor of activity level at 3 months, 1 year, and 2 years, as well as at the final follow-up. The survivorship was 77% in the PTO group at an average of 7.2 years and 94% in the UKA group at an average of 5.8 years (p < 0.01). The average time to failure was 98 months (range, 38 to 169 months) in the PTO group and 42 months (range, 2 to 123 months) in the UKA group (p < 0.01). Conclusions: In this comparative cohort study of young patients with isolated unicompartmental arthritis, those treated with UKA reached a higher level of activity early after surgery and it persisted at mid-term follow-up. The UKA group had earlier, but less frequent, revision to total knee arthroplasty. Level of Evidence: Therapeutic Level III . See Instructions for Authors for a complete description of levels of evidence.

Background/aim To determine if Boys Bantam and Peewee and Girls U14 sustain fewer concussions, head hits, ‘other injuries’ and penalties in hockey tournaments governed by intensified fair play (IFP) than non-intensified fair play (NIFP). Methods A prospective comparison of IFP, a behaviour modification programme that promotes sportsmanship, versus control (non-intensified, NIFP) effects on numbers of diagnosed concussions, head hits without diagnosed concussion (HHWDC), ‘other injuries’, number of penalties and fair play points (FPPs). 1514 players, ages 11–14 years, in 6 IFP (N=950) and 5 NIFP (N=564) tournaments were studied. Results Two diagnosed concussions, four HHWDC, and six ‘other injuries’ occurred in IFP tournaments compared to one concussion, eight HHWDC and five ‘other injuries’ in NIFP. There were significantly fewer HHWDC in IFP than NIFP (p=0.018). However, diagnosed concussions, ‘other injuries’, penalties and FPPs did not differ significantly between conditions. In IFP, a minority of teams forfeited the majority of FPPs. Most diagnosed concussions, HHWDC, and other injuries occurred to Bantam B players and usually in penalised teams that forfeited their FPPs. Conclusions In response to significant differences in HHWDC between IFP and NIFP tournaments, the following considerations are encouraged: mandatory implementation of fair play in regular season and tournaments, empowering tournament directors to not accept heavily penalised teams, and introducing ‘no body checking’ in Bantam.