Levy, Bruce A.

Abstract Treatment of femoroacetabular impingement through an arthroscopic approach has gained widespread popularity in recent years. Although outcomes are generally favorable, one of the most common reasons for failure is incomplete resection of cam lesions of the femoral neck. As a result, the T‐capsulotomy has been introduced as a method for improving access to the femoral head‐neck junction, which is not always visible through a standard interportal capsulotomy. The T‐capsulotomy has the benefits of improving arthroscopic visualization of the femoral neck, reducing overall fluoroscopy exposure for the patient and surgeon, and facilitating capsular plication. We present a reliable and efficient method for creating and repairing the T‐capsulotomy. We routinely perform this technique in patients with cam lesions that are too large or too distal to safely visualize and decompress through an interportal capsulotomy. Technique Video video See video under supplementary data.

Abstract Capsulotomy is typically performed during arthroscopic treatment for femoroacetabular impingement. As the frequency of hip arthroscopy continues to expand rapidly, increased attention is being paid to the implications of interportal capsulotomy and the need for repair. To minimize the risk of postoperative instability, capsular closure has been recommended to restore the anatomy and biomechanical function of the capsule. We present a reliable, efficient, and effective method for arthroscopic closure of the interportal capsulotomy after hip arthroscopy. Technique Video video See video under supplementary data.

Abstract Posterior cruciate ligament (PCL) injuries account for nearly 20% of knee ligament injuries. PCL injuries can occur in isolation or, more commonly, in the setting of multiligamentous knee injuries. Isolated PCL disruptions are commonly treated nonoperatively; however, symptomatic grade III injuries, as well as PCL injuries found in multiligamentous injuries, are frequently treated surgically. Several reconstructive techniques exist for the treatment of PCL deficiency without a clear optimal approach. We describe our preferred operative technique to reconstruct the PCL using an all‐inside arthroscopic approach with a quadrupled tibialis anterior or peroneus longus allograft with both tibial and femoral suspensory fixation. Technique Video video See video under supplementary data.

Background: Patients with patellar instability and a tibial tubercle–trochlear groove (TT-TG) distance ≥20 mm may be candidates for distal tubercle realignment surgery. Although this variable has proven valuable in predicting recurrent dislocations, it is not individualized to patient size, bony structure, or patellofemoral mechanics. Purpose: To develop a patellar instability ratio (PIR) that predicts the risk of recurrent instability based on the TT-TG distance to patient-specific anatomy. Study Design: Cohort study; Level of evidence, 3. Methods: On magnetic resonance imaging scans of 59 knees with patellar instability, the TT-TG distance, tibial tubercle–posterior cruciate ligament (TT-PCL) distance, sagittal patellar length (PL), sagittal trochlear length (TL), axial patellar width (PW), and axial trochlear width (TW) were calculated by 2 observers in a blinded and randomized fashion. Patients were divided into 2 groups: those with a single dislocation and those with multiple (≥2) dislocations. The ability of the TT-TG and TT-PCL distances as well as the 8 different ratios to predict recurrent instability was assessed by calculating odds ratios (ORs), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for each measure. Results: Twelve knees (20%) experienced a single dislocation, while 47 (80%) sustained multiple dislocations. A TT-TG distance ≥20 mm was predictive of recurrent instability, with a sensitivity, specificity, and PPV of 21%, 100%, and 100%, respectively. The OR for TT-TG ≥20 mm was 5.38 ( P = .29), while the highest ORs for recurrent instability were noted for a TT-TG/PW ≥0.4 (OR, 7.37; P = .02) and a TT-TG/TW ≥0.5 (OR, 8.88; P = .04). The sensitivity, specificity, and PPV of a TT-TG/PW ≥0.4 were 62%, 83%, and 94%, respectively, while those of a TT-TG/TW ≥0.5 were 45%, 92%, and 95%, respectively. Conclusion: Two novel PIRs (TT-TG/TW and TT-TG/PW) were identified and found to more effectively predict recurrent instability than the TT-TG distance alone. Each ratio takes into account patient-specific anatomy and can be measured in an accurate and reliable fashion by clinicians. These PIRs are a step toward overcoming some of the limitations of using the TT-TG distance in isolation. Further investigation into the clinical applications and utility of the TT-TG/TW is warranted.

Objectives: Patients with patellar instability and tibial tubercle to trochlear groove (TT-TG) distances ≥ 20 mm are often considered to have pathologic lateralization of the tibial tubercle (TT) relative to the femoral trochlea and may be candidates for TT realignment surgery. Although this variable has proven valuable in predicting risk of recurrent dislocations, the measure is limited in that it is not individualized to the patient and does not consider patient size, bony morphology, or patellofemoral mechanics. The purpose of this study was to develop a Patellar Instability Ratio (PIR) that predicts the risk of recurrent instability based on the TT-TG distance relative to patient specific anatomy. Methods: Fifty-nine patients with a clinical diagnosis of patellar instability were included in the study. A number of measures were calculated on MRI by two observers (one staff orthopedic surgeon and one senior resident) in a blinded and randomized fashion. Variables analyzed included: TT-TG, tibial tubercle to posterior cruciate ligament distance (TT-PCL), sagittal patellar length (PL), sagittal trochlear length (TL), axial patellar width (PW), axial trochlear width (TW), sagittal patellofemoral engagement (SPE; SPE=TL/PL) and axial patellofemoral engagement (APE; APE=PW/TW). Patients were divided into two groups: those with ≤ 1 dislocation and those with ≥ 2 dislocations. Using these groups, the ability of TT-TG, TT-PCL, and 12 different ratios with cut-off values to predict recurrent instability was assessed by calculating odds ratios (OR), C-statistics, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for each of the ratios. P values < 0.05 were considered significant. Results: Mean follow up was 6.6 years. Twelve patients (20%) did not experience recurrent instability while 47 (80%) sustained two or more dislocations during the study period. Pathologic TT-TG (≥ 20 mm) was predictive of recurrent instability with an OR of 5.38 (p=0.29). The highest OR's for recurrent instability were noted for TT-TG/PW ≥ 0.40 (OR > 7.37, p=0.02) and TT-TG/TW ≥ 0.5 (OR 8.88, p=0.04)(Table 1). PW was defined as the largest medial to lateral width of the articular surface of the patella. TW is defined as the largest articular width of the trochlea that is articulating with the patella and can be conceptualized as the “jump distance” that must be overcome for the patella to dislocate. The sensitivity and specificity of TT-TG/PW ratio ≥ 0.40 was 61% and 83% and that of TT-TG/TW ≥ 0.50 was 59% and 92%, respectively. PPV for TT-TG/PW ≥ 0.40 was 94% while that of TT-TG/TW ≥ 0.49 was 97%. Conclusion: Two distinct PIR (TT-TG/PW and TT-TG/TW) have been identified that predict recurrent patellar instability more accurately than TT-TG ≥ 20 mm with PPV's ≥ 94% and OR's > 7 for both measures. These ratios take into consideration the TT-TG relative to patient specific size and anatomy. Further research is needed to determine if patients with increased PIR benefit from TT osteotomy.

Objectives: Female athletes are an at-risk population for anterior cruciate ligament (ACL) injury, with rates of injury significantly higher than the general population. Few studies have reported on a second ACL injury for female athletes. The purpose of this study was to (1) report the rate of subsequent ACL injury (ACL graft rupture or contralateral ACL tear) in competitive female soccer players, (2) compare these rates to those of other female athletes of similar competitive level, and (3) determine risk factors for second ACL injury in this athletic population. Methods: The medical record at our institution was reviewed for patients treated with primary ACL reconstruction between 1998 and 2013. Female patients injured during a competitive athletic event were included for further review, and followed for an average of 7.1 years postoperatively (range 1.0 - 17.7 years). Chi-square analysis was used to compare rate of graft rupture and contralateral ACL injury based on pre-operative Tegner score, graft type, and injury side for soccer players vs. other female athletes. Wilcoxon rank-sum test was used to compare rate of subsequent ACL injury to patient age. Results: 337 patients met our inclusion/exclusion criteria (90 soccer players; 247 non-soccer playing athletes) with a mean age of 24.0 years. Of the 337 athletes, 6 were injured during professional competition, 43 were injured during collegiate athletics, 184 during high school play, and 136 during recreational athletics. No patient demographical differences were found at baseline between the soccer group and non-soccer group, including Tegner scores. Overall, 21 soccer players (23.3%) sustained a second ACL injury compared to 21 (8.5%) non-soccer athletes (P < 0.001). Soccer players had significantly more graft ruptures (10.0% vs. 2.0%, P = 0.003) and more contralateral ACL tears (13.3% vs. 6.5%, P = 0.04). Risk factors for ACL graft tear included young age (mean 16.0 vs 24.3 years; P < 0.0001) and higher Tegner Activity Level scores (mean 8.1 vs. 7.4, P = 0.049), but were not risk factors for contralateral ACL injury. Graft selection and injury side showed no statistical significance on graft rupture or contralateral injury. Conclusion: Female soccer players treated with ACL reconstruction had an increased rate of second ACL injury, including graft tear and contralateral ACL injury, compared to a similar group of non-soccer female athletes (23.3% vs. 8.5%, P < 0.001). In addition, young age and higher activity level were risk factors for graft rupture in this population.

Objectives: Arthroscopic treatment for femoroacetabular impingement (FAI) generally has good results, but the outcome can worsen with increasing arthritis, largely defined as joint space narrowing. There remains a subset of hips with maintained joint space but with subchondral edema and cystic change of the acetabulum on MRI, with unknown outcome. Therefore, the purpose of this study was (1) to determine if postoperative outcome was worse for patients with subchondral edema on preoperative MRI compared to a matched control group and (2) to identify risk factors on MRI leading to inferior outcome. Methods: A review of patients who underwent arthroscopic hip surgery for FAI between 2007 and 2013 identified 530 patients. Of these, 39 patients had evidence of subchondral edema or subchondral cyst on the preoperative MRI with maintained joint space on preoperative radiographs. Lesions were characterized by grade, location and type using an established radiographic MRI classification system. A matched cohort of patients without evidence of subchondral edema or cyst was identified, and 2:1 matching was based on patient age, surgeon, surgery year within one year, surgical procedure, Tegner activity score, and Tonnis grade osteoarthritis changes on preoperative x-ray. Outcome was compared using the Hip Outcome Score and Modified Harris Hip Score between the two groups. Risk factors were then analyzed for type of lesion, grade and location of bone marrow lesion. Results: Thirty-nine patients (20 males, 19 females) showed preoperative MRI evidence of subchondral edema, and were followed for a mean of 23 months postoperatively (range, 12 - 60 months). Patients had an average age of 41 years (range, 19 - 67) and a preoperative Tegner score of 3.8 (range, 1 - 9). 87% of patients with subchondral cystic change had evidence of a grade IV full thickness cartilage lesion at the time of hip arthroscopy. Average Modified Harris Hip Scores were inferior for the subchondral group (79.9 ± 18.7) compared to the control group (86.6 ± 12.5; p = 0.027). In addition, Sport sub-scales of the Hip Outcome Score showed significantly lower scores for the subchondral group for both score (69.1 ± 27.0 vs. 79.5 ± 21.4, p = 0.018) and rating (66.7 ± 27.5 vs. 78.0 ± 20.8, p = 0.044). No radiographic risk factors, including type, location, or size of lesion were significant predictors of outcome. Two patients in the subchondral group were later converted to a total hip replacement. Conclusion: The presence of a subchondral acetabular cyst on MRI is indicative of a full thickness cartilage lesion at the time of arthroscopy. These patients have inferior outcomes for arthroscopic treatment of FAI compared to patients with similar age and activity level without MRI subchondral changes. Therefore, we recommend caution with consideration of hip arthroscopy in this patient subset.

Purpose To (1) determine the incidence of symptomatic venous thromboembolic events (VTEs) after knee arthroscopy and arthroscopy‐assisted procedures at a single institution and (2) determine associated risk factors for VTEs in these patients. Methods The records of patients who underwent knee arthroscopy at a single institution between 1988 and 2008 were reviewed. Chemoprophylaxis was not routinely used. Confirmed VTEs occurring within 4 weeks after the index arthroscopy procedure were included. A 2:1 matched control group was generated to include patients in whom knee arthroscopy was performed by the same surgeon either on the same day or immediately before each case resulting in a VTE. Preoperative and perioperative data were collected with respect to demographic data, medical history, medications, and surgical and anesthesia data. Univariate and multivariate analyses were performed. Results During the study period, 12,595 patients underwent knee arthroscopy. Among these patients, 43 cases of VTEs (35 deep venous thromboses [DVTs], 5 pulmonary embolisms [PEs], and 3 DVTs that progressed to PEs) occurred, resulting in an incidence of 0.30% (95% confidence interval [CI], 0.22% to 0.41%) for DVT, 0.06% (95% CI, 0.03% to 0.12%) for PE, and 0.34% (95% CI, 0.25% to 0.46%) for VTEs overall. Factors associated with an elevated risk of symptomatic postoperative VTEs included a history of malignancy ( P = .01; odds ratio [OR], 6.3), a history of VTEs ( P = .02; OR, 5.2), or the presence of more than 2 classic risk factors for VTEs ( P = .01; OR, 13.6). Conclusions In this study, symptomatic VTEs were rare and occurred infrequently, with an incidence of 0.34% (95% CI, 0.25% to 0.46%), after knee arthroscopy and arthroscopy‐assisted cases in the absence of routine chemoprophylaxis. Patients with a history of VTEs, a history of malignancy, or 2 or more classic risk factors are at increased risk of VTEs after knee arthroscopy, and chemoprophylaxis should be considered in these select patients. Level of Evidence Level III, case‐control study.

Objectives: Intra-articular (IA) anesthetic injection is commonly performed as a diagnostic test in the setting of femoroacetabular impingement (FAI). Currently, there is a paucity of data correlating post-injection pain relief and outcomes after hip arthroscopy for FAI. The purpose of this study is to determine whether the amount of pain relief after IA injection predicts clinical and functional outcomes following hip arthroscopy. We hypothesize that increased pain relief (>50%) will correlate with better outcomes after surgery. Methods: The records of patients undergoing hip arthroscopy for FAI at our institution between April 2007 and April 2012 were reviewed. We identified patients who underwent IA injection and subsequent hip arthroscopy. Inclusion criteria were: ultrasound or fluoroscopic guided intra-articular anesthetic injection performed at our institution, prospectively documented pre- and post-injection Numerical Rating Scale (NRS) pain scores, no prior ipsilateral hip surgery, and minimum 1 year follow-up. Pre-operative radiographs were reviewed and degree of osteoarthritis was determined using the Tonnis classification system. Outcomes were assessed with Modified Harris Hip Score (MHHS) and Hip Outcome Score (HOS). Univariate and multivariate models were performed to assess whether percent pain relief correlated with outcome scores. Results: 99 hips in 96 patients met our inclusion criteria and included 71 females (74%) and 25 males (26%) with a mean age of 37.5 ± 14.0 years. 26 patients had Tonnis grade 0, 55 had grade 1, and 18 had grade 2 (0 grade 3). Mean pain relief after IA injection was 73.6 ± 36.1 (range 0-100) percent. 26 patients (26%) had ≤50% pain relief while 73 (74%) had >50% pain relief. Outcome scores were obtained at a mean 14.9 months. Mean MHHS, HOS-ADL, and HOS-Sport scores were 79.2 ± 17.3, 82.6 ± 17.3, and 67.4 ± 28.2, respectively. There was no statistical correlation between percent pain relief and MHHS, HOS-ADL, or HOS-Sport scores. There was no significant difference in outcome scores between those with ≤50% and >50% pain relief. While patients that received >50% pain relief by IA injection were more likely to achieve positive results (MHHS >70) with a likelihood ratio of 1.23 (95% CI 0.92-1.53), this was not significant. Multivariate regression analysis demonstrated no significant predictors of outcome, including age, gender, Tonnis grade, percent relief with IA injection, number of months postoperatively, or type of surgery. Conclusion: In patients undergoing hip arthroscopy for FAI, our data indicates that the amount of pain relief from IA injection is a poor predictor of short term outcome, even when adjusting for chondral degeneration. While anesthetic injections can be an important diagnostic tool in select patients, correlation of the clinical history, physical examination, and imaging findings are fundamental. In addition, outcome following hip arthroscopy remains multifactorial.