Abstract
Keywords
All that is human must retrograde if it does not advance.
Many years ago, I asked the late Howard Sweeny to see one of our Chicago varsity athletes who seemed to have an intra-articular hip problem. Together, the player and I made the pilgrimage from Chicago to Evanston where Dr. Sweeny, a pioneer of arthroscopic surgical techniques and education, positioned the patient on a conventional hip fracture table and conducted his masterful arthroscopic examination of the hip. At the time, indications for hip arthroscopy seemed few and far between. Dr. Sweeny was one of a handful of surgeons in the United States who were known for their expertise in the technique.
That was before Reinhold Ganz and his Bernese colleagues described femoroacetabular impingement (FAI) as a cause of what had hitherto been deemed idiopathic osteoarthritis of the hip. 7 The rest, as the saying goes, is history. Although Dr. Ganz treated the condition via open surgical dislocation of the hip, arthroscopic surgeons worldwide quickly recognized FAI as a prime candidate for arthroscopic treatment. Since then, the number of practitioners skilled in hip arthroscopy has grown rapidly, while the complexity and sophistication of the hip procedures that can be accomplished arthroscopically has evolved at a similar frenetic pace.
As an editor, I was quickly presented with a dilemma. Eager to be the first to publish their experience with a new technique, hip arthroscopists were submitting clinical studies with only 6 months of follow-up. I knew that these early series were likely to become landmark publications, but I also felt strongly that 6 months was much too short a time to appreciate the outcome of a novel surgical intervention. In fact, while 2 years is usually considered adequate to demonstrate the initial efficacy of a procedure, much longer follow-up is needed to assess the ultimate outcome, especially in the case of an intervention designed to delay or prevent the onset of osteoarthritis in a young patient. Two decades after Dr. Ganz's publication, hip arthroscopy studies with 10 or more years of follow-up are appearing with increasing frequency.2,3,5,9,15,16
In this month's American Journal of Sports Medicine (AJSM), Lamba and colleagues 12 share with us their institution's early experience in the realm of hip arthroscopy. They report the results of the arthroscopic debridement of acetabular labral tears from 1996 to 2010 with a minimum 13-year follow-up. Given the rapid pace at which arthroscopic surgery of the hip has progressed since the turn of the 21st century, it is inevitable that the techniques that they describe do not reflect the current protocols of most orthopaedic hip preservation specialists, including the authors themselves. Criteria for inclusion in the study were a labral tear confirmed by preoperative magnetic resonance imaging, age >18 years, Tönnis grade <3, and lateral center-edge angle (LCEA) >25°. All patients underwent isolated debridement of their labral tear without resection of any cam or pincer lesions or repair of their interportal capsulotomy. Any patient who underwent subsequent surgery on the same hip or judged their hip function to be abnormal or severely abnormal was classified as having experienced a surgical failure.
In a prior report on many of the same patients after 5 years of follow-up, 10 the authors classified 45% as having failed results: 20% who had further surgery on their operative hip, of whom 12% were total hip arthroplasties (THAs), and 25% who reported abnormal or severely abnormal function. In the current study, the authors were able to evaluate the status of 48 hips in 47 patients a mean of 17 years after surgery. At that point, 19 (40%) of the hips had received additional surgery, of which 16 (33%) were THAs, and an additional 5 (10%) were rated abnormal or severely abnormal for a total failure rate of 50%. Although the cohorts in the 2 studies may not have been identical, a comparison of the results suggests that most of the ultimate failures were evident at the 5-year mark, with the overall failure rate only demonstrating a modest 5% increase from 45% at 5 years to 50% at 17 years. The percentage with abnormal or severely abnormal function decreased from 25% to 10% during this interval, while the percentage receiving THAs increased from 12% to 33%, implying that many of the patients enduring abnormal function in the early postoperative years elected to seek improvement via a THA. Notably, the 50% of patients in the cases not deemed failures were not doing too badly: Their functional ratings on the modified Harris Hip Score (mHHS), Hip Outcome Score–Activities of Daily Living (HOS-ADL), and Hip Outcome Score–Sports Subscale (HOS-SS) were 82.2, 81.9, and 79.9, respectively. These scores were not significantly different from their 5-year ratings, suggesting that the outcomes in this lucky subset were fairly durable.
Differences between the patient selection guidelines and surgical protocols used in this study and current standards are evident. The authors included patients with radiographic Tönnis grade 2 arthritis in their cohort, while contemporary practice would usually consider this degree of degeneration to be a contraindication to arthroscopic treatment. 17 Indeed, although they could not show that the Tönnis grade 2 patients did worse as a group, their univariate failure analysis demonstrated that patients who went on to have subsequent surgery had higher mean acetabular International Cartilage Regeneration & Joint Preservation Society chondral and acetabular labrum articular disruption grades at the time of surgery compared with patients who did not have subsequent surgery. They had not considered performing femoral or acetabular osteoplasty in any of their patients, pointing out that the landmark paper of Ganz and colleagues 7 describing FAI was not published until 2003, halfway through their collection period. Retrospective analysis, however, showed that 43% of the cohort had cam morphology, 8% had pincer morphology, and 38% had mixed cam and pincer morphology, suggesting that many of their patients would be candidates for osteoplasty if operated on today. Indeed, a 15-year follow-up study published last year in AJSM reported that patients with symptomatic FAI who underwent femoral osteoplasty achieved significantly higher mHHS scores and lower rates of reoperation and THA conversion than those who did not. 16 Similarly, the surgical protocol in the current study did not include labral repair or capsular closure, techniques that many contemporary hip preservation specialists would perform routinely.4,8,11,13,14,18 Nevertheless, their long-term experience is valuable. As the authors state, their results provide a benchmark against which these subsequent innovations can be judged and demonstrate a pattern in the postoperative evolution of outcomes that might still be relevant.
In contrast with this study, another paper in this month's AJSM reports 10-year follow-up on 94 patients with FAI syndrome (FAIS) who were treated with what the authors describe as more contemporary surgical techniques. 1 The procedures in this series included labral repair in 96%, femoral osteochondroplasty in 97%, acetabular rim trimming in 73%, and capsular repair in 100%. Notably, patients with Tönnis grade >1 osteoarthritis or LCEA <20° were excluded. Outcome measures included the HOS-ADL, HOS-SS, mHHS, and visual analog scale (VAS) for pain and patient satisfaction. The International Hip Outcome Tool (iHOT-12) was added to the outcome menu 2 years after surgery. By the first postoperative year, all preoperative outcome scores had significantly improved. The HOS-ADL, HOS-SS, mHHS, and iHOT-12 showed little further change at 2, 5, and 10 years postoperatively, while the mean pain VAS fluctuated: further improvement at 2 years, worsening at 5 years, and then improving again at 10 years. Because 9 patients underwent revision arthroscopy a mean of 5 years postoperatively and 6 were converted to THA a mean of 4 years postoperatively, the apparent improvement in pain VAS at 10 years may be due to these patients being censored from patient-reported outcome measure (PROM) collection after their subsequent surgery. The minimal clinically important difference (MCID) and Patient Acceptable Symptom State (PASS) were assessed for each outcome measure using data from the overall cohort. At 10 years, 91% achieved MCID and 80% achieved PASS on at least 1 outcome measure (73%-89% and 71%-77%, respectively, for individual PROMs). Although there was no contemporaneous comparison group and follow-up was limited to just >70%, this report, like that of Lamba et al, 12 gives us a benchmark that will be useful for comparison with other studies.
Tönnis grade 1 radiographic changes are not generally considered an absolute contraindication to arthroscopic treatment of FAIS or labral tears, but are they associated with a worse ultimate prognosis? Another long-term study in this month's AJSM from the same institution compares the outcome between patients classified as Tönnis grades 1 and 0 at a minimum of 10 years after surgery. 6 The authors of this study were able to achieve follow-up on 31 patients with Tönnis grade 1 radiographs who underwent arthroscopic treatment for FAIS between 2012 and 2013. For analysis, these patients were matched by age, sex, and body mass index to 62 patients with Tönnis grade 0 radiographs who underwent similar surgery during the same period. Surgery comprised treatment of cam and pincer deformities, labral repair, and capsular plication and closure. Outcome measures included HOS-ADL, HOS-SS, mHHS, iHOT-12, and VAS for pain and patient satisfaction. The MCID and PASS were assessed for each outcome measure and progression to revision surgery or THA was recorded.
At the 10-year follow-up, there were no significant differences between the Tönnis groups on any of the outcome measures. Although the differences were not statistically significant, the Tönnis grade 0 group did score consistently better on all measures; some of these differences were arguably clinically important in magnitude and may have proven statistically significant in a larger cohort. The rates of achieving MCID and PASS exhibited a similar pattern: no significant differences were identified, but the Tönnis grade 0 group consistently achieved these milestones at a higher rate, with some of the differences potentially clinically important. In the most extreme example, the mean iHOT-12 PASS rate was 71.2% for the Tönnis grade 0 group and 44.4% for the Tönnis grade 1 group. Additionally, although overall rates of secondary surgery were not significantly different, the rate of reoperation for the Tönnis grade 1 group was twice that of the Tönnis grade 0 group (29% vs 14.5%). Most of the reoperations in the Tönnis grade 1 group were conversions to THA, and the difference in THA conversion between the groups was both statistically and clinically significant, 25.8% compared with 4.8%.
Given these findings, it is not surprising that the authors state anecdotally that they have decreased their overall rate of primary hip arthroscopy on patients with Tönnis grade 1 osteoarthritis. As they note, the Tönnis classification is a radiographic one that can miss variations in articular degeneration that are readily visible during arthroscopy. Indeed, the Tönnis grade 1 patients had significantly worse femoral head and acetabular cartilage damage than the Tönnis grade 0 patients. Furthermore, within the Tönnis grade 1 group, the patients who went on to THA conversion had significantly worse femoral head damage than those who did not, even given the small numbers available for this subgroup analysis.
In view of the burgeoning importance of the treatment of hip disorders in orthopaedic sports medicine, Andrea Spiker joins the AJSM team of editors this month as associate editor for hip and thigh. In 2024, surgeons skilled in hip arthroscopy continue to push the limits of their craft. When labral repair is no longer feasible, labral reconstruction fills the gap. Looking beyond cam and pincer lesions, surgeon-scientists are investigating the role that factors such as femoral and acetabular version, other pelvic prominences, and the spinopelvic interface may play in various forms of impingement. Arthroscopists now venture outside the confines of the hip joint itself to repair damage to the gluteal or hamstring tendons. If hip preservation surgery follows the pattern of other areas of orthopaedic innovation, many of the new techniques will prove to be therapeutic advances, while some will be discarded as ineffective or worse. While theory and laboratory investigation can indicate potentially valuable paths to innovative treatments, clinical studies, especially long-term ones, will be the ultimate arbiter of the value of new techniques.