Predicting Outcomes in Hip Arthroscopy for Femoroacetabular Impingement Syndrome

Abstract

Purpose of Review

Arthroscopic treatment of femoroacetabular impingement syndrome (FAIS) continues to rise in incidence, and thus there is an increased focus on factors that predict patient outcomes. The factors that impact the outcomes of arthroscopic FAIS treatment are complex. The purpose of this review is to outline the current literature concerning predictors of patient outcomes for arthroscopic treatment of FAIS.

Recent Findings

Multiple studies have shown that various patient demographics, joint parameters, and surgical techniques are all correlated with postoperative outcomes after arthroscopic FAIS surgery, as measured by both validated patient-reported outcome (PRO) scores and rates of revision surgery including hip arthroplasty.

Summary

To accurately predict patient outcomes for arthroscopic FAIS surgery, consideration should be directed toward preoperative patient-specific factors and intraoperative technical factors. The future of accurately selecting patient predictors for outcomes will only improve with increased data, improved techniques, and technological advancement.

Introduction

Femoroacetabular impingement (FAI) pathology is described as a mechanical deviation that can compromise hip joint function and mobility. FAI is characterized by atypical articulation between the femoral head and acetabulum, predisposed by cam and pincer morphology, respectively, and is considered the most common pre-arthritic condition of the hip [1–8]. These morphologies can contribute to subsequent disruption and tearing of the acetabular labrum [1, 6, 9, 10••, 11]. The resulting symptoms can lead to debilitating pain for patients and are most predictably alleviated by arthroscopy [2, 8, 12, 13, 14•, 15, 16•, 17] with osteoplasty of the cam and/or pincer lesions and repair, reconstruction, or debridement of the labrum when indicated [3, 8, 13, 18]. Femoroacetabular impingement diagnosis and arthroscopic treatment have exponentially increased in incidence since the term was first coined by Ganz et al. [9] in 2003 [3–9, 11, 13, 18, 19••, 20•, 21–23]. The Warwick Agreement in 2016 better defined pathologic FAI as a syndrome (FAIS) which includes a triad of symptoms, clinical signs, and radiographic imaging that is now understood to be the standard, modern definition of the pathology [3, 6].

As arthroscopic treatment of FAIS has proven to consistently improve patients’ symptoms clinically over the last two decades, it has also led to the development of more advanced surgical techniques and improvement in predicting patient outcomes [3–8, 10••, 11, 13, 18, 20•, 21–24, 25••, 26]. With an increase in diagnosis and treatment of FAIS, there is interest in establishing predictors of patient outcomes following surgery. Reliable predictive outcomes have been shown to aid in the selection of patient populations and lead to appropriate rehabilitation expectations [19••, 25••, 27•, 28]. Inherent in considering outcomes for arthroscopic FAIS surgery are preoperative patient-specific factors and intraoperative surgical techniques. Our goal in the article is to review current commonly accepted predictors, group them into distinct categories, and discuss how technological advances are improving the efficiency and reliability of outcomes to aid in physician-patient shared decision-making.

Predicting patient outcomes following hip arthroscopy for FAIS is a challenging task that considers several factors and relies heavily on patient-reported outcome (PRO) scores and whether minimal clinically important difference (MCID) is achieved [24, 29, 30]. There is no consensus on which PRO tool should be used over others [30]. Early (within 2 years) revision hip arthroscopy or conversion to total hip arthroplasty (THA) after hip arthroscopy has also been used as a proxy for poor outcome [24]. It is imperative to consider factors relating to the patient’s personal demographics, surgical history, physical exam, and current parameters of their femoroacetabular joint. All play a crucial role in determining if patients improve or worsen after surgery.

Patient-Specific Factors that Influence Postoperative Outcome

Several studies have identified patient factors that may lead to worse outcomes following arthroscopic treatment of FAIS. These include but are not limited to the presence of osteoarthritis, older age, increased duration of symptoms, lower preoperative outcome score, and dysplasia [10••, 13, 19••, 25••, 26, 27•, 31•]. Other patient predictors that should be acknowledged, though not as clearly correlated to worse outcomes, include lower activity level, history of previous hip arthroscopy, higher body mass index (BMI), female sex, history of mental illness, alcohol consumption, tobacco use, opioid use, worker’s compensation status, and lower back pain [10••, 25••, 29, 32•, 33, 34].

The literature agrees that the degree of osteoarthritis is a main predictor of poor outcomes. Patients with Tonnis grade greater than 2 or joint space narrowing less than 2 mm are frequently associated with failure of FAIS arthroscopy [10••, 13, 25••, 31•, 35–39]. There is strong consensus that patients with Tonnis grade 2 or 3 consistently have lower PRO scores and are up to 4–5 times more likely to require conversion to THA within 2 years [25••, 35, 39, 40]. There is also evidence that joint space narrowing less than 2 mm is an absolute contraindication for arthroscopy [10••, 37]. Patients who undergo arthroscopy for FAIS and have greater than 2 mm of joint space narrowing have more than 5 times the survivorship rate when compared to those with less than 2 mm [38]. Domb et al. [39] compared FAIS arthroscopy patients who had Tonnis grade 1 to Tonnis grade 0 after a minimum of 5 years of follow-up and found the Tonnis grade 1 group had a survivorship rate of 69% compared to 88% for the Tonnis grade 0 cohort, which suggests that the degree of arthritis negatively affects outcomes no matter how mild it may be.

Older age is frequently cited as one of the dominant risk factors for inferior outcomes and failure of arthroscopy, e.g., revision arthroscopy, progression of osteoarthritis, and conversion to THA [10••, 13, 19••, 25••, 27•, 35–38, 40–44]. It has been a common observation that performing arthroscopy in each decade of life after the age of 40 results in increased rates of conversion to THA. Yao et al. [40] specifically compared rates of conversion to THA within 2 years of FAIS arthroscopy to decades of interest. They found that patients under 35 had a conversion rate of less than 1%, while those aged 35–44, 45–54, and greater than 55 had conversion rates of 6%, 13%, and 18%, respectively. Domb et al. [45] found that patients over the age of 50 with a Tonnis grade of 0 or 1 had an 83% survivorship rate at 2 years after hip arthroscopy, compared to survivorship of 98% for the younger control group that was 30 or under. There have been some important exceptions noted for patients 40 years of age and older who have superior outcomes in the literature. Saks et al. [41] noted that patients ≥40 who participated in an organized sport within 1 year prior to surgery had better outcomes following arthroscopy when compared to an age-matched cohort. Research has also shown that patients >40 have better postoperative outcomes and reach clinically significant improvement when treated with osteoplasty and labral repair/reconstruction versus debridement [24, 32•, 37]. These studies suggest that patients ≥40 require extensive screening to allow for superior postoperative outcomes, specifically noting the amount of arthritis present pre- and intraoperatively as well as patient activity level.

Duration of patient symptoms has also been noted in several studies as being an indication of worse patient outcomes [10••, 13, 25••, 27•, 43, 46]. Research has demonstrated that patient outcomes can be negatively impacted if symptoms are present for greater than 6 months [13, 27•], while others state negative effects were not seen unless symptoms were present for at least 2 years [32•, 43, 46]. Although there does not appear to be a consensus on a quantifiable duration of symptoms, it is apparent that the longer a patient waits to pursue arthroscopy, the more likely their outcomes will be affected. An explanation for worse outcomes could be that the longer a patient goes without surgery, the harder it is to repair/reconstruct the labrum, or that there is arthritic progression or an effect of aging itself. Another explanation could be that chronic pain negatively affects the mental health and resilience of a patient, and poorer mental health and resilience have both been associated with poor outcomes after surgery for FAIS [25••, 29].

Preoperative PRO scores have been cited as a strong predictor of postoperative PRO scores [10••]. Most PRO tools have significant overlap and contain standardized questions attempting to quantify patients’ quality of life, perception of pain, activity level, and functional ability [47]. Some tools have been shown to be applicable to the general population, while others are tailored to specific psychometric properties or demographics including athlete status, age, and activity level, suggesting that a combination of tools may lead to more accurate outcome prediction [30, 47]. Literature shows that higher pre- and postoperative PRO scores are associated with better patient outcomes, while lower postoperative PRO and function/satisfaction scores are associated with worse outcomes [10••]. Caution should be taken when looking at PRO tools independently as they do not incorporate important patient characteristics such as radiographic findings, physical examination, duration of symptoms, and BMI [30]. Even though PRO scores show great utility in predicting patient outcomes, they should be used concomitantly with other patient and surgical factors for a more complete picture.

Other patient demographic variables have also been shown to correlate with outcomes, depending on the study. Domb et al. [19••] found that patients with 5-year follow-up data who were aged >45 years, BMI >30, or had a revision arthroscopy performed were more likely to have worse outcomes and be converted to THA. Their 2-year follow-up data also showed the female sex as being another predictor of worse outcomes; however, this finding did not persist into the 5-year follow-up data. Minkara et al. [10••] also revealed in their systematic review and meta-analysis on outcomes after arthroscopic FAIS treatment that greater age, female gender, and increased duration of symptoms led to what they classified as a failure of arthroscopy. This study attempted to find an association between a higher BMI and inferior outcomes but was unable to find the strength of evidence needed to support the claim. One suggestion regarding the connection between increased BMI and lower postoperative outcome scores cites the increased complication risk associated with diabetes [25••]. Research has also shown that female patients can achieve equivalently high rates of MCID following FAIS arthroscopy [34]. There is variable evidence in the literature regarding BMI and female gender as negative predictors, and further analysis is needed before they can be considered dependable factors.

Failure of a completed primary arthroscopy for FAIS can leave patients with persistent pain and worse outcomes that require revision arthroscopy or conversion to THA. Revision arthroscopy for FAIS is an increasingly more common phenomenon as the incidence of primary arthroscopy rises [19••, 27•, 33, 48–50]. Haeberle et al. [48] summarized patient factors to look for prior to initial arthroscopy for FAIS that could increase the risk of a revision and lead to worse outcomes. Among these factors included worse preoperative PRO scores, older age, and increased BMI. Shapira et al. [33] found that there was a risk of a secondary arthroscopy if lower patient PROs were seen at 1 year following initial surgery. The study also discussed certain risk factors that predispose patients for revision, most notably open workers’ compensation claims, female sex, pain medication use, antalgic gait, and lateral labral tear. Perez-Carro et al. [50] listed several causes of primary arthroscopy failure which include osteoplasty under/over-resection, recurrent labral pathology, degenerative joint disease progression, and coinciding disorders affecting the hip. There is hope for patients facing revision arthroscopy as some advanced techniques have the potential to improve patient symptoms and prognosis. Some of these procedures include labral reconstruction/augmentation, capsule reconstruction, ligamentum teres reconstruction, and remplissage of the femoral head-neck junction [49, 50]. Regardless of the reason for revision arthroscopy, preoperative planning and expectation management should be strongly considered before proceeding with surgery, as literature supports the decline of patient outcomes and the likelihood of reaching MCID when undergoing revision surgery [19••, 26, 27•, 33, 48–50].

The parameters of the femoroacetabular joint can be reliable predictors of postoperative outcomes. Large cam morphology has been shown to be more likely to have cartilage damage at the time of hip arthroscopy which can thus impact outcomes due to the correlation of poor cartilage with outcomes [10••, 19••, 31•, 36]. Hip dysplasia can be identified radiographically by a number of parameters which include but are not limited to anterior center edge angle (ACEA) under 20°, lateral center edge angle (LCEA) under 20°, Tonnis angle or acetabular index greater than 10°, and measurements on cross-sectional imaging (CT) [25••, 31•, 51]. True dysplasia is frequently identified in the literature as a predictor of inferior patient outcomes after hip arthroscopy [13, 25••, 31•, 36, 51]. This condition has been reported as a reason for postoperative dislocations or subluxation even when closure of the capsule is performed [25••, 31•]. There is consensus that patients with true hip dysplasia should not undergo isolated hip arthroscopy but could undergo arthroscopic labral repair (with possible cam osteoplasty) when in conjunction with periacetabular osteotomy to address the deficient acetabulum [25••, 51–53]. There is debate regarding the best treatment for patients with mild or borderline hip dysplasia (ACEA or LCEA 20–25°). Atzmon and Safran [51] found that hips with mild or borderline hip dysplasia (LCEA 20–25°) that underwent arthroscopic FAIS treatment had significantly better postoperative outcomes when compared to true hip dysplasia. Some studies have identified good outcomes with careful arthroscopic treatment that includes robust capsular plication and minimal acetabular osteoplasty [54], while other studies show a greater benefit after periacetabular osteotomy [52, 53]. It is likely that a combination of patient factors besides center edge angle—such as acetabular and femoral version, standing pelvic tilt, and ligamentous laxity—must be considered before the recommendation of an appropriate surgical course.

The femoral version is another parameter of interest. It has been repeatedly observed that patients with severe femoral anteversion (>25°) have worse postoperative outcomes after hip arthroscopy when compared to those patients with normal femoral version (0–25°) [25••, 55]. This may be due to instability of the hip and load on the hip flexor tendon, which cannot be solely addressed with hip arthroscopy. Boschung et al. [56] discussed how absolute femoral retroversion (<0°) can lead to extra-articular subspine impingement between the anterior inferior iliac spine (AIIS) and proximal femur. It is possible that arthroscopy cannot sufficiently decompress this impingement and symptoms may persist after surgery, leading to worse postoperative outcomes. Some authors thus recommend derotational osteotomy for patients with abnormal version (retro- or anteversion) rather than an arthroscopic-alone approach. Further investigation is required regarding the exact joint parameters that can predict patient outcomes.

Intraoperative/Technical Factors that Influence Postoperative Outcome

With more surgeons performing hip arthroscopy for FAIS each year, there has been particular interest in seeing what intraoperative factors and technical skills can most reliably predict patient outcomes. Components of interest include labral repair/reconstruction versus debridement, osteoplasty, capsular closure, microfracture, perineal post-use, operative time, pain management, and suture anchor placement. Although certain factors have been implicated as predictors of better or worse outcomes, the literature consistently suggests that arthroscopic treatment of FAIS provides superior patient outcomes when compared to conservative management [2, 8, 12, 13, 14•, 15, 16•, 17, 38].

The labrum is vital for the integrity of the femoroacetabular joint. It provides stability, prevents distraction of the femoral head from the acetabulum through negative pressure, and protects the underlying cartilage [20•, 21, 57–59]. As knowledge, data, and technical skills have become more refined over the last few decades, surgeons have moved away from labral debridement and are more often performing repair or reconstruction [18]. Labral repair is now widely regarded as a gold standard treatment for FAIS [20•], and there is substantial evidence showing that patients routinely have better outcomes when repair is performed rather than debridement or labrectomy [5, 10••, 60]. A primary labral reconstruction with auto- or allograft has superior outcomes over labrectomy when there is a deficiency or extensive/irreparable tearing [49, 57–59].

Femoral and acetabular osteoplasty are used to treat cam and pincer lesions, respectively, and are of interest when predicting patient outcomes [61]. Both procedures have been cited as contributors to improved patient outcomes and reduced risk of reoperation when compared to matched cohorts without osteoplasty [5, 62, 63]. Nepple et al. [64•] followed patients for a minimum of 15 years who had arthroscopic femoral osteoplasty and compared them to those who had arthroscopy for FAIS alone. They found that the femoroplasty group had significantly better PRO scores and reoperation-free survivorship; the non-osteoplasty group was almost 3 times more likely to convert to hip replacement within the 15-year follow-up. Ayeni et al. [65••] completed a randomized control trial in patients with FAIS, performing arthroscopic osteoplasty and labral repair in one cohort and labral repair alone in the other. Both groups demonstrated significantly improved outcomes by 1 year; however, the osteoplasty group showed a significantly lower reoperation rate at 2 years.

Most studies identify the need for femoroplasty when indicated in treating FAIS and the correlation between cam deformities and osteoarthritic progression [63, 64•, 66–68]. However, there is controversy regarding acetabuloplasty, as some studies demonstrate decreased impingement with pincer resection [69], while others argue that acetabular coverage should be preserved to avoid degenerative progression, and that pincer morphology is not associated with the development of osteoarthritis [66]. Johannsen et al. [62] studied the relationship between the degree of pincer resection and postoperative outcomes in patients with a minimum of 5-year follow-up. The study found that patients with larger resections and an LCEA outside the normal reference range had satisfactory outcome scores similar to those patients with smaller resections and normal LCEA. There are arguments that reference over- and under-resection of cam and pincer deformities as common reasons for revision arthroscopy, or even complications such as femoral neck fracture [10••, 19••, 49, 56]. In summary, femoroplasty procedures favor improved patient outcomes and can allow patients to regain ROM, while more research may be needed to confirm the similar efficacy of acetabuloplasty.

There has been a shift in literature over the last several years regarding hip capsule management and its relation to patient outcome prediction. It was long considered standard to leave the hip capsule unrepaired after arthroscopy for FAIS, under the assumption that the capsule will heal over time and not significantly affect patient outcomes or symptoms [21, 70, 71]. More recent evidence supports capsular closure as a significant positive predictor for better outcomes [5, 19••, 20•, 49, 72, 73, 74•, 75]. It is possible that those patients who do not undergo capsule closure in studies may have more arthritis or other confounders that impact postoperative outcomes. The principal argument for capsular closure emphasizes restoring the hip’s native biomechanics and stability and preventing iatrogenic complications [5, 20•, 72, 74•, 75, 76]. Looney et al. [74•] completed a systematic review and meta-analysis that analyzed 3427 repaired capsules versus 1705 unrepaired capsules. The study revealed significantly higher postoperative PRO scores and improvement in those patients treated with repair. Even studies that argue against the routine use of capsular closure still recommend closure in special circumstances, i.e., revision arthroscopy, borderline dysplasia, joint laxity, athletes, and extensive capsulotomies [21, 71]. Although more research is needed to confidently rely on capsular closure to improve outcomes, the current literature is trending in that direction.

Intra-articular cartilage defects and their subsequent treatment can be frustrating for surgeons and patients alike. Microfracture has been shown, at least in the short- and mid-term, to provide improvement to patient outcomes, but with potential for high failure rates [21, 27•, 77, 78]. Domb et al. [79] followed patients for 5 years following acetabular microfracture and found sustained statistically significant improvement in outcomes, but a 28% conversion rate to THA. They strongly suggest carefully selecting patients to avoid THA conversion. Other studies discuss the variability in patient symptoms following microfracture and suggest augmenting with hyaline autograft or autologous matrix-induced chondrogenesis to help improve outcomes [21, 77]. Much of the literature brings into question the reliability of microfracture as a treatment option and shows evidence that short-term improvements can quickly decline to an inferior outcome [10••, 19••, 78]. Although microfracture and its newer techniques can be an option for patients with significant cartilage defects, patients must be counseled about the difficulty of treating these defects and the poor prognosis they may have regardless of surgical treatment.

The use of a perineal post to distract the hip during arthroscopy for FAIS has become a topic of interest to affect postoperative patient outcomes. Complications of post-use are well documented and include compression-related soft tissue injuries, transient neurapraxia, and even permanent neurologic dysfunction [20•, 80•, 81–85]. Performing post-less arthroscopy has proven to create adequate traction for arthroscopic instruments and eliminate pudendal nerve and surrounding perineal soft tissue injury [80•, 82, 83]. Mei-Dan et al. [82] prospectively followed 1000 hips that underwent post-less hip arthroscopy and found no evidence of nerve or soft tissue complication. Some studies have noted that overall postoperative pain does not significantly change with or without the use of a perineal post; however, post-less arthroscopy patients had expedited discharge times, which could contribute to better surgical experiences [84]. There are also arguments stating that the use of a perineal post during arthroscopy should not be relied upon to predict patient outcomes, as most complications resolve within 3 months and both cohorts improve similarly [81, 86]. If a post is used during arthroscopy, the discussion regarding short-term patient outcomes should shift to traction time, force, and surgeon experience [20•, 81]. The innovation of post-less hip distraction for arthroscopy is a compelling alternative to traditional arthroscopy techniques and supports the short-term improvement of patient outcomes.

Machine Learning for Patient Predictors

Hip arthroscopy, especially in relation to the treatment of FAIS, is constantly evolving as surgeons and researchers find new ways to improve outcomes, techniques, and use of technology. Currently, there is significant interest in the application of machine learning (ML) and how it can improve FAIS treatment. Some of these applications include reading pelvic/hip radiographs, estimating the cost of the procedure, and predicting overnight admissions rates, patient follow-up loss, need for revision arthroscopy, and patient outcomes [42, 48, 87–90]. A primary goal of ML is to supplement evidence-based medicine and create personalized treatment options that can predict and produce reliable patient outcomes [87]. The best utilization of ML for patient outcomes in hip arthroscopy is by inputting preoperative, demographic and surgical data into an algorithm to predict the likelihood of achieving MCID [29, 32•, 54, 88, 91–94]. Although ML appears promising, it is still in its infancy, and many algorithms created have been based on a single surgeon’s data or a limited region, and thus require external validation before a more universal application is available for everyday clinical use [87, 91–94]. The future and utilization of ML algorithms for predicting patient outcomes in arthroscopic FAIS surgery has the potential to make data collection and analysis more efficient, and could eventually provide instant preoperative odds of achieving MCID [91]. Adoption of this technology will likely come with time if the evidence continues to show its potential to contribute to surgical outcome prediction.

Conclusion

As arthroscopic treatment of FAIS continues to improve and the number of hip arthroscopy cases increases worldwide, it is important to understand the factors associated with improved postoperative outcomes. There are currently no official guidelines for what predictors should be considered over others, but research does shed light on many patient-specific and technical factors that consistently lead to reliable surgical outcomes. Agreed-upon patient factors that negatively impact outcome after isolated primary hip arthroscopy for FAIS include the presence of osteoarthritis, increasing patient age, increased duration of symptoms, poor preoperative outcomes scores, and presence of true hip dysplasia. Agreed-upon surgical factors that lead to better outcomes include labral repair over debridement and adequate osteoplasty, particularly of the femur. Although this review article does not include all potential factors, it provides an approach that attempts to group them into distinct and comprehensive categories. What remains consistent across studies is that thorough preoperative evaluation, shared decision-making, and expectation management are all imperative to guide outcomes and patient satisfaction.

Author contributions

AS - wrote the main manuscript MH - provided guidelines for the manuscript, resource recommendations, and extensive revisions

Declarations

Competing interests

The authors declare no competing interests.

Conflict of Interest

The authors declare no competing interests.

Human and Animal Rights and Informed Consent

The article does not contain any studies with human or animal subjects performed by any of the authors.