Factors Associated With Failure of Hip Arthroscopy in Patients With Hip Dysplasia

• Updated on: 2018-04-07
• Read original article here

Feature Article 
Factors Associated With Failure of Hip Arthroscopy in Patients With Hip Dysplasia
Andrew J. Bryan, MD; Kirsten Poehling-Monaghan, MD; Aaron J. Krych, MD; Bruce A. Levy, MD; Robert T. Trousdale, MD; Rafael J. Sierra, MD
Orthopedics. 2018;41(2):e234-e239
Abstract
Abstract
The purpose of this study was to compare a group of dysplastic hips treated successfully with hip arthroscopy with a group of dysplastic hips treated unsuccessfully with hip arthroscopy to determine (1) preoperative patient characteristics and radiographic parameters and (2) intraoperative findings and treatment associated with outcome. The authors retrospectively reviewed a prospective database of 20 adult patients (17 female, 3 male) with hip dysplasia who underwent primary hip arthroscopy between January 2009 and February 2013. Modified Beck scores to quantify cartilage damage as well as preoperative and postoperative radiographic measurements (including alpha, Tonnis, and lateral center edge angles [LCEAs]) were compared between patients who failed hip arthroscopy (11 patients) and those who did not (9 patients). Failure after hip arthroscopy was defined as a modified Harris hip score of less than 80 or the need for subsequent hip arthroscopy, arthroplasty, or periacetabular osteotomy. The mean follow-up for the successful patients was 58 months (range, 37–82 months), with an average modified Harris hip score of 93 at most recent follow-up. Preoperative radiographs showed a lower mean LCEA (18.0° vs 21.3°; P=.02) in the failure group, and all successes occurred with a LCEA of 17° or greater. The failure group was more likely to have rim resection of greater than 3 mm performed (hazard ratio, 3.53; P=.04). Among the hips with dysplasia undergoing arthroscopic treatment, patients with a poor outcome were more likely to have an LCEA of less than 17° and intraoperative rim resection of greater than 3 mm. Furthermore, the labral repair group did substantially better than the labral debridement group. [Orthopedics. 2018; 41(2):e234–e239.]
Full Text
Abstract
The purpose of this study was to compare a group of dysplastic hips treated successfully with hip arthroscopy with a group of dysplastic hips treated unsuccessfully with hip arthroscopy to determine (1) preoperative patient characteristics and radiographic parameters and (2) intraoperative findings and treatment associated with outcome. The authors retrospectively reviewed a prospective database of 20 adult patients (17 female, 3 male) with hip dysplasia who underwent primary hip arthroscopy between January 2009 and February 2013. Modified Beck scores to quantify cartilage damage as well as preoperative and postoperative radiographic measurements (including alpha, Tonnis, and lateral center edge angles [LCEAs]) were compared between patients who failed hip arthroscopy (11 patients) and those who did not (9 patients). Failure after hip arthroscopy was defined as a modified Harris hip score of less than 80 or the need for subsequent hip arthroscopy, arthroplasty, or periacetabular osteotomy. The mean follow-up for the successful patients was 58 months (range, 37–82 months), with an average modified Harris hip score of 93 at most recent follow-up. Preoperative radiographs showed a lower mean LCEA (18.0° vs 21.3°; P=.02) in the failure group, and all successes occurred with a LCEA of 17° or greater. The failure group was more likely to have rim resection of greater than 3 mm performed (hazard ratio, 3.53; P=.04). Among the hips with dysplasia undergoing arthroscopic treatment, patients with a poor outcome were more likely to have an LCEA of less than 17° and intraoperative rim resection of greater than 3 mm. Furthermore, the labral repair group did substantially better than the labral debridement group. [Orthopedics. 2018; 41(2):e234–e239.]
Hip arthroscopy for chondrolabral pathology and femoroacetabular impingement has become widely accepted during the past decade, providing a reliable and minimally invasive treatment. 1 Proponents of the technique cite equivalent outcomes when compared with open procedures for correction of bony impingement, 2 as well as excellent pain relief and return to function in patients treated for a variety of intra-articular conditions. 3 The role of hip arthroscopy is less well defined for patients with symptomatic labral tear secondary to underlying hip dysplasia. Analyses of unsatisfactory patient outcomes after arthroscopy largely focus on the presence or absence of residual bony impingement, 4 with little mention of the potential for unrecognized preoperative dysplasia. 5
Several outcomes-based studies report both promising 6 and poor 7 postoperative function and pain scores associated with arthroscopic labral treatment in patients with dysplastic hips. However, data addressing the measurable factors contributing to success or failure are limited. Therefore, the authors compared a group of successfully treated dysplastic hips with a group of unsuccessfully treated dysplastic hips to examine pre- and postoperative radiographic parameters and intraoperative procedures associated with poor outcome. They hypothesized that unrecognized dysplasia treated with rim resection and labral debridement (as opposed to repair) would be significant risk factors for failure of hip arthroscopy.
Materials and Methods
This study received institutional review board approval. The authors retrospectively reviewed the charts of patients with hip dysplasia undergoing hip arthroscopy. All radiographs were reviewed by 2 independent reviewers (K.P.M., A.J.K.). The authors initially identified 46 patients with dysplasia who had undergone hip arthroscopy either at their institution by a fellowship-trained sports surgeon (A.J.K., B.A.L.) or at an outside hospital. Borderline dysplasia was defined as a lateral center edge angle (LCEA) of 20° to 25°, and moderate dysplasia was defined as an LCEA of less than 20°. 8 A total of 26 patients were excluded: 7 for previous periacetabular osteotomy (PAO), 3 for arthroscopy or surgery related to trauma, 14 for incomplete outside medical records, and 2 who declined to participate in research. Patients were considered a “failure” if they had a modified Harris hip score of less than 80 postoperatively 9 or had had a subsequent procedure (repeat hip arthroscopy, PAO, or hip arthroplasty). Outcomes were considered a “success” if patients were without failure at most recent follow-up. All patients were evaluated between January 2009 and February 2013 by 1 of 3 hip arthroscopists (A.J.K., B.A.L., R.J.S.) at the authors' institution and had minimum 1-year follow-up. Patients whose index hip arthroscopy was performed at an outside hospital had outside documents (magnetic resonance images, preoperative radiographs, and operative reports) scanned into their medical record and documented release of information for these records to be viewed. Of the 15 patients who underwent surgery at the authors' institution, only 4 (27%) of the 15 underwent capsular repair. The authors were unable to determine whether the 5 patients who had surgery elsewhere underwent capsular repair.
Patient charts were reviewed for preoperative diagnosis and operative procedures performed. Rim resection was defined as removal of at least 3 mm of acetabular bone reported in the operative note. The authors determined that this was an appropriate cutoff for patients based on outcome in this study. Pre- and post-arthroscopy radiographic measurements were recorded using computer-assisted calculation tools by 2 independent reviewers (A.J.B., K.P.M.). These values included alpha angle, Tonnis angle, and LCEA, all per previously described methods on weight-bearing radiographs. 10 The modified Beck scale was used to quantify magnetic resonance imaging and intraoperative findings of cartilage damage. 11
Statistical Analysis
Patient outcomes included a comparison of pre- and postoperative radiographic parameters as well as intraoperative findings and procedures performed. Patient demographics, baseline clinical data, surgical factors, and patient outcomes that are measured on a continuous scale (eg, time from injury to surgery, length of procedure, and months to failure) were compared using 2-sample t tests or nonparametric Wilcoxon rank sum tests. Demographics, baseline and surgical data, and outcomes defined as binary or multilevel categorical variables (eg, type of surgery, change in radiographic measurements, procedure performed, and intraoperative complications) were evaluated using chi-square tests. Differences between groups were evaluated using Cox proportional hazards regression models. If differences in patient demographics, baseline data, or surgical factors were identified, the study outcomes were analyzed using multivariable models to adjust for the between-group differences. When used, linear regression was applied for the continuous outcomes, logistic regression for the binary outcomes, and Cox proportional hazards regression for the time-to event outcomes. All statistical tests were 2-sided. P<.05 was considered significant.
Results
Of the 20 patients who met inclusion criteria, 11 were included in the failure group and 9 were included in the success group. The mean follow-up for the success group was 58 months (range, 37–82 months). Fifteen of the 20 patients had their hip arthroscopy performed at the authors' institution (9 of 9 in the success group; 6 of 11 in the failure group). The average modified Harris hip score at most recent follow-up in the success group was 93 (range, 82–100). The mean time to failure was 23 months (range, 18–75 months) (Table 1 ). There were no significant demographic differences between the success and the failure groups, with a mean age at surgery of 36.3 and 30.4 years, respectively, and more females in both groups (8 in the success group and 9 in the failure group) (Table 1 ). Only 35% (7 of 20) had hip dysplasia recognized preoperatively, and this most commonly resulted from inadequate radiographic measurements. Seven of the failures (77%) went on to PAO, whereas 2 underwent total hip arthroplasty, 1 had a modified Harris hip score of less than 80 with no subsequent procedure, and 1 underwent repeat hip arthroscopy with revision labral repair for failure of a previously repaired labrum.
Table 1:
Patient Demographics
There was no significant difference in the mean cartilage damage between the success and failure groups as measured by Beck score (1.7 vs 2.0; P=.3). However, the failure group had significantly more full-thickness, Beck grade 5 lesions compared with the success group (5 vs 1; P<.01).
In the failure group, the mean preoperative LCEA was lower (18.0° vs 21.3°; P=.02) and the mean Tonnis angle was greater but did not reach statistical significance (14.8° vs 11.1°; P=.11) (Table 2 ). In the success group, the LCEA was consistently greater than 17°. However, it was not predictive of success, as shown by the failure group. Five patients (45%) with an LCEA of greater than 17° failed hip arthroscopy, suggesting that factors other than LCEA are predictive of the success of hip arthroscopy in patients with dysplasia. The risk of failure was increased for each LCEA degree below 17° (hazard ratio, 1.13; 95% confidence interval, 0.94–1.35; P=.05). The mean postoperative LCEAs (22.1° success; 18.5° failure) and Tonnis angles (11.4° success; 13.7° failure) were essentially unchanged from preoperative measurements in both groups (Table 3 ).
Table 2:
Table 3:
Postoperative Radiographic Findings
Labral repair other than debridement was most commonly performed in patients who had successful procedures. Labral debridement was performed in 1 (11%) of the 9 successes and 9 (82%) of the 11 failures, whereas labral repair was performed in 8 successes (89%) and 2 failures (18%). At 1 year, patients who underwent rim re-section had a 71.4% failure-free survivor-ship compared with 92.3% for those who did not have rim resection of greater than 3 mm. At 2 years, this decreased to 53.5% compared with 76.9% (P<.03) (Table 4 ).
Table 4:
Developmental Dysplasia of the Hip: Survivorship Free of Failure
Addressing the femoral head–neck junction abnormality was not associated with a better result. The mean alpha angles were similar preoperatively between the success and failure groups (53.9° vs 58.4°; P=.45) (Table 2 ). Intraoperatively, 6 (67%) of the 9 patients in the success group had an identifiable abnormality of the femoral head–neck junction that was removed compared with 9 (82%) of the 11 patients in the failure group. Postoperative mean alpha angles were decreased compared with preoperative mean alpha angles but were not significantly different between the success and failure groups (40.0° and 42.5°; P=.42) (Table 3 ).
Discussion
Hip arthroscopy can be a valuable tool for the relief of pain and the preservation of the hip. As the indications have evolved, understanding of the limitations in the context of individual hip pathoanatomy has improved. In this study, the authors found that moderate acetabular dysplasia (as measured by a preoperative LCEA of <17°) and concomitant intraoperative rim resection negatively affected outcomes after hip arthroscopy.
Each patient in this study was identified as dysplastic based on several sets of radiographs showing abnormal center edge and Tonnis angles preoperatively, yet only 35% (7 of 20) were formally diagnosed as being dysplastic before their procedure. The ANCHOR Investigators recently evaluated 30 PAOs performed after failed hip arthroscopy. They found that most of the failures involved young females with high preoperative functional limitations. Twenty-two percent of those patients also underwent a second hip arthroscopy before dysplasia was recognized and eventually treated with a PAO. 12 Similar to the current series, the ANCHOR Investigators concluded that identifying dysplasia preoperatively was of utmost importance. The current authors recommend routine systematic review of the radiographs, including a well-positioned anteroposterior pelvic radiograph, of every hip patient. 10 Mandatory measurement of the LCEA is necessary to reliably screen patients for the presence of dysplasia. The biggest problem is that measurements vary from one surgeon to the next. The authors occasionally use computed tomography scans with 3-dimensional reconstructions to help determine the level of anterior and lateral coverage in mildly dysplastic hips because they believe that a review of the 3-dimensional coverage of the hip is paramount in addition to noting the femoral version. Hips with borderline dysplasia but increased femoral version may indicate that instability is causing the patient's hip pain. In addition, the computed tomography scan may help determine whether the patient lacks anterior coverage, sometimes obscured by a normal false profile view, if the patient has a degree of cranial retroversion. Areas laterally and anteriorly of incongruency can also indicate hip dysplasia and should be noted.
The consequences of unrecognized dysplasia on hip arthroscopy extend beyond bony anatomy. Haene et al 13 assessed the tear patterns and quadrants in 59 dysplastic hips undergoing hip arthroscopy and found that dysplastic hips were more likely to have a radial tear. In a retrospective review of 72 symptomatic dysplastic hips, Tamura et al 14 used high-resolution computed tomography arthrography to document the location of labral tears in dysplastic hips compared with hips with femoroacetabular impingement. They found that dysplastic hips more commonly had tears through the body of the labrum, as opposed to the base. The current study found higher survivorship in dysplastic hips undergoing labral repair compared with labral debridement; however, this was confounded by the increased incidence of acetabular rim resection in those hips undergoing labral repair. Interestingly, the 2 labral repair patients who failed also had lower LCEAs (16° and 20°) preoperatively. This finding is similar to that of previous reports suggesting that, in the absence of anterosuperior coverage, dysplastic hips rely on the intact labrum to contain the articulation, and that resection might thereby worsen outcomes. 15 This was likely worsened by concomitant rim resection accompanying labral repair, as the current authors found a higher failure rate in this scenario. They considered rim resection to be the removal of at least 3 mm of bone from the acetabular rim. Dysplastic hips underwent rim resection to create a bed for labral repair and to decrease the risk of impingement. This does correlate with the overall under-recognition of the dysplasia, as this procedure made the dysplasia and microinstability worse. Interestingly, the authors observed no change in the LCEA following rim resection. This likely represents removal of anterior rim instead of lateral bone. It would not be detected on an anteroposterior radiograph and would also correspond to the most common location of labral pathology in a dysplastic hip. Notably, the need for labral resection or repair did not correlate with age or center edge angle.
The recognition of preoperative dysplasia can impact the surgeon's treatment algorithm and should warrant caution when considering hip arthroscopy. Most studies reporting on the outcomes of labral resection, repair, or osteoplasty specifically exclude patients with dysplasia from their analyses. 16 There is a paucity of data assessing the true risk of performing hip arthroscopy on dysplastic hips, and despite modern treatment algorithms primarily aimed at dysplastic hips, the reported results are fair at best. 17
This retrospective study had several limitations. First, some of the patients defined as failures underwent revision surgery. Revision surgery such as PAO may be performed in cases of mild dysplasia, and patient selection is subject to bias of the treating surgeon. However, all of the patients were symptomatic enough to warrant a second procedure. Second, this was a relatively small cohort of patients, which limited the authors' ability to obtain certain outcomes measures. Many outside records did not contain false profile views for calculating anterior center edge angles, and operative reports varied in documenting the presence or absence of capsular repair at the end of the case. Of the 15 patients who underwent surgery at the authors' institution, only 4 (27%) of the 15 underwent capsular repair. This was particularly detrimental to the authors' analysis, as studies have shown that capsular plication with labral preservation may be a viable arthroscopic treatment for mildly dysplastic hips. 17 However, the authors found significant differences between the failure and success groups. To the authors' knowledge, no previous studies have defined risk factors for failure following hip arthroscopy in a cohort with dysplasia. Furthermore, as a result of patients who had a prior hip arthroscopy at an outside institution being included, the surgical techniques and rehabilitation for each patient were likely different. There was also an inherent bias in the failure group, as hip arthroscopy has a high likelihood of failure when being performed with Beck grade 5 lesions.
Recognition of the degree of not only preoperative dysplasia but also joint degeneration can impact the surgeon's treatment algorithm, and both should warrant caution when considering hip arthroscopy. This is key for 3 main reasons noted in this study: (1) Labral debridement is a poor predictor of success. If the labrum cannot be repaired, labral techniques other than repair may be needed. (2) The degree of cartilage damage was a predictor of failure as well. This should be noted on the preoperative magnetic resonance image. Regardless of age, patients with more severe forms of disease should not be treated arthroscopically. (3) If dysplasia is not diagnosed preoperatively, the surgeon may remove rim to either repair the labrum or treat acetabular cartilage defects. This was also a poor prognostic factor in this study. The data from this study are extremely useful because most studies reporting on the outcomes of labral resection, repair, or osteoplasty specifically exclude dysplastic patients from their analyses. 16 There is a paucity of data assessing the true risk of performing hip arthroscopy on dysplastic hips, and despite modern treatment algorithms primarily aimed at dysplastic hips, the reported results are fair at best. 17
Conclusion
Among the hips with dysplasia undergoing arthroscopic treatment, patients with poor outcomes were more likely to have an LCEA of less than 17° and intra-operative rim resection of greater than 3 mm. Furthermore, the labral repair group did substantially better than the labral debridement group. For those hips doing well at short to mid-term follow-up, careful longer-term follow-up is needed to determine whether these improvements in pain and function are maintained.
References
Gupta A, Redmond JM, Stake CE, Dunne KF, Domb BG. Does primary hip arthroscopy result in improved clinical outcomes? 2-year clinical follow-up on a mixed group of 738 consecutive primary hip arthroscopies performed at a high-volume referral center. Am J Sports Med. 2016; 44(1):74–82. doi:10.1177/0363546514562563 [CrossRef]
Büchler L, Neumann M, Schwab JM, Iselin L, Tannast M, Beck M. Arthroscopic versus open cam resection in the treatment of femoroacetabular impingement. Arthroscopy. 2013; 29(4):653–660. doi:10.1016/j.arthro.2012.12.009 [CrossRef]
Montgomery SR, Ngo SS, Hobson T, et al. Trends and demographics in hip arthroscopy in the United States. Arthroscopy. 2013; 29(4):661–665. doi:10.1016/j.arthro.2012.11.005 [CrossRef]
Heyworth BE, Shindle MK, Voos JE, Rudzki JR, Kelly BT. Radiologic and intraoperative findings in revision hip arthroscopy. Arthroscopy. 2007; 23(12):1295–1302. doi:10.1016/j.arthro.2007.09.015 [CrossRef]
Fayad TE, Khan MA, Haddad FS. Femoroacetabular impingement: an arthroscopic solution. Bone Joint J. 2013; 95-B(11)(suppl A):26–30. doi:10.1302/0301-620X.95B11.33016 [CrossRef]
Jayasekera N, Aprato A, Villar RN. Hip arthroscopy in the presence of acetabular dysplasia. Open Orthop J. 2015; 9:185–187. doi:10.2174/1874325001509010185 [CrossRef]
Hetaimish BM, Khan M, Crouch S, et al. Consistency of reported outcomes after arthroscopic management of femoroacetabular impingement. Arthroscopy. 2013; 29(4):780–787. doi:10.1016/j.arthro.2012.11.011 [CrossRef]
Tannast M, Hanke MS, Zheng G, Steppacher SD, Siebenrock KA. What are the radiographic reference values for acetabular under- and overcoverage?Clin Orthop Relat Res. 2015; 473(4):1234–1236. doi:10.1007/s11999-014-4038-3 [CrossRef]
Aprato A, Jayasekera N, Villar RN. Does the modified Harris hip score reflect patient satisfaction after hip arthroscopy?Am J Sports Med. 2012; 40(11):2557–2560. doi:10.1177/0363546512460650 [CrossRef]
Clohisy JC, Carlisle JC, Beaulé PE, et al. A systematic approach to the plain radiographic evaluation of the young adult hip. J Bone Joint Surg Am. 2008; 90(suppl 4):47–66. doi:10.2106/JBJS.H.00756 [CrossRef]
Beck M, Kalhor M, Leunig M, Ganz R. Hip morphology influences the pattern of damage to the acetabular cartilage: femoroacetabular impingement as a cause of early osteoarthritis of the hip. J Bone Joint Surg Br. 2005; 87(7):1012–1018. doi:10.1302/0301-620X.87B7.15203 [CrossRef]
Ross JR, Clohisy JC, Baca G, Sink EANCHOR Investigators. Patient and disease characteristics associated with hip arthroscopy failure in acetabular dysplasia. J Arthroplasty. 2014; 29(9)(suppl):160–163. doi:10.1016/j.arth.2014.03.054 [CrossRef]
Haene RA, Bradley M, Villar RN. Hip dysplasia and the torn acetabular labrum: an inexact relationship. J Bone Joint Surg Br. 2007; 89(10):1289–1292. doi:10.1302/0301-620X.89B10.17319 [CrossRef]
Tamura S, Nishii T, Takao M, Sakai T, Yoshikawa H, Sugano N. Differences in the locations and modes of labral tearing between dysplastic hips and those with femoroacetabular impingement. Bone Joint J. 2013; 95-B(10):1320–1325. doi:10.1302/0301-620X.95B10.31647 [CrossRef]
Henak CR, Ellis BJ, Harris MD, Anderson AE, Peters CL, Weiss JA. Role of the acetabular labrum in load support across the hip joint. J Biomech. 2011; 44(12):2201–2206. doi:10.1016/j.jbiomech.2011.06.011 [CrossRef]
Byrd JW, Jones KS. Hip arthroscopy in the presence of dysplasia. Arthroscopy. 2003; 19(10):1055–1060. doi:10.1016/j.arthro.2003.10.010 [CrossRef]
Domb BG, Stake CE, Lindner D, El-Bitar Y, Jackson TJ. Arthroscopic capsular plication and labral preservation in borderline hip dysplasia: two-year clinical outcomes of a surgical approach to a challenging problem. Am J Sports Med. 2013; 41(11):2591–2598. doi:10.1177/0363546513499154 [CrossRef]
Table 1