Knee Surg Sports Traumatol Arthrosc DOI 10.1007/s00167-014-2885-9
Hip arthroscopy with labral repair for femoroacetabular impingement: short-term outcomes Christian Dippmann • Kristian Thorborg • Otto Kraemer • Søren Winge • Henrik Palm Per Ho¨lmich
Received: 3 November 2013 / Accepted: 26 January 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose The purpose of this study was to examine the progression of clinical outcomes 3, 6 and 12 months after hip arthroscopy with labral repair for femoroacetabular impingement (FAI). Methods From May 2009 to December 2011, 87 consecutive patients [55 females (median age 38, range 17–63) and 32 males (median age 38, range 15–59)] underwent hip arthroscopy and labral repair, by the same experienced surgeon. Standardised, but unstructured, post-operative rehabilitation instructions were provided. Function and pain were evaluated using modified Harris Hip Score (mHHS) and visual analogue scale (VAS) pre-operatively and post-operatively at 3, 6, and 12 months. Results A main effect of time was seen for mHHS and pain (VAS) at 3, 6, and 12 months (p \ 0.001). Mean (SE) mHHS improved significantly from 59.9 (1.9) pre-operatively to 74.5 (1.9) after 3 months (p \ 0.001), and again from 3 to 6 months from 74.5 (1.9) to 80.1 (1.9) (p = 0.004), with no additional changes from 6 to 12 months [80.1 (1.9)–78.7 (1.9), (n.s.)]. Also, the mean
Electronic supplementary material The online version of this article (doi:10.1007/s00167-014-2885-9) contains supplementary material, which is available to authorized users. C. Dippmann (&) K. Thorborg O. Kraemer P. Ho¨lmich Arthroscopic Center Amager, Copenhagen University Hospital Hvidovre, Copenhagen, Denmark e-mail: [email protected]
S. Winge Copenhagen Private Hospital, Copenhagen, Denmark H. Palm Department of Orthopedic Surgery, Copenhagen University Hospital Hvidovre, Copenhagen, Denmark
(SE) VAS improved significantly from pre-operative 57.9 (2.6) to 30.0 (2.6) after 3 months (p \ 0.001), and from 30.0 (2.6) to 22.6 (2.6) (p = 0.017), after 6 months with no additional changes from 6 to 12 months [22.6 (2.6)–27.9 (2.6), (n.s.)]. Conclusions Improvements in function (mHHS) and pain (VAS) were seen in patients after hip arthroscopy with labral repair for FAI at 3, 6, and 12 months. While significant improvements occurred from 3 to 6 months, no further improvements were seen from 6 to 12 months. Level of evidence Prospective case series, Level IV. Keywords Hip arthroscopy 1 Patient reported outcome 2 Early results 3 Labral repair Femoroacetabular impingement 4
Introduction Over the last decade, hip arthroscopy has become a wellaccepted procedure to address early onset hip osteoarthritis and/or non-arthritic hip pain [10, 13, 21]. Femoroacetabular impingement (FAI) as a result of pincer and/or cam morphology has been proposed as a cause of osteoarthritis of the hip [6, 16]. Different ways of addressing the FAI pathologies have been reported including osteoplasty, acetabular rim trimming, and labrum repair [1, 10, 25]. Although the function of the acetabular labrum and surgical management is not fully understood, midterm results (1–3 years followup) indicate that outcomes after labral repair seem to be superior to labral resection [4, 9, 12, 14, 15, 23]. The only outcome studies collecting data during the first post-operative months are the studies by Larson et al. [14, 15]. They investigated the outcome between labral repair and labral debridement and found significant pain and
Knee Surg Sports Traumatol Arthrosc
functional improvements immediately within the first couple of months [14, 15]. A literature search found no other studies investigating the outcomes after hip arthroscopy with labral repair in management of FAI within the first year after surgery. The purpose of this study was therefore to study the outcome after hip arthroscopy with labral repair evaluating function using the modified Harris Hip Score (mHHS) and pain using a visual analogue scale (VAS) at 3, 6, and 12 months after surgery. The hypothesis was that significant improvements in function and pain could be seen within the first 3 months after hip arthroscopy with labral repair.
performed. Patients responding with temporary pain relief and/or increased active and passive range of motion were scheduled for a hip arthroscopy with labral repair, as the response to the injection indicated the presence of an intraarticular lesion. Pre-operatively, function and pain were assessed with the modified Harris Hips Score (mHHS) (0–100) and a VAS. These outcome measures have previously been used in this population [14, 15, 27]. Pain was evaluated on a 100-mm line, where 0 indicates no pain and 100 mm indicates worst imaginable pain, during the past month. These questionnaires were completed again at the follow-up consultations at 3, 6, and 12 months after surgery. According to Danish law, no ethical approval is required for the observational studies of this kind.
Materials and methods Surgical procedure From May 2009 to December 2011, 104 consecutive patients with FAI underwent hip arthroscopy for the first time and were eligible for inclusion. Fourteen of these patients were not included due to the following reasons: (a) patients did not fill in the primary evaluation form (mHHS) before surgery (n = 3), (b) hip arthroscopy with labral repair was not performed (n = 9) (instead, labral resection (n = 7), psoas tenotomy (n = 1), and the removal of loose bodies (n = 1) were performed). Furthermore, five patients were lost to follow-up as they had a hip arthroscopy re-operation, or a total hip replacement (THR) within the first year. Thus, 87 patients, including 55 females [median age 38 (15–59)] and 32 males [median age 38 (range 17–63)] met the inclusion criteria and were enrolled in the study. In all but seven patients, femoral osteoplasty was also performed. The decision whether femoral osteoplasty should be performed or not was based on the pre-operative radiographic examination and the dynamic examination of the hip during surgery under visual and radiological control. None of the patients underwent microfracture. FAI was determined by obtaining a profound medical history of the patient and by a full clinical examination, including localisation of pain, range of motion, and test for anterior impingement . Furthermore, all patients had anterior– posterior (AP) and cross-table lateral radiographs of the pelvis and the hip joint to identify radiological signs of cam and/or pincer morphology. Hips with an a-angle [55° and/ or a positive crossover sign were considered to have FAI. The joint space was measured centrally, medially, and laterally as described by Jacobsen et al. . A joint-space width of 2 mm or less was considered pathological in terms of osteoarthritis, and patients were therefore not eligible for hip arthroscopy. MRI was not performed as a routine, but was available when needed. In cases where radiographic or MRI findings were inconclusive, an ultrasound-guided, intra-articular injection with a local anaesthetic was
We used the surgical technique described by Philippon et al.  and Larson et al.  with the patient in the supine position and with a standard two portal technique (anterolateral and inferior mid-anterior). Under temporary traction, the labrum was inspected and labral tears and/or chondrolabral separations were debrided and/or partially detached. Depending on the pre-operative CE angle and the intra-operative situation (sub-labral osteophytes, sub-spinal impingement, etc.), a full acetabular rim trimming or just a smoothening/levelling of the acetabular rim was performed. Depending on the size of the labral lesion, 2–4 suture anchors were used to reattach the labrum. Chondral defects were debrided and left in a stable status combined with microfracture if indicated. Cam morphology was treated by osteoplasty without traction, and sufficient resection was confirmed under dynamic testing. The patients were discharged later the same afternoon or at the latest day after surgery. All patients those received femoral osteoplasty used crutches with partial weight-bearing (i.e. 10 kg) for 4 weeks. Flexion beyond 90 degrees and abduction more than 25 degrees were not allowed until after 2 weeks. Braces were not used. The post-operative rehabilitation was supervised by the local community physiotherapies, since no structured rehabilitation programme was provided at our clinic. The patients were discharged with a standardised rehabilitation manual (Appendix 1), based upon the rehabilitation progression model described by Stalzer et al. . The 26-page training manual included illustrates relevant exercise options including a time schedule for post-operative mobilisation and levels of progression for all patients. The local community physiotherapists supervised the rehabilitation in the first 8–12 weeks after the operation according to these guidelines. All patients were guided
Knee Surg Sports Traumatol Arthrosc
through four phases  (mobility, stability, strength, and return to sport/competition) by their local physiotherapist. The strength and coordination exercises were progressively increased under careful supervision by the physiotherapist guided by the patients’ pain and functional improvement. Specific adherence to rehabilitation was not possible to control for due to patients being rehabilitated at different physiotherapy clinics. Statistics All 87 patients originally included in the study were included in the statistical analyses: 55 women and 32 men, with a mean age of 38 (15–63) years. The primary outcome measure was mHHS (0–100 points). The secondary outcome measure was VAS pain scores (0–100 mm). Statistical analysis was performed using SPSS version 19. The data were analysed by fitting separate linear mixed models for mHHS and VAS. Missing data from the 87 patients during the 1-year follow-up were assumed to be missing at random. Mean and standard deviations were given for baseline calculations. The mean of each outcome was estimated along with its corresponding standard error at all observation times. Different time points and gender were included as factors in the repeated measure analysis, and age was included as a covariant. Post hoc comparisons between the main effects of all pairs of points in time were performed separately for each model, corresponding to the primary (mHHS) and secondary outcomes (VAS). Bonferroni adjustments were applied for pairwise comparisons to account for multiple testing. The significance level was set at 0.05, and all differences with p values below this level were considered to be statistically significant.
Fig. 1 Modified Harris Hip Score (mHHS) within 1 year after hip arthroscopy with labral repair. Asterisks denotes a statistically difference in mHHS from pre-surgery to this time point
Fig. 2 VAS pain score within 1 year after hip arthroscopy with labral repair. Asterisks denotes a statistically difference in VAS from presurgery to this time point
Modified Harris Hip Score (mHHS)
The linear mixed models’ analysis revealed that there was a primary effect over time mHHS and VAS (p \ 0.001). Significant changes were seen in between baseline (presurgery) and 3, 6, and 12 months for mHHS and VAS (p \ 0.001), Figs. 1 and 2. Mean (SE) mHHS improved from pre-operatively to 3 month post-operative [59.9. (1.9)–74.5 (1.9), (p \ 0.001)] and from 3 to 6 months [74.5 (1.9)–80.1 (1.9), (p = 0.004)], with no additional changes from 6 to 12 months [80.1 (1.9)–78.7 (1.9), (n.s.)].
A total of 348 mHHS evaluations for the study population of 87 patients were planned. Thirteen of these evaluations (4 %) were missing in 11 patients (nine patients missing one and two patients missing two). These 11 patients were compared to the 76 patients with complete data regarding pre-operative differences. The mean (SD) age of the 11 patients with missing data was 40 (6) years, and it was 38 (12) years for the group with complete data. Five of the eleven patients with missing data were males, as compared to 27 of the 76 in the complete data group. The pre-operative mHHS values were similar. The mean (SD) preoperative mHHS was 54 (14) in the missing data group and 60 (14) in the complete data group (n.s.).
Visual analogue scale As with the mHHS, the linear mixed models’ analysis revealed that there was a primary effect of time
Knee Surg Sports Traumatol Arthrosc
(p \ 0.001). Mean (SE) VAS score improved from preoperatively to 3 month post-operative [57.9 (2.6)–30.0 (2.6) (p \ 0.001)] and from 3 to 6 months [30.0 (2.6)–22.6 (2.6), (p = 0.017)] with no additional changes from 6 to 12 months [22.6 (2.6)–27.9 (2.6), (n.s.)].
Discussion The most important finding in the present study is that clinically and statistically relevant improvements in function and pain could be seen already 3 months after hip arthroscopy with labral repair for FAI. The purpose of this study was to report the 3- to 12-month outcomes in patients having hip arthroscopy with labral repair combined with acetabular rim trimming and/or osteoplasty. The present study shows that function and pain improve significantly within the first year after surgery at all time points. Significant improvements from pre-operatively to 3 months and again from 3 to 6 months were seen for both function and pain. Interestingly, no further significant and clinical improvement in mHHS and VAS scores was seen from 6 to 12 months. Our hypothesis was therefore confirmed, and hip arthroscopy with labral repair in patients with FAI leads to significant improvements already within the first 3 months after surgery. The mean improvement in mHHS from 3 to 6 months is smaller (5.5 points) than from pre-operatively to 3 months (14.5 points). It can be discussed whether 5.5 points on the mHHS is a clinically relevant change. Previously, it has been reported by Larson et al.  that 6 points constitute a clinically relevant change, without any scientific justification for this. However, based upon the suggestions of Norman et al.  where 0.5 SD at baseline has been shown to be a relevant estimate of the minimal important change, 6–7 points seems a reasonable suggestion for a clinically relevant improvement in the mHHS, as the SD of the present study was 14 at baseline. Therefore, while there are clearly relevant changes occurring from baseline (preoperatively) to 3, 6, and 12 months, the significant change from 3 to 6 months may not necessarily be clinically relevant, thus indicating that the effect of hip arthroscopy mainly occurs within the first 3 months after surgery, as also previously indicated by Larson et al. We have no information about when, during the first 3 months until the first visit in the outpatient clinic, the improvement took place. Furthermore, as all but seven patients underwent hip arthroscopy with labral repair and femoral osteoplasty, we are unable to say whether the combination of labral repair and femoral osteoplasty, or one or the other, improved the clinical outcome. In a systematic review by Kemp et al. , three prospective studies for hip arthroscopy without osteoplasty
and nine prospective studies for hip arthroscopy with osteoplasty were analysed. These studies included very inhomogeneous groups and used different primary outcome measures and inclusion criteria. Among them, there were only two studies documenting the outcome after labral repair/refixation by Philippon et al. [21, 23]. In both studies, mHHS, as the primary outcome measure, improved significantly from 58 to 84 [mean difference = 24 (95 % CI, 19–28)] (n = 112)  and from 70 (range 57–100) to 95 (range 74–100) (n = 28) , respectively. The surgical procedure used in our study was similar to the one used in these studies. The average follow-up period was 24 months  and 2.3 years , respectively, but no data were published regarding the outcomes within the first year after surgery. In a recently published randomised study by Krych et al. , significant improvements were found in a group treated with labral repair compared to a group undergoing labral debridement. The Hip Outcome Score (HOS) ADL and HOS Sport were used as primary outcome score . The average follow-up was 32 months (range 12–48) but no data were published regarding the outcomes within the first year after surgery. The aim of the two retrospective studies including the same cohort by Larson et al. [14, 15] was to compare the clinical outcome after hip arthroscopy with labral repair to HA with labral debridement after 1 year  and after [3.5 years . However, both studies also indicated clinical relevant improvements in mHHS and VAS pain 3 months after surgery, but since no exact figures were reported in their paper, a direct comparison with our results is not possible. The results in this study indicate that a significant improvement in outcomes can be expected within the first 3 months after surgery. As a consequence of this, we believe that a revision of diagnoses and/or a modification of the rehabilitation programme should be considered if improvement is not observed during this period. Whether this improvement was the result of the post-operative physiotherapy or a direct result of the surgical intervention remains unclear, as there are no comparative studies investigating conservative and surgical treatment in patients with hip and groin pain. The literature provides little evidence on rehabilitation after hip arthroscopy. Depending on the surgical procedure performed and the surgeon’s preferences, the degree of mobilisation and weight-bearing varies. Larson et al.  allowed toe-touch weight-bearing during the first two postoperative weeks after hip arthroscopy with labrum refixation, whereas Philippon et al.  only allowed partial weight-bearing (5, 4 kg for 8–12 h/day) during the first 4 weeks. While some authors advocated the use of crutches only, others combined it with a hip brace . We followed the surgical technique described by Philippon et al.
Knee Surg Sports Traumatol Arthrosc
, and as a consequence, we also followed his protocol for weight-bearing post-operatively. All patients operated with femoral osteoplasty were instructed to use crutches to reduce weight-bearing to 10 kg for the first 4 weeks after surgery. However, the significance of weight-bearing or not has so far not been the subject of scientific research. The rehabilitation protocols, although not described in detail, published in recent years share some common principles in terms of the rehabilitation time schedule and the structure of the rehabilitation programmes (phases and main focus) [2, 3, 22, 24, 28]. Stalzer et al.  describe a four-phase rehabilitation programme with the main focus on protection and mobility (phase 1, 1–4 weeks after surgery), controlled stability (phase 2, 4–7 weeks after surgery), strengthening (phase 3, 7–9 weeks after surgery), and return to sport (phase 4, 9–24 weeks after surgery). The progression through the different aspects of the training and proceeding from one phase to the next is depending on the patients’ performance and response to the rehabilitation. However, these rehabilitation concepts are based on the authors’ experiences and have not been tested scientifically. Since there is no evidence available in the literature regarding the post-operative rehabilitation, we adopted the ideas of the rehabilitation programme by Stalzer et al.  focusing on mobility, stability, strength, and return to sport. Depending on their progress, the patients were able to move linear through the phases or train simultaneously in up to three phases. The local community physiotherapists who in most cases supervised the rehabilitation used this programme as a guide in the rehabilitation. The role of rehabilitation after hip arthroscopy in relation to the outcome is not known; further research in this field is needed. It is evident from the current literature that the documentation and evaluation of the outcome after hip arthroscopy needs to become better standardised, as too many different measurement tools are in use. The only patient-reported outcome scores (PROs) validated to detect changes in the patient outcome after hip arthroscopy over time are the HOS [5, 11, 17, 26], international Hip Outcome Tool (iHOT) , and the Copenhagen Hip and Groin Outcome Score (HAGOS) . As the HAGOS and iHOT did not exist, and the HOS was not validated in the Danish language at the time of our study, we used mHHS and VAS pain score to assess the patients’ outcome. The mHHS was developed for hip arthroplasty and might not be sufficient for measuring subtle clinical changes in patients after hip arthroscopy with labral repair. The systematic reviews by Hetaimish et al.  and Kemp et al. , including 29 studies of arthroscopic FAI treatment, found variations in the reporting of the clinical outcomes, and we agree with their recommendations that
there is a need for consistent outcome reporting after arthroscopic FAI surgery . Despite these limitations, we found that the mHHS and VAS pain score were able to detect changes in the patients’ outcomes over time. These changes were, as hypothesised, statistically significant during the first months after surgery. The findings presented in this study are of high clinical relevance for surgeons following patients after hip arthroscopy with labral repair for FAI. In lack of improvement, the treatment should be reconsidered already 3 months after surgery.
Conclusion Improvements in function (mHHS) and pain (VAS) were seen in patients after hip arthroscopy with labral repair for FAI at 3, 6, and 12 months. While significant improvements occurred from 3 to 6 months, no further improvements were seen from 6 to 12 months. Acknowledgments The authors gratefully acknowledge the help of Physiotherapist, MSc., Thomas Linding Jakobsen, Department of Physiotherapy, Copenhagen University Hospital, Amager-Hvidovre, for the development of the rehabilitation manual. The authors gratefully acknowledge the statistical expertise and assistance of Ditte Sæbye from Institute of Preventive Medicine, Copenhagen University Hospital, Frederiksberg.
References 1. Byrd JWT, Jones KS (2011) Arthroscopic management of femoroacetabular impingement: minimum 2-year follow-up. Arthroscopy 27(10):1379–1388 2. Edelstein J, Ranawat A, Enseki KR, Yun RJ, Draovitch P (2012) Post-operative guidelines following hip arthroscopy. Curr Rev Musculoskelet Med 5(1):15–23 3. Enseki KR, Martin R, Kelly BT (2010) Rehabilitation after arthroscopic decompression for femoroacetabular impingement. Clin Sports Med 29(2):247–255 4. Espinosa N, Rothenfluh DA, Beck M, Ganz R, Leunig M (2006) Treatment of femoro-acetabular impingement: preliminary results of labral refixation. J Bone Joint Surg Am 88(5):925–935 5. Fitzpatrick R, Davey C, Buxton MJ, Jones DR (1998) Evaluating patient-based outcome measures for use in clinical trials. Health Technol Assess 2(14):i–iv, 1–74 6. Ganz R, Parvizi J, Beck M, Leunig M, No¨tzli H, Siebenrock KA (2003) Femoroacetabular impingement: a cause for osteoarthritis of the hip. Clin Orthop Relat Res 417:112–120 7. Hetaimish BM, Khan M, Crouch S, Simunovic N, Bedi A, Mohtadi N, Bhandari M, Ayeni OR (2013) Consistency of reported outcomes after arthroscopic management of femoroacetabular impingement. Arthroscopy 29(4):780–787 8. Jacobsen S, Sonne-Holm S, Søballe K, Gebuhr P, Lund B (2004) The relationship of hip joint space to self reported hip pain. A survey of 4.151 subjects of the Copenhagen City Heart Study: the Osteoarthritis Substudy. Osteoarthr Cartil 12(9):692–697 9. Kelly BT, Weiland DE, Schenker ML, Philippon MJ (2005) Arthroscopic labral repair in the hip: surgical technique and review of the literature. Arthroscopy 21(12):1496–1504
Knee Surg Sports Traumatol Arthrosc 10. Khanduja V, Villar RN (2007) The arthroscopic management of femoroacetabular impingement. Knee Surg Sports Traumatol Arthrosc 15(8):1035–1040 11. Kemp JL, Collins NJ, Makdissi M, Schache AG, Machotka Z, Crossley K (2012) Hip arthroscopy for intra-articular pathology: a systematic review of outcomes with and without femoral osteoplasty. Br J Sports Med 46(9):632–643 12. Krych AJ, Thompson M, Knutson Z, Scoon J, Coleman SH (2013) Arthroscopic labral repair versus selective labral debridement in female patients with femoroacetabular impingement: a prospective randomized study. Arthroscopy 29(1):46–53 13. Larson CM, Giveans MR (2008) Arthroscopic management of femoroacetabular impingement: early outcomes measures. Arthroscopy 24(5):540–546 14. Larson CM, Giveans MR (2009) Arthroscopic debridement versus refixation of the acetabular labrum associated with femoroacetabular impingement. Arthroscopy 25(4):369–376 15. Larson CM, Giveans MR, Stone RM (2012) Arthroscopic debridement versus refixation of the acetabular labrum associated with femoroacetabular impingement: mean 3.5-year follow-up. Am J Sports Med 25(4):1015–1021 16. Leunig M, Beck M, Dora C, Ganz R (2006) Femoroacetabular impingement: trigger for the development of coxarthrosis. Orthopade 35(1):77–84 17. Martin RL, Kelly BT, Philippon MJ (2006) Evidence of validity for the hip outcome score. Arthroscopy 22(12):1304–1311 18. Mohtadi NGH, Griffin DR, Pedersen ME, Chan D, Safran MR, Parsons N, Sekiya JK et al (2012) The development and validation of a self-administered quality-of-life outcome measure for young, active patients with symptomatic hip disease: the International Hip Outcome Tool (iHOT-33). Arthroscopy 28(5): 595–605
19. Norman GR, Sloan JA, Wyrwich KW (2003) Interpretation of changes in health-related quality of life: the remarkable universality of half a standard deviation. Med Care 41(5):582–592 20. Philippon MJ, Maxwell RB, Johnston TL, Schenker M, Briggs KK (2007) Clinical presentation of femoroacetabular impingement. Knee Surg Sports Traumatol Arthrosc 15(8):1041–1047 21. Philippon MJ, Briggs KK, Yen YM, Kuppersmith DA (2009) Outcomes following hip arthroscopy for femoroacetabular impingement with associated chondrolabral dysfunction: minimum two-year follow-up. J Bone Joint Surg Br 91(1):16–23 22. Philippon MJ, Christensen JC, Wahoff MS (2009) Rehabilitation after arthroscopic repair of intraarticular disorders of the hip in a professional football athlete. J Sport Rehabil 18(1):118–134 23. Philippon MJ, Weiss DR, Kuppersmith DA, Briggs KK, Hay CJ (2010) Arthroscopic labral repair and treatment of femoroacetabular impingement in professional hockey players. Am J Sports Med 38(1):99–104 24. Stalzer S, Wahoff M, Scanlan M (2006) Rehabilitation following hip arthroscopy. Clin Sports Med 25(2):337–357 25. Sta¨helin L, Sta¨helin T, Jolles BM, Herzog RF (2008) Arthroscopic offset restoration in femoroacetabular cam impingement: accuracy and early clinical outcome. Arthroscopy 24(1):51–57 26. Thorborg K, Roos EM, Bartels EM, Petersen J, Ho¨lmich P (2010) Validity, reliability and responsiveness of patient-reported outcome questionnaires when assessing hip and groin disability: a systematic review. Br J Sports Med 44(16):1186–1196 27. Thorborg K, Ho¨lmich P, Christensen R, Petersen J, Roos EM (2011) The Copenhagen Hip and Groin Outcome Score (HAGOS): development and validation according to the COSMIN checklist. Br J Sports Med 45(6):478–491 28. Wahoff M, Ryan M (2011) Rehabilitation after hip femoroacetabular impingement arthroscopy. Clin Sports Med 30(2):463–482