Clinical outcomes and causes of arthroscopic hip revision surgery

Hip arthroscopic surgery has become a common technique during the past decade, leading to an increased number of arthroscopic hip revision surgeries. This study aimed to evaluate the clinical outcomes in a series of revision hip arthroscopies to analyse the causes of reoperation in the short to medium-term follow-up. We retrospectively analysed 22 patients who underwent arthroscopic hip revision surgery in our institute. All patients accepted a detailed physical examination and obtained radiographs to measure the centre edge angle, the alpha angle and the offset. Three-dimensional computed tomography was used to evaluate the deformities. The modified Harris Hip Score, visual analogue scale and patient self-reported satisfaction were collected and analysed preoperatively and postoperatively. Our results demonstrated that the modified Harris Hip Score improved from 52.8 to 81.6, and the visual analogue scale decreased from 5.0 to 1.1. Patient self-reported satisfaction was 8.5. In conclusion, patients who underwent arthroscopic hip revision surgery exhibited significant improvement in patient-centred outcomes in the short- to medium-term follow-up, and the outcomes indicated the effectiveness of revision surgery. Residual femoroacetabular impingement and extra-articular impingement are common reasons for arthroscopic hip revision surgery. Misdiagnosis of osteoid osteoma and relapse of synovial chondromatosis are also important reasons for revision.

Over the past decade, hip arthroscopic surgery has developed rapidly, becoming a common technique. According to one meta-analysis involving 6134 mostly adult patients and a second meta-analysis involving 333 patients, the reoperation rates of arthroscopic hip revision surgery were 5.0% and 6.3%, respectively 1,2 . With the increase in arthroscopic hip revision surgery, the clinical outcomes and causes of this revision surgery have become a major current focus. Sardana et al. 3 performed a systematic review of revision hip arthroscopy and concluded that revision hip arthroscopy was a successful treatment to improve functional outcomes and relieve symptoms, and the main common reasons for revision hip arthroscopy included residual femoroacetabular impingement (FAI), labral tears, and chondral lesions. Newman et al. 4 performed a matched-cohort study to compare outcomes in patients who required arthroscopic hip revision surgery with patients undergoing primary hip arthroscopic surgery and concluded that arthroscopic hip revision surgery provided significant improvement in Hip Outcome Score-Activities of Daily Living subscale (HOS-ADL, increased from 65 to 79), modified Harris Hip Score (mHHS, increased from 62 to 79), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC, improved from 30 to 18), Hip Outcome Score-Sports subscale (HOS-Sports, increased from 40 to 62), and the 12-Item Short Form Health Survey (SF12, increased from 39 to 47). Other studies have proven the effectiveness of arthroscopic hip revision surgery, but studies of reasons for reoperation exhibited different results [5][6][7][8][9] . In addition, all of these studies targeted Caucasian populations. Dudda et al. 10 found that the morphometry of impingement and asphericity were more common in Caucasian compared with Chinese hips in a study of elderly women and Caucasians may be at increased risk of hip osteoarthritis compared with Chinese because of morphologic findings that predispose them to FAI. Chinese people had quite a different lifestyle compared with Caucasians. Morphological differences and different lifestyles may influence the outcomes of revision surgery.

Discussion
The results of this study indicated that arthroscopic hip revision surgery could provide good clinical outcomes for patients who need reoperation. The improvement of mHHS and VAS is statistically significant. Previous report demonstrated that capsulolabral adhesions, failure of labral debridement, and failure to address FAI were the most common reasons precipitating revision hip arthroscopic surgery 11   Stiffness of joint 0 (0) 1 (4.5) Adhesive capsulitis 0 (0) 1 (4.5) Iliopsoas impingement 0 (0) 1 (4.5) Anterior inferior iliac spine/ subspine impingement 0 (0) 2 (9.1) Femoral neck bone cyst 1 (4.5) 1 (4.5)  13 reported the results of a 2-year follow-up of 47 revisions by hip arthroscopy, and FAI that was unaddressed or inadequately addressed was present at the time of revision in 66% of hips. Gupta et al. 12 reported that 41 (59%) and 36 (51.4%) patients presented with persistent cam and pincer abnormalities before revision surgery. Newman et al. 11 reported 21 (50%) patients had cam and pincer impingement, 9 (21.4%) had cam impingement, 3 (7.1%) had pincer impingement, and 9 (21.4%) had painful adhesions preoperatively in the revision group. Aprato et al. 6 evaluated a total of 63 patients who underwent arthroscopic hip revision surgery and reported that chondral lesion associated with labral re-injury was the most common finding at revision surgery, and only 31% of patients underwent revision  for persistent FAI. Underlying residual bone, which was not corrected at the index surgery, is the reason for such a high incidence of labral tears and is also an important reason for reoperation. Compared with the above studies, we find that the main reasons for revision surgery do not differ between Chinese and Caucasian individuals. In addition to residual FAI, extra-articular hip impingement was also an important reason leading to revision surgery. Extra-articular hip impingement can be classified into iliopsoas impingement, anterior inferior iliac spine/subspine impingement, ischiofemoral impingement, and greater trochanteric/pelvic impingement 14 . To date, extra-articular hip impingement has not been well recognized. In this study, one patient was diagnosed with FAI before the index surgery and experienced persistent hip pain after surgery. Iliopsoas impingement was identified before revision surgery of this patient, and symptoms were relieved after iliopsoas release. Two cases of anterior inferior iliac spine/subspine impingement were identified in this study, and symptoms were also relieved after arthroscopic focal subspinal decompression. Ricciardi et al. 7 evaluated 147 patients who underwent revision hip preservation surgery and found that residual or unaddressed intra-and extra-articular impingement were the most common causes.
It is important to note that misdiagnosis of osteoid osteoma is also a reason easily overlooked for revision surgery. Osteoid osteoma of the hip is a relatively rare diagnosis, which can closely mimic symptomatic presentation of FAI, and arthroscopic surgery can be especially effective in patients with concomitant FAI 15 . Osteoid osteoma can be diagnosed by MRI. Osteoid osteoma should be carefully excluded before surgery in case of misdiagnosis and revision surgery. Synovial chondromatosis is also a relatively rare disease involving cartilaginous metaplasia of synovial tissue, and treatment via synovectomy and loose body removal results in high revision rates 16 . Given the high recurrence rate, synovial chondromatosis is also an important reason for revision surgery.
In this study, the mHHS improved from 52.8 preoperatively to 81.6 postoperatively, and VAS improved from 5.0 to 1.1. Domb et al. 13 reported that mHHS improved from 55 to 72 and VAS improved from 7.3 to 3.9. Larson et al. 9 reported that mHHS improved from 62.1 to 79.9 and VAS improved from 4.4 to 3.0. Philippon et al. 5 reported that mHHS improved from an average of 56 to 77 after one-year follow-up. Aprato et al. 6 reported that mHHS improved from 54 to 59 after 3-year follow-up. Ricciardi et al. 7 reported that mHHS improved from 56.6 to 72.1 after 15-month follow-up. Gupta et al. 12 reported that mHHS improved from 57.8 to 73.7 after 2-year follow-up. Newman et al. 11 reported that mHHS improved from 55.3 to 74.3 after a minimum 2-year follow-up. Only one study did not achieve postoperative mHHS equal to the other studies. Aprato et al. 6 reported that the presence of chondral lesions was associated with reduced postoperative mHHS. It is notable that one patient did not fare as well as others with mHHS (52) and VAS (5) after revision surgery. This patient had the lowest mHHS (33) and the   highest VAS (7) and the severest joint degeneration before revision surgery. This finding suggests that we might need to choose patients strictly, and patients with low mHHS, high VAS and severe joint degeneration may have poor operative effect after revision surgery. Compared with the above studies, we believe that the outcomes of revision surgery reveal no differences between Chinese and Caucasian individuals. Given the lack of a similar study among Chinese or Asians, further studies are needed. Preoperative 3-dimensional computed tomography (3D CT) is helpful for identifying pathological areas preoperatively and especially for residual FAI and extra-articular impingement 12 . Here, 3D CT can compensate for the limitation of radiography and the limited vision of arthroscopy to increase accuracy and success rate of surgery and decrease reoperation 17,18 . In addition, 3D CT can improve diagnostic assessment of patients with suspected FAI or residual FAI and is recommended uncertainty is noted 12,19 . We used 3D CT to help identify pathological areas and demonstrate the amount of bony osteoplasty.
In conclusion, patients who underwent arthroscopic hip revision surgery exhibited significant improvement in patient-centred outcomes in the short to medium-term follow-up, and the outcomes indicated the effectiveness of revision surgery. Residual FAI and extra-articular impingement are common reasons for arthroscopic hip revision surgery. In addition, misdiagnosis of osteoid osteoma and relapse of synovial chondromatosis are also important reasons for revision surgery. The inclusion criteria were as follows:(1) the index surgery was arthroscopic hip surgery; (2) hip joint space >2 mm in the affected side and Tönnis grade was I or II; (3) persistent hip pain and limited activity that was aggravated or aggravated after remission with pain mainly in the frontal groin area, and conservative treatment was ineffective; or (4) an obvious cause of their pain was discovered. The following exclusion criteria were employed: (1) acetabular fracture, (2) avascular necrosis, (3) Legg-Calve-Perthes disease, (4) Ehlers-Danlos syndrome, and (5) pigmented villonodular synovitis. surgical procedure. Arthroscopic hip reversion surgery was performed in the supine position with a standard fracture table and custom perineal post. Traction was applied to the operative extremity to ensure that the operative side hip joint space was 8 to 10 mm under fluoroscopic guidance. In addition, the hip joint was fully adducted, and internal rotation was performed. According to Philippon and Sehenker's method, a spinal needle was inserted under fluoroscopic guidance to establish the anterolateral portal. Then, the anterior portal was made under direct visualization by viewing from the anterolateral portal 20 . Arthroscopic knife or radio frequency ablation device was used to complete capsulotomy, and a detailed inspection of the central compartment was performed to assess the acetabular rim, acetabular labrum, articular cartilage and ligamentum teres. Labral repair or labral debridement was performed according to the injury. We tried to suture the damaged acetabular labrum instead of resection because over-resecting the acetabular labrum will destabilize the hip joint. Depending on the size, location, and Outerbridge classification system, cartilage damages were treated with arthroscopic debridement for partial-thickness lesions or arthroscopic microfracture for full-thickness lesions 21 . If a cam bump in the head-neck junction or acetabular overcoverage was identified, femoral osteoplasty or acetabuloplasty was performed. Iliopsoas release was performed in patients with a positive iliopsoas impingement test result. Debride trochanteric bursa and footprint of gluteus medius muscles was performed in patients who had trochanteric tenderness or inflammation in the footprint of gluteus medius muscles as demonstrated by MRI before surgery. Focal subspinal decompression, loose body removal, synovectomy, excision of osteoid osteoma, synovial chondromatosis debridement and bone cystectomy were performed when necessary. postoperative protocol. Begin Ankle pump, quadriceps strengthening and other isometric exercises were initiated 1 or 2 days after surgery. Hip passive range of motion (ROM) exercise as tolerated was initiated 3 or 4 days after surgery. Partial weight bearing with crutches was initiated 3 to 7 days after surgery, and passive ROM exercise and active ROM exercises were performed as tolerated after 4 weeks postoperatively. Patients should advance to full weight bearing by 6 weeks and restore symmetrical hip ROM 6 weeks after surgery. Patients can begin jogging and advance to running 3 to 6 months after surgery. outcome measures. A detailed physical examination was conducted on all hips before surgery during a routine physical exam without anaesthesia. A half-circle bodied metal goniometer with a moveable arm and scale marked in 5° increments was used 22 . Internal and external rotation were measured while the patient was in the supine position with both the hip and knee flexed at 90° or 0°2 2,23 . Abduction and flexion were also recorded 22 . FADIR or FABER tests are considered positive if hip or groin pain was elicited when the hip was placed in 90° of flexion and then adduction and internal rotation or flexion, abduction and external rotation applied 23 . Supine anteroposterior (AP) pelvis, cross-table lateral radiographs and Dunn view (45 degrees) radiographs were obtained on all patients to measure the ACE angle 24,25 , the alpha angle 26,27 , the joint space, and the degree of femoral head-neck offset. The ACE angle was measured from radiographs using the method described by Ömeroglu and Wiberg et al. 28 . The alpha angle was measured from radiographs using the method described by Barton and Notzli et al. 26,27 . The joint space was measured from radiographs using the method described by Reis et al. 29 . The offset was evaluated on cross-table lateral radiographs using the method described by Notzli et al. 27 .
Scientific RepoRts | (2019) 9:1230 | https://doi.org/10.1038/s41598-018-37708-y Three-dimensional computed tomography (3D CT) was used to evaluate the deformities and provide intuitive view. One observer performed all measurements. The mHHS, VAS and patient self-reported satisfaction with surgical outcomes (on a scale of 0 to 10 with 10 being the best) were collected and analysed retrospectively. statistical analysis. The 2-tailed paired t-test was used to evaluate significance between preoperative and postoperative groups. P-values of <0.05 were considered statistically significant. Statistical analysis was performed using the SPSS software (version 11; SPSS Inc.).

Data Availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.