The Clinical Outcomes of Ultrasound-Guided Hip Joint Injection in The Treatment of Persistent Pain After Hip Arthroscopy

Background: To evaluate the short-term and long-term clinical outcomes of ultrasound-guided hip joint injection in the treatment of persistent pain after hip arthroscopy. Methods: Patients who had persistent pain after hip arthroscopy and underwent ultrasound-guided hip injection for treatment between January 2016 and February 2019 were involved. Pre-injection patient-reported outcomes (PROs) and post-injection PROs 1 month after injection and at nal follow-up were obtained, including visual analog scale (VAS), modied Harris Hip Score (mHHS), Hip Outcome Score – Sport Specic Subscale (HOS - SSS) and Hip Outcome Score - Activity of Daily Living (HOS – ADL). VAS 10 minutes after injection was also recorded. Patient satisfaction with outcome of injection (graded as excellent, good, fair or poor) was documented at the end of follow-up. Results: A total of 33 patients were involved in this study. The mean follow-up time after injection was 24.1 months. The mean time between surgery and injection was 10.7 months. The VAS, mHHS, HOS-SSS and HOS-ADL improved from 5.6 ± 1.4, 56.4 ± 10.7, 59.1 ± 8.5 and 44.2 ± 17.1 to 3.3 ± 2.1, 67.3 ± 12.7, 69.1 ± 14.7 and 57.2 ± 23.0, respectively at 1 month after injection and improved to 2.4 ± 2.2, 76.4 ± 11.7, 80.3 ± 14.3 and 69.6 ± 23.0 at nal follow-up. All results demonstrated statistically signicant difference between different time point (P < 0.05). Three (9.1%) patients thought the outcome of ultrasound-guided hip injection was excellent, 8 (24.2%) thought the outcome was good, 9 (27.3%) thought the outcome was fair, and 13 (39.4%) thought the effect of injection was poor. Conclusion: to intra-articular injection.


Introduction
Over the past decade, hip arthroscopic surgery has developed rapidly and became a common technique. Persistent postoperative pain is becoming an evolving domain and may in uence the postoperative outcomes. [1] There are many studies on early postoperative pain management in hip arthroscopy. [1][2][3] Several methods of pain management have been described, such as femoral nerve block, lumbar plexus block, fascia iliaca block, intra-articular injections, soft tissue surrounding surgical site injection. [3] However, there are few studies on the treatment of persistent postoperative pain. Zhang et al. [4] retrospectively analyzed a total of 36 patients who had hip pain after arthroscopic repair of acetabular labral tears and underwent ultrasound-guided drug injection and concluded that ultrasound-guided drug injection can effectively reduce hip pain, improve hip activity, and promote hip functional reconstruction for patients with postoperative hip pain. Common causes of persistent pain included under-resected femoroacetabular impingement (FAI), residual labral tear, in ammatory reaction, recurrent or missed structural pathology. [5,6] Some patients continue to present with pain and poor functional outcomes and may be candidates for a revision procedure. [7] Ultrasound-guided injection can be a feasible treatment for FAI. [8] Postoperative in ammatory reaction and adhesion may be a reason for persistent postoperative pain, which results in a threshold that blocks the recovery, and ultrasound-guided injection could be useful for such problems. So we hypothesized that ultrasound-guided hip joint injection would be effective in the treatment of persistent pain after hip arthroscopy.
The purpose of this study was to evaluate the short-term and long-term effects of ultrasound-guided hip joint injection in the treatment of persistent pain after hip arthroscopy.

Methods
We evaluated consecutive patients who had persistent pain after hip arthroscopy and underwent ultrasound-guided hip injection for treatment between January 2016 and February 2019. The inclusion criteria were as follows: (1) patients who underwent arthroscopy in our hospital and had persistent pain after surgery, (2) patients had ineffective conservative treatment for more than 3 months, (3) patients underwent ultrasound-guided hip joint injection for treatment of persistent pain. Persistent pain was de ned as unrelieved or new-onset pain at rest, with activity, or with motion in speci c planes after hip arthroscopy with ineffective conservative treatment for more than 3 months. The exclusion criteria were as follows: (1) patients who had other hip surgeries except for the primary hip arthroscopy were excluded from this study, (2) obvious residual FAI, labral tear or other obvious indication of revision arthroscopy identi ed by ultrasound, X-rays, computed tomography (CT) or MRI, (3) bony development problems like hip instability, femoral anteversion and femoral retroversion were also excluded. The Ethics Committee of our hospital approved this study.

Ultrasound examination and ultrasound-guided injection
Ultrasound examination and ultrasound-guided injections were performed by a single radiologist, who specializes in musculoskeletal disorders with more than 10 years of experience. Ultrasonic examination was performed as described by Gao et al [9]. Patients were excluded if residual FAI or labral tear were identi ed by ultrasound, X-rays, CT or MRI. After ultrasound examination, skin preparation was undertaken using povidone iodine solution and then the area was draped. Under real-time ultrasound guidance, a 22-gauge spinal needle was advanced into the hip joint from anterolateral side to superomedial side targeting the anterior surface of the junction of femoral neck and head ( Figure 1).
Once the tip of the needle contacted the cortex within the anterior recess, intra-articular position of the needle was further con rmed by injecting a mixture containing 2 ml of 2% lidocaine and 5 mg of compound betamethasone injection (Diprospan, Schering-Plough) diluted to 10 ml with normal saline. Then a spot image was taken to document location.

Outcome measures
Pre-injection patient-reported outcomes (PROs) and post-injection PROs 1 month after injection and at nal follow-up were obtained, including modi ed Harris Hip Score (mHHS), Hip Outcome Score -Sport Speci c Subscale (HOS -SSS) and Hip Outcome Score -Activity of Daily Living (HOS -ADL). Preinjection visual analog scale (VAS) for pain and post-injection VAS 10 minutes after injection, 1 month after injection and at nal follow-up were recorded. The mHHS of patients before primary surgery was also recorded. Patient satisfaction with nal outcome of injection (satis ed or poor) was documented at the end of follow-up. Patients who reported "excellent" or "good" were divided into "satis ed" group and patients who reported "fair" or "poor" were divided into "unsatis ed" group.

Statistics
The two-tailed paired t test was used to evaluate signi cance between pre-injection and post-injection PROs. Continuous variables with a normal distribution in the baseline data between groups were examined using the independent samples t test. Percentages were compared using the Chi-square test. P values <.05 were considered statistically signi cant. All statistical analyses were performed with SPSS Statistics, version 22 (IBM).

Results
As shown in Table 1, a total of 33 patients (mean age, 34.9 years; age range, 17-53 years; 19 males and 14 females) were involved in this study. The mean body mass index (BMI) was 23.3 (range, 18.9-29.4).
The mean follow-up time after injection was 24.1 months (range, 12-41). The mean time between surgery and injection was 10.7 months (range 3-24). No side effects were observed. One patient underwent revision arthroscopy. Thirty-three (4.4%) in 750 patients who underwent hip arthroscopic surgeries during this period in our hospital had persistent postoperative pain.
As shown in Table 2, arthroscopy con rmed the presence of cam impingement in all patients, pincer impingement in 25 (75.8%) patients, labral tear in all patients, and ischiofemoral Impingement syndrome (IFI) in 1 (3.0%) patients. The surgeries patients underwent were also shown in Table 2. There were 2 (6.1%) patients who had Outerbridge I or II femoral cartilage damages, 5 (15.2%) patients who had Outerbridge II acetabular cartilage damages, 6 (18.2%) patients who had Outerbridge III acetabular cartilage damages and 6 patients (18.2%) who had Outerbridge IV acetabular cartilage damages.
As shown in Figure 2, the VAS improved from a mean of 5.6 ± 1.4 to 2.0 ± 1.7 10 minutes after injection, changed to 3.3 ± 2.1 at 1 month after injection and improved to 2.4 ± 2.2 at nal follow-up. The mean mHHS before primary surgery was 67.4 ± 8.1. The mHHS improved from a mean of 56.4 ± 10.7 to 67.3 ± 12.7 at 1 month after injection and improved to 76.4 ± 11.7 at nal follow-up. The HOS-ADL improved from a mean of 59.1 ± 8.5 to 69.1 ± 14.7 at 1 month after injection and improved to 80.3 ± 14.3 at nal follow-up. The HOS-SSS improved from a mean of 44.2 ± 17.1 to 57.2 ± 23.0 at 1 month after injection and improved to 69.6 ± 23.0 at nal follow-up. All results demonstrated statistically signi cant difference between different time point (P < 0.05).
Three (9.1%) patients thought the outcome of ultrasound-guided hip injection was excellent, 8 (24.2%) thought the outcome was good, 9 (27.3%) thought the outcome was fair, and 13 (39.4%) thought the effect of injection was poor. There were 11 (33.3%) patients in "satis ed" group and 22 (66.7%) patients in " unsatis ed " group. The mean BMI of " satis ed " group was 25.1 and the mean BMI of "unsatis ed" group was 22.0 (P 0.05). The mean age of " satis ed " group was 41.1 and the mean age of "unsatis ed" group was 31.9 (P 0.05). As shown in Table 3, there is no signi cant difference in mHHS before primary surgery, gender, chondral damage and time between surgery and injection between "satis ed" group and " unsatis ed" group. There was no signi cant difference in VAS, mHHS, HOS-ADL and HOS-SSS between " satis ed " group and "unsatis ed" group before injection. There was signi cant difference in VAS, mHHS, HOS-ADL and HOS-SSS between " satis ed " group and "unsatis ed" group 1 month after injection and at nal follow-up ( Figure 3 and Figure 4). The mean VAS 10 minutes after injection in "satis ed" group and "unsatis ed" group was 1±1 and 2.6±1.7, respectively (P 0.05). The mean VAS 10 minutes after injection and 1 month after injection in "satis ed" group had no signi cant difference (P 0.05). The mean VAS 10 minutes after injection and 1 month after injection in "unsatis ed" group increased from 2.6±1.7 to 4.4±1.7 and had signi cant difference (P 0.05).

Discussion
In our study, three (9.1%) patients thought the outcome of ultrasound-guided hip injection was excellent,  Gao et al. [11] evaluated 21 patients who underwent revision arthroscopy and concluded that misdiagnosed extra-articular impingement, osteoid osteoma and synovial chondromatosis can also be the indication of revision arthroscopy. Mansor et al. [12] indicated that over-resection may also be an important reason of revision arthroscopy and actually cause more profound problems than under-resection. In our study, patients with obvious indication of revision arthroscopy were excluded. Bony development problems like hip instability, femoral anteversion and femoral retroversion were also excluded.
Ultrasound-guided hip joint injection is a kind of safe diagnostic and therapeutic method and can also be a feasible treatment for FAI. [8,13,14] Abate et al. [15] evaluated 20 patients with mild FAI and underwent intra-articular ultrasound-guided injection of hyaluronic acid and reported that pain decreased from 6.7 ± 1.3 to 3.7 ± 1.8 and to 1.7 ± 1.8 after 6 and 12 months, respectively and the mean Harris Hip Score improved from 83.3 ± 6 before treatment to 88.2 ± 4.7 at 12 months. Lee et al. [16] evaluated 30 patients with FAI clinically and radiologically and underwent hip injection using steroid or hyaluronic acid and reported that intra-articular hip injection may be effective in FAI, with faster effect of pain improvement by steroid and more delayed effect of function improvement by hyaluronic acid. Zhang et al. [4] reported that ultrasound-guided drug injection can effectively reduce hip pain, improve hip activity, and promote hip functional reconstruction for patients with hip pain after arthroscopic repair of acetabular labral tears.
But patients in that study didn't have diagnosis of FAI. However, there are some studies that reported limited therapeutic effect of intra-articular injection for patients with FAI. [17,18] In our study, the mHHS, HOS-ADL and HOS-SSS improved as time went up after injection. There was signi cant improvement between mHHS before primary surgery and at nal follow-up. This proved the effect of surgery and accurate diagnosis. VAS increased one month after injection and improved at the last follow-up. Although VAS increased one month after injection, it was signi cantly improved compared with that before injection. In this study, there were 22 (66.7%) patients in " unsatis ed " group. The mean VAS before injection and 10 minutes after injection had signi cant difference in " unsatis ed " group. The signi cant improvement after injection showed that the cause of persistent pain should be intra-articular pathology. But the mean VAS 10 minutes after injection and 1 month after injection in "unsatis ed" group increased from 2.6±1.7 to 4.4±1.7 and had signi cant difference (P 0.05). For patients in "unsatis ed" group, injection didn't provide sustained effect. Although they have temporally relief after injection, the injection didn't solve the problem. Previously published short-and midterm studies have indicated the presence and severity of chondral degradation was predictor of worse clinical outcomes. [19] Chondral degeneration, chronic nonspeci c in ammation, muscle strength not recovered, reasons clinical and radiologic evaluation didn't found could be the causes of unrelieved postoperative pain. On the other hand, the mean VAS 10 minutes after injection and VAS 1 month after injection in "satis ed" group had no signi cant difference (P 0.05). This proved that the ultrasound-guided hip injection in "satis ed" group has a sustained good effect for treatment of persistent pain after hip arthroscopy. One of the reasons for the improvement is that there may be chronic nonspeci c in ammatory blocking recovery. Hip injection could be effective for chronic nonspeci c in ammatory. And recovery of muscle strength and function rehabilitation cannot enter the virtuous circle due to chronic pain. The ultrasound-guided hip joint injection may help these patients to enter the virtuous stage of functional recovery for a period of time and improve the clinical outcomes nally. So we think it will help patient to cross the thresholds of recovery and help to delay or avoid revision surgery.
One patient in our study underwent revision arthroscopy 18 months after primary surgery because of unrelieved postoperative pain. Ultrasound examination and MRI before revision surgery didn't identify obvious residual FAI, labral tear or other pathology. However, this patient was diagnosed with labral tear and residual FAI in revision arthroscopy and underwent labral repair, femoral osteoplasty and acetabuloplasty. Inconspicuous labral tear and residual FAI may be misdiagnosed by ultrasound or MRI. Ultrasound-guided hip joint injection can also be a diagnostic tool for persistent postoperative pain.
Patients in "unsatis ed" group may also have labral tear or residual FAI that ultrasound and MRI could not identify. For patients with transient relief but recurring pain after hip injection, ineffectiveness of injection in the long term may be one of the indications of revision surgery.
It should be noted that older patients and patients with higher BMI had better treatment satisfaction in this study. The mean BMI of " satis ed " group was 25.1 and the mean BMI of "unsatis ed" group was 22.0 (P 0.05). Previous studies also reported that obesity may in uence the outcome of hip arthroscopy. [20,21] The mean age of " satis ed " group was 41.1 and the mean age of "unsatis ed" group was 31.9 (P 0.05). The elderly patients and the patients with high BMI may not able to complete postoperative rehabilitation very well, leading to unsatis ed postoperative results. So ultrasound-guided injection may be more effective in these patients. We also found that the mean VAS 10 minutes after injection in "satis ed" group and "unsatis ed" group was 1±1 and 2.6±1.7, respectively (P 0.05). This proved that patients who were sensitive to intra-articular injection may have a better long-term clinical outcome. Ultrasound-guided hip joint injection may better help these patients enter the virtuous circle of rehabilitation.
This study has several limitations. First, this study has a relatively small size, because the number of patients with persistent postoperative pain is relatively small. Second, this is not a randomized control study and a control group is lacking. A placebo effect and improvement due to the passage of time cannot be ruled out. However, the persistent good short-term outcomes in "satis ed" group can prove the good effect of ultrasound-guided hip joint injection, not just the effect of passage of time. Another limitation is that HOS-ADL, HOS-SSS and VAS before primary surgery are not recorded.
In conclusion, ultrasound-guided hip joint injection would be a feasible treatment method of persistent pain after hip arthroscopy, especially in older patients, patients with higher BMI and patients who are sensitive to intra-articular injection.