Can intra-articular hip injections predict arthroscopy outcomes for femoroacetabular impingement syndrome? A systematic review

Abstract

Background

Femoroacetabular Impingement (FAI) syndrome represents a prevalent aetiology of hip discomfort observed among both adolescent and adult populations. It is initially managed conservatively with oral anti-inflammatories and physiotherapy; some patients proceed to receiving an intra-articular (IA) hip injection, but ultimately, the gold-standard treatment is hip arthroscopy.

Study design

Systematic Review.

Purpose

To systematically investigate the relationship between response to IA anaesthetic or steroid hip injections and arthroscopy outcomes for FAI syndrome.

Methods

A systematic search of PubMed, Medline, CINAHL, SCOPUS, and Cochrane was conducted in line with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines.

Results

Seven studies (Level II-IV) were identified that met our inclusion and exclusion criteria. These studies collectively included 637 patients, demonstrating an average age of 37.5 years (and a range of 14–72 years). Two of the seven studies reported a statistically significant positive correlation between response to IA injections and arthroscopy outcomes. The remaining five studies found that although a positive IA injection response increased the odds of a good outcome post arthroscopy (defined across various studies as a post-operative modified Harris Hip score of >70 points, >79 points or an improvement by 8 or more points), this correlation was not statistically significant.

Conclusion

IA hip injections can be a useful prognostic tool, though they are not a consistently reliable predictor of which patients will have good arthroscopic outcomes.

1. Introduction

Femoroacetabular impingement (FAI) syndrome is now recognised as a prevalent cause of hip pain amongst athletes, adolescents, and adults, contributing to a significant amount of morbidity in those populations.¹ FAI syndrome is characterised by an incongruent interaction between the head of the femur and the acetabulum, causing friction and repetitive microtrauma that ultimately leads to early onset joint degeneration and pain.^2,3 Current surgical interventions include arthroscopy: targeting specific morphological pathology and underlying soft tissue injuries to achieve pain free range of motion and improved overall function.4, 5, 6

Surgery is often considered after failure of non-operative treatment (i.e., when oral anti-inflammatories or physiotherapy have failed to provide adequate symptom relief or return to baseline activities). Surgery may also be considered if diagnostic imaging and clinical examination suggest that the patient is unlikely to respond to conservative treatment.7, 8, 9

A meta-analysis of three methodologically robust randomized control trials (RCT) found that surgical intervention was superior to conservative management, resulting in an increase in hip outcome scores for activities of daily living subscale (HOS ADL) by 10 points (P < 0.001).¹⁰ Minkara et al. assessed the return to sport after surgical intervention for FAI syndrome and found rates as high as 87.7 % (95 % CI, 82.4%–92.9 %; P < 0.001).¹¹ Sansone et al. showed statistically and clinically significant improvement in all post-operative outcome measures including international Hip Outcome Tool (iHOT-12), EuroQol (EQ-5D) and visual analogue scale (VAS) for global hip function (P < 0.05).¹² Despite such apparently promising results, the data on clinically meaningful postoperative outcomes and patient satisfaction rates are quite variable, with one RCT by Palmer et al. suggesting that only 51 % of patients achieved the minimum clinically important difference (MCID) in outcomes after arthroscopy and 48 % achieved the patient acceptable symptomatic state (PASS) for the Hip Outcome Score.¹³

Intra-articular (IA) hip injections have been studied for use as a predictive tool for hip arthroscopy outcomes.¹⁴ It has been postulated that patients who achieve adequate symptom relief from IA hip injections would hypothetically respond similarly to arthroscopy. Thus, the purpose of this review is to systematically investigate the relationship between response to IA anaesthetic or steroid hip injections and arthroscopy outcomes for FAI syndrome. Our hypothesis is that those who respond positively to IA hip injections are at increased odds of achieving good outcomes post arthroscopy for FAI syndrome.

2. Material and methods

This systematic review was carried out in line with the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) guidelines.¹⁵ The protocol for the review was registered on PROSPERO on May 6, 2022.

A systematic review of Level I-IV studies investigating the correlation between response to IA hip injections and arthroscopy outcomes was performed using the Ovid MEDLINE, PubMED, SCOPUS, CINAHL, and Cochrane databases from inception to August 2023. The following search terms were used: “steroid* or corticosteroid* or intraarticular*or anaesthetic* or local) adj2 inject*” AND “hip adj3 (Arthrit* or arthros* or arthralgia or joint pain* or degenerat* or pain)” OR “femoroacetabular impingement* or femoral impingement or FAI” AND “((arthroscopic or operative) adj2 (management* or treatment* or procedure)”. The full search algorithm can be viewed in Appendix 1.

3. Inclusion and exclusion criteria

The following predetermined inclusion criteria were established and subsequently applied to the selection process of the study: a) diagnosis of FAI syndrome, b) arthroscopic intervention preceded by an IA hip injection, c) patient-reported outcomes after injection and after arthroscopy. Studies were excluded if they were: a) not in English, b) performed on a paediatric population (i.e., <12 years old), c) studies reporting either surgical outcomes alone or IA injection outcomes alone, d) studies that are not on the hip joint, e) studies where the surgical intervention was total hip replacement (given that this is not usually the treatment of choice for FAI syndrome) and f) systematic reviews/meta-analyses/commentaries or book chapters. If any studies were published by the same research group on the same set of patients, the most recent study was included whilst older versions were excluded to avoid duplication of results/over-inflation of true patient numbers.

4. Study selection

One independent reviewer and one librarian devised and performed the literature search. Articles were imported into Covidence systematic review software, Veritas Health Innovation, for screening. Two authors independently conducted the initial screening of titles and abstracts, excluding articles deemed irrelevant to the question. Discrepancies were discussed and resolved before proceeding to the full text screening. Again, the full-text screening was performed by two independent authors and discrepancies were discussed before a consensus was reached on the final set of papers that satisfied both the inclusion and exclusion criteria (Fig. 1). The citations within the included articles were reviewed to identify any potentially pertinent studies that might not have been captured in the initial search.

Fig. 1.

PRISMA flowchart.

5. Data extraction and quality assessment

The authors gathered the following data from the included studies: study design, level of evidence, journal and year of publication, the number of participants and their demographics, content and volume of the IA hip injection used, patient response to the injection (and time at which this was assessed) and response to arthroscopy. Furthermore, data on the number of responders and non-responders to the IA hip injections was collected, and from amongst those two groups, the number of responders and non-responders to subsequent arthroscopy was also extracted (with information on how the various studies defined a responder versus a non-responder).

Due to the level of evidence of included studies (II-IV), as well as inter-study variability and heterogeneity (e.g. variability in the definition of a positive injection or arthroscopy outcome and the time at which that assessment was made), a meta-analysis was deemed inappropriate to perform.¹⁶ Table 1, Table 2, and Fig. 2, present synthesised data on the relationship between IA response and arthroscopy response. The comparator groups were injection responders and injection non-responders. Categorisation of responders and non-responders varied: some studies dichotomised results into ‘pain relieved’ or ‘not relieved’ post injections or set a 50 % improvement in pain as the cut-off for ‘positive’ or ‘negative’ response to the injection. Similarly, studies had different criteria for determining a positive arthroscopic response. Two studies used a post-operative mHHS of >70 as a threshold for a positive outcome^17,18; one study used a mHHS of >79.¹⁹ Two studies^20,21 reported a ‘positive outcome’ when a patient achieved the minimum clinically important difference (MCID) in mHHS score, being determined as an improvement of 8 points or more.^22,23

Table 1.

Study details, patient demographics, injection content and outcomes.

Study, year, and journal.	No. of patients	No. F/M	Mean age	Study design, Level of Evidence	Hip Injection content	Response to injection	Response to arthroscopy
Ayeni et al. 201417	52	30/22	37	Prospective Cohort, Level II	5 mL of 0.25 % bupivacaine and 40 mg of methylprednisolone acetate	42 = positive response (any pain relief)	28/42 positive responder had good operative outcome (mHHS>70)
Knee Surgery, Sports Traumatology, Arthroscopy						10 = negative response (no pain relief)	5/10 negative responders had good operative outcome (mHHS>70)
Chinzei et al. 20207	49	27/22	36.4	Retrospective Cohort, Level III	5 mL of 1 % lignocaine	30 = positive response (>50 % pain relief)	Pre-op mHHS: positive injection responders 65, negative injection responders 55 (p = 0.048)
European Journal of Orthopaedic Surgery & Traumatologie						19 = negative response (<50 % pain relief)	Post op mHHS: positive injection responders 90, negative injection responders 85 (p = 0.026)
Gao et al. 202120	78	41/37	36.5	Case Series, Level IV	2 mL of 2 % lidocaine and 5 mg of betamethasone	50 = positive response (>50 % pain relief)	33 positive responders had arthroscopy and 3 negative responders had arthroscopy (total 36)
Arthroscopy						28 = negative response (<50 % pain relief)	34/36 patients undergoing arthroscopy met the MCID of 8 points and had significant improvement in mHHS and VAS scores.
							Of the 3 negative injection responders, 1 had improvement in mHHS score from 30 to 84. The other 2 patients had improvement in mHHS from 67 to 79 to 72 and 84, respectively.
Krych et al. 201618	97	71/25	37.6	Case Series, Level IV	5 mL of 0.5 % ropivacaine or 0.25 % bupivacaine or 1 % lidocaine ± betamethasone. No statistically significant difference was noted based on injection volume or content.	70 = positive response (pain relief >50 %)	55/69 positive injection responders who had arthroscopy had a good outcome (mmHS>70)
Arthroscopy						26 = negative response (<50 % pain relief)	17/25 negative injection responders who had arthroscopy had a good outcome (>70 mmHS)
Ladd et al. 201619	93	62/31	38.8	Retrospective Cohort, Level III	10–15 mL of a solution containing 5 mL of 1 % lidocaine, 5 mL 0.5 % ropivacaine, 5 mL of sterile saline.	48 = positive responders (61–100 % pain relief)	28/48 positive IA responders who had arthroscopy had a good outcome (mmHS>79)
American Journal of Roentgenology						45 = poor/moderate responders (poor: 0–30 %, moderate 31–60 % pain relief)	24/45 poor or moderate IA responders who had arthroscopy had a good outcome (mmHS>79)
							No positive or negative correlation was found between the response to IA injections and the outcome at 1 year after hip arthroscopy.
Li et al. 202126	60	33/27	38.6	Prospective Cohort, Level II	4 mL of 2 % lidocaine and 4 mL 1 % ropivacaine.	Pre-injection iHOT: 52.1±23.2 Post injection: 84.1±18.1 (p < 0.001)	Significant correlation was detected between iHOT-12 scores after IA injections and one year after arthroscopy (r = 0.784, P < 0.001)
Orthopaedic Surgery
Mujahed et al. 202121	208	99/109	37.9	Retrospective Cohort, Level III	2 mL of 3 mg/mL betamethasoneone and 5 mL of 1 % lidocaine.	120 = positive responder (any improvement in pain)	No statistically significant difference in baseline mHHS scores between injection responders and non-responders. Post-op mHHS for responders: 81.48 ± 7.53 vs 76.6 ± 8.43 for non-responders (p < 0.01)
Orthopaedic Journal of Sports Medicine						88 = negative responders (no improvement in pain)	Percentage of patients achieving MCID in responder vs non-responder group was 94 % vs 89 % (p = 0.24). Percentage achieving PASS in responder vs non-responder group was 85.8 % vs 62.5 % (p < 0.01)

Table 2.

Time of measuring IA response, type of arthroscopy and duration of follow-up.

	Injection outcome measures	Follow-up and post operative scores	Example of types of arthroscopic procedures performed.
Ayeni et al.17	Daily pain diary for two weeks post injection.	mHHS score administered 6 months after arthroscopy.	Femoral head/neck osteochondroplastly, labral debridement, acetabular rim trimming.
Chinzei et al.7	Pain assessment two weeks after injection was given.	mHHS scores administered at 1-year follow up.	No information.
Gao et al.20	Percentage pain relief 10 min after injection.	mHHS and VAS scores, average of 22.5±9.7 months follow-up post operatively (range, 12–38 months).	Labral repair, labral debridement, femoral osteoplasty and acetabuloplasty.
Krych et al.18	Patients were contacted by telephone within 2 weeks of their injection and asked to report the most significant amount of pain relief they experienced within the initial 24 h following the injection.	mHHS and Hip Outcome Score (HOS) 12 months post arthroscopy.	Labrum repair/debridement, femoral osteoplasty and acetabular osteoplasty.
Ladd et al.19	Pain scores using the visual analogue scale before and immediately after the Injection.	mHHS after 1 year from arthroscopy.	Not reported.
Li et al.26	The patient's pain level was assessed before the injection and 30-min after its administration.	iHOT-12, mHHS, and the patient's satisfaction 12 months after arthroscopy.	Labral repair/debridement, acetabular rim trimming, femoral osteochondroplasty, and limited synovectomy were performed.
Mujahed et al.21	Response to the injection was defined as any subjective improvement reported by the patient at first follow-up post injection.	2-year postoperative patient-reported outcomes (mHHS and Hip Outcome Score–Activities of Daily Living).	Labral repair, femoral/acetabular osteoplasty, and capsular closure.

Fig. 2.

The odds of a good arthroscopic outcome post a positive response to the IA injection (full data in appendixTable 1).

Two authors evaluated the methodological quality of the seven included papers using the Cochrane Risk of Bias in Non-Randomized Studies—of Interventions (ROBINS-I) tool.²⁴ Disagreements in scoring were resolved through discussion between the two authors. A third author was not needed to resolve any of the disagreements.

6. Results

A total of 1971 papers were identified using the outlined search strategy across the five databases. Those papers were imported into Covidence whereby 452 duplicates were automatically removed by the software, leaving 1519 articles to screen using title and abstract. One study was excluded for not being in English. Subsequently, 50 articles progressed to full text screening of which a total of 8 studies were eligible for inclusion. One of these studies²⁵ was repeated¹⁸ by the same study group to include a larger set of patients from the same pool by expanding the duration of the study period by an extra year. Only the larger study was included to avoid duplication (Fig. 1).

7. Study, patient, local injection, and arthroscopy characteristics

The included studies comprised of prospective and retrospective cohort studies and case-series. (Table 1). All studies were on patients diagnosed with FAI syndrome through a combination of clinical history, clinical examination, and radiological findings. Gao et al. specifically included patients diagnosed with FAI who had atypical symptoms.²⁰ A total of 637 patients were included across the seven studies, with a mean age of 37.5 years (range 14–72) and 57 % of participants were female. All studies reported the content of the IA injection: volumes varied from 3 mL to 15 mL, composed of isolated agents or combinations of local anaesthetics and steroid agents. (Table 1). Four studies used ultra-sound guidance when injecting the joint^18,20,21,26; two studies used fluoroscopy,^7,19 and one study did not comment on whether any radiological modality was used during hip injections.¹⁷ All surgical interventions were arthroscopic (Table 2): no open surgical procedures were performed, and no total joint replacements were performed. The quality of studies ranged from moderate to serious risk of bias according to the Cochrane ROBINS-I tool (Table 3).

Table 3.

Assessment of methodological quality.

Study	Bias due to confounders	Bias in selection of participants	Bias in classification of interventions	Bias due to deviations from intended interventions	Bias due to missing data	Bias in measuring outcomes	Bias in selection of reported results	Overall study bias
Ayeni et al.	Serious	Low	Low	Low	Low	Moderate	Low	Moderate
Chinzei et al.	Serious	Low	Low	Low	Low	Moderate	Low	Moderate
Gao et al.	Moderate	Low	Low	Serious	Low	Moderate	Low	Serious
Krych et al.	Moderate	Low	Low	Low	Moderate	Moderate	Low	Moderate
Ladd et al.	Moderate	Serious	Low	Low	Moderate	Moderate	Low	Serious
Li et al.	Moderate	Low	Low	Low	Low	Moderate	Low	Moderate
Mujahed et al.	Moderate	Serious	Low	Low	Low	Moderate	Low	Serious

8. Outcome measures and follow-up

The most common post-operative outcome score used was the modified Harris Hip Score (mHHS).²⁷ Two studies reported on the percentage of patients achieving the MCID and PASS^20,21 and one study used the international hip-outcome tool (iHOT-12) in addition to mHHS.²⁶ The follow-up period after arthroscopy ranged from 6 months to 2 years (Table 2). Response to injection was assessed differently across the seven studies: some assessed for response within minutes after the injection,^19,20,26 another in the first 24-h post injection¹⁸; one study assessed pain scores at the two-week mark,⁷ and another required patients to fill in a pain diary daily for the first two weeks after the injection¹⁷; one study assessed for improvement at the ‘first follow-up’ post injection without specifying the exact timing.²¹

9. Predictive value of intra-articular injections

Two of the seven studies reported a statistically significant correlation between a positive response to the IA injection and a positive arthroscopy outcome.^7,26 The study by Li et al. showed a strong positive correlation between iHOT-12 scores after IA injections and 12 months post arthroscopy (r = 0.784, p < 0.001). Chinzei et al. found that injection responders had, on average, higher mHHS scores post-operatively (5 points higher, p = 0.026) compared to injection non-responders, though they also had higher mHHS scores pre-operatively (average of 10 points higher, p = 0.048). Mujahed et al. showed that although there was no statistically significant difference in baseline mHHS scores, the post-operative mHHS for injection responders was 81.46 compared to 76.6 for non-responders (p < 0.01). The remainder of the studies reported better surgical outcomes for positive injection responders overall, however, that data was not statistically significant.17, 18, 19^,21 Krych et al. and Ayeni et al. show that injection responders were 1.15 times more likely to have a positive surgical outcome (mHHS >70), however the 95 % CI was 0.844–1.57 and 0.85–1.56 respectively.^17,18 Gao et al. concluded that a positive response to IA injections predicted greater likelihood of positive arthroscopic outcomes, however, they do not comment on the statistical significance of this finding. Given that their patient demographic was mostly patients with atypical FAI syndrome symptoms, negative injections responders were not recommended to proceed to surgery, instead, an alternative source for their hip pain was searched for. Only three negative IA responders proceeded to receive arthroscopy. As such, a comparison of arthroscopy outcomes between IA responders and non-responders was difficult to ascertain.

10. Methodological quality

The main limitations of the included studies are their small sample size (range 49–208 patients), the presence of confounders that are unaccounted for, lack of randomisation, and lack of blinding. Another key limitation is the disproportionate number of injection responders and non-responders proceeding to surgery, which is ultimately going to affect calculations of sensitivity and specificity and introduce a selection bias. Gao et al. had 66 % of positive injection responders proceed to arthroscopy whereas only 10.7 % of non-responders received arthroscopy.²⁰ The nature and presence of a rehabilitation process, which has potential to impact post operative outcomes, is poorly reported in most studies. Patient baseline function, duration of disease and other comorbidities that may affect response to various treatments was infrequently explored. Five of the seven studies report that arthroscopies were performed by a single surgeon and injections by a single radiologist, thereby limiting inter-operator factors.^7,17, 18, 19^,21

Additionally, given that all primary outcomes were patient-reported, bias in reporting results is almost inevitable given the subjective nature of such outcomes. A summary of the methodological quality assessment for each study is provided in Table 3.

11. Discussion

This systematic review indicates that whilst there may be a positive correlation between a good response to IA hip injections and a positive outcome after arthroscopy, this correlation was not consistently statistically significant. Although some studies have reported that on average, IA responders have better post-arthroscopic outcome measure scores, it's difficult to establish how clinically meaningful those differences are. The diversity in injections used and method of measuring response to them and arthroscopic outcomes is likely to contribute to the varying results reported by the included studies.

Literature on hip arthroscopy suggests that the key to achieving good post-operative outcomes is careful patient selection.^28,29 Various studies have thus proposed that response to IA injections should be part of the surgeon's repertoire of pre-operative data when selecting arthroscopy candidates.^28,30,31 This systematic review's findings suggest that a patient's response to IA injections needs to be cautiously interpreted and not used alone to include or preclude patients from receiving arthroscopy.

In terms of prognostic utility, Ayeni et al. have suggested that a negative response to the IA injection is more useful than a positive one.¹⁷ They postulate that a negative response to the hip injection takes a pre-injection probability of 80 % of having a good arthroscopic outcome down to 69.5 % (negative LR = 0.57). Contrastingly, a positive response to the injection takes the likelihood of a positive outcome to 82 % (positive LR = 1.15).

One potential reason for the discrepancy between response to IA injections and arthroscopy is that both treatment modalities target different aspects of the disease. Arthroscopy aims to target and reverse the structural abnormalities within the hip joint whereas IA injections, especially ones containing a steroid component, work on modulating the inflammatory response within the hip joint. It might be that the presence of FAI syndrome, being the result of genetic, developmental, and activity-related factors, affects cartilage turnover despite correction of the structural defect, resulting in chronic inflammation that makes certain patients more likely to benefit from a local anaesthetic/steroid injection compared to arthroscopy.^17,32,33

Our findings indicate that although IA injections have a role to play in the overall clinical picture and subsequent decision to operate or not, they are not necessarily a conclusive predictor of how a patient will respond to arthroscopy.

12. Strengths and limitations

This review systematically evaluated the available literature on the relationship between response to IA hip injections and post-operative outcomes for FAI syndrome. Although most papers were internally consistent in measuring outcomes and limited proceduralist variability, there was significant inter-study variability. The included studies were all observational in nature (i.e., lacked a control group) and included a relatively small number of patients. Negative injection responders were less likely to proceed to arthroscopy, introducing a selection bias that was not accounted for. Finally, most studies had a follow-up period of 6 months to 2 years, limiting insight into the long-term outcomes after arthroscopy.

13. Conclusion

This systematic review evaluating the relationship between IA hip injections and arthroscopic outcomes for FAI syndrome found that injections were not a consistent, reliable predictor of which patients will achieve satisfactory outcomes post arthroscopy. As such, injections results should be interpreted cautiously and amongst other relevant patient factors before deciding to operate or not.

RE patient consent

Consent was not required for this manuscript as it did not directly obtain any new patient information or data.

Ethics statement

Ethics approval was not required for this manuscript as it is a systematic review of published papers that have sought ethics approval from their respective committees/departments.

Statement of funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Funding and conflict of interest

No funding was received for this project and there are no conflicts of interest to declare.

CRediT authorship contribution statement

Antony Takla: Methodology, Literature search, Data analysis, Writing – original draft. Karin Gunatilake: Literature search, Data analysis, Contributions to writing manuscript. Norine Ma: Data analysis, Writing – review & editing. Ash Moaveni: Conceptualization, Writing – review & editing.

Appendix. Search Strategy

1.1 Ovid Medline Search

#	Searches	Results
1	((steroid* or corticosteroid* or intra-articular* or intraarticular* or intra articular* or articular* or anaesthetic* or anaesthetic* or local or diagnostic) adj3 inject*).mp.	28,431
2	intra articular injections.mp. or Injections, Intra-Articular/	9211
3	Anesthetics, Local/or local anaesthetic.mp.	39,625
4	1 or 2 or 3	64,626
5	(hip adj3 (Arthrit* or arthros* or osteoarthrit* or osteoarthro* or arthralgia or joint pain* or degenerative arthriti* or osteoarthriti* or pain)).mp.	19,173
6	femoroacetabular impingement.mp. or Femoracetabular Impingement/	2917
7	(Femoracetabular Impingement* or femoro-acetabular impingement* or femoro acetabular impingement* or femoral impingement*).mp.	2403
8	5 or 6 or 7	20,297
9	total hip replacement.mp. or Arthroplasty, Replacement, Hip/	35,676
10	(Hip* adj3 (surger* or replac* or arthro*)).mp.	53,227
11	((arthroscopic or operative) adj2 (management* or treatment* or procedure)).mp. [mp = title, abstract, original title, name of substance word, subject heading word, floating sub-heading word, keyword heading word, organism supplementary concept word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier, synonyms]	31,720
12	(operative adj2 (outcome* or success or result*)).mp. [mp = title, abstract, original title, name of substance word, subject heading word, floating sub-heading word, keyword heading word, organism supplementary concept word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier, synonyms]	13,545
13	9 or 10 or 11 or 12	95,755
14	4 and 8 and 13	240

1.2 PubMed Search

#	Searches	Results
1	hip arthrit* OR hip arthros* OR hip osteroarthrit* or hip osteroarthro* OR hip arthralgia OR hip joint pain* OR hip degenerative arthriti* OR hip pain	39,313
2	femoroacetabular impingement* OR femoro-acetabular impingement* OR femoro acetabular impingement* OR femoral impingement*	4568
3	1 or 2	41,027
4	hip replac* OR hip surger* OR hip arthro* OR total hip replacement OR arthroscopic management* OR arthroscopic treatment* OR arthroscopic procedure* OR operative management* OR operative treatment* OR operative procedure* OR operative outcome* OR operative success OR operative result*	3,939,25
5	steroid* injection* OR corticosteroid* injection* OR intra-articular* injection* OR anaesthetic* injection* OR anaesthetic* injection* OR local injection* OR diagnostic injection* OR anaesthetic arthrogram OR anaesthetic arthrogram	390,256
6	3 and 4 and 5	1119
7	3 and 4 and 5 filters: English	1036
8	3 and 4 and 5 filters: English, Humans	880

1.3 SCOPUS Search

#	Searches	Results
1	(steroid* OR corticosteroid* OR intra-articular* OR intraarticular* OR intra AND articular* OR articular* OR anaesthetic* OR anaesthetic* OR local OR diagnostic) W/3 inject*	15,037
2	(anaesthetic OR anaesthetic) W/2 (arthrogram OR arthrography)	39
3	1 or 2	15,068
4	hip W/3 (arthrit* OR arthros* OR osteoarthrit* OR osteoarthro* OR arthralgia OR joint AND pain* OR degenerative AND arthriti* OR osteoarthriti* OR pain)	67,514
5	femoracetabular AND impingement* OR femoro-acetabular AND impingement* OR femoro AND acetabular AND impingement* OR femoral AND impingement	1695
6	4 or 5	67,984
7	total AND hip AND replacement	137,775
8	hip* W/3 (surger* OR replac* OR arthro*)	78,013
9	(arthroscopic OR operative) W/2 (management* OR treatment* OR procedure)	116,448
10	operative W/2 (outcome* OR success OR result*)	29,600
11	7 or 8 or 9 or 10	298,768
12	3 and 6 and 11	685

1.4 CINAHL Search

#	Searches	Results
1	((steroid* or corticosteroid or intra-articular* or intraarticular* or intra articular* or articular* or anaesthetic* or anaesthetic* or local or diagnostic) N3 inject*) OR anaesthetic arthrogram OR intraarticular injection	9873
2	(Hip N3 (Arthrit* or arthros* or osteoarthrit or osteoarthro* or arthralgia or joint pain* or degenerative arthriti* or osteoarthriti* or pain)) OR (femoroacetabular impingement or fai or hip impingement) OR hip arthritis	14,543
3	(Total hip replacement or total hip arthroplasty or hip replacement surgery) OR ((arthroscopic or operative) N2 (management* or treatment* or procedure)) OR (operative N2 (outcome* or success or result*))	29,828
4	1 and 2 and 3	68

1.5 Cochrane Search

#	Searches	Results
1	(steroid* or corticosteroid* or intra-articular or intraarticular* or intra articular* or articular or anaesthetic or anaesthetic* or local or diagnostic) near/3 inject*	11,101
2	(Hip near/3 (Arthrit* or arthros* or osteoarthrit* or osteoarthro* or arthralgia or joint pain* or degenerative arthriti* or osteoarthriti* or pain)) OR (femoro-acetabular impingement*) OR (Femoracetabular Impingement* or femoro-acetabular impingement* or femoro acetabular impingement or femoral impingement)	5572
3	(Hip* near/3 (surger* or replac* or arthro*)) OR (*total hip arthroplasty* OR total hip replacement) OR ((arthroscopic or operative) near/2 (management* or treatment* or procedure)) OR (operative near/2 (outcome* or success or result*)	20,736
4	1 and 2 and 3	100

Appendix Table 1.

number of responders to IA injections and number of responders to arthroscopy

		Positive Surgical Outcome	Negative Surgical Outcome	Definition of positive surgical outcomes.
Ayeni et al. 201417 Knee Surgery, Sports Traumatology, Arthroscopy	Positive LA Response	28	14	mHHS >70 at 6 months post arthroscopy
	N = 42
	Negative LA Response	5	5
	N = 10
Krych et al. 201618Arthroscopyrowhead	Positive LA Response	55	14	mHHS >70 (minimum 1 year follow-up)
	N = 70
	Negative LA Response	17	8
	N = 26
Ladd et al. 201619 American Journal of Roentgenology	Positive LA Response	28	20	mHHS >79 or improvement by 20 points after 1 year of arthroscopy
	N = 48
	Negative LA Response	24 b	21 c
	N = 45 a
Mujahed et al. 202121 Orthopaedic Journal of Sports Medicine	Positive LA Response	113	7	MCID (improvement by > 8 points in mHHS) 2 years after arthroscopy
	N = 120
	Negative LA Response	79	9
	N = 88

Table legend.

^aOut of the 45 negative injection responders, 29 had a ‘poor’ injection response and 16 had a ‘moderate’ injection response’.

^bOf the 24 participants, 17 had a ‘poor’ injection response and 7 had ‘moderate’ injection response.

^cOf the 21 participants, 12 had a ‘poor’ injection response and 9 had a ‘moderate’ injection response.