What is the long-term survivorship, complication profile, and patient reported outcomes after Birmingham hip resurfacing?

Abstract

Introduction

When indicated, Birmingham Hip Resurfacing (BHR) presents a viable alternative to total hip arthroplasty (THA), but there remain questions about the long-term outcomes of BHR. Therefore, we asked: 1) what are the long-term patient-reported outcomes and 2) survivorship rates following BHR; 3) what are the causes for revision surgery after BHR? and 4) how have these outcomes compared to THA at long-term follow-up.

Methods

A query of PubMed, MEDLINE, Scopus, and Cochrane in September 2023 was performed. Articles were included if they reported BHR survivorship rates at ≥10 years. Survivorship was defined as an all-cause revision of any BHR component. This review encompasses 26 articles, totaling 13,103 hips. Mean follow-up ranged from 6.0 to 20.9 years, but each study had at least a subgroup analysis for ≥10-year follow-up.

Results

Five studies compared preoperative and postoperative PROs for BHR, with four reporting improvement in at least one PRO after 10-year follow-up. Overall, survivorship rates ranged from 83% to 100% across diverse long-term timeframes, with 25 of 26 studies reporting 10-year survivorship rates greater than 87%. The primary reasons for revisions were implant loosening (22%), adverse reactions to metal debris (21.2%), and fractures of any kind (17.2%). In the six studies that compared BHR to THA, long-term survivorship was similar while BHR exhibited slightly superior activity levels.

Conclusion

The findings from this study suggested favorable long-term survivorship and postoperative outcomes of BHR. In studies comparing long-term BHR and THA, survivorship was comparable, with BHR potentially providing enhanced postoperative activity levels.

1. Introduction

The Birmingham hip resurfacing arthroplasty (BHR) is utilized to treat arthritis in young patients with good bone stock.1, 2, 3 Proponents of BHR sought to have it become a viable alternative for total hip arthroplasty (THA) in the young and active population.^2,3 The modern era of hip resurfacing started in 1991 and revolved around a metal-on-metal device with hybrid fixation which was favorable in pilot studies. BHR was first available in 1997 and utilized an advantageous design with cast-in beads on the acetabular cups’ outer surface and became available in the United States in 2006.4, 5, 6

BHR is indicated for young active patients with end-stage degenerative hip disease.^5,7,8 BHR has been reported to be contraindicated in patients with femoral head osteonecrosis, structural deformities, and bone stock deficiencies of the femoral neck or head.^5,7 The benefits of BHR are that it provides an accurate restoration of native hip biomechanics, preserves the femoral bone stock, and allegedly has better long-term functional outcomes.^2,9, 10, 11, 12, 13 BHR has also come under criticism due to adverse reactions to metal debris and pseudotumor formation.¹⁴ At short and midterm follow-up BHR has shown promising survivorship with favorable clinical outcomes; however due to its metal-on-metal design and associated pseudotumor risk, the durability of these positive outcomes in the long term is unclear.^15,16 A recent systematic review of 12 articles reporting on BHR outcomes at 10-year follow-up showed satisfactory 96.4% survivorship with a satisfactory complication rate. However, this study did not report specific indications for revisions, included limited details about patient-reported outcome measures, and did not contextualize these results in light of femoral head diameter and other surgical parameters nor compare outcomes to THA.¹⁷ Therefore, a systematic review exploring the long-term (>10 year) outcomes of BHR in light of surgical parameters was conducted to evaluate the long-term functional outcomes and survivorship of BHR compared to THA.

Specifically, we asked: 1) what are the long-term patient-reported outcomes and 2) survivorship rates following BHR. 3) What are the rationales for revision surgery after BHR? 4) How have these long-term outcomes compared to THA in similar timeframes?

2. Methods

2.1. Search strategy

A query of PubMed, MEDLINE, Scopus, and Cochrane electronic databases in September 2023 was performed with the following keywords: (Hip Resurfacing OR Birmingham resurfacing OR BHR) AND (outcomes OR complications OR revision OR patient-reported outcomes OR follow-up) AND (long term or 10 year or ten year) using guidelines from Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) (Table 1). Study protocol was registered in Prospero, CRD42023464567.

Table 1.

Article searches in PubMed, cochrane, and scopus.

Database	Specific Search
PubMed	(Hip Resurfacing OR Birmingham resurfacing OR BHR) AND (outcomes OR complications OR revision OR patient reported outcomes OR follow-up) AND (long term or 10 year or ten year)
Cochrane	1. (Hip Resurfacing): ti,ab,kw OR (Birmingham resurfacing): ti,ab,kw OR (BHR): ti,ab,kw 2. (outcomes): ti,ab,kw OR (complications): ti,ab,kw OR(revision): ti,ab,kw OR(patient reported outcomes): ti,ab,kw OR(follow up): ti,ab,kw 3. (ten year): ti,ab,kw OR (10 year): ti,ab,kw OR (long term): ti,ab,kw 4.{And #1-#3}
Scopus	(TITLE-ABS-KEY (hip AND resurfacing) OR TITLE-ABS-KEY (Birmingham AND resurfacing) OR TITLE-ABS-KEY (BHR)) AND (TITLE-ABS-KEY (outcomes) OR TITLE-ABS-KEY (complications) OR TITLE-ABS-KEY (revision) OR TITLE-ABS-KEY (patient AND reported AND outcomes) OR TITLE-ABS-KEY (follow AND up)) AND (TITLE-ABS-KEY (long AND term) OR TITLE-ABS-KEY (ten AND year) OR TITLE-ABS-KEY (10 year))

Ti; title. Abs; abstract. Ab; abstract. Key; keywords. Kw; keywords.

2.2. Study selection

Authors (SF) and (AKS) used Rayyan software to screen articles independently.¹⁸ Inclusion requirement for the studies were a.) reporting of long-term BHR survivorship rates at a timeframe of 10 years longer b.) full-text articles were written in English. Duplicate studies and non-original research such as biomechanical studies, case reports, opinion articles, review articles, and technique articles were excluded. Initially, the titles and abstracts of all 808 unique articles retrieved were screened to ensure they met the inclusion criteria, followed by a full-text review of the 46 studies that satisfied these requirements. A thorough review of the references section of each included article yielded no additional articles. A total of 26 studies reporting on 13,103 hips were included (Fig. 1). This study was deemed exempt from review by the ethical review board due to all included data being publicly available and lacking any protected health information.

Fig. 1.

PRISMA diagram outlining the study selection process.

2.3. Risk of bias assessment

Studies were independently scored using Methodologic Index for Non-randomized studies (MINORS) tool by two authors (SF) and (AKS).¹⁹ This validated tool scores comparative studies from 0 to 24 among 12 criteria about the rigor of the study design, statistical analyses, outcomes assessed, and adequate follow-up. For each domain, the scores are as follows: 0 for not reported, 1 for reported but insufficient, and 2 for documented and sufficient. As higher scores signify more rigorous research quality. Any scoring disputes were settled with a third independent reviewer (CJH). The mean MINORS score was 13.4 ± 3.0. Level of evidence for each study was documented.

2.4. Data extraction and analysis

For all included studies, two authors independently extracted data (FS, AKS). Only PROs reported in three or more studies were reported. Study author and year, institution, study type, study period, number of surgeons, number of hips, sex, average follow-up, mean age, BMI, surgical approach, cup/head diameter, cup inclination, cup anteversion, preoperative characteristics/surgical indications, Harris Hip Score(HHS), EuroQol 5 Dimension 5 Level(EQ-5D), Oxford Hip Score(OHS), University of California Los Angeles activity scale (UCLA), Western Ontario and McMaster Universities Osteoarthritis Score (WOMAC), survivorship timeframes, survivorship(with number of males and females), total number of revisions, femoral head diameter of revised patients, as well as time until and reasons for revision were all extracted (Table 2). Data was not pooled due to risk for bias from low levels of evidence, and was rather presented as a narrative synthesis.²⁰

Table 2.

Characteristics of studies included in the final analysis.

Study	Institution	Study Type	Study Period	# of Surgeons	Hips(n)	Sex(% Male)	Average Follow-up, years	Age, years	BMI, kg/m2	Averaged MINORS
Hunter et al. 201821	University Hospitals of North Midlands NHS Trust, UK	Case Series	2004–2005	1	121(111 patients)	63.1	Minimum 10	52.5 (53.0 M and 48.8 F)	NR	10.5
Jonas et al. 201913	Avon Orthopaedic Centre, UK	Comparative Study	1991–2001	1	54	75.9	17.6 ± 0.53	49.8 (18–67)	25.7(19.7–35.1)	19
Dhawan et al. 202322	Melbourne Orthopaedic Group	Case Series	1991–2001	3	217	65.4	20.9 (19.3–22.4)	52 (18–68)	27(16.2–45.3)	12.5
Coulter et al. 201223	Melbourne Orthopaedic Group	Case Series	1999–2001	3	213	65.7	10.4 (9.6–11.7)	52.1 (18–82)	27(16.2–45.3)	12.5
Bourget-Murray et al. 202224	University of Calgary, Cumming School of Medicine, Rockyview General Hospital & Alberta Hip and Knee Clinic	Retrospective Cohort Study	2003–2009	1	127	77.6	12.37	49.1 ± 7.3	28.27 ± 5.2	11.5
Treacy et al. 201125	Royal Orthopaedic Hospital, UK	Case Series	1997–1998	1	144	74.3	10.8 (10.2–12.2)	52 (17–76)	NR	12
Mehra et al. 201526	Royal Orthopaedic Hospital, UK	Case Series	1997–2001	1	120	52.5	10.8 (10.0–14.0)	50 (28–63)	NR	11.5
Moroni et al. 201727	Vita-Salute San Raffaele University, Italy	Case Series	2001–2004	1	100(96 patients)	58.3	10.8 ± 1.6	48.9 ± 11.5 (16–73)	NR	12
Hartmann et al. 201228	University Hospital Carl Gustav Carus Dresden, Germany	Case Series	1998–2001	4	100(95 patients)	49.4	10 (9.3–10.5)	52 (28–69)	27.3 (19–41)	11.5
Samuel et al. 202229	CCF	Case Series	2006–2009	1	433(371 patients)	73	11.3	53 {48, 58}	30{26.1, 35}	11.5
Stoney et al. 202030	Australian Orthopaedic Association National Joint Replacement Registry	Retrospective Cohort Study	1999–2018	N/A	4790	100	11.9	52 ± 7.8	NR	18.5
Pietiläinen et al. 202231	Turku University Hospital, Finland	Retrospective Cohort Study	2003–2010	NR	274 (233 patients)	M: 68	14 (0.6–17)	53 (18–76)	NR	15
Uemura et al. 201732	Osaka University, Japan	Case Series	1998–2007	2	136	M: 47.3	12 (2–18)	53 (19–85)	NR	11
Van Der Straeten et al. 201333	Ghent University Hospital, Ghent, Belgium	Case Series	1998–2001	1	250(238 patients)	M: 70	9.7	50.6 (17–76)	NR	11
Reito et al. 201434	Coxa Hospital for Joint Replacement	Case Series	2001–2004	4	261	67.8 M	10.4 ± 0.8 (8.7–12.4)	53.7 M, 53.6 F	NR	10.5
Daniel et al. 201435	The McMinn Centre, Birmingham, UK	Case Series	1997–2000	1, designer	1000	66.5	13.7 (12.3–15.3)	53 (15–84)	NR	13
Su et al. 202136	HSS, Anderson Orthopedic Institute	Case Series	2006–2009	5	280	73.6	9.0 ± 2.5	51.3 ± 7.1 (22–72)	27.8 ± 4.4(19.4–41.6)	12.5
Azam et al. 201637	Malabar Orthopaedic Institute	Case Series	1999–2004	1	244	68.9	12.1	58.28 ± 10.06 M, 55.06 ± 10.43 F	NR	11
Scholes et al. 201938	joint orthopaedic Centre, Sydney, Australia	Case Series	1999–2006	1	238	79.8	12 {10,13}	45 {41, 47.3}	NR	13.5
Matharu et al. 201339	royal orthopaedic Hospital, Birmingham, UK	Case Series	1997–2006	1	447	59.7	10.1 (5.2–14.7)	41.5 (14.9–49.9)	NR	13
Holland et al. 201240	Freeman Hospital, Newcastle-Upon-Tyne	Case Series	1998 – NR	1	100	74	11.5 (10–13)	51.3 (21–68)	NR	17.5
Murray et al. 201241	Nuffield Orthopaedic Centre, Oxford, United Kingdom	Case Series	1999–2009	4	646	41.3	8	51.9 (16.5–81.5)	NR	11
Haddad et al. 201542	University College London Hospitals	Prospective Comparative Study	1999–2002	1	49	NR	12.1 (10–14)	NR	NR	21.5
Ford et al. 201843	Barnes-Jewish Hospital	Matched Cohort Analysis	2006–2011	3	324	86.4	7.2	52.0 ± 7.5 (22.8–81.6)	27.0 ± 3.5(17.4–40.0)	19
Pailhe et al. 201444	The Royal Orthopaedic Hospital, Birmingham, UK	Case Series	1997–2012	1, designer	180	61.7	6.0 (1.0–14.4)	69.2 (65.0–82.7)	NR	13
Seppänen et al. 201645	Department of Orthopaedics and Traumatology, Turku University Hospital, Turku	Retrospective Cohort Study	2001–2013	N/A	2141	NR	NR	NR	NR	12.5

Values reported as value(%) or average ± SD (range) or [CI]. {}, interquartile range. Underline denotes median. M, male. F, female. NR, not reported. N/A, not applicable.

2.5. Study characteristics

A total of 13,103 hips (8760 males and 2013 females were recorded) were included across the 26 studies.^13,21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45 Mean follow-up ranged from 6.0 to 20.9 years.^22,44 Mean patient age ranged from 41.5 to 69.2 years.^39,44 BMI which was reported in ten studies ranged from 25.7 to 30.^13,29 A total 24 studies reported at least one surgical characteristic(approach, surgical/implant characteristics, preoperative characteristics or surgical indications). A total of 21 studies documented patient-reported outcomes.^13,21, 22, 23, 24, 25, 26, 27, 28, 29^,33, 34, 35, 36^,38, 39, 40, 41, 42, 43, 44 All 26 studies reported survivorship rates and 24 reported cause of revision^13,21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45 Six studies compared to outcomes of THA to that of BHR.^{13,30,31,42,43,45} The most common surgical approach was a standard posterior approach, followed by modified or posterolateral approaches(Table 3). Femoral head component diameters ranged from 48.2 mm–50.8 mm^26,29 Mean cup inclination ranged from 42 to 49° while mean cup anteversion ranged from 11 to 23°.^34,40,43

Table 3.

Surgical approach, surgical and implant characteristics of included studies.

Study	Approach	Cup/Head Diameter	Cup Inclination (Degrees)	Cup Anteversion	Preoperative Characteristics/Surgical Indications n(%)
Hunter et al. 201821	Posterior	Femoral:	NR	NR	NR
		M: 52 mm
		F: 46 mm
Jonas et al. 201913	Posterior	NR	NR	NR	NR
Dhawan et al. 202322	Posterior	Acetabular: 48 mm to 64 mm	NR	NR	NR
		Femoral: 6 mm–8 mm less than acetabular
Coulter et al. 201223	Posterior	Acetabular: 48 mm to 64 mm	NR	NR	NR
		Femoral: 6 mm–8 mm less than acetabular
Bourget-Murray et al. 202224	Posterior	Acetabular: 48 mm to 64 mm	M: 45.5 ± 6.0 (34.6–57.2)	15–20	NR
		Femoral: 6 mm–8 mm less than acetabular	F: 44.6 ± 5.9 (29.0–58.9)
Treacy et al. 201125	Posterior	Mean Femoral: 49.3 mm	NR	NR	NR
		NR: 8 (6%)
Mehra et al. 201526	Posterior	Femoral: 48.2 mm	45	NR	NR
Moroni et al. 201727	Posterior	Femoral: 49.5 mm	NR	NR	Osteoarthritis: 66(66)
					Congenital dysplasia: 19(19)
					Avascular necrosis: 6(6)
					Acetabular fracture: 4(4)
					Rheumatoid arthritis: 2(2)
					Perthes: 2(2)
					Slipped capital femoral epiphysis: 1(1)
Hartmann et al. 201228	Posterior	Femoral: 50 mm (42–58)	NR	NR	Osteoarthritis: 26(26)
					Developmental dysplasia: 65(65)
					Avascular necrosis: 6(6)
					Tilt deformity: 3(3)
Samuel et al. 202229	Anterolateral	Femoral: 50.8 mm	NR	NR	Osteoarthritis: 382(88.2)
					Rheumatoid arthritis: 1(0.2)
					Juvenile rheumatoid arthritis: 4(0.9)
					Slipped capital femoral epiphysis: 10(2.3)
					Developmental dysplasia: 24(5.5)
					Avascular necrosis: 12(2.8)
Stoney et al. 202030	NR	Femoral:	NR	NR	NR
		All ≥50 mm
Pietiläinen et al. 202231	NR	NR	NR	NR	NR
Uemura et al. 201732	Posterolateral	NR	NR	NR	Osteonecrosis: 31(24)
					Osteoarthritis secondary to developmental dysplasia: 67(52)
					Osteoarthritis without developmental dysplasia: 29(22)
					Rheumatoid arthritis: 3(2)
Van Der Straeten et al. 201333	NR	NR	NR	NR	Osteoarthritis: 202(80.8)
					Avascular necrosis: 23(9.2)
					Congenital dysplasia: 11(4.4)
					Rheumatoid arthritis: 10(4)
					Trauma: 2(0.8)
					Neurometabolic: 2(0.8)
Reito et al. 201434	modified posterior approach	Femoral: M: 54 mm46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58 F: 50 mm42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54	49 ± 6.7	21 ± 6.6	NR
Daniel et al. 201435	posterior approach	NR	43 ± 5.1	20	Osteoarthritis: 763(76%)
					Osteonecrosis: 40(4%),
					Hip dysplasia: 103(10%), Destructive arthritis: 37(4%)
					Inflammatory arthritis: 18(2%)
					Slipped epiphysis and/or Perthes' disease: 25(3%)
					Post-traumatic: 14(1%)
Su et al. 202136	posterolateral 251(89.6%), direct anterior 29(10.4%)	Femoral: 49.1 mm	NR	NR	Osteoarthritis: 265(94.6%)
					Developmental dysplasia: 8(2.9%)
					Osteonecrosis: 5(1.8%)
					Traumatic arthritis: 2(0.7%)
Azam et al. 201637	standard poster-lateral approach in the lateral position	NR	NR	NR	Osteoarthritis
Scholes et al. 201938	posterior approach	NR	NR	NR	Osteoarthritis
Matharu et al. 201339	posterior approach	Femoral: 49.0 mm	47.1(31–67)	NR	Osteoarthritis: 304(68%) Developmental Dysplasia: 46(10.3%)
					Avascular necrosis: 41(9.2%)
					Inflammatory arthritis: 21(4.7%)
					Slipped upper femoral epiphysis: 13(2.9%)
					other causes: 22(4.9%)
Holland et al. 201240	Standard posterior approach	Femoral mm: 50 mm(42–58)	46.2 (34–59)	11.0(0–30)	Osteoarthritis: 79(79%)
					slipped upper femoral epiphysis: 3(3%), post-traumatic OA: 2(2%)
					Osteoarthritis with cysts: 2(2%), sepsis: 2(2%)
					Ankylosing spondylitis: 1(1%)
					Dysplasia: 3(3%)
					Osteo-arthritic dysplasia: 1(1%)
					Osteo-arthritic protrusion: 1(1%)
					Human Immunodeficiency Virus Osteoarthritis: 1(1%)
Murray et al. 201241	extended posterior approach	Femoral (n):	NR	NR	Osteoarthritis: 526(81%)
		<46 mm: 82
		46 mm: 197			Osteoarthritis secondary to dysplasia: 85(13%)
		50 mm: 184			Avascular necrosis: 21(3%)
		>50 mm:183			Other: 14(2%)
Haddad et al. 201542	posterior approach	Femoral:	NR	20(16–22)	Osteoarthritis
		48.6 mm
Ford et al. 201843	extensile posterior approach	Femoral: 50.1 ± 3.3 mm(42–58)	42.0 ± 5.2(25–60)	23.0 ± 7.2(-5-52)	Osteoarthritis: 304(93.8%) Rheumatoid arthritis: 1(0.3%)
					Head destruction: 3(0.9%)
		Acetabular: 56.8 ± 3.1 mm(48–66)			Osteonecrosis: 10(3.1%)
					Posttraumatic arthritis: 6(1.9%)
Pailhe et al. 201444	posterior approach	Mean Femoral: 49.7 mm	For n = 105(59%): 43(28.4–60.2)	NR	Osteoarthritis: 180(100%)
Seppänen et al. 201645	NR	NR	NR	NR	NR

Values reported as value(%) or average ± SD (range) or [CI]. {}, interquartile range. Underline denotes median. M, male. F, female. NR, not reported. N/A, not applicable.

3. Results

3.1. Long-term BHR patient-reported outcomes

HHS, OHS and UCLA were reported in the majority of studies, with a total of 11 studies each(Table 4).^13,21, 22, 23, 24, 25, 26, 27, 28, 29^,33, 34, 35, 36^,39, 40, 41, 42, 43, 44 EQ-5D and WOMAC were reported in four studies each.^{13,22,23,36,38,40,42} Five studies compared preoperative and postoperative PROs, two reported statistically significant improvement with OHS, two with UCLA, one with WOMAC, one with HHS, and one reported non-significant improvement with HHS.^27,28,38,43

Table 4.

Patient-reported outcomes.

Study	Hips(n)	Follow-up, years	PROS	HHS		EQ-5D		OHS		UCLA		WOMAC
Hunter et al., 201821	121	10	OHS	NR		NR		p-value <0.01//difference from preoperative to postoperative & no difference from postoperative to 10-year scores		NR		NR
Jonas et al. 201913	35	18	UCLA, OHS, EQ-5D	NR		Postoperative	73/100 {30–70}	Postoperative	13 {12–18}	Postoperative	8 {6–10}	NR
Dhawan et al. 202322	108	Minimum 10	OHS, EQ-5D	NR		Postoperative	82.6(50–100)	82.6 (50–100)	4529, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48	NR		NR
Coulter et al. 201223	197(184 patients)	10.4	OHS, HOOS WOMAC	NR		NR		4528, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48		NR		Postoperative HOOS WOMAC
												Pain: 0.8(0–8)
												Motion: 0.6 (0–4)
												Function: 4.0 (0–30)
												Total: 5.4 (0–41)
Bourget-Murray et al. 202224	127(107 patients)	12.4 ± 1.4	HHS, UCLA	Postoperative	M: 93.9	NR		NR		Postoperative	M: 8.2	NR
					F: 93.6						F: 7.2
Treacy et al. 201125	124(111 patients)	Minimum 10	OHS, UCLA	NR		NR		4.2 {0–19}		7 {5–8}		NR
Mehra et al. 201526	98(83 patients)	10.810, 11, 12, 13, 14	HHS	Preoperative	44.5(28–60)	NR		NR		NR		NR
				Postoperative	84.0 (19–100)
Moroni et al. 201727	100	Minimum 10	OHS, HHS, UCLA	Preoperative	58.0 ± 14.1 (25.0–88.0)	NR		Preoperative	29.1 ± (6.2 9.0–39.0)	Preoperative	4.9 ± 1.2 (2.0–7.0)	NR
				Postoperative	98.6 ± 2.3 (90.0–100.0)			Postoperative	47.4 ± 1.3 (42.0–48.0)	Postoperative	8.2 ± 1.1 (6.0–10.0)
				p-value	P < 0.001			p-value	P < 0.001	p-value	P < 0.001
Hartmann et al. 201228	80	10 (9.3–10.5)	HHS, UCLA	Preoperative	50.5 (18–78)	NR		NR		Postoperative	6 (5–7)	NR
				Postoperative	94 (48–100)
				p-value	P = 0.063
Samuel et al. 202229	389(350)	11.3	HHS	Postoperative	84.0{80–86.0}	NR		NR		NR		NR
Van Der Straeten et al. 201333	190	19.8(10.0–13.6)	HHS	Postoperative	97.7(65–100)	NR		NR		NR		NR
Reito et al. 201434	223	10.4 ± 0.8(8.7–12.4)	HHS	Preoperative	56(20–86)	NR		NR		NR		NR
				Postoperative	100(52–100)
Daniel et al. 201435	325	Minimum 10 years	HHS, OHS, UCLA	Postoperative	85 ± 9.8	NR		Postoperative	15 ± 5.3	Postoperative	7.8 ± 1.1	NR
Su et al. 202136	Variable	10/pre-revision	HHS, EQ-5D	Preoperative(n = 280)	59{49–65}	Preoperative(n = 278)	74{60-85	NR		NR		NR
				Postoperative(n = 205)	99{96–100}	Postoperative(n = 215)	90{85–95}
Scholes et al. 201938	238	12	WOMAC	NR		NR		NR		NR		WOMAC	Preoperative (n = 117)			Postoperative (n = 130)		p-value
												Pain	9.5[9–10.5]			1[0.5–1.5]		<0.001
												Motion	4.5[4.0–4.5]			1[1–1.4]		<0.001
												Function	34[32–36.5]			5[3.5–6.5]		<0.001
Matharu et al. 201339	447	10.1(5.2–14.7)	UCLA, OHS	NR		NR		Preoperative	60%{46–73}	Postoperative	6.0{5,8} men: 7 women: 5	NR
									48-point OHS: 19.2{13.0–25.9}
								Postoperative	4.2%{0–24}
									48-point OHS: 46.0{36.5–48.0}
Holland et al. 201240	100	10	WOMAC, HHS, UCLA	Preoperative	48.6(10–92)	NR		NR		Postoperative	7(2–10)	WOMAC		Preoperative			Postoperative
				Postoperative	96.0(26–100)							Pain		17.7(0–25)			5.0(5–20_
												Stiffness		8.4 (4–56)			2.0(2–10),
												Physical function		59.2(6–85)			19.00(17–76)
Murray et al. 201241	540	Minimum 10	OHS, UCLA	NR		NR		Postoperative	43.0 ± 8	Postoperative	6.4 ± 2	NR
Haddad et al. 201542	49	10	EQ-5D, HHS, OHS, UCLA, WOMAC	Preoperative	53.9 ± 13.9	Preoperative	0.32 ± 0.32	Preoperative	19.1 ± 7.8	Preoperative	5 ± 2	Preoperative			46.7 ± 19.2
				Postoperative	97.1 ± 5.1	Postoperative	0.85 ± 0.05	Postoperative	40.1 ± 0.4	Postoperative	8.12 ± 1	Postoperative			3.24 ± 15
Ford et al. 201843	283	Latest Follow-up Minimum 10	UCLA	NR		NR	NR	Preoperative	7.0 ± 2.5(2–10)	NR
								Postoperative	8.0 ± 2.0(2–10)
								p-value	<0.001
Pailhe et al. 201444	180	Minimum 10	OHS	NR		NR	Preoperative	50.0%{37.5–68.8}	NR	NR
							Postoperative	4.4%{0–10.4}

EQ-5D, EuroQol 5 Dimension 5 Level. OHS, Oxford Hip Score. UCLA, University of California Los Angeles activity scale. WOMAC, Western Ontario and McMaster Universities Osteoarthritis. HHS, Harris Hip Score. HOOS WOMAC, elements of HOOS drawn from the Western Ontario and McMaster Universities Osteoarthritis Index.

3.2. Long-term BHR survivorship

Overall, survivorship rates ranged from 83 to 100% across ≥10 year timeframes ranging from ten to 20 years, with 25 of 26 studies reporting 10-year survivorship rates greater than 87%(Table 5).^13,21, 22, 23, 24, 25, 26, 27, 28, 29, 30^,32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45 Six studies reported survivorship at timeframes of 15 years or longer, which ranged from 83 to 96.8%.^13,22,30, 31, 32^,35,38 Four studies observed survivorship in cohorts with greater than 500 hips with long-term survivorship rates ranging from 87.1 to 95.8%.^30,35,41,45 Further, in the five studies where mean or median femoral head diameter was greater than or equal to 50 mm, survivorship ranged from 88 to 96.9%.28, 29, 30^,34,40 In studies where sex-specific survivorship rates were recorded long-term male survivorship ranged from 93 to 99%, whereas female survivorship ranged from 73.9 to 91.9%. Three studies reported long-term survivorship rates male cohorts with greater than 50 mm femoral head diameter, the survivorship rates for these studies were 97, 93, 93.3% respectively.^21,28, 29, 30^,34,41,44

Table 5.

Survivorship and revision characteristics.

Study	Hips(n)	Survivorship Time Frame, years	Survivorship(%)	Number of Revisions, n (%)	Femoral Head Diameter of Revised Patients, total, M and F	Time to Revision	Cause of Revision n (%)
Hunter et al. 201821	121(111 patients)	10	Overall: 91 M: 97 F: 80	11 (9)	<46 mm(64% of failures, not statistically significant)	NR	1(11.1) - femoral neck fracture 1(11.1) - infection'9(81.2) - pain and/or loosening
Jonas et al. 201913	54	18	88[75.9–94.6]	6 (11.1)	NR	NR	Femoral component failure
Dhawan et al. 202322	217	20	Overall: 91.9 (19) M: 135 (95.1) F: 63 (85.1)	Total: 19(8.1) M:7 F:12	All 12 F revisions were <46 mm, all 3 M revisions were ≤50 mm	12.3 (4.9–20.8) F: 10.2 years (5–17.8) M: 15.8 years (10–21)	6(31.6) – unexplained groin pain 6(31.6) - adverse reaction to metal debris 2(10.5) – avascular necrosis 2(10.5) – elevated Co/Cr 1(5.3) - femoral neck fracture 1(5.3) - loosening 1(5.3) - instability
Coulter et al., 2012	213	10	Overall: 94.5 [90.1–96.9] M: 97.5[92.4–99.2] F: 89.1[79.2–94.4]	Total: 11 M: 3 F: 8 10 Revised to THA	femoral head component head size was ≤46 mm in 9/11 revisions	NR	2(18.2) – avascular necrosis 2(18.2) - loosening 1(9.1) - osteolysis 2(18.2) - unexplained pain 1(9) - elevated Co/Cr 1(9.1) – spinal tumor 2(18.2) – adverse reaction to metal debris
Bourget-Murray et al. 202224	127	13	Overall: 91.9 [86.9–97.2] M: 93.8[87.9–100] F: 87[77.8–97.3]	Total: 11 M: 5 F: 6	Both: 49.1 mm, M: 54.4 mm, F: 44.7 mm	M: 98.6 ± 47 months (8.2 yr; range 48–160 mo) F: 92.8 ± 19 months (7.7 yr; range 62–110 mo)	2(18.2) - pseudotumor 3(27.2) - loosening 1(9.1) – avascular necrosis 1(9.1) - acetabular impingement 1(9.1) - elevated Co/Cr 1(9.1) – periprosthetic fracture 1(9.1) - ongoing symptomatology
Treacy et al. 201125	144	10.9 (10.2–12.2)	93.5[89.2–97.6]	10//all to THA	NR	NR	3(30) – avascular necrosis 1(10) – femoral neck fracture 3(30) - deep infection 1(10) – developmental dysplasia of the hip 2(20) - dislocation
Mehra et al. 201526	120	10	94.2[88.8–98.7]	Total: 6 M: 2 F: 4	both: 44.7 mm, M: 48 mm, F: 43 mm	1.9–9.1 years	4(66.7) - loosening 1(16.7) – femoral neck fracture 1(16.7) - heterotopic ossification
Moroni et al. 201727	100(96 patients)	10	96	Total: 4 M: 0 F: 4	45 mm	5–88.6 months	2(50) – femoral neck fracture 2(50) - varus femoral shift
Hartmann et al. 201228	100(95 patients)	10	Overall: 88[77.7–93.1] M: 93[80.3–98.9] F: 84[65.9–91.6]	Total: 10 M: 3 F: 7	NR	NR	2(20) - infection 1(10) – femoral neck fracture 3(30) - pseudotumor 1(10) - psoas impingement 3(30) - loosening
Samuel et al. 202229	433 (371 patients)	10	Overall: 96.9[94.9–98.6] M: 99.0[97.8–100] F: 90.9[84.2 to 96.4]	12	both: 47.5 mm, M: 50 mm, F: 46.7 mm	6.4 ± 3.1 years	1(8.3) – femoral head collapse, pain 1(8.3) - femoral neck fracture 4(33.3) – Elevated Co/Cr levels, pain 1(8.3) - femoral component subsidence 1(8.3) - loosening 3(25) - cup malposition 1(8.3) – acetabular fracture
Stoney et al. 202030	4790	17	93.30	215	NR	NR	60(27.9) - loosening 45(20.9) - fracture 32(14.8) - metal-related pathology 21(9.7) - lysis 17(7.9) - infection 12(5.6) - pain 10(4.6) – osteonecrosis 5(2.3) - malposition 13(6) - other
Pietiläinen et al. 202231	274 (233 patients)	16	83	40	NR	NR	10(25) – adverse reaction to metal debris 7(17.5) – periprosthetic fracture 12(30) - loosening 4(10) - mechanical impingement 2(5) - infection 2(5) - malalignment 1(2.5) - pain 1(2.5) - elevated Co/Cr 1(2.5) – limb length discrepancy
Uemura et al. 201732	136	10, 15	Overall: 10-year: 96.5 [90.9–98.7] 15-year: 93.6[83.4–97.7]	Total: 6 M: 2 F: 4	both: 46.7 mm, M: 50 mm, F: 45 mm	NR	3(50) - Loosening 2(33.3) - Infection 1(16.7) - Neck Fracture
Van Der Straeten et al. 201333	250(238 patients)	13.2	92.4[90.8–94.0]	10	both: 46.4 mm, M: 56 mm, F: 44 mm	6.8 (0.1–13.4)	1(10) - Fracture 1(10) - Metal sensitivity 2(20) - Loosening 1(10) - Malpositioning 2(20) - Impingement 2(20) - Osteolysis 1(10) - Pain
Reito et al. 201434	261	10	91	Total: 23(8.8) M: 10 F: 13	NR	NR	11(47.8) - Adverse reaction to metal debris 4(17.4) – Avascular Necrosis 4(17.4) – Femoral neck fracture 1(4.3) – infection 1(4.3) – impingement 1(4.3) - pain
Daniel et al. 201435	1000	10, 15	10-year: 97.4 15-year: 95.8	Total: 38(3.8) M: 13 F: 25	NR	8.7 years	4(10.5) – femoral neck fracture 13(34.2) – femoral head collapse 2(5.3) – cup loosening 3(7.9) – unexplained groin pain 7(18.4) – Adverse reaction to metal debris 7(18.4) - infection
Su et al. 202136	280	10	92.9[89.8–96.1]	Total: 20(7.1) M: 8 F: 12	Total: 46.8, F 43.5 M 51.75 F	5.4 ± 2.9 years	5(25) – femoral loosening 3(15) – femoral neck fracture 3(15) – pseudotumor 2(10) – osteolysis 1(5) – acetabular loosening 6(30) – combination of pain, noise, or metal levels
Azam et al. 201637	244	Average Follow-up Time ≥12.05	Overall: 93.7 M: 95.43 F: 89.96	16(6.5)	Total: 45.9 mm, F 42.7 mm, M 50 mm	NR	8(50) – aseptic loosening and varus collapse of femoral component 5(31.3) – metal allergy 2(12.5) – femoral neck stress fractures 1(6.3) – acetabular implant loosening
Scholes et al. 201938	238	15	96.8	6 M: 2 F: 4	NR	NR	2(33.3) – loosening 1(16.7) – lysis 1(16.7) – malposition 1(16.7) – fracture 1(16.7) - osteonecrosis
Matharu et al. 201339	447	10, 14	10-year: 96.3[93.7–98.3] 14-year: 94.1[84.9–97.3]	16(3.6%) M: 5 F: 11	46 total, F 44.2, M 50 mm	6.2 years(0.5–11.8)	1(6.3) – aseptic acetabular loosening 3(18.8) - deep infection 1(6.25) – femoral neck fracture 3(18.8) – undiagnosed pain 6(37.5) – aseptic femoral loosening 1(6.3) – avascular necrosis 1(6.3) – implant fracture
Holland et al. 201240	100	10	Overall: 92[86.7–97.3] M: 94.6[89.4–100] F: 84.6[70.7–98.5]	8	NR	NR	6(75) – femoral component failure 2(25) – adverse reaction to metal debris
Murray et al. 201241	646	10	Overall: 87.1[83–91.2] M: 94.7[92.0–97.4] F: 73.9[63.6–84.2]	54(8.4)	NR	NR	26(48.1) – pseudotumor 12(22.2) – fracture or avascular necrosis 8(14.8) – painful impingement 6(11.1) – loose component 1(1.9) – infection 1(1.9) – recurrent dislocation
Haddad et al. 201542	49	10	100	0	NR	NR	NR
Ford et al. 201843	324	10	93.8[88.8–96.7]	14(4.3) M: 9 F: 5	47.9 mm, M 49.8 mm, F 44.4 mm	4.1 years	1(7.14) – femoral neck fracture 2(14.3) – undiagnosed pain 2(14.3) – fibrous ingrowth of cup 4(28.6) – adverse local tissue reaction and/or pseudotumor 1(7.1) – persistent dislocation 1(7.1) - femoral component loosening 3(21.4) - infection
Pailhe et al. 201444	180	10	Overall: 96.4[90.3–100] M: 98.9[94.8–100 F: 91.9[77.0–100]	3(1.7) F: 2 M: 1	total: 46 mm, M: 50 mm, F: 44 mm	4.2 years	1(33) – deep infection 1(33) – subtrochanteric fracture(high-energy) 1(33) – adverse reaction to metal debris
Seppänen et al. 201645	2141	10	91[89–92]	NR	NR	NR	NR

Values reported as value(%) or average ± SD (range) or [CI]. {}, interquartile range. Underline denotes median. M, male. F, female. NR, not reported. N/A, not applicable.

3.3. BHR revision characteristics

There were 569 revisions recorded in total(Table 5).^13,21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41^,43,44 The primary reasons for revisions reported upon pooling the data were implant loosening which accounted for 22.0% (n = 125), metal-related pathology or pseudotumor which accounted for 21.2% (n = 121), fractures of any kind which accounted for 17.2%(n = 98), pain which accounted for 10.4%(n = 59), infection which accounted for 7.5%(n = 43) and avascular or osteonecrosis which accounted for 6.3%(n = 36).21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41^,43,44 Average time to revision ranged from 4.1 to 12.3 years.^22,43 Average femoral head diameter of hips that were revised ranged from 44.7 mm to 49.1 mm(42.7 mm–46.7 mm in females and 49.8 mm–56 mm in males).

3.4. BHR long-term comparison to THA

Survivorship, revisions, satisfaction, and findings on postoperative activity between BHR and THA were compared(Table 6). Two were retrospective studies analyzing registry data, and four were single-institution studies comparing contemporary cohorts.^{13,30,31,42,43,45} Two studies performed comparative analyses with matched cohorts according to age, sex, BMI, and preoperative activity level or UCLA.^13,43 Survivorship for THA ranged from 74 to 100, whereas in BHR it ranged from 83 to 100%. Two studies reported patient satisfaction, one showed significantly increased patient satisfaction in BHRs compared to THA and the other did not report a difference.^13,43 Three long-term follow-up studies compared postoperative activity levels, revealing that two studies showed an improved return to activities such as sports, heavy manual labor, and running.^13,42 Additionally, one study found a higher proportion of BHR patients with high preoperative UCLA scores^9,10 remained significantly active postoperatively compared to those undergoing THA.⁴³

Table 6.

THA vs. BHR subgroup analysis.

Study	Group	Hips(n)	Timeframe, years	Survivorship(%)		Revisions, n(%)	Satisfaction		Activity
Jonas et al., 2019	BHR	54	18	88[75.9–94.6]	p-value = 0.067	6(11.1)	91%(32/35) delighted or pleased	p-value = 0.002	More BHR patients ran (p-value = 0.007), played sports(p-value = 0.01) and performed heavy manual labor(p-value = 0.002) compared to THA patients
	THA	54	18	74[59.4–84.0]		14(26.0)	56%(15/26) delighted or pleased
Stoney et al., 2020	BHR	4790	17	93.3		6.7[5.8–7.9]	NR		NR
	THA	2696	17	96.4		3.4[2.5–4.6]	NR
Pietiläinen et al., 2022	BHR	233	16	83		NR	NR		NR
	THA	38	12	87		NR	NR
Haddad et al., 2015	BHR	49	10	100		0(0.0)	NR		Running	BHR: 26(53.1) THA: 16(51.6)	p-value = 0.1
	THA	31	10	100		0(0.0)	NR		Sports involvement	BHR: 42(85.7) THA: 16(51.6)	p-value = 0.09
									Heavy manual labor	BHR: 10(20.4) THA: 4(12.9)	p-value = 0.19
Ford et al., 2018	BHR	159	At final follow-up, minimum 5	NR		NR	5(extremely satisfied)	p-value = 0.99	0.4 ± 2.6-point mean postoperative UCLA advantage for BHR compared to THA(p = 0.040)
									In patients with preoperative UCLA scores of 9–10, decrease in postoperative UCLA was smaller for BHR compared to THA(0.0 for BHR, 1.0 for THA, p < 0.001).
	THA	159				3(1.9)	5(extremely satisfied)		Percentage of highly active patients(UCLA 9–10) remaining highly active postoperatively: BHR: 61% THA: 20% p-value<0.001
Seppänen et al., 2016	BHR	2141	10	91[89–92]		NR	NR		NR
	THA	6485		92[91–92]		NR	NR

Values reported as value(%) or average ± SD (range) or [CI]. Underline denotes median. {}, interquartile range. NR, not reported. BHR, Birmingham Hip Resurfacing. THA, Total Hip Arthroplasty. UCLA, University of California Los Angeles activity scale.

4. Discussion

Though BHR serves as an viable alternative procedure to treat younger, active THA candidates, it has faced criticism for pseudotumor formation and adverse reaction to the metal-on-metal hardware.^{14,24,26,29,33,37,43} At present, there is a lack of consensus about the long-term durability of this implant beyond ten years. The primary purpose of this study was to aggregate findings from the literature to better understand the long-term patient-reported outcomes, survivorship, and revision surgical characteristics following BHR and to compare long-term postoperative outcomes with THA. Postoperative outcomes were found to be favorable with notable improvements across multiple metrics. Survivorship was shown to be satisfactory at the long-term. Indications for revision were variable, but mostly due to implant loosening and metal-related pathology. Further, the comparison between the BHR and THA cohorts suggested possible long-term activity benefits of BHR over THA. Ultimately, these findings encourage that with proper patient selection, BHR implants yield favorable outcomes and may sustain patients at timeframes of 10 years and beyond even when compared to the THA.

4.1. Patient-reported outcomes

This review showed promising postoperative PRO scores at timeframes of ten years and beyond with statistically significant improvements in WOMAC, UCLA, OHS and HHS.^21,27,38,43 Unfortunately, many studies included lacked preoperative values for PROs and therefore did not perform more robust statistical analysis. Nevertheless, in the studies that did include both preoperative and postoperative PROs, many displayed marked improvement that persisted in the long-term. For instance, Singh et al. conducted a cohort analysis of 2667 THAs and calculated Minimally Clinically Important Improvement (MCII) for HHS at 5-year post-THA of 15.9 for and 39.6 for moderate improvement.⁴⁶ Using these derived thresholds, for comparison, all the studies in this review that documented preoperative and postoperative HHS would have met criteria for moderate improvement.26, 27, 28^,34,36,40,42 Future studies with both preoperative and postoperative PROs as well as clinical benefit thresholds such as Minimally Clinically Important Differences(MCID) or Patient Acceptable Symptom State(PASS) may better contextualize surgical results.

4.2. Long-term survivorship

Overall BHR survivorship at ten years and beyond was promising ranging from 83 to 100% across all studies and greater than 87% in 25 out of 26 studies.^13,21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45 When follow-up length was longer than 15 years, survivorship continued to remain favorable ranging from 83 to 96.8%.^13,22,30, 31, 32^,35,38 Prior literature shows better outcomes associated with male sex and larger femoral head diameter.^22,23,25 In fact, based on Smith & Nephew guidelines, U.S. Food and Drug Administration (FDA) recommendations and existing outcome studies, the current consensus is to restrict BHR use to males under the age of 65 with a femoral head diameter ≥48 mm.^30,36,43,47 This review aligned with their statement, finding that long-term male survivorship was notably higher ranging from 93 to 99% across studies compared to females which ranged from 73.9 to 91.9%.^{28,29,34,41,44} Further in the three studies that reported it, long-term survivorship rates for males with greater than 50 mm femoral head diameter was high ranging from 93.3 to 97%.^21,30,34 In addition, Coulter et al. employed structural equation modelling and derived a similar finding that women were 1.4 times more likely to experience failure than men and that each millimeter increase in femoral component size yielded 19% lower chance of failure.²³ Cox model analysis from Treacy et al. indicated a 1.14 times annual revision risk increase with every 4 mm decrease in femoral head diameter and 5.78 time increase associated with being a female patient ²⁵. Mechanisms of these discrepancies remain unclear; however it has been proposed that reduced acetabular coverage in smaller femoral heads, which are associated with being female, may cause edge loading and subsequent increased wear, metal debris and risk for failure.48, 49, 50, 51 Despite, this with proper patient selection BHR appeared to retain high survivorship in the long-term.

4.3. Revision characteristics

This review also found that primary revision reasons at long-term follow-up were implant loosening, accounting for 22.0%(n = 125) and metal-related pathology or pseudotumor formation 21.2%(n = 121) of 569 revisions. It is important to note that the sample size distributions between study may bias these findings, as data from Stoney et al. alone accounted for over one third of the hips in this study.³⁰ The leading cause for revision appeared to be implant loosening, this was consistent with the fact that BHR is preferentially performed in younger patients with increased activity levels, a risk factor.^{13,42,43,52,53} In addition, this study found that at long-term, adverse reactions to metal debris, and pseudotumor formation secondary to prosthetic wear remained a concern for long-term failure.⁵⁴ Metal ions from the cobalt and chromium alloy used in implants may trigger inflammatory reactions leading to subsequent periprosthetic bone loss and aseptic loosening.^55,56 In their study on 129 BHRs, Bisschop et al. reported a 28% incidence of pseudotumor formation by 41 month follow-up, noting that increases in pseudotumor size were correlated with pain and discomfort.⁵⁷ Though adverse reaction to metal debris(AMRD) remains a significant cause of BHR revision, its incidence varied broadly across studies due to potential mediating risk factors such as varying sex distribution, surgical alignment of implants or femoral diameter58, 59, 60

4.4. Comparison with THA

Excluding the large discrepancy reported by Jonas et al., survivorship rates between the BHR and THA were comparable across the studies, with any differences being within a 4 percent margin.¹³ Stoney et al. compared BHR with three conventional THA implants having the lowest revision rates and found significantly higher revision rates in the BHR cohort after the first 2.5 years.³⁰ On the other hand, Jonas et al. noted that 18-year BHR survivorship rate of 88 was nearly significantly higher than 74% of THA(p-value = 0.067).¹³ The heterogeneity of implant choices in THAs performed as well as discrepancies in patient selection for both procedures likely contributed to these differences. More rigorous studies involving matched cohorts to compare the long-term survivorship of the most widely utilized THA implants with BHR could elucidate potential differences in outcomes. Further, the studies reviewed also compared long-term activity levels between THA and BHR, in which all three studies reporting found improved activity in BHR patients compared to THA.^13,42,43 Patients who underwent BHR experienced enhanced engagement in sports and heavy labor after surgery, alongside the benefit of a lesser reduction in postoperative UCLA activity score change.^13,42,43 These results suggest long-term activity benefits of BHR. Additionally, as mentioned by Ford et al., self-selection may also influence the observed increase in activity levels among patients who undergo BHR, since individuals who opt for BHR may possess a higher motivation to engage in high activity levels. Thus the satisfaction metric may provide insight on whether or not postoperative activity levels met patient expectations, however in Jonas et al., the increased activity level in BHR patients did not result in a higher satisfaction.¹³ Currently, the literature does not indicate whether the potential benefits of enhanced activity levels associated with BHR surpass the risks related to metal debris. However, this review has shown that, with careful patient selection, BHR can offer comparable long-term survivorship and improved activity levels compared to the THA.

4.5. Potential limitations

There were multiple limitations with this systematic review. Most studies reviewed were low level of evidence studies with limited sample sizes contributing to increased risk for bias. In addition, though patient-reported outcomes were recorded in 21 studies, statistical significance between preoperative and postoperative scores was not calculated in 16 studies, which precluded quantitative analysis of such metrics. In addition, data comparing BHR and THA was limited and heterogeneous. There were likely baseline patient differences between performing surgeons and selection bias associated with the decision to recommend BHR over THA, as BHR is usually performed by high-volume surgeons experienced with the technique. This may limit the validity and generalizability of the subgroup analysis. This review covered a wide range of surgical techniques, performed by both designer and non-designer surgeons, and did not account for surgical advancements that occurred over the extended study period. This heterogeneity may further confound findings.

5. Conclusion

This review showed that at ten years and beyond BHR exhibited improved patient-reported outcomes and satisfactory survivorship. Implant-loosening and metal-related pathology or pseudotumor formation remained a concern for revision at ten years and beyond. Female sex and small femoral head diameters may continue to be associated with revision risk at long-tern. In studies comparing long-term BHR and THA survivorship was comparable, with BHR potentially providing enhanced postoperative activity levels. The findings from this study suggested favorable longevity and postoperative outcomes of BHR under the pretense of careful patient selection. Surgeons should continue to remain cautious of the leading causes for revisions and their potential risk factors to accurately identify optimal candidates for BHR.

Ethical committee approval

Ethical approval was waived as our analysis does not contain human data.

Registration

PROSPERO registration of the study protocol: CRD42023464567.

Guardian/patient's consent

No consent was needed as no patients or human data were included in this study.

Ethical committee approval

Ethical approval was waived by Western Institutional Review Board as our analysis does not contain human data as defined in Section §164.514(b)(1) of the HIPAA Privacy Rule.

Funding

No funding was received for this study.

CRediT authorship contribution statement

Scott Fong: Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing. Aakash K. Shah: Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing. Christian J. Hecht: Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing. Atul F. Kamath: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.

Declaration of competing interest

A.F.K. reports the following disclosures: paid presenter or speaker (Zimmer Biomet), paid consultant (Zimmer Biomet, BodyCad, Ortho Development, United Ortho), stock or stock options (Zimmer Biomet, Johnson & Johnson, and Procter & Gamble), IP royalties (Innomed), and board or committee member (AAOS, AAHKS, and Anterior Hip Foundation). SF, AKS, and CJH report no disclosures.