Periacetabular osteotomy with and without concomitant arthroscopy: a systematic review of evidence on post-operative activity levels and return to sport

ABSTRACT The purpose of this systematic review is to (i) compare post-operative activity levels after periacetabular osteotomy (PAO) versus PAO + HA (concomitant PAO and hip arthroscopy) using patient-reported outcomes that specifically assess activity and sports participation [Hip Disability and Osteoarthritis Outcome Score—Sport and Recreation subscale (HOOS-SR), University of California Los Angeles (UCLA) activity score, Hip Outcome Score—Sport-Specific Subscale (HOS-SSS)] and (ii) compare post-operative return to sport (RTS) data between PAO and PAO + HA groups. A systematic review of literature was conducted on 1 June 2023, utilizing PubMed, Cochrane and Embase (OVID). Articles were screened for inclusion using specific inclusion and exclusion criteria. Twenty-six out of 1610 articles met all inclusion criteria, without meeting any exclusion criteria. In the 12 studies containing only subjects who underwent PAO alone, 11 demonstrated an average score improvement in UCLA, HOOS-SR or HOS-SSS post-operatively (P < 0.05). In the three studies containing subjects who underwent PAO with concomitant HA, significant improvements were seen in the HOS-SS and UCLA scores (P < 0.05). In the five studies that directly compared UCLA, HOS-SSS and HOOS-SSS scores between PAO groups and PAO + HA groups, all demonstrated statistically significant improvement post-operatively (P < 0.05). The rate of RTS ranged from 63% to 90.8% among PAO studies and was found to be 81% in the single PAO + HA study that assessed RTS. When performed in patients with intra-articular pathology, concomitant PAO + HA may provide similar sport-related outcomes to PAO alone in patients without intra-articular pathology.


IN TRODUCTION
Periacetabular osteotomy (PAO) is a well-studied method to treat patients with symptomatic hip dysplasia.This surgical technique aims to conserve joint cartilage within the hip via acetabular reorientation and femoral stabilization [1][2][3].Prior studies have shown that the primary demographic group for hip dysplasia consists of young females in the age range of 14-18 years and that the condition is common within the female athletic community [4,5].As hip pain and groin pain have shown to be prevalent obstacles in athletes [6,7], PAO is an effective tool in improving sports participation and activity levels in hip dysplasia patients [8][9][10][11][12][13][14][15][16][17].
Although PAO effectively corrects for malalignment within the hip joint, an increasing number of studies have demonstrated a high prevalence of intra-articular pathology within the dysplastic hip that may require arthroscopic intervention [18].While not recommended as an isolated treatment in moderate and severe cases of hip dysplasia [19], arthroscopic techniques have been used in select patients with mild dysplasia, but they remain controversial in their efficacy [20].Patients with untreated but identified concomitant intra-articular pathology often have worse surgical outcomes following a PAO and may undergo arthroscopic intervention in the future [21][22][23].Furthermore, there is evidence that a previously failed HA may worsen outcomes in patients who later undergo PAO [24].These discoveries have created the need for further research on the effectiveness of a combined PAO and HA (PAO + HA) in managing patients with hip dysplasia-particularly in young and active groups.Favorable activity-related outcomes have been found in these patients [25][26][27][28][29][30].
Several prior systematic reviews have assessed outcomes of PAO and/or PAO + HA.One study found that the majority of patient-reported outcomes (PROs) improved post-operatively when included and that conversion rates to total hip arthroplasty (THA) were lower in arthroscopy alone compared with PAO and PAO with arthroscopy [31].However, since their publication, at least eight more studies have investigated outcomes of concomitant PAO and HA, analyzing the impact of this surgery Periacetabular osteotomy with and without concomitant arthroscopy • 99 on sports participation and the activity level, and have overall seemed to find no significant differences in PROs between PAO and PAO + HA [26-28, 30, 32-35].Additionally, this study failed to include PROs such as University of California Los Angeles (UCLA) activity score or hip disability and osteoarthritis outcome score-sports and recreation (HOOS-SR) or data regarding return to sport (RTS) in athletic populations.Six more systematic reviews have looked at outcomes of PAO [36][37][38][39][40][41], with one assessing RTS outcomes without the inclusion of PROs [37].This particular study found that PAO is an effective procedure in athletes to improve sports participation post-operatively.The other studies have overall found good clinical outcomes in a variety of categories related to hip survivorship after an isolated PAO [37][38][39][40].However, none of these studies compared results to groups receiving PAO + HA [36][37][38][39][40][41].Finally, another systematic review assessed outcomes and survivorship after PAO + HA, finding favorable outcomes overall, without comparing their data to an isolated PAO [42].
The purpose of this systematic review is to (i) compare post-operative activity levels after PAO versus PAO + HA using PROs that specifically assess activity and sports participation [(HOOS-SR, UCLA activity score and Hip Outcome Score-Sport-Specific Subscale (HOS-SSS)] and (ii) compare postoperative RTS data between PAO and PAO + HA groups.Our research question is as follows: in subjects with dysplastic hips, do concomitant PAO and HA, when compared to an isolated PAO, result in improved post-operative sports and activity outcomes?Our hypothesis is that sports and activity outcomes will be similar in patients who underwent PAO alone or PAO with concomitant arthroscopy.

M ETHODS Search strategy
This study is a systematic review of studies published prior to 1 June 2023 that have investigated RTS and/or postoperative activity levels in subjects who have undergone PAO or PAO + HA.This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for reporting systematic reviews [43].One author searched PubMed/Medline, Cochrane and Embase (OVID) databases on 1 June 2023, using the following terms:(((hip OR acetabular OR' Acetabulum'[MeSH]) AND (dysplasia OR dysplastic))OR 'Hip Dislocation' [MeSH])AND (periacetabular AND (osteotomy OR'Osteotomy'[MeSH])).

Summary of the screening process
Duplicates and papers written in languages other than English were excluded.Two authors independently screened all articles returned from the initial search using pre-determined inclusion and exclusion criteria.Each study was first screened for relevance by title and abstract.Full text for studies deemed 'relevant' was then reviewed for inclusion/exclusion in this systematic review.Conflicts were resolved by a neutral third author.
Inclusion criteria are as follows: (i) human studies that either report outcomes in patients who underwent PAO alone or report outcomes in patients who underwent concomitant PAO and arthroscopy, (ii) studies that report outcomes at least 1 year postoperatively, (iii) surgeries must be indicated for hip dysplasia and (iv) studies that report RTS data or sport/activity-related outcomes according to any of the following three PRO measures: HOS-SSS, Hip Disability and Osteoarthritis Outcome Score-Sport and Recreational Activity (HOOS-SR) and University of California Los Angeles (UCLA) activity scale.The HOS-SSS is a validated sub-scale used to assess RTS/activity after hip surgery [44,45], including within the context of hip dysplasia [46,47].The HOOS-SR has been validated and used in the post-THA population [48,49] as well as in patients who have undergone PAO [50].The UCLA activity score serves as a general estimate of patients' post-operative activity levels that have been validated for monitoring physical activity levels in those with hip and/or knee osteoarthritis [51]; it has also been used in studies on hip dysplasia [52,53].
Exclusion criteria are as follows: case reports, review articles, conference abstract presentations, non-English studies, studies that do not report outcomes at least 1 year post-operatively, studies that do not explicitly state whether a PAO in isolation or a combined PAO/arthroscopy was performed, studies that contain both PAO and PAO + HA subjects but do not provide separate analyses for both populations, non-human studies, any indication for surgery other than hip dysplasia, studies investigating only modified PAO techniques, studies that did not report on RTS or any of the following three PRO measures: HOS-SSS, HOOS-SR and UCLA activity score.

Quality assessment
Two authors assessed each included study for quality and risk of bias.For case-control studies and cohort studies, the Newcastle-Ottawa Scale (NOS) was used [54].Case series studies were assessed using the National Institute of Health (NIH) Quality Assessment Tool for case series studies and prospective designs with before-after studies without a control group [55].

Outcomes of interest and data extraction
Outcomes of interest included RTS data, HOS-SSS scores, HOOS-SR scores and UCLA activity scores.Data were extracted from each of the included studies by two authors independently using a standardized data extraction table created by one of the authors.Relevant information related to each study's methods and subject demographics was collected in addition to outcomes of interest.

R E SULTS Search results
The initial search returned 1610 articles.After 630 duplicates were removed, there were 980 studies left to be screened by title and abstract.Of these, 790 were excluded based on inclusion/exclusion criteria.There were 190 full-text articles sought for retrieval, all of which were successfully retrieved.Of these, a total of 165 articles were excluded based on inclusion/exclusion criteria, leaving a total of 26 studies that met inclusion criteria without any exclusion criteria [8-14, 16, 17, 26-28, 30, 32-35, 50, 53, 56-62].The inclusion/exclusion process is depicted in Fig. 1.

Study design and quality assessment
Thirteen of the 26 included studies were case series design [8-14, 16, 26, 27, 32, 35, 56], three were case-control design [10,28,58], six were retrospective cohort design [30,33,34,53,59,60] and four were prospective cohort design [50,57,61,62].Eleven out of 13 case series studies were determined to be of 'good' quality with at least seven out of nine NIH criteria being met.One case series met only four out of nine NIH criteria due to poorly described statistical methods and lack of detailed description of pre-operative injury and functional status of their subjects [9].All three case-control studies met eight out of eight NOS criteria and were considered to be of 'good' quality [10,28,58].Eight of the 10 cohort studies met at least eight out of eight NOS criteria and were considered to be of 'good' quality.Two prospective cohort studies lacked a control group and therefore were evaluated with the NIH quality assessment tool for beforeafter studies without the control group.They met nine out of nine criteria and were also determined to be of 'good' quality [50,57].

Periacetabular osteotomy with and without concomitant arthroscopy • 101 Table I. Outcomes of interest in studies where subjects received PAO alone PAO-alone study Design (n, average length of followup) UCLA activity score (average) HOOS-SR (average) HOS-SSS (average) RTS and other activity-related data
Heyworth et al.

PRO measures
In the 12 studies containing only subjects who underwent PAO alone, 11 demonstrated an average score improvement in UCLA, HOOS-SR or HOS-SSS post-operatively.The remaining study demonstrated successful return to baseline in UCLA score [8].Details regarding the included PAO-alone studies can be found in Table I.In the three studies containing subjects who underwent PAO with concomitant HA, significant improvements were seen in the HOS-SSS [27,35] and UCLA [26] scores.Details regarding studies containing only PAO + HA subjects can be found in Table II.In the five studies that directly compared UCLA, HOS-SSS and HOOS-SSS scores between PAO groups and PAO + HA groups, all demonstrated statistically significant improvement post-operatively [28,32,34,58,61].It is worth mentioning that one of the most frequently used PRO measures, the International Hip Outcome Tool (iHOT), was not included in the primary analysis of the present study [63].Although six of the included studies reported iHOT scores [11,14,26,[60][61][62], the authors chose not to include these data in the primary analysis because the iHOT scores did not report sport/recreationspecific subscores, which is the primary focus of this study.Therefore, it was not included in the primary analysis of the present study.These details regarding these and other included comparative studies can be found in Table III.

RTS
For the purposes of reporting, any RTS data reported by the included studies were included in the analysis of this study.Three PAO-alone studies reported RTS data [8,14,56].In all three, RTS was defined as resumption of organized sport (i.e., performance or competition) at any capacity.The average rate of RTS among these three studies was 77.93% (range, 63-90.8%) with an average time to RTS of 8.9 months.The lowest rate (63%) was found in a study of female dancers [56].One study found a significant increase in low-impact sport participation post-operatively (31.7-52%;P = 0.001) while also demonstrating a non-significant decrease in high-impact sport participation post-operatively (42.3-36.6%;P = 0.361) [14].
One PAO + HA study reported an 81% rate of RTS at a final follow-up of 2.83 years [27].Interestingly, they reported a higher rate of RTS in recreational athletes (85.6%) compared to competitive athletes (77.8%).Another study comparing triple PAO to PAO found no significant differences in visual analog scale (VAS) (0-10 scale with 10 being the worst outcome) fitness level and hours of sport participation per week post-operatively between the two groups (P > 0.05) [59].

DISCUSSION
The limited number of available studies included in this review appears to affirm the hypothesis that no significant difference in sports participation and activity levels between subjects who received PAO and PAO + HA exists.Across all variables assessed in this review, patients in both the PAO and PAO + HA groups demonstrated greater activity and sports participation levels in their post-operative evaluations.However, there was no identifiable trend in differences in the outcomes of interest between patients who underwent one of the two procedures.One comparative study found a statistically significant increase in UCLA score in the PAO + HA group compared to PAO alone [58].Although the heterogeneity of the studies makes direct comparative analysis difficult, the overall results support the use of both procedures to improve sports and activity outcomes in patients with hip dysplasia without a significant variation in their success to do so.
(2017) [8]   Retrospective cohort (n = 93; (2023) [9]   Retrospective cohort (n = 60; (2022) [10]   Retrospective These review findings are agreeable with Lodhia et al., in their conclusion that PRO scores do not vary significantly between PAO and PAO + HA groups in that both cohorts demonstrate improved outcomes post-operatively [31].Although they demonstrated similar PRO scores between groups in their study, they did not report on specific sport-related outcome measures such as the HOS-SSS, HOOS-SR or UCLA activity score.Similar to our study, they reported a much larger proportion of subjects who received PAO alone (n = 703 hips) to subjects who received PAO + HA (n = 17 hips).Over the 7 years since publication of the study by Lodhia et al., more studies have investigated PAO + HA groups, providing the current systematic review with a much larger, albeit still unequal, proportion of PAO + HA to PAO (n = 355 hips versus n = 1544 hips, respectively).In a recent systematic review on outcomes after concomitant PAO + HA, Lee et al. similarly demonstrated excellent post-operative PRO scores [42].However, they also did not report on sport-specific subscales.Our study builds on the work of these two prior systematic reviews, providing evidence that PAO + HA, when performed on dysplastic hips with intra-articular pathology, leads to similar post-operative sport-related PRO scores compared to subjects who appropriately received PAO without intra-articular intervention [31,42].
Our findings are in agreement with findings of Curley et al. that PAO can provide good outcomes regarding RTS [36].Unlike the present review, the authors did not report on subjects with PAO + HA or on sport-specific PROs.However, the rate of RTS found in the present systematic review (63-90.8%) is similar to the rates reported by Curley et al. [36].This is likely due to the fact that both reviews share multiple PAO studies in common [8,10,27,56,59].In addition to these shared studies, our systematic review also reveals data from the study by Leopold et al. that demonstrated a significant decrease in high-impact sports participation from pre-operative to post-operative (42.3% to 36.6%;P = 0.361) and an increase in low-impact sports participation from pre-operative to post-operative (31.7% to 52%; P = 0.001) [14].This brings up an interesting point that sports participation after PAO or PAO + HA will largely depend on the target sport/activity.Heyworth et al. similarly found that recreational athletes return to prior level of competition at a higher rate (89%) compared to competitive athletes (college, high school and professional levels) (58%) after PAO [8].Therefore, when considering RTS expectations in athletic populations, it is important to consider the goal sport and level at which it is hoped to be played post-operatively.
These findings coincide with current supportive evidence for the use of PAO + HA in the case of hip dysplasia with concomitant intra-articular pathology.Prior studies have shown that patients with intra-articular pathology at the time of surgical intervention tend to have longer hip survival when these concomitant issues are addressed in their initial corrective procedure and are less likely to require further surgical treatment later on [21][22][23].One study found that advanced intra-articular lesions not addressed during an osteotomy caused a high rate of progression to osteoarthritis in the dysplastic hip, worsening longevity of the procedure [23].Moreover, several studies have reported low but not insignificant rates of conversion to other surgical procedures, including THA and arthroscopy, in hips that underwent PAO [21,22,31,38].Without evidence to suggest that PAO + HA will worsen athletic outcomes in cases of active patients with dysplastic hips, the aforementioned results help affirm the use of this procedure in cases with identified intra-articular pathology.
This study is not without limitations.One limitation is the failure of some studies to report any procedures performed concurrently with PAO.This creates the possibility that some cases analyzed as an isolated PAO may actually be properly categorized as PAO + HA (or some other combination of procedures), which could bias comparison between the two procedures.However, this was not standard across all the included studies as many did report concomitant surgeries in both PAO and PAO + HA groups [28,32,58,61].Heterogeneity in the degree of RTS (i.e., return to the pre-operative level versus return to any level) also limits our conclusions regarding RTS outcomes.Additionally, patients undergoing isolated PAO may have presented with asymptomatic intra-articular pathologies, while those undergoing PAO + HA had symptomatic intra-articular changes.Additionally, several of the included studies did not meet criteria to be considered of 'good' quality according to the risk of bias assessment tools used by the authors.Lastly, the studies included in the present systematic review are inherently heterogeneous with varying study protocols, reporting methods and outcome measures.For this reason, the authors determined that a metaanalysis was not feasible.Despite these limitations, this systematic review successfully demonstrated comparable sport-and activity-related outcomes in subjects who undergo PAO and those who undergo PAO + HA.

CONCLUSION
Based on the heterogeneous studies reported in this systematic review, PAO with and without concomitant HA appears to provide similar sport-and activity-related outcomes.However, the lack of uniform identification of pre-operative intra-articular across studies makes it difficult to conclude that outcomes are truly similar between groups.More high-quality, controlled studies are needed to assess hip survivability in order to help surgeons identify patients who would benefit most from concomitant surgery.

Fig. 1 .
Fig. 1.PRISMA flow diagram.There were 32 studies included in this systematic review.

(
not report; Δ, change in score; HOOS-SR, Hip disability and Osteoarthritis Outcome Scale-Sport and Recreation subscale; MCID, minimal clinically important difference.104 • P. Wyatt et al.

of interest in the included comparative studies Comparative Study Design (n, average length of follow- up) Groups compared UCLA activity score (average)
Δ, change in score; DNR, did not report; HOOS-SR, Hip disability and Osteoarthritis Outcome Scale-Sport and Recreation subscale; MCID, minimal clinically important difference; PASS, patient acceptable symptom state.Periacetabular osteotomy with and without concomitant arthroscopy • 105 Table III.Outcomes Hip disability and Osteoarthritis Outcome Scale-Sport and Recreation subscale; MCID, minimal clinically important difference; PASS, patient acceptable symptom state; TPO, triple pelvic osteotomy.