Periacetabular osteotomy of the hip: an 8-year follow-up of 96 consecutive cases

doi:10.21203/rs.3.rs-8058147/v1

Download PDF

Research Article

Periacetabular osteotomy of the hip: an 8-year follow-up of 96 consecutive cases

https://doi.org/10.21203/rs.3.rs-8058147/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Introduction

A periacetabular osteotomy (PAO) is a joint-preserving surgical option for treatment of acetabular dysplasia. The procedure aims to prevent, or at least delay, the development of osteoarthritis, and subsequent need for total hip arthroplasty (THA). The conversion rate to THA differs widely in the literature, but most of the studies have few patients, and the follow-up time is often short for THA as an endpoint. The aim of this study was to evaluate the long-term outcome after PAO surgery with the rate of conversion to THA as the primary outcome.

Materials and methods

Patients ≥18 years that underwent a PAO operation at the Karolinska University Hospital in Stockholm, Sweden from 2006 to 2022 were included. Radiological signs of hip osteoarthritis, and the lateral center-edge angle (LCEA) was calculated on pre- and postoperative radiographs or CT-scans. The national Swedish Arthroplasty Register was used to find cases who had a secondary operation with THA.

Results

The number of cases included was 96. Median age was 30 (18–46) years, and 84% (n = 81) were females. Median follow-up time was 99 (17–227) months (8 years). A total of 21 (22%) cases had a secondary THA. Cox regression analyses identified that age ≥ 30 years was associated with THA in both uni- (HR 2.8, CI 1.1–7.3) and multivariable (HR 5.2, CI 1.5–18) analyses. Preoperative osteoarthritis (Tönnis grade 2) was associated with THA in multivariable (HR 24, CI 2.1–247) analysis. A total of 49 (51%) cases were reoperated due to other reasons than THA. The most common reason was extraction of screws (n = 39, 41%). Forty-four (43%) patients had an adverse event. The most common was a transient nerve injury.

Conclusions

The PAO procedure is a suitable option in young patients with symptomatic dysplasia of the hip in order to avoid, or at least delay, hip arthroplasty.

Periacetabular osteotomy

Bernese osteotomy

Hip dysplasia

Joint-preserving surgery

Total hip arthroplasty

Acetabular dysplasia is a well-known risk factor for early-onset hip osteoarthritis in younger patients [1]. A periacetabular osteotomy (PAO) is a joint-preserving surgical option for treatment of acetabular dysplasia [2]. The rationale for this surgical method is that the shallow dysplastic acetabulum causes mechanical instability and increased focal pressure on the cartilage of the hip joint, and left untreated often result in early development of osteoarthritis [3, 4]. A nowadays commonly used surgical method for PAO was described in 1988 by Ganz [5] and is often referred to as the Bernese PAO. In this operation, the acetabulum is reoriented to increase the anterolateral coverage of the femoral head and to achieve medialization of acetabulum, thereby increasing the cartilage joint contact surface and mechanical stability of the joint [5]. The procedure aims to prevent, or at least delay, the development of osteoarthritis, and subsequent need for total hip arthroplasty (THA) [6]. The PAO procedure has been associated with relatively high patient satisfaction in general, but also an inversed correlation between severity of dysplasia and patient satisfaction, with more dysplastic hips having lower rates of postoperative satisfaction [7]. Furthermore, preoperative patient evaluation and selection seems to be crucial as postoperative functional recovery varies widely [8, 9]. Postoperative outcome after surgical procedures can be measured in several ways, and in the case of PAO the rate of conversion to THA is of major interest. In a recently published systematic review and meta-analysis with a mean follow-up of 54 months, it was reported that 6% of the patients underwent a secondary THA after a PAO [6]. However, most of the included studies had few patients, and the follow-up time was short if THA surgery is to be used as an endpoint. The aim of this study was to evaluate the long-term outcome after Bernese PAO surgery with the rate of conversion to THA as the primary outcome.

All patients ≥18 years that underwent a PAO at the Karolinska University Hospital in Stockholm, Sweden from 2006 to 2022 were included in this retrospective cohort study. Patients were identified through the local surgical database and all medical records including radiographs were manually reviewed. A letter with information about the study with the option to opt-out was sent to all patients.

Collected demographic variables included patient age, gender and indication for surgery. Radiological signs of hip osteoarthritis on preoperative radiographs or CT-scans were classified according to Tönnis [10]. The lateral center-edge angle (LCEA) was calculated on pre- and postoperative radiographs or CT-scans [11] and classified as; normal (> 25°), borderline (20–25°) or dysplastic (< 20°) [12]. The indication for the surgery was symptoms from the hip (pain) and radiographs indicating dysplastic anatomy of the hip.

To find cases who had a secondary operation with a total hip arthroplasty, cross-referencing was performed with the national Swedish Arthroplasty Register (SAR). The SAR started in 1979 and collects detailed data on all primary and secondary hip arthroplasties performed in Sweden in patients with a valid Swedish personal identification number. The completeness in SAR for primary total hip arthroplasty procedures has been > 98% over the last 10 years (www.slr.registercentrum.se).

Other follow-op variables included any other reoperation, including cause and type, any adverse event not requiring surgical treatment (nerve injury, pneumonia, pulmonary embolism, deep venous thrombosis, urinary tract infection, sepsis, kidney failure, superficial wound infection). In addition, mortality was recorded. All patients were followed for a minimum of 2 years after the surgery.

All surgeries were performed in a supine position under general anesthesia with intraoperative fluoroscopy. The surgical incision used was a shortened ileo-inguinal, and the fixation of the osteotomy was normally done with 3 fully threaded 3.5 mm cortical screws. Perioperative intravenous antibiotic prophylaxis was given to all patients, as well as postoperative low-molecular-weight heparin for prevention of blood clots. Postoperatively, patients were instructed to partial weight-bear the operated side using crutches for the first 6–8 weeks.

The study was approved by the Swedish Ethical Review Authority with reference number: 2023-067811-01.

Statistical methods

Patients with bilateral operations were analyzed as two separate cases. Numerical data was presented as median (range). Categorical data was presented as frequency with percent distribution. Nominal variables were tested with the Fisher’s exact test. The Wilcoxon signed rank test or the Mann-Whitney U test were used for comparison of scale variables. All tests were two-sided. Cox regression analysis was performed to evaluate factors associated with THA reoperation. Age (< 30 or ≥ 30 years (median age)), preoperative Tönnis grade (0–3), preoperative LCEA (dysplasic, borderline, normal) was included in the analyses. First, crude association for each variable was tested in univariable models. Second, a multivariable model was used to study the adjusted associations. The associations were presented as hazard ratios (HRs) with 95% confidence intervals (CIs). The follow-up time was defined as the time from the date of PAO surgery to the date of THA surgery, death or December 31, 2024. The results were considered significant at p < 0.05. The statistical software used was IBM SPSS Statistics, Version 31 for Windows (SPSS Inc., Chicago, Illinois).

A total of 84 patients were identified. Two patients had emigrated abroad, and two patients replied with an opt-out of the study, leaving 80 patients to be included. Of those, 16 patients were operated in both hips, all at different occasions. Thus, the total number of cases/hips included in the analysis was 96. Median age at the operation was 30 (18–46) years, and 84% (n = 81) were females (Table 1). Median follow-up time was 99 (17–227) months (8 years). All patients were alive at the closure of the study (December 31, 2024).

Table 1
Patient characteristics including comparisons between patients that underwent subsequent THA surgery (THA), and those who did not (No THA)
Variable	All cases (n = 96)	THA (n = 21)	No THA (n = 75)	p-value
Age; Median (range)	30 (18–46)	39 (18–44)	28 (18–46)	0.02
Female gender; n= (%)	81 (84)	20 (95)	61 (81)	0.2
Operated hip left side; n= (%)	44 (46)	9 (43)	35 (47)	0.8
Smoker; n= (%)	4 (4.2)	3 (14)	1 (1.3)	0.03
Diagnosis; n= (%) Primary dysplasia Secondary dysplasia due to Perthes	94 (98) 2 (2.1)	21 (100) 0	73 (97) 2 (2.7)	1.0
THA = total hip arthroplasty

Radiographic analyses

For one patient, preoperative radiographs could not be retrieved. In the remaining cases, 77 (81%) had Tönnis grade 0, 16 (17%) cases had grade 1 and 2 cases (2%) had grade 2 radiological signs of hip osteoarthritis. Preoperative, 75 (78%) cases had a dysplastic (< 20°), 14 (15%) cases had a borderline (20–25°), 6 (6.3%) cases had a normal (> 25°) LCEA. The median preoperative LCEA was 13° (0–29°), and the median postoperative LCEA was 36° (14–58°) (p < 0.001).

Secondary operation with THA

After cross-referencing with the SAR, a total of 21 (22%) cases were identified that had a secondary operation with a THA. The median age of these patients at the time of the THA surgery was 43 (28–54) years, and 20 were females (Table 1). An uncemented arthroplasty was the most commonly used type of THA (n = 17), followed by a reverse hybrid type (cemented cup and uncemented stem) (n = 3) or a hybrid type (uncemented cup and cemented stem) (n = 1) (Fig. 2). The median time to the THA operation was 57 (17–214) months. Cox regression analyses identified that age ≥ 30 years at the time of the PAO operation was associated with THA reoperation in both uni- (HR 2.8, CI 1.1–7.3) and multivariable (HR 5.2, CI 1.5–18) analyses. Preoperative osteoarthritis (Tönnis grade 2) was associated with THA reoperation in multivariable (HR 24, CI 2.1–247) analysis (Table 2).

Reoperations

A total of 49 (51%) cases were reoperated due to other reasons than a secondary THA. The most common reason for reoperation was extraction of screws (n = 39, 41%), followed by nonunion (n = 6, 6.3%), infection (n = 2, 2.1%), postoperative bleeding (n = 1, 1.0%) and nerve exploration (n = 1, 1.0%). Three patients underwent multiple reoperations: one patient with nonunion was reoperated three times (plate fixation x 2, later extraction of implants), another patient with nonunion was also reoperated three times (plate fixation, replacement of screw penetrating the joint, later extraction of implants) and one patient with nonunion was reoperated twice (extraction of screws, plate fixation) (Fig. 3). The median time to reoperation with extraction of screws was 17 (2-113) months. The median time to the first reoperation due to nonunion was 13 (6–21) months.

Other adverse events

Forty-four (43%) patients had an adverse event. The most common was an injury to the lateral femoral cutaneous nerve (n = 31, 32%), followed by urinary tract infection (n = 4, 4.2%), vascular injury needing vascular intervention (n = 4, 4.2%), superficial infection (n = 1, 1.0%) and femoral nerve injury (n = 1, 1.0%). In all but three of the patients with nerve injury, the symptoms resolved within 2–8 months. The four patients with vascular injury that needed vascular intervention were: three patients with occlusion of the femoral artery that were treated with endovascular thrombectomy on the same (n = 2) or the following (n = 1) day, and one patient with a perioperative injury to the femoral artery that was treated with open surgical repair during the PAO operation.

The main finding of this study was that about one fifth (22%) of the patients were subsequently operated with THA, and the median time to this procedure was close to 5 years (57 months).

Our number of conversion rate to THA was substantially higher than the 6% reported in a systematic review and meta-analysis by Tan [6]. This difference might be explained by their shorter follow-up time; mean 54 months compared to our median 99 months. Accordingly, they reported that longer follow-up time (beyond 6 years) was a negative prognostic factor. Interestingly, they reported that their mean time to secondary THA was 4.7 years, which was actually longer than the follow-up time for the whole cohort, and a number close to ours. This somewhat notable result was most likely caused by the wide range (1-336 months) in their follow-up time, an expected consequence of the systematic review and meta-analysis design of the study. In a later meta-analysis by Ahmad et al., the 5-year survival (regarding THA) after PAO was 96%, and the 10-year survival was 91% [13]. Although encouraging numbers, no results were presented regarding the time to the secondary THA or the actual number of patients that were eligible to be included in the survival analysis at the different time points, making their results somewhat difficult to interpret and compare. In recent single center study by West et al., they reported a somewhat longer time (8 years versus our 5 years) to secondary THA after PAO in a retrospective series of 103 hips [14]. Their mean age at the time of THA surgery was similar to our patients (40 versus 43 years). Moreover, they reported excellent implant survival and good clinical outcomes, concluding that prior PAO does not seem to compromise outcomes in secondary THA. However, contradictive results were reported in a previous systematic review by Shapira et al., who concluded that PAO may entail challenges on subsequent THA surgery, illustrated by higher intraoperative blood loss, lower consistency in cup positioning and compromised patients reported outcomes [15]. All the above presented results highlight the need for long follow-up times if THA surgery is to be used as an outcome. Although not very recently published, Lerch et al. reported 2016 in an impressive long-term follow up that the survivorship was approximately 30% after 30 years in their first 75 patients [16].

We found that high age (≥ 30 years) and preoperative osteoarthritis (Tönnis grade 2) were associated with an increased rate of secondary THA surgery. Although the association with osteoarthritis must be considered as weak (only 2 patients), these findings were not surprising, and these factors are probably considered as relative contraindications for PAO surgery in many institutions. As an example, the inclusion criteria for the meta-analysis by Ahmad et al. stated that only patients aged < 40 years and with Tönnis grade < 3 were included [13]. However, choosing between PAO and THA in young and active patients with hip dysplasia is challenging. In a systematic review and meta-analysis on this topic by Kim and Kim, they reported that the incidences of postoperative complications and revision surgery were not different between THA and PAO groups [17]. However, postoperative pain was less in the THA group, and the activity score was higher in the PAO group. Somewhat contradictive results were reported by Parilla et al. in a retrospective cohort study comparing patients < 40 years after PAO due to dysplasia or THA due to secondary osteoarthritis [18]. The patients were followed for around 10 years, and they found no differences in complications, reoperations or functional scores between the groups. One could still argue that a PAO operation is joint-preserving, and can be followed by a THA if necessary at a later stage in life, whereas an arthroplasty is an end-stage procedure when it comes to the hip joint. Especially aseptic loosening has been a major concern for young patients after THA, and therefore a delay in arthroplasty surgery could be beneficial for the individual patient [19]. In the process of decision making, it is of utmost importance that the individual patient is well informed about specific risks associated with each of the options, for example periprosthetic fractures and dislocations in the case of THA, and osteoarthritis and non-union in the case of PAO.

A total of 6 patients had an LCEA of > 25° on preoperative radiographs. Although the LCEA is commonly used for screening, it can be difficult to measure and the exact value for individual patients must be interpreted with some caution [20]. We found that half (51%) of the patients were reoperated due to other reasons than subsequent THA. This is a high number, but since the absolute majority of these cases were extraction of implants/screws, which might be considered as a minor or even expected event, the rate was probably acceptable. The same was for the other adverse events which affected 43% of the patients, and where the by far most common one was a transient nerve injury of the lateral femoral cutaneous nerve. A slightly less invasive surgical approach was reported 2008 by Troelsen et al. [21], and this might decrease the risk for some of the adverse events that we had in our cohort. Although there is no universal standard on how to report reoperations and adverse events making comparisons between studies difficult, our numbers seem to be in the same magnitude as many other studies [6].

A major strength of this study was the large number of included patients. Another strength was the relatively long follow-up time, allowing for the capture of late as well as early reoperations and adverse events. Furthermore, the extraction of data from the national Swedish Arthroplasty Register using the unique Swedish personal identification number ensured that data on the secondary THA operations was collected with high reliability. There were several limitations with the study, and its retrospective design being an obvious one. In addition, although 8 years follow-up time is relatively long in comparison to most other studies, it could still be of great interest with longer follow-ups as an increasing number of patients could be expected to suffer from complications and subsequent THA operations over time. Another limitation was the possibility that some adverse events could have been missed if they were treated at other hospitals. However, since the Karolinska University hospital is the only hospital in the region treating these patients, the likelihood for this remains limited. Another limitation was the lack of standardized follow-ups with functional outcomes. Finally, this was a consecutive series of patients (except 2 patients who preferred to opt-out). The consecutive design could be both a strength, with no selection of patients, and a limitation as it means that also outlier patients are included, who might skew the overall results. However, we think that our results are highly generalizable and represents a typical large university hospital clinic.

In this eight-year follow-up of 96 consecutive patients the rate of conversion to THA was about one fifth, and in these cases the time to the THA surgery was around five years. Taken together with the relatively low number of serious complications, the PAO procedure is a suitable alternative in young patients with symptomatic dysplasia of the hip in order to avoid, or at least delay, hip arthroplasty.

CI=confidence interval, HR=hazard ratio, LCEA=lateral center-edge angle, PAO=periacetabular osteotomy, THA=total hip arthroplasty

Funding

The study was funded by grants from the Ulla & Gustaf af Ugglas foundation, and the Ulla Hamberg Angeby & Lennart Angeby foundation.

Pun S (2016) Hip dysplasia in the young adult caused by residual childhood and adolescent-onset dysplasia. Curr Rev Musculoskelet Med. Dec;9(4):427-434
Dhaliwal AS, Akhtar M, Razick DI, Afzali A, Wilson E, Nedopil AJ (2023) Current Surgical Techniques in the Treatment of Adult Developmental Dysplasia of the Hip. J Pers Med. Jun 1;13(6):942
Cooperman D (2013) What is the evidence to support acetabular dysplasia as a cause of osteoarthritis? J Pediatr Orthop. Jul-Aug;33 Suppl 1:S2-7
Kosuge D, Yamada N, Azegami S, Achan P, Ramachandran M (2013) Management of developmental dysplasia of the hip in young adults: current concepts. Bone Joint J. Jun;95-B(6):732-7
Ganz R, Klaue K, Vinh TS, Mast JW (1988) A new periacetabular osteotomy for the treatment of hip dysplasias. Technique and preliminary results. Clin Orthop Relat Res. Jul;(232):26-36
Tan JHI, Tan SHS, Rajoo MS, Lim AKS, Hui JH (2022) Hip survivorship following the Bernese periacetabular osteotomy for the treatment of acetabular dysplasia: A systematic review and meta-analysis. Orthop Traumatol Surg Res. Jun;108(4):103283
Bloom DA, Herrero CP, Blaeser A, Castañeda PG. Factors Associated with Patient Satisfaction After Periacetabular Osteotomy (2024) Bull Hosp Jt Dis. Dec;82(4):261-265
Cohen D, Ifabiyi M, Mathewson G, Simunovic N, Nault ML, Safran MR, Ayeni OR (2023) The Radiographic Femoroepiphyseal Acetabular Roof Index Is a Reliable and Reproducible Diagnostic Tool in Patients Undergoing Hip-Preservation Surgery: A Systematic Review. Arthroscopy. Apr;39(4):1074-1087
Du P, Gu Y, Jin W, Li S, Yue Y, Sun H, Yan X (2024) Construction of a predictive nomogram for functional recovery after Bernese periacetabular osteotomy. Front Surg. Jul 26;11:1343823
Tönnis D, Heinecke A (1999) Acetabular and femoral anteversion: relationship with osteoarthritis of the hip. J Bone Joint Surg Am. 81:1747–1770
Ogata S, Moriya H, Tsuchiya K, Akita T, Kamegaya M, Someya M (1990) Acetabular cover in congenital dislocation of the hip. J Bone Joint Surg Br. 1990 Mar;72(2):190-6
Schmitz MR, Murtha AS, Clohisy JC (2020) ANCHOR Study Group. Developmental Dysplasia of the Hip in Adolescents and Young Adults. J. Am. Acad. Orthop. Surg. 28, 91–101
Ahmad SS, Giebel GM, Perka C, Meller S, Pumberger M, Hardt S, Stöckle U, Konrads C (2023) Survival of the dysplastic hip after periacetabular osteotomy: a meta-analysis. Hip Int. Mar;33(2):306-312
West C, Inclan P, Laboudie P, Labbott JJ Sierra RT Trousdale R, Beaulé P, Thornton T, Thapa S, Pashos G, Clohisy JC (2024) Total Hip Arthroplasty After Peri-Acetabular Osteotomy Results in Significant Improvement in Hip Function With Low Revision Rates at Mid-Term Follow-Up. Iowa Orthop J. 44(1):73-78
Shapira J, Annin S, Rosinsky PJ, Maldonado DR, Lall AC, Domb BG (2021) Total hip arthroplasty after pelvic osteotomy for acetabular dysplasia: A systematic review. J Orthop. Apr 21;25:112-119
Lerch TD, Steppacher SD, Liechti EF, Siebenrock KA, Tannast M (2016) Periazetabuläre Osteotomie nach Ganz : Indikationen, Technik und Ergebnisse 30 Jahre nach Erstbeschreibung [Bernese periacetabular osteotomy. : Indications, technique and results 30 years after the first description]. Orthopade. Aug;45(8):687-94
Kim CH, Kim JW (2020) Periacetabular osteotomy vs. total hip arthroplasty in young active patients with dysplastic hip: Systematic review and meta-analysis. Orthop Traumatol Surg Res. Dec;106(8):1545-1551
Parilla FW, Freiman S, Pashos GE, Thapa S, Clohisy JC (2022) Comparison of modern periacetabular osteotomy for hip dysplasia with total hip arthroplasty for hip osteoarthritis-10-year outcomes are comparable in young adult patients. J Hip Preserv Surg. Jul 5;9(3):178-184
Zampogna B, Ferrini A, Zampoli A, Talesa GR, Giusti S, Papalia GF, Vorini F, Papalia R (2025) Total hip arthroplasty in patients under 35 years: a systematic review of the last 2 decades studies. Hip Int. Jan;35(1):92-101
Carreira DS, Emmons BR (2019) The Reliability of Commonly Used Radiographic Parameters in the Evaluation of the Pre-Arthritic Hip: A Systematic Review. JBJS Rev. Feb;7(2):e3
Troelsen A, Elmengaard B, Søballe K (2008) A new minimally invasive transsartorial approach for periacetabular osteotomy. J Bone Joint Surg Am. Mar;90(3):493-8

Table 2 is available in the Supplementary Files section.

No competing interests reported.

Table2RegressionReopTHA251107.docx

Download PDF

Editorial decision: Revision requested
28 Dec, 2025
Reviews received at journal
26 Dec, 2025
Reviews received at journal
22 Dec, 2025
Reviewers agreed at journal
22 Dec, 2025
Reviews received at journal
15 Dec, 2025
Reviewers agreed at journal
02 Dec, 2025
Reviewers agreed at journal
01 Dec, 2025
Reviewers invited by journal
28 Nov, 2025
Editor assigned by journal
13 Nov, 2025
Submission checks completed at journal
13 Nov, 2025
First submitted to journal
07 Nov, 2025

You are reading this latest preprint version

Periacetabular osteotomy of the hip: an 8-year follow-up of 96 consecutive cases

Status:

Version 1

Abstract

Figures

Background

Patients and methods

Statistical methods

Results

Radiographic analyses

Secondary operation with THA

Other adverse events

Discussion

Strengths and limitations

Conclusion

Abbreviations

Declarations

References

Table 2

Additional Declarations

Supplementary Files

Status:

Version 1