Functional Outcomes of Dual-Mobility versus Fixed-Bearing Primary Total Hip Arthroplasty in Ankylosing Spondylitis Patients with Hip Involvement: A Retrospective Cohort Study

Introduction

For most people, movements like walking, sitting, or climbing stairs happen without a second thought. For patients with AS, however, these simple actions can become daily struggles. AS is a chronic, progressive inflammatory disease that primarily affects the spine and sacroiliac joints. Over time, the inflammation may extend to the hips, causing stiffness, severe pain, and in advanced cases, complete fusion of the joint.^1,2 When hip fusion occurs, whether bony or fibrous, the joint loses all mobility, forcing patients to adapt with awkward, exhausting movements that severely impact their independence and quality of life.^3,4

Between 30% and 50% of individuals with AS experience hip involvement, with bilateral hip involvement in 47% to 90%.^2,5,6 THA becomes the most effective option to restore movement, reduce pain, and improve daily function.⁷ Yet, performing THA in the context of AS presents unique technical challenges. Years of inflammation and abnormal posture can alter pelvic alignment, restrict surgical exposure, and increase the risk of postoperative complications, particularly instability and dislocation.^8–10 Research indicates that. When the spinopelvic alignment is fixed, the likelihood of hip dislocation increases.¹¹ In such cases, selecting the most suitable implant is critical for long-term success.

DM implants have been developed to address instability by incorporating a larger effective head size and an additional articulation, which can increase the range of movement and reduce the risk of dislocation.^12,13 These features may be especially valuable in AS patients, who often have stiff spines and altered biomechanics. A systematic review found that DM cups had a strikingly low dislocation rate of 0.2%, compared with 7.1% for unipolar (UP) cups, which share a similar design and function to FB cups.¹⁴ In contrast, FB implants remain a widely used and trusted option, supported by decades of reliable outcomes in standard THA populations.^15,16

While both designs have theoretical advantages, Direct comparative evidence on the use of DM versus FB implants in AS patients with hip fusion remains limited. Most existing studies focus on general arthroplasty populations, where anatomical and biomechanical factors differ substantially. This gap in the literature leaves uncertainty about which implant type offers the most favorable outcomes in this particularly complex group of patients. The present study aims to address this gap by comparing functional outcomes, radiographic parameters, and early complication rates for primary THA between DM and FB with bony or fibrous hip fusion. Through this analysis, provide clearer evidence to guide implant selection and improve surgical planning for one of the most challenging scenarios in hip arthroplasty.

Materials and Methods

This retrospective cohort study included AS patients with hip fusion who underwent primary THA at our institution between January 2019 and December 2023. Hip fusion was diagnosed based on medical history, X-ray, and physical examination, confirming a restricted range of motion, and was further classified as either bony or fibrous fusion.

Inclusion criteria: a confirmed diagnosis of AS, the presence of either bony or fibrous hip fusion, stable AS disease activity at the time of surgery, underwent THA with FB or DM implants, and a minimum follow-up period of six months. Exclusion criteria: follow-up period less than six months, incomplete medical or radiographic records, inadequate imaging quality preventing accurate measurement of cup implantation, previous spine or hip surgeries, and urgent surgeries due to trauma or fracture.

A total of 51 patients met the inclusion criteria. The mean age was 42.3 ± 11.6 years, and the majority were male (43 patients, 84.3%) compared with female (8 patients, 15.7%). In terms of the operated side, 29 patients (56.9%) were on the left side and 22 patients (43.1%) on the right side. Unilateral procedures were performed in 21 patients (41.2%), while 30 patients (58.8%) underwent bilateral procedures. Osteoporosis was documented in 16 patients (31.4%), and the mean number of comorbidities per patient was 2.4 ± 1.3. Patients with severe spinal and hip stiffness were advised to consider a DM prosthesis, but the final choice of implant type was made by the patient after discussing both options. This approach combined clinical guidance with patient autonomy in the prosthesis selection process. All procedures were performed with the patient in the lateral position, using a posterolateral surgical approach. Cementless acetabular cups and stems were implanted in all cases. Intraoperative radiographs were obtained to verify cup positioning and depth. Patients were divided into two groups according to implant type: the DM group (22) and the FB group (29). Subgroups: 20 patients presented with bony fusion, and 31 patients with fibrous fusion.

Clinical and radiographic assessments were performed preoperatively, immediately after surgery, and during follow-up visits. Most imaging evaluations were conducted using standard anteroposterior pelvic X-rays. Functional outcomes were assessed preoperatively and at three and six months postoperatively using, SF-12 and HHS. Disease activity was measured using the BASDAI, which captures several parameters influencing surgical outcomes in AS patients. Continuous variables were expressed as median (variance). Data distribution was assessed, and group comparisons were performed using one-way analysis of variance and independent sample t-tests for normally distributed variables. Statistical significance was defined as p < 0.05. All analyses were performed using SPSS Statistics version 26 (IBM Corp., Armonk, NY, USA). The study was approved by the Institutional Ethics Committee.

Results

The baseline characteristics of the FB and DM groups were broadly similar, with no statistically significant differences observed in any variable. In the FB group, 22 patients (75.9%) were male and 7 (24.1%) were female, while in the DM group, 21 patients (95.5%) were male and 1 (4.5%) was female (p = 0.12). The mean age in the FB group was 44.2 ± 10.7 years compared with 39.7 ± 12.5 years in the DM group (p = 0.09). Regarding the surgical side, left-sided accounted for 58.6% and right-sided for 37.9% in the FB group, compared with 50.0% left-sided and 45.5% right-sided in the DM group (p = 0.49). For procedure type, bilateral procedures were performed in 69.0% and unilateral in 31.0% of FB patients, compared with 45.5% bilateral and 50.0% unilateral in the DM group (p = 0.16). Osteoporosis was present in 27.6% of FB patients and 31.8% of DM patients (p = 0.66). The mean number of comorbidities was 3.0 ± 2.3 in the FB group and 2.5 ± 2.1 in the DM group (p = 0.39). These findings indicate that the two groups were well matched at baseline.

In this study, the DM and FB groups were assessed at separate time points. At baseline, there was no meaningful difference between the two groups. Regarding this study scoring system, lower scores indicate better function and recovery, while higher scores reflect poorer outcomes. This is important to clarify, as it differs from other studies. The figures presented results were designed specifically to reflect this scoring direction, making it visually clear that lower values correspond to improved patient performance over time. This approach ensures consistency with the way the data were collected and analyzed in this study. Additionally, according to the data, there were no readmissions within the follow-up period for either group due to dislocations or infections.

Tables 1 and 2 compare various surgical and clinical parameters across different patient groups. In most measures, such as leg length, pelvic tilt, anteversion, inclination, blood loss, surgery duration, cup size, head size, and femoral stem size, there were no significant differences between the groups. However, in Table 2, the only statistically significant difference was in the stiffness of the spine, which was more common in the bony fusion group. Overall, these findings suggest that while most surgical parameters were similar between groups, a few differences in implant sizing and spine stiffness were observed. Table 3 summarizes the baseline demographic and clinical characteristics.

Table 1 The Initial Phase is Comparing DM and FB Implants. This Was Done by Analyzing the Mean Values of Clinical and Radiographic Variables Used to Assess Hip Involvement in AS Patients

Table 2 To Evaluate Hip Involvement in AS Patients, We Determined the Mean Clinical and Radiographic Scores and Subsequently Used Them to Compare Cases of Bony Fusion Against Fibrous Fusion

Table 3 Baseline Demographic and Clinical Characteristics of the Study Cohort

Functional Outcomes

Preoperatively, SF-12, HHS and OHS scores were compared between the groups. At 3 months, DM patients showed significantly better results, with lower median HHS (14.0 [12.0–15.8] vs 18.0 [15.0–20.0]; p = 0.001), OHS (17.0 [15.0–21.0] vs 30.0 [21.0–32.8]; p < 0.001), and SF-12 scores (26.5 [25.0–29.0] vs 30.0 [29.0–32.8]; p < 0.001) compared with FB. At 6 months, this advantage persisted, with DM maintaining lower HHS (12.5 [12.0–13.0] vs 15.5 [13.0–16.8]; p = 0.001), OHS (14.0 [12.8–18.0] vs 21.5 [20.0–26.5]; p < 0.001), and SF-12 scores (19.0 [17.0–21.0] vs 26.0 [24.0–29.8]; p < 0.001), indicating sustained functional superiority.

BASDAI scores were similar between groups at baseline (89.0 [59.0–93.5] vs 92.5 [41.0–151.8], p = 0.934). At 3 months, both groups improved, with the FB group showing a slightly lower median BASDAI score (40.0 [22.3–55.8] vs 52.0 [13.5–73.5]), suggestive of better early symptom control, though not statistically significant (p = 0.055). By 6 months, BASDAI scores continued to improve in both groups, with no meaningful difference between DM (32.5 [8.3–75.3]) and FB (44.0 [26.8–80.3], p = 0.507).

Figures 1 and 2 show pre- and post-operative X-rays used in this study to measure clinical parameters for each patient. Image (A) in both figures displays the pelvis pre-operative, showing bony fusion of the hips. Image (B) shows the post-operative view following bilateral THA, using either DM implants (Figure 1) or FB implants (Figure 2). These radiographs were used to evaluate surgical outcomes and to compare results between the two implant types.Across all follow-up periods, both the DM and FB groups showed marked functional improvement after surgery. However, DM consistently achieved better outcomes, with HHS, OHS, and SF-12 scores at 3 and 6 months, indicating faster recovery and greater functional gains. BASDAI scores improved in both groups, with a slightly greater early reduction in the FB group at 3 months, although this difference did not reach statistical significance, and by 6 months, the scores were similar (Figures 3–6). When comparing fusion types, patients with bony fusion generally experienced greater improvements in OHS, HHS, and SF-12 scores than those with fibrous fusion, suggesting better functional recovery. In contrast, BASDAI scores improved in both groups, but the fibrous fusion group showed a slightly greater reduction over time, indicating better perceived symptom relief, although the difference was not significant (Figures 7–10).

Figure 1 (A) An anterior-posterior pelvic radiograph from a typical case, taken pre-operatively, reveals bony fusion. (B) Subsequent radiological follow-up after the patient underwent bilateral DM total Hip arthroplasty.

Figure 2 (A) An anterior-posterior pelvic radiograph from a typical case, taken pre-operatively, reveals bony fusion. (B) Subsequent radiological follow-up after the patient underwent bilateral FB total Hip arthroplasty.

Figure 3 Comparative analysis of HHS scores across multiple time points demonstrated significant improvement in the DM group.

Figure 4 The DM group showed a more significant improvement in OHS scores than the other group when measured across various time intervals.

Figure 5 A comparison of SF-12 scores at different time periods revealed a more substantial improvement, particularly in the DM group.

Figure 6 The DM group showed a greater improvement in BASDAI scores across the follow-up periods compared to the other group.

Figure 7 A comparison of OHS scores at various time intervals revealed significantly greater improvement in the bony fusion group compared to the fibrous fusion group.

Figure 8 A comparison of HHS scores at various time intervals revealed a more significant improvement in the bony fusion group compared to the fibrous fusion group.

Figure 9 A comparison of SF-12 scores at different time periods revealed a more substantial improvement, particularly in the bony fusion group.

Figure 10 A comparison of BASDAI scores at various time intervals revealed greater improvement in the fibrous fusion group compared to the bony fusion group.

Discussion

AS may seem uncommon, but it affects people worldwide, typically between 0.1% and 1.4% of the population, depending on region and ethnicity. AS usually starts in early adulthood, often between ages 15 and 40.^17,18 This variation is largely due to how common the HLA-B27 gene is in different populations, which influences who is more likely to develop AS. A recent long-term study in South Korea found that AS cases climbed from about 27 per 100,000 people in 2010 to almost 82 per 100,000 in 2023, that’s a threefold rise over just over a decade.¹⁹ This increase is likely driven by improved disease detection, growing public and medical awareness, and the cumulative effect of new cases each year. Older estimates suggest continent-level differences. For instance, Europe saw 24 cases per 10,000 people, while Asia had about 16–17, North America had approximately 32, Latin America around 10, and Africa about 7 per 10,000.^20,21 Although men are more frequently diagnosed, new studies show that the gap between men and women may be narrowing, possibly due to better recognition and diagnosis of the disease in women.^17,21

As the prevalence of THA continues to rise in the AS population, studies confirm that even patients with pre-existing hip fusion can achieve excellent functional outcomes post-surgery.^22,23 According to a study by Hamilton et al, the primary factor determining post-operative functional outcome after joint replacement is a reduction in limb discomfort.²⁴ Some scientists further argue that patient satisfaction following THA is primarily determined by improvements in claudication gait, the remission of residual pain, and overall functional enhancement.^25,26 This study compared functional and clinical outcomes between DM and FB THA in patients with AS and hip fusion, a surgical scenario known for its technical complexity and elevated risk of instability. Both implant designs produced substantial improvements in mobility, pain relief, and quality of life over the six-month follow-up period. However, Patients receiving DM implants demonstrated significantly higher HHS, OHS, and SF-12 scores at both 3 and 6 months postoperatively compared with those receiving FB implants. While the theoretical biomechanical advantages of DM constructs may apply broadly across THA populations, these benefits may be particularly relevant in patients with AS due to their characteristic spinal rigidity and fixed pelvic alignment. However, given the absence of a comparative non-AS cohort in the present study, disease-specific mechanisms should be interpreted cautiously. Importantly, this interpretation is based on functional outcome measures and does not imply superior long-term implant performance. A recent multi-center investigation comparing DM and FB prostheses found that patients receiving DM implants experienced an enhanced range of motion, along with a reduced incidence of dislocation, readmission, and revision compared to those with FB implants.²⁷ A recent publication supports the use of DM bearings in younger patients, who typically require higher-performing implants due to their greater physical activity levels and longer life expectancy.²⁸ Historical data consistently shows the superiority of DM implants over FB designs, with advantages in cost-effectiveness, dislocation rates, and overall survivorship.²⁹

Our findings are consistent with previous reports suggesting that DM implants may reduce the risk of instability and support improved early functional recovery, particularly in patients with altered biomechanics or fixed spinopelvic alignment. In patients with AS, spinal stiffness and abnormal pelvic tilt restrict normal compensatory motion, which may increase susceptibility to impingement and functional instability following THA.³⁰ Within this biomechanical context, the design characteristics of DM implants may contribute to improved early functional stability; however, these potential advantages are not exclusive to AS and should be interpreted within the broader framework of implant biomechanics. This interpretation focuses on functional performance And patient satisfaction rather than dislocation prevention alone and remains consistent with the outcomes measured in the present study. While FB implants remain a reliable choice in standard THA. Interestingly, BASDAI scores improved similarly in both groups, with the FB group showing a slightly greater early reduction at 3 months, although this was not statistically significant. This may reflect variations in subjective symptom perception rather than true differences in joint function, as objective functional scores consistently favoured DM. The lack of difference in BASDAI at six months suggests that both implant types ultimately achieve comparable control of disease-related symptoms over time. When stratified by fusion type, patients with bony fusion demonstrated greater improvements in functional scores than those with fibrous fusion. This could be due to the more predictable restoration of joint mechanics after replacing a fully fused hip, compared with the subtler biomechanical changes in fibrous fusion. Conversely, patients with fibrous fusion reported slightly better improvements in BASDAI, possibly because they began with more residual mobility and experienced greater subjective relief from pain and stiffness.

This study has several strengths, including a well-defined patient cohort, direct comparison of two widely used implant types, and the use of multiple validated functional outcome measures. However, limitations should be acknowledged. The retrospective design introduces potential selection bias, particularly as patients were involved in implant choice after surgical counselling. The sample size, while adequate for detecting major functional differences, limits subgroup analyses. Additionally, the six-month follow-up captures only short-term outcomes; longer follow-up is needed to determine whether the early advantages of DM implants persist and translate into reduced long-term complication or revision rates.

Future prospective studies with longer follow-up are warranted to confirm these findings and to explore whether DM’s early benefits lead to improved long-term implant survival and patient satisfaction.

Conclusion

Both DM and FB prostheses led to substantial functional improvement in AS patients after THA. DM implants showed consistently better HHS, OHS, and SF-12 scores at 3 and 6 months, suggesting faster recovery and greater functional gains. BASDAI scores improved in both groups, with no significant long-term difference. Bony fusion patients achieved greater functional recovery, while fibrous fusion patients reported slightly better symptom relief.