The DM design embraces the following concepts: low friction and minimal Charnley wear, a large head to restore anatomy and increase stability proposed by McKee and Farrar, meaning that this design includes a joint that minimises wear issues and a large joint to prevent instability [21]. In fact, with the increase in the "head to neck ratio" the "jump distance" will also rise, thus reducing the risk of dislocation.
The objectives of this design are to reduce wear and loosening forces, to use a physiological system and to increase joint range without compromising intra-prosthetic stability [21,22].
A review of comparative results, where the main parameter to be evaluated is stability, showed that dual mobility cups have excellent short and medium term results compared to standard implants in pTHR [22]. In a prospective cohort study of 143 dual mobility versus 130 standard implants at 4-year FU, Epinette established a statistically significant difference in the dislocation rate in favour of dual mobility (0% vs. 5.4%) [23]; in all cases, the same stem and a cementless cup were fitted with a 28 mm head; there were no cases of mechanical cup loosening in any of the cohorts. Similarly, in a case-control study comparing 105 dual mobility and 215 standard prostheses with a 22 mm head in pTHR, Caton et al. observed a statistically significant difference in both the dislocation rate (0.9% versus 12.9% respectively) and the revision rate (2.1% versus 12.9% respectively) at 10-year [24]. Prudhon et al. found no significant differences in aseptic loosening, infection, or periprosthetic fracture between the two cohorts. The main significant difference was the higher revision rate due to dislocation on the standard bearing cups (17.7%) when compared to dual mobility (4.7%) [25].
Most of the literature on DM is based in France, where its use is more common. In fact, from January 1st 2006 to December 31st 2019, for a total of 45397 primary THAs, in 39.3% of cases, a DM cup was implanted [26]. Data on DM from other global national registers is more limited, even if we are witnessing a progressive increase in its use, as recently stated by the American Joint Replacement Registry (AJRR). Actually, the 2021 AJRR Annual Report reported a statistically significant increase in the use of DM cups for elective pTHR when comparing 2012 to 2020, especially in young patients (< 50 years) [27]. In addition, encouraging early results have emerged from some European national joint registries: comparing 620 dual mobility prostheses with 2,170 standard cemented cups with a 28 mm head in the “Lithuanian Arthroplasty Register”, the 5-year cumulative revision rate was 3.9% in the dual mobility group and 5.2% in the cemented standard prosthesis group [28]. In addition, in the “Dutch Arthroplasty Register”, analysis of 3,038 dual mobility and 212,915 standard hip replacements showed that 0.2% of hip replacements with dual mobility prostheses underwent a revision for dislocation at 5-year FU compared to 0.5% in the standard primary prostheses group [29]. Our cohort, even if limited in number, seems to confirm the very low dislocation rate of dual mobility THA, since after 2 years follow-up we did not recorded even one.
The results of contemporary DM THA have been even more interesting in high-risk patient populations, such as obese patients (defined as a BMI greater than 30 kg/m2). Hernigou et al. reported a statistically significant difference between obese patients who underwent a pTHR: at 7-year FU the dislocation rate of THAs with DM (or constrained liners) was 2%, whereas that of THAs with standard cups was 9% [30]. Furthermore, the use of DM was more effective than preoperative bariatric surgery in reducing the risk of dislocation (dislocation rate of 14% at 7-year FU) [30]. Patients with cerebral palsy or other neurological diseases are also at high risk of instability after THR. This is likely to be secondary to persistent coxa valga, increased femoral anteversion, and associated unbalanced forces generated by the adductor, internal rotator and hip flexor muscles. The review published by Raphael et al. about the use of THR with standard support in patients with cerebral palsy showed a dislocation rate of 14% with a mean FU of 9.7 years [31]. Subsequently, DM was used with promising short-term results: Sanders et al. reported no dislocations in dual mobility hip joint replacements for patients with cerebral palsy at a mean FU of 39 months [32]. Similarly, Morin et al. report that there were no aseptic loosening or dislocations in dual mobility THR performed in 40 patients with cerebral palsy at a mean FU of 5 years [33].
Dislocation represents one of the most insidious and feared complications for neurological patients [34]. Within our cohort, there were 4 patients with neurological disease; 3 of them suffered from Parkinson’s disease, while 1 had a diagnosis of multiple sclerosis. Despite this issue, we have not reported cases of dislocation within our cohort. Moreover, these patients score lower both on preoperative and postoperative HHS, however reporting a substantial improvement in the score.
Several studies testified to the greater risk of postoperative instability among THRs for femoral fracture compared to those implanted for other causes [36–38]. This could be secondary to a combination of muscle failure and a propensity for recurrent falls, representing a life-threatening complication [38]. Consequently, several centres started to employ DM system THAs for femoral neck fractures, with promising initial results. In a population of 105 patients, Tarasevicius et al. reported a statistically significant reduction in the dislocation rate of dual mobility THAs (0/42) compared to standard cups (8/56) during the first postoperative year [39]. Similarly, in a prospective multicentre study of 214 femoral neck fractures treated with DM THAs, Adam et al. reported a dislocation rate of only 1.4% at 9-month FU with 70% of patients returning home without increased dependency [40].
Dual mobility hip replacements have also performed favourably compared to hemiarthroplasty (HA). Bensen et al. retrospectively compared 171 bipolar HAs with 175 dual mobility arthroplasties performed in patients with displaced femoral neck fractures [41]. There was a statistically significant difference in the rate of dislocation, with an incidence of 14.6% among bipolar HAs and of 4.6% for DM arthroplasties [41]. Patient outcome studies are also promising: in the cross-sectional study by Tabori-Jensen et al. 89% of the 124 patients with dual mobility arthroplasties after a femoral neck fracture were satisfied with their surgical outcome, using health-related quality of life questionnaires comparable to the population norm with a mean FU of 2.8 years [42].
In a few cases, we have not achieved a clear improvement in the HHS score, or in any case, an improvement not comparable to the other subjects of the cohort. This was especially due to other osteoarticular pathologies that several patients were suffering from, such as severe osteoarthritis of the contralateral hip or knee osteoarthritis. Whereby, despite having a hip replacement and greatly improving their quality of life, hip replacement alone did not lead to complete resolution of the patients' pain and disability, and they remain with an assisted gait or were unable to walk for a prolonged time, or perform other activities such as walking up and down the stairs unassisted. Therefore we are expecting to achieve lower postoperative HHS in these patients when compared to patients who prior to surgery only suffer from osteoarthritis.
We found that younger patients achieve better results in their postoperative outcome. This issue is not surprising and indeed goes to support what previous studies had already underlined [43,44].
In the study there are two important limitation: the small number of patient and the surgical approach. All the prostheses included in the cohort were implanted with the direct lateral approach of Hardinge. Previous studies have shown that this surgical approach is less prone to postoperative dislocations than others: the posterolateral approach, with or without re-attachment of the short external rotators and/or the posterior joint capsule, is characterized by a higher dislocation rate [11,45].