How to prevent dislocation after revision total hip arthroplasty: a systematic review of the risk factors and a focus on treatment options

The number of revision surgeries is directly related to dislocation rate. Carter et al. [7] found a dislocation rate of 32% in cases of repeated rTHA compared to 15% in cases of first-time rTHA. Similarly, Kosashvili et al. [9] found increased dislocation rates of 35, 46 and 383% after second, third and fourth or more rTHA when compared with first-time rTHA. Moreover, Wetters et al. [4] indicated that patients undergoing revision for instability after a previous rTHA had a 19% (25 of 129) rate of dislocation after the procedure. Recently, Jo et al. [10] found an increased hazard ratio for dislocation after rTHA (1.46) in patients with a history of ≥ 2 previous hip surgeries.

Abductors at the time of revision surgery are estimated and classified as either intact or deficient. The latter is due to abductor post-surgical trochanteric non-union, detachment or myopathy [4]. Alberton et al. [12] reported 7 cases of dislocation in 9 trochanteric non-unions after rTHA from 1548 rTHAs in 1405 patients. Wetters et al. [4], in a retrospective review, identified abductor deficiency as a stronger predictor risk factor for dislocation with an odds ratio (OR) of 2672. Recently, Yoshimoto et al. [13] reported a re-dislocation rate of 18.2% after rTHA (16 re-dislocations in 88 rTHAs), finding a correlation between the rate of re-dislocation and joint laxity, including necrosis of the gluteus muscles, which was identified in 7 of 16 hips.

rTHA performed in a dislocating hip or in patients suffering from previous dislocation has a higher risk of further dislocations.

Wetters et al. [4] found that an unstable implant presented a history of instability in 46% of cases (52 of 113 patients) compared to a stable implant which presented a previous history of instability in 24.2% of cases (251 of 1039 patients).

Similar results were reported by Yoshimoto et al. [11] who found that 20% of patients who had dislocation after rTHA (16 of 178) had a previous dislocation before revision surgery, which was significantly higher than 5.4% patients who did not have a dislocation after rTHA.

Yoshimoto et al. found that a diagnosis ONFH at the time of primary surgery was also an independent risk factor for dislocation after rTHA [11]. The authors suggest that the dislocation could have been induced by the reduced soft-tissue stiffness, which can cause a higher range of motion (ROM) [13], in addition to the primary causes that could have led to osteonecrosis (alcohol and use of corticosteroid).

In cases of severe bone loss, the revision procedure is complicated and troublesome and proper component orientation is not always possible to achieve leading to an increased dislocation rate. Wetters et al. [4] stated that acetabular defect complexity (Paprosky classification of ≥ 3) represents a risk factor because of the difficulties in restoring the centre of rotation of the hip. Proper cup position may also be difficult in these patients due to the altered anatomy.

The effects of age on dislocation have not been fully understood in the available literature. Wetters et al. [4] found that younger patients present a higher risk of dislocation after rTHA, probably due to enhanced activity levels. Patients who had a dislocation subsequent to rTHA were an average of 3 years younger (62 vs 65) than patient who did not dislocate; however, these values did not reach statistical significance.

Conversely, Jo et al. [10] found no differences in dislocation rate after rTHA per 10-year increment; however, again without statistically significant differences.

Furthermore, two multicentre studies [11, 13] found a statistically significant 2.9-fold rise in the risk of dislocation after rTHA for each 10-year age increase. This was related to decreased muscle strength and the increased risk of falling [11].

The size of the femoral head component has been widely accepted as a risk factor for dislocation after rTHA with an increased risk in cases of smaller femoral head diameter.

Alberton et al. [12] found a significant increase in the number of unstable hips after rTHA using a 22-mm head compared to a 26-mm head or larger sizes. Similarly, Wetters et al. [4] found that unstable implants present more frequently with a smaller head size (mean 34.2 mm) compared to stable implants (mean 36.0 mm). Garbuz et al. [16] reported a dislocation rate of 1.1% in the group with a larger head diameter (36 or 40 mm) and 8.7% in cases with a smaller head diameter (32 mm). Recently, Yoshimoto et al. [13] found that 12 of 16 patient (75%) who had a dislocation after rTHA, were revised using a femoral head < 32 mm and that 30 of 72 patients (42.3%) who did not dislocate were revised using a larger femoral head.

In cases of rTHA, the procedure could involve one or both components. A higher dislocation rate was found in the literature in cases of single-component revision. Kosashvili et al. [17] found a significantly higher dislocation rate (10.28%) in cases of isolated acetabular revision compared to revision of both components or femoral stem-alone revision (6.61 and 0%, respectively); however, no statistically significant differences were found.

Conversely, Jo et al. found an increased rate of dislocation after rTHA when the acetabular cup is retained. Other authors [2, 4, 8, 10‐12] found a trend toward an increased dislocation rate when rTHA results in combined component malposition. However, none of those studies reached statistical significance. We did not retrieve any study evaluating the relationship between the combined anteversion techniques and rTHA. Since the combined anteversion technique is an important topic for stability in primary hip arthroplasties [19‐22], further studies addressing this argument could be useful.

Alberton et al. [12] reported a 3.8% dislocation rate after acetabular components revised with an elevated rim liner compared to 8.4% of dislocations after acetabular components revised with a standard rim. Similar results were achieved by Kosashvili et al. [9] although were not statistically significant.

Based on the retrieved literature, it is advisable to use a 36-mm head diameter or larger when performing rTHA.

A jumbo femoral head has a maximal head to neck ratio and this minimizes implant impingement [30].

Larger heads need to be coupled with larger acetabular components; however, the latter potentially produce impingement with iliopsoas muscle or tendon, giving anterior hip pain. When using an increased head size, the polyethylene liner thins to accommodate it, but only to a point (the liner has a certain minimum thickness). Therefore, femoral heads > 38 mm require larger acetabular components so that a liner of at least minimum permissible thickness can be used. Theoretically, the surgeon can decide to ream for a larger cup to accommodate the larger head, creating more acetabular bone loss.

In cases of rTHA with moderate to extensive acetabular bone loss, a jumbo acetabular component represents an effective solution. Several studies reported a high rate of success with up to 90% 10-year survival. However, long-term studies have shown late loosening of ‘first-generation’ porous surfaces and wear associated with periprosthetic osteolysis, caused by standard polyethylene liners. ‘New generation’ polyethylene, as highly X-linked polyethylene and vitamin E-doped polyethylene reduce the rate of wear [31, 32].

In addition, use of larger metal heads have raised concerns regarding potential adverse local tissue reactions (ALTRs), secondary to corrosion and metal release at the head–neck taper junction. Increasing the head size generates larger torsional forces at the trunnion–head junction, and significantly increases the maximal principal stress in the neck medial area, regardless of the material used for the head. These torsional forces enhance tribocorrosion and could lead to ALTRs.

Therefore, the use of large diameter femoral heads is advisable because the biomechanical and clinical data support the finding of increased stability and increased ROM advantages despite the increased bearing surface wear, which may affect implant longevity.

When approaching revision surgery, the liner design must be taken into consideration. An elevated rim increases the area of contact with the femoral head and theoretically guarantees a higher ROM before the head dislocates.

However, a not well-positioned elevated rim liner may increase impingement against the neck of the prosthesis at high degrees of extension/flexion and internal/external rotation, and this could counteract hip stability.

Dual mobility cups are widely reported to reduce the risk of dislocation after rTHA [33‐40]. Dual mobility cups are designed with a 22- or 28-mm diameter femoral head component moving inside a larger polyethylene liner. The liner moves freely in a metal shell with an inner diameter corresponding to the outer diameter of the polyethylene liner. Dual mobility components have been extensively used in Europe for > 25 years [33, 34, 41, 42]. They provide greater ROM and jump distance than a single femoral head. This is due to the additional articular surface. Furthermore, a metal dual mobility component could be cemented into a well-fixed retained acetabular shell.

Some concerns regarding dual mobility implants still remain. Some implants in the past showed intraprosthetic dislocation [43‐45], a specific way of failure due to dislocation between the small femoral head and the mobile polyethylene liner. The dislocation seems to be related to the excessive motion granted by dual mobility implants that may result in impingement of the femoral neck or of the femoral component itself against the large outer polyethylene bearing, leading to polyethylene wear and loss of its inner retentive function.

An additional concern of dual mobility components is whether postoperative motion will continue to occur at both articulations or only at the polyethylene–metal interface.

Despite these issues, dual mobility has a great utility in rTHA. The design provides a low risk of recurrent instability without increasing mechanical complications, particularly when compared with constrained or tripolar cups; namely, a dual mobility liner seems to decrease the risk of dislocation without increasing the risk of loosening [39]. Dual mobility cups also have the advantage of decreasing the need for stem exchange in complex revisions, as they can be used with 22- or 28-mm heads, especially if the head–neck junction is properly designed to interact with dual mobility components. This is shown in several studies in which dual mobility components were used specifically to treat recurrent dislocation [35‐38]. Furthermore, dual mobility implants have also shown good reliability in unstable hemiarthroplasty revision converted to THA [40].

A constrained bearing insert represents another option in cases of multiple failures for dislocation [46, 47]. A constrained bearing insert consists of a polyethylene liner, which includes a 22- or 28-mm femoral head in the inner-diameter concave surface with a locking ring. This liner is inserted in a polished CoCr shell. The shell articulates with another polyethylene liner (the outer bearing) that can be inserted into a standard acetabular shell. However, some studies report a high rate of impingement between the femoral neck and acetabular cup and reduced ROM when a constrained liner is used [47, 48].

Similarly, Jo et al. [10] reported that the use of a constrained liner could be protective against re-dislocation but was not associated with a lower re-revision rate.

In cases of multiple failures, with possible deteriorating health conditions of the patient, surgical options could be limited. An excision arthroplasty of the hip, also known as a Girdlestone procedure [49], should be taken into account despite the severe limb shortening and limp.