Tripolar Constrained and Dual Mobility Liners Are Both Successful When Cemented Into Tantalum Acetabular Revision Shells in Complex Revision Total Hip Arthroplasty.

BACKGROUND: Instability is a common complication after revision total hip arthroplasty. Tripolar constrained (TC) and dual mobility (DM) liners cemented into tantalum acetabular revision (TM) shells are established alternatives that reduce instability risk. This study compares outcomes of TC and DM liners cemented into TM shells in complex revision hip replacements.
METHODS: Fifty cases using a TM shell and a TC or DM cemented liner with at least 12 months of follow-up were identified. There were 25 TC and 25 DM liners. Clinical, radiographic, and patient-reported outcomes were collected. Reasons for re-revision and failures were analyzed.
RESULTS: The average age was 73 years in the TC group and 75 years in the DM group. The median follow-up duration was 50 months (23 - 96) and 13 months (12 - 21) for the TC and DM groups, respectively. The mean Harris Hip Score improved from 37 to 61 for the TC group and from 51 to 73 in the DM group at the last review. Two TC cases required revision for wear and liner failure. One DM insert was changed at reoperation for periprosthetic femur fracture. There were no failures of cement fixation in either group and no revisions for cup loosening.
CONCLUSIONS: In complex revision hip replacement with a history of instability or judged to have a high risk of instability postoperatively, TC and DM liners cemented into tantalum revision shells are both successful, giving acceptable results. The improved arc of movement with DM liners, minimizing impingement, makes them preferable except when there is a global abductor deficiency or severe joint laxity.

Introduction

Although total hip arthroplasty (THA) is extremely successful [1,2], dislocation remains a problem, with an incidence from 0.2% to 10% in primary THA and from 10% to 28% after revision THA (rTHA) [3]. Berry et al. [4] reported that in primary hip arthroplasty, the cumulative risk of dislocation is 1% at 1 month and 1.9% at 1 year and rising at a constant rate of approximately 1% every 5 years to 7% at 25 years. Many are single episodes that do not require a revision. Surgery is indicated when recurrent, or where a surgically correctable error has been identified.

Instability is the most common indication for re-revision after rTHA (Re-rTHA), accounting for 20%-35% of all re-revisions [3,[5], [6], [7]]. The Australian Orthopedic Association National Joint Replacement Registry 2020 annual report identifies dislocation as the most common reason for re-revision comprising 33.7% of second revisions [8]. Multiple factors are related to dislocation after revision hip replacement. These include age [9], number of prior revisions [3,9], reason for current revision, component orientation, and status of the abductor mechanism [9,10].

Large heads, constrained liners, unconstrained tripolar constructs, tripolar constrained (TC) liners, and dual mobility (DM) liners have all been used in an attempt to minimize the risk of instability after rTHA [[11], [12], [13]].

Tantalum trabecular metal (TM) acetabular revision shells, with and without augments, have performed well, particularly in moderate to severe acetabular defects [[14], [15], [16], [17]].

A reduction in dislocation risk is seen after rTHA using DM and TC liners [18,19]. We have cemented them into TM shells in complex revision surgeries since 2003. To our knowledge, there are no studies directly comparing TC and DM cemented liners in either rTHA or Re-rTHA.

The aim of this single-surgeon retrospective observational cohort study was to evaluate and compare clinical and radiographic outcomes, as well as reoperation and revision of TC and DM liners cemented into TM shells in complex revision hip replacement.

Material and methods

This is a single-surgeon (N.R.B.), single-institution retrospective case series study. An analysis of prospectively collected data from our Human Research Ethics Committee-approved outcome database was performed. Informed consent was obtained from every patient. Between November 2003 (first cemented TC liner) and January 2019, 52 hip revision procedures (in 50 patients) were identified that had used a combination of a TM shell, with or without augmentation (Trabecular Metal Acetabular Revision System; Zimmer Biomet, Warsaw, IN) and either a cemented DM liner (Avantage; Zimmer Biomet, Warsaw, IN) (Fig. 1) or a cemented TC liner (Trident Constrained Acetabular Insert; Stryker, Mahwah, NJ) (Fig. 2) and had ≥1 year of follow-up. Two patients were excluded from analysis. One died before the 12-month review, from an unrelated cause. The other patient was lost to follow-up before 12 months but died 10 years later with the implants unrevised.

Figure 1

Avantage DM liner.

Figure 2

Tripolar constrained liner.

All patients underwent revision of the acetabular cup, liner and head, with or without revision of the femoral component. The outcome database (http://socratesortho.com), medical records, and radiographs were reviewed. Age at surgery, side, indication for revision, previous surgery, body mass index, and Charnley class were recorded. The Harris Hip Score (HHS) [20] was used as a hip-specific outcome. Data were collected preoperatively, at each postoperative review, and at the latest follow-up. All radiographs were reviewed for this analysis.

Operative details included the side and duration of surgery, components revised, and implants used. Acetabular bone defects were categorized according to Paprosky classification [21]. After the acetabular reconstruction was performed, empty holes and screw heads were filled with bone wax. Antibiotic-loaded (gentamicin) acrylic cement was hand-mixed in all cases and inserted early into the shell followed by the acetabular liner (TC or DM). Liner position was adjusted during cementation to accommodate any minor shell malposition (Figure 3, Figure 4, Figure 5, Figure 6, Figure 7).

Figure 3

Implanted Trabecular Metal Acetabular Revision System (TMARS).

Figure 4

Cement into cup.

Figure 5

Cemented DM liner.

Figure 6

DM reconstruction.

Figure 7

TC reconstruction.

We identified postoperative complications, reoperations, and revision procedures. The Australian Orthopedic Association National Joint Replacement Registry was queried to identify patients who may have undergone unidentified re-revision THA elsewhere, or who were deceased.

A history of instability, the presence of abductor deficiency (soft tissue or bony), as well as global capsular deficiency were definite indications for cemented TC or DM inserts. Two-stage revision for infection has a higher incidence of instability after reimplantation [22], and TC or DM liners were often used.

Summary characteristics are presented using mean (minimum, maximum), median (lower and upper quartiles), and number (percent). Between-group comparisons of continuous variables using unpaired t test or difference in medians according to distribution and of categorical variables using likelihood ratio Chi-squared test were performed. Results are presented as point estimate of difference (95% confidence interval [95% CI]) and the associated P value. Significance level was two-sided and set at 0.05. No adjustment was made for within-subject correlation where multiple procedures were performed on the same patient, nor for multiple comparisons. Stata v1 6 statistical software was used to perform the analysis (StataCorp. 2019. Stata Statistical Software: Release 16.; College Station, TX) and the bpmedian program [23].

Results

Fifty complex revision hip procedures using a TM shell and a TC or DM cemented liner were analyzed. Two patients had bilateral, but staged, revision procedures: One received TC liners (interval 7 months), and one received DM liners (interval 3 months). There were 25 cemented TC and 25 cemented DM liners that completed a minimum of 12 months of follow-up. The median follow-up duration was 50.0 (23.0 – 96.3) months in the TC group and 13.2 (12.0 – 21.4) months in the DM group; median difference 36.9 (95% CI 8.2 to 65.6, P = .02) months. Follow-up was significantly longer in the TC group.

Nine unrelated deaths occurred during the follow-up period, 6 in the TC and 3 in the DM group. None had been Re-revised. Patient characteristics are presented in Table 1. There was no significant difference between groups in sex, age, and body mass index or number of previous arthroplasty procedures. A trend toward a higher number of arthroplasty procedures was seen in the TC group, although not significant.

Table 1

Preoperative patient characteristics.

Patient characteristics	Tripolar constrained	Dual mobility	P value
	N = 25	N = 25
Age (y)	73 (42, 91)	75 (43-88)	.57
Gender (male)	13 (52.0%)	12 (48.0%)	.78
Body Mass Index (kg/m2)	31.6 (21.1-51.6)	28.8 (17.8-66.6)	.32
Normal	6 (26.1%)	5 (20.0%)
Over weight	3 (13.0%)	10 (40.0%)
Obese	12 (52.2%)	7 (28.0%)
Morbidly obese	2 (8.7%)	2 (8.0%)
Super obese	0 (0.0%)	1 (4.0%)
Previous arthroplasty procedures	3 (2-4)	2 ([1], [2], [3])	.07
1	6 (24.0%)	9 (36.0%)
2	5 (20.0%)	8 (32.0%)
3	7 (28.0%)	5 (20.0%)
4	3 (12.0%)	1 (4.0%)
5	2 (8.0%)	1 (4.0%)
7	0 (0.0%)	1 (4.0%)
8	1 (4.0%)	0 (0.0%)
12	1 (4.0%)	0 (0.0%)

Data are given as mean (minimum, maximum), median (lower quartile – upper quartile), and number (%).

The acetabular component was revised in all 50 hips. The femoral component was revised in only 29 cases (58%); 17 of 25 (68%) in the TC group and 12 of 25 (48%) in the DM group. Indications for revision surgery included a history of instability in 8 (16%) and infection in 15 (30%). Table 2 summarizes the surgical details.

Table 2

Surgery details.

Surgical variables	Tripolar constrained	Dual mobility	P value
	N = 25	N = 25
Duration (mins)	275 (235 - 345)	230 (210 - 270)	.10
Side (right)	16 (64.0%)	19 (76.0%)	.35
Charnley Class
A	3 (12.0%)	2 (8.0%)
B1	1 (4.0%)	3 (12.0%)
B2	2 (8.0%)	6 (24.0%)
C1	14 (56.0%)	0 (0.0%)
C2	1 (4.0%)	9 (36.0%)
C3	4 (16.0%)	5 (20.0%)
Components replaced
Partial	8 (32.0%)	13 (52.0%)	.15
All	17 (68.0%)	12 (48.0%)
Diagnosis at revision
Infection	9 (19.1%)	6 (10.2%)
Cup loosening	15 (31.9%)	11 (18.6%)
Dislocation	4 (8.5%)	4 (6.8%)
Fracture	2 (4.3%)	2 (3.4%)
Implant breakage	2 (4.3%)	6 (10.2%)
Pain	9 (19.1%)	13 (22.0%)
Subsidence	3 (6.4%)	8 (13.6%)
Wear, no loosening	3 (6.4%)	9 (15.2%)
Acetabular defect (Paprosky)
1	0 (0.0%)	1 (4.0%)
2a	6 (24.0%)	9 (36.0%)
2b	1 (4.0%)	0 (0.0%)
2c	1 (4.0%)	3 (12.0%)
3a	9 (36.0%)	6 (24.0%)
3b	8 (32.0%)	6 (24.0%)
Acetabular shell (mm)	66 (56, 78)	66 (58, 80)	.80
Acetabular augments	13 (52%)	12 (48%)	.78
Cup-cage construct	6 (24%)	4 (16%)	.48

Data are given as mean (minimum, maximum), median (lower quartile – upper quartile), and number (%).

All patients had improved HHS at follow-up. Raw data means (see Table 3) and associated pointwise 95% confidence limits are presented in Figure 1. The HHS were different preoperatively, with a mean difference of 14 (95% CI -1 to 27, P = .03) higher in the DM group. Unadjusted for baseline scores, the mean postoperative HHS score at 12 months was 13 (4 to 21, P = .005) points higher in patients who received a DM liner.

Table 3

Outcomes by liner type.

Outcome	Tripolar constrained	Dual mobility	P value
Reoperation			.83
Minor	2 (8.0%)	2 (8.0%)
Revision	2 (8.0%)	1 (4.0%)
Loosening	0	0
HHS
Preoperative	37 (21, 78)	51(21, 98)	.03
12-mo Postoperative	61 (31, 92)	73 (77, 98)	.005
Deceased	6 (24.0%)	3 (12.0%)	.26

HHS, Harris Hip Score.

Data are given as mean (minimum, maximum), median, (lower quartile – upper quartile), and number (%).

The restricted cubic spline regression model indicates that the estimated mean HHS is 14.1 (95% CI 2.2 to 26.0, P = .02) points higher in patients with DM liners (Fig. 2).

Two patients in each group underwent early reoperation: 1 for acute infection with a new organism and another for a sterile hematoma in the TC group and 2 for sterile hematomas in the DM group.

Three patients underwent rerevision of components. One DM case had a polyethylene insert change at revision for periprosthetic femur fracture. Two patients in the TC group required late revision for wear and liner failure after 7 and 9 years. One of these had subluxation without dislodging the constraining ring. Radiographic review showed no cup migration, no mechanical breakage, and no progression of radiolucent lines. There was no cemented liner loosening.

Discussion

Instability is the most frequent postoperative complication after revision THA [6] with a cumulative risk after 15 years of up to 35% [7,8]. With increasing numbers of primary procedures, and increasing life expectancy, revision surgery numbers are expected to increase over time [24]. Despite all improvements, dislocation continues to be a devastating complication after rTHA.

Some risk factors have been identified. Abductor deficiency (soft tissue or bony), capsular laxity or deficiency, number of previous surgeries, history of instability, infection, and Paprosky acetabular defect type 3 have all been implicated [7,10,22,25]. This informs choice at rTHA. Eight patients (16%) in our study had instability as the indication for revision, whereas 15 (30%) cases were 2-stage revisions for periprosthetic infection. This latter has a high risk of instability, with a 9% probability at 1 year [22].

The use of large femoral heads, constrained liners, unconstrained tripolar articulations, and DM and TC liners has been reported to address instability risk. Each liner type has its advantages and disadvantages [[11], [12], [13],18,19,26].

Constrained acetabular liners (CAL) capture the femoral head [13]. This results in greater forces transmitted to the implant-bone interface, increasing loosening. CAL have a reduced arc of movement. As a result, impingement, resulting in mechanical damage to polyethylene and the locking ring, is another failure mechanism [27]. Poor results and catastrophic failures of the CAL [13,[28], [29], [30]] prompted the TC acetabular liner (Fig 3a and 3b), first reported by Capello et al. [31,32]. A bipolar component is constrained in a polyethylene liner [33]. This results in a larger diameter articulation and an increased range of motion.

DM implants have a larger range of motion than TC bearings. They have emerged as an alternative in the management and prevention of hip prosthesis instability (Fig 4a and 4b). The DM liners have a larger diameter (>36 mm). Cementing into the TM shell allows adjustment to accommodate shell position. Restoration of offset and tension in the abductor mechanism are facilitated [33]. In our DM group, there were no dislocations and no re-revisions. This corroborates the findings of Brüggemann A. et al. [34].

This study demonstrates that both DM and TC bearings are successful at short and midterm follow-up, with low risk of complications and a very low risk of instability (2%), providing good alternatives for addressing cases of complex rTHA and Re-rTHA where there is a high risk of instability. The TM revision acetabular shell is a reliable substrate into which these liners are cemented. No loosening or migration of the cup was seen even in patients with failed liners.

Both groups had similar characteristics. Their follow-up, however, was significantly different (P = .002). The TC group had a mean follow-up duration of 59.36 months (12 - 132) compared with 17.32 months (12 - 45) in the DM group. A cemented DM option was not readily available in the early years, so the TC liner was used. As cemented DM liners appeared, they were used more frequently. As a result, the TC follow-up is longer.

The HHS is a reliable tool for assessing clinical outcome in rTHA [35]. Both groups improved, but there was no significant difference, between groups, in that improvement. The DM group had higher preoperative HHS than the TC group. The difference in HHS between groups is entirely accounted for by the baseline (preoperative) difference.

Both groups showed a mean Minimal Clinically Important Improvement in their mean HHS at 12 months compared with the mean baseline scores, according to Singh et al. [36]. Five patients in each group had moderate improvement. These findings suggest that both TC and DM liners in complex revision THA allow a significant clinical improvement.

The higher preoperative HHS in the DM group was examined. There were fewer previous revisions, and less complex revision surgeries was needed in the DM group (Table 1) TC cases were more likely to have a global abductor deficiency. This affects adversely the HHS functional score.

TC liners are successful at 10 years in revision THA for patients with a high risk of instability [37]. Longer follow-up was possible in this group. TC cases had, overall, more complex surgeries. There were 2 late revisions because of wear and liner failure at 7 and 9 years. Although numbers are small, we suggest that TC liners cemented into TM revision cups are reliable.

This is the only report published to date attempting to compare DM to TC liners in complex revisions. Several limitations are acknowledged. The small number of patients in each group makes it impossible to make some statistically significant conclusions. The decision on bearing type used was made on a case-by-case basis. TC patients undergo usually more complex surgeries, and this introduces a selection bias to the disadvantage of the TC group. There was a preference for TC liners in cases with severe abductor deficiency or generalized capsule laxity.

With a longer and similar follow-up for each group, data on more complications, specifically dislocations, may have been collected.

Conclusion

In conclusion, this study supports the use of both DM and TC liners cemented into trabecular metal acetabular revision shells as safe alternatives for patients undergoing complex revision hip replacement.

Institutional Review Board statement

This study was approved by the local institutional review board.

Conflicts of interest

N. R. Bergman is the President of Arthroplasty Society of Australia and the Chair of AOA National Joint Replacement Registry Committee.