Similar Outcomes Achieved Between Anterior and Posterior Approach Total Hip Arthroplasty Using Dual Mobility Implants

Abstract

Background

Dual mobility (DM) bearings for total hip arthroplasty (THA) have been proposed to reduce the risk of instability in high-risk patients; however, their utility in primary THA remains relatively unexplored. No previous reports have described whether surgical approach influences outcomes associated with DM implant systems. This study aims to compare patient reported outcomes and post-operative groin pain between patients undergoing anterior approach versus posterior approach following primary THA with DM implants.

Methods

We retrospectively reviewed all patients who underwent primary THA and received a DM implant between 2011-2021. Patients were stratified into two cohorts based on surgical approach (anterior vs. posterior approach). Primary outcomes included the presence of substantial postoperative groin pain as well as readmission and revision rates. Demographic differences were assessed using chi-square and independent sample t-tests. Outcomes were compared using multilinear and logistic regressions.

Results

Of the 495 patients identified, 55 (11%) underwent THA via the anterior approach and 440 (89%) via the posterior approach. Surgical time (100.24 vs. 109.42 minutes, p=0.070), length of stay (2.19vs.2.67 days,p=0.072), discharge disposition (p=0.151), and significant postoperative groin pain (1.8%vs.0.7%,p=0.966) did not statistically differ between the cohorts. 90-day readmission (9.1%vs.7.7%,p=0.823) and revision rate (0.0%vs.3.0%,p=0.993) did not significantly differ as well. Specifically, readmission (p=0.993) and revision (p=0.998) for instability did not significantly differ between the cohorts. We found no statistical difference in HOOS, JR (p=0.425), VR-12 PCS (p=0.718), and VR-12 MCS (p=0.257) delta score improvement from preoperative to 1-year follow-up between the two groups.

Conclusion

Comparable outcomes following implantation of DM constructs may be achievable irrespective of the surgical approach employed. The incidence of iliopsoas injections for groin pain did not significantly differ between anterior and posterior approaches. Future investigation is needed to determine whether surgical approach influences long-term outcomes in patients receiving DM implants.

Level of Evidence: III

Introduction

Postoperative dislocations represent a significant complication after total hip arthroplasty (THA). Approximately 0.2-7% of the patients experience instability following primary THA and 10-25% after revision THA.¹ Previous reports suggest that the risk increases by approximately 1% every five years following the initial procedure.^2,3 As a result, there have been many improvements in surgical techniques and implant designs in an effort to help prevent hip instability. Dual mobility (DM) bearings have demonstrated excellent short and midterm outcomes compared to standard bearing implants in reducing dislocation rates, especially in higher-risk patients.^4-8 However, a recent meta-analysis found that DM articulations were not associated with lower overall revision rates compared to fixed bearing constructs.⁷ DM constructs comprise of two articulations: one between the femoral head and a large-diameter polyethylene (PE) outer head bearing and a second between the PE outer head bearing and the acetabular shell or acetabular liner, thus providing a greater range of motion and a greater jump distance before dislocation occurs.⁸ Although the use of these implants was initially limited due to concerns of excessive polyethylene wear, progressively improved designs have conferred excellent long-term results.¹⁰ However, controversy regarding the use of modular DM implants remains as French et al.⁹ found that modular DM implants can cause adverse reactions due to metal corrosion between the metal liner and the metal shell in up to 1.8% of cases, though it only occurred in 0.3% of cases on average.

The direct anterior approach (DAA) and posterior approach (PA) are two commonly used surgical techniques in THA. Previous studies have shown that pairing the DAA with DM bearings does not increase complications or worsen implant positioning¹¹ and is a safe and effective strategy in increasing postoperative stability and reducing dislocations in patients undergoing THA.^12-16 Similar studies employing the PA with DM implants show comparable results, such as increased postoperative stability and reduced dislocation rates.^17-19 However, some studies have raised concerns about higher rates of postoperative groin pain following the use of DM implants via DAA, which may be attributed to the removal of the capsule and irritation of the iliopsoas muscle due to the use of larger implant heads.^14,16

Although previous studies have described that both surgical approaches with DM implants have positively influenced patient outcomes in those with a high risk of instability, there is a paucity in the literature describing whether the surgical approach affects outcomes in patients receiving DM acetabular implants. In their systematic review of outcomes of DM components in THA, Darrith et al.²⁰ proposed that further study was required to determine whether the surgical approach affects the outcomes and rates of dislocation after DM use in THA. Although the use of larger heads decreases the rate of dislocation, particularly in patients in whom a PA is used, their systematic review could not identify an independent effect of surgical approach without patient-level data, which provides the impetus for the present study.

Therefore, the purpose of this study is to evaluate outcomes between patients undergoing primary THA with DM implants via DAA versus PA. Outcomes of interest include the incidence of requiring iliopsoas injections for postoperative groin pain as well as readmission and revision rates. We hypothesize that the differences between surgical approaches will be negligible during the early postoperative period.

Methods

We retrospectively queried all patients over the age of 18 who underwent primary THA with DM constructs between June 2011 and February 2021 at a single urban institution, which comprises a large academic center and a tertiary orthopedic specialty hospital. Patients were separated into two cohorts based on surgical approach: DAA vs. PA. Patients undergoing revision THA, as well as THA performed for non-elective or oncologic reasons were excluded from this analysis. Femoral stems were used at the discretion of the surgeon. Of the 30 surgeons who contributed at least one case to our analysis, 24 completed an additional year of fellowship training in adult reconstructive surgery. Of the 495 DM cases identified, 145 (30%) were skirted stainless steel (POLARCUP; Smith and Nephew, Memphis, Tennessee), 40 (8%) were anatomic cobalt-chromium (Anatomic Dual Mobility [ADM]; Stryker Corporation, Mahwah, New Jersey), 210 (42%) were cylindrospheric (Modular Dual Mobility [MDM]; Stryker Corporation, Mahwah, New Jersey), and 100 (20%) were subhemispheric. Of the subhemispheric implants, 58 were manufactured from cobalt-chromium (CoCr) alloy (G7; Zimmer Biomet, Warsaw, Indiana), and the remaining 42 were manufactured from zirconium (OR3O; Smith and Nephew, Memphis, Tennessee). The skirted stainless steel DM cup offers the option of screw fixation via a superior tab (not utilized in our study population); however, the anatomic cobalt-chromium implants lacked this capability thus both implants were only used without any screw fixation.

All patients included in this study participated in our institutional-wide comprehensive total joint pathway program, which encompasses uniform standardized protocols for all aspects of perioperative care. In addition, a standard institutional postoperative rehabilitation protocol, as well as a standard postoperative pain protocol was followed for all patients. Patient records and data were de-identified as part of our institutional quality improvement program; however, human-subjects review by our Institutional Review Board (IRB) was obtained prior to this study.

Data Collection

Patient demographic data including age, gender, race, body mass index (BMI; kg/m²), American Society of Anesthesiology (ASA) classification, Charlson Comorbidity Index (CCI), and smoking status were collected. In addition, clinical data including the length of stay (LOS; days), surgical time (minutes), discharge disposition, whether the patient required a postoperative iliopsoas injection to manage substantial groin pain, as well as 90-day all-cause readmissions and revisions were collected from our electronic patient medical record system (Epic Caboodle. version 15; Verona, WI) using Microsoft SQL Server Management Studio 2017 (Redmond, WA). Characteristics of the surgery, including the use of DM articulations, were gathered from the review of operative reports and implant logs. LOS was evaluated in days spent in the hospital following surgery and surgical time was derived from calculating the time difference between initial skin incision and skin closure. All patients were either discharged home with self-care or home services or to an acute or subacute rehabilitation facility.

As part of our institutional standard of care, patients were registered for an electronic patient rehabilitation application (EPRA; Force Therapeutics, New York, NY) at the time of surgical scheduling by clinical care coordinators. The EPRA is a mobile and web-based technology that wirelessly delivers digital patient reported outcome measure (PROM) surveys to patients at pre-defined time intervals. This application was used to collect Hip dysfunction and Osteoarthritis Outcome Score for Joint Replacement (HOOS, JR) and Veterans RAND 12 Physical and Mental components (VR-12 PCS & VR-12 MCS) scores preoperatively, three months postoperatively, and one year postoperatively.

Outcome Measures

The primary outcome evaluated included the incidence of requiring iliopsoas injections to treat considerable postoperative groin pain. Patients were deemed as having groin pain if they had an iliopsoas injection for anterior groin pain that was ordered at least 3-months but no more than 6-months postoperatively. Any patient who had a documented interventional radiology (IR) procedure coded as a “hip injection” was manually chart reviewed to determine if this was an iliopsoas injection; if documentation of an iliopsoas injection was present in the electronic medical record the patient was deemed as having groin pain.

Key secondary outcomes included 90-day all-cause readmission and revision rates, 90-day readmission and reoperation rates for dislocation, and PROMs, as assessed by the HOOS, JR, VR-12 PCS, and VR-12 MCS. Additional outcomes evaluated included perioperative data, such as surgical time, LOS, discharge disposition.

Statistical Analysis

All data were organized and collected using Microsoft Excel software (Microsoft Corporation, Richmond, WA). A binary variable was created to identify patients who underwent THA via the DAA and PA. Demographic and clinical baseline characteristics of study participants were described as means with standard deviations (SD) for continuous variables and frequencies with percentages for categorical variables. Statistical differences in continuous and categorical variables were detected using independent samples t-tests and chi-squared (χ2) tests, respectively. Multivariate linear and logistic regressions were performed to control for potential confounding variables. These regression models were used to compare our primary and secondary outcomes measures between the two cohorts. A p-value of less than 0.05 was considered to be significant. All statistical analyses were performed using SPSS v25 (IBM Corporation, Armonk, New York).

Results

A total of 485 DM THA cases were included in this analysis. Of which, 55 (11%) underwent THA via the DAA and 440 (89%) via the PA. The two cohorts did not statistically differ with regards to age (p=0.202), gender (p=0.652), race (p=0.563), BMI (p=0.168), ASA classification (p=0.620), CCI (p=0.098), and smoking status (p=0.204). Full demographic information for both cohorts is summarized in Table 1.

Table 1.

Demographics of Included Patients Stratified by Approach

	Anterior Approach (n=55)	Posterior Approach (n=440)	P-value
Age (years)	61.20±12.12	63.50±12.62	0.202
Male- no. (%)	15 (27.3)	133 (30.2)	0.652
Race- no. (%)			0.563
White	42 (76.4)	297 (67.5)
African American	5 (9.1)	58 (13.2)
Asian	1 (1.8)	6 (1.4)
Other	7 (12.7)	79 (18.0)
BMI (kg/m2)	27.68±5.93	29.05±6.71	0.168
ASA Class- no. (%)			0.620
I	4 (7.3)	37 (8.4)
II	33 (60.0)	257 (58.4)
III	16 (29.1)	140 (31.8)
IV	2 (3.6)	6 (1.4)
CCI	3.27±2.38	3.81±2.26	0.098
Smoking Status- no. (%)			0.204
Never	34 (61.8)	242 (55.0)
Former	20 (36.4)	161 (36.6)
Current	1 (1.8)	37 (8.4)

ASA, American Society of Anesthesiologists; BMI, body mass index; CCI, Charleston Comorbidity Index no., number.

Primary Outcomes

The incidence of iliopsoas injections to treat groin pain did not significantly differ between the DAA and PA cohorts (1.8% vs. 0.7%, p=0.966). Ninety-day all-cause readmission (9.1% vs. 7.7%, p=0.823) and revision (0.0% vs. 3.0%, p=0.997) rates did not differ between groups. The full clinical outcome comparison between the two groups is shown in Table 2.

Table 2.

Clinical Outcomes of Included Patients

	Anterior Approach (n=55)	Posterior Approach (n=440)	P-value
Surgical Time (min)	100.24±42.37	109.42±36.86	0.070
LOS (days)	2.19±2.52	2.67±2.16	0.072
Discharge Disposition- no. (%)			0.151
Home	47 (85.5)	346 (78.6)
Other Facility	8 (14.5)	94 (21.4)
Groin Pain- no. (%)	1 (1.8)	3 (0.7)	0.966
90-day Readmission- no. (%)	5 (9.1)	34 (7.7)	0.823
Dislocation	1 (1.8)	3 (0.7)	0.993
90-day Revision- no. (%)	0 (0.0)	13 (3.0)	0.997
Dislocation	0 (0.0)	3 (0.7)	0.998

LOS, hospital length of stay; min, minutes; no., number.

HOOS, JR scores preoperatively (p=0.294), three months (p=0.115), and one year postoperatively (p=0.262) did not statistically differ between the two groups. Similarly, VR-12 PCS scores preoperatively (p=0.898), three months (p=0.471), and one year postoperatively (p=0.092) as well as VR-12 MCS scores preoperatively (p=0.162), three months (p=0.205), and one year postoperatively (p=0.185) did not statistically differ between the two cohorts. Delta improvement from baseline to one year follow-up was statistically similar between the two groups with regards to the HOOS, JR (p=0.425), VR-12 PCS (p=0.718), and VR-12 MCS (p=0.257). Full details regarding the patient reported outcome measure comparisons are highlighted in Tables 3 and 4.

Table 3.

Patient-reported Outcome Measures For Included Patients

	Anterior Approach (n=55)	Posterior Approach (n=440)	P-value
HOOS, JR
Preop	51.42±17.06 (n=17)	47.09±15.88 (n=141)	0.294
3m	83.57±16.62 (n=12)	75.10±14.36 (n=124)	0.115
1y	79.59±24.35 (n=11)	80.11±17.88 (n=80)	0.262
VR-12 PCS
Preop	30.28±5.94 (n=17)	30.01±8.25 (n=140)	0.898
3m	45.22±7.05 (n=13)	42.45±10.63 (n=127)	0.471
1y	41.50±12.07 (n=12)	44.72±10.11 (n=96)	0.092
VR-12 MCS
Preop	43.38±13.45 (n=17)	47.74±11.91 (n=140)	0.162
3m	49.12±13.80 (n=13)	52.12±9.70 (n=127)	0.205
1y	49.43±12.72 (n=12)	53.33±9.20 (n=96)	0.185

1y,1 year; 3m, 3 months; HOOS, JR, Hip Dysfunction and Osteoarthritis Outcome Score for Joint Replacement; MCS, mental component score; PCS, physical component score; preop, preoperative; VR-12, The Veterans RAND 12 Item Health Survey.

Table 4.

Delta Improvements in Patient-reported Outcome Measures From Baseline to 1-year Postoperatively

	Anterior Approach (n=55)	Posterior Approach (n=440)	P-value
HOOS, JR: Preop to 1y	28.17±14.81 (n=10)	31.08±20.35 (n=76)	0.425
VR-12 PCS: Preop to 1y	11.22±8.14 (n=11)	14.46±10.06 (n=92)	0.718
VR-12 MCS: Preop to 1y	6.05±8.30 (n=10)	4.52±11.28 (n=92)	0.257

1y, 1 year; 3m, 3 months; HOOS, JR, Hip Dysfunction and Osteoarthritis Outcome Score for Joint Replacement; MCS, mental component score; PCS, physical component score; Preop, preoperative; VR-12, The Veterans RAND 12 Item Health Survey.

Secondary Outcomes

Surgical time (p=0.070), LOS (p=0.072), and discharge disposition (p=0.151) were found to be statistically similar between patients who underwent DM THA via the DAA and PA (Table 2). Additionally, the 90-day readmission (1.8% vs. 0.7%, p=0.993) and revision (0.0% vs. 0.7%, p=0.998) rate due to dislocation did not statistically differ between both cohorts.

Discussion

Previous studies have suggested that the implantation of DM implants with either DAA or PA has favorable results, such as reducing the risk of instability.^11-17,21 However, no study to date has compared the difference in postoperative outcomes between these two surgical approaches in primary THA with DM bearings. Despite the success of these implants in high-risk patients, this study found no statistical differences in clinical or patient-reported outcomes between DAA and PA in primary THA performed using DM bearings. Furthermore, we found similar results regarding the use of iliopsoas injections for the management of substantial groin pain, readmission rates, and revision rates.

Overall, the literature remains inconclusive as to which surgical approach yields more favorable outcomes. A plethora of previous reports suggests no significant difference in patient outcomes between the DAA and PA, such as postoperative function and dislocation rates.^22-27 However, some studies have found that the DAA may yield superior early results when compared to PA. Sheth et al.²⁸ and Tsukada et al.²⁹ reported that the DAA approach led to lower dislocation rates without increasing the risk of revision surgeries.

With regards to short-term outcomes, Martin et al.²⁷ noted that DAA patients had a shorter LOS and earlier mobilization. In addition, Wang et al.²⁶ found that the DAA provided better early functional recovery as assessed by postoperative Harris Hip Scores (HHS) and lower Visual Analog Scale (VAS) pain scores. Jia et al.³⁰ and Rodriguez et al.³¹ also recorded superior postoperative early function and pain relief in DAA patients. An earlier study performed at our institution found that DAA patients expressed higher satisfaction as assessed via the Forgotten Joint Score (FJS-12) 12 weeks postoperatively; however, controlling for surgeon volume nullified this difference.³² Notably, none of these studies stated if they included DM bearings. The integration of DM bearings may eliminate the variance seen in outcomes between the DAA and PA in primary THA.

Studies evaluating DM constructs via the DAA have shown promising results in decreasing dislocation rates, particularly in high-risk patients. For instance, Homma et al.¹⁵ and Ochi et al.¹² found a reduction in dislocation rates in patients with displaced femoral neck fractures that underwent THA via DAA using DM bearings. Also, Jinnai et al.¹³ reported no dislocations, early postoperative functional recovery, and a low one-year mortality rate in their patient population following DM bearing implantation using the DAA. They hypothesized that the DAA and the increased range of motion provided by the DM bearings could have contributed to the early postoperative functional recovery and the low dislocation rates, respectively. However, notable complications of DM bearing use and the DAA include groin pain and perioperative femoral fracture. A retrospective study by Homma et al.¹⁴ using DM bearings for primary THA reported two cases of persistent postoperative groin pain in the group that received DM bearings via the DAA. The diagnosis of iliopsoas muscle impingement was determined based on pain with active flexion, hip flexion against resistance, and passive hyperextension. In another similar study, Batailler et al.¹⁶ compared the outcomes of 201 primary THA using a DM bearing via DAA to 101 performed via a posterolateral approach. They reported three cases of postoperative groin pain due to impingement in the DAA group. Although previous literature has associated postoperative groin pain with DM bearings in patients undergoing THA via the DAA, the present study found no significant difference in the rates of requiring iliopsoas injections for groin pain between the two approaches.

Previous studies have also shown beneficial results in patients that undergo THA via PA using DM bearings. Bouchet et al.¹⁸ and Tarasevicius et al.¹⁹ reported no dislocations in their study population when employing the PA in tandem with DM constructs. Furthermore, Guyen et al.¹⁷ reported maintaining hip stability in 51 (94%) of the 54 patients who received DM bearings with the PA. In addition, they reported no loosening or osteolysis in their patients on follow-up.

This study is not without limitations. The retrospective nature of the study may have introduced biases. Although we reviewed the patient demographics and demonstrated similarities regarding patient age, sex, and BMI, there is still the possibility of subtle differences in patient characteristics between the two groups not adequately captured by those criteria. Additionally, the use of DM implants was left to the operating surgeons’ discretion; therefore, we could not determine if its use was predetermined or carried out due to intraoperative instability. Thus, the incidence of these reasons may differ across groups; however, we attempted to minimize this confounding variable by controlling for demographic differences between the two cohorts. In addition, only 55 patients (11%) in this study underwent DAA THA compared to 440 (89%) who underwent PA THA; this could introduce sampling bias. It is important to note that in a recent polled survey of members attending the 2018 American Association of Hip and Knee Surgeons (AAHKS) annual meeting, the average utilization of the DAA use in THA was 56.2%.³³ Multiple implants were used in this study, which may have some confounding effects on our results. To further elucidate the effectiveness of each approach, prospective studies with longer follow-up are required. Notably, we did not examine component positioning, especially in regards to being related to groin pain, though this likely did not differ between the groups as evidenced by the similar outcomes observed between the two. Our analysis included THA patients from 2011-2021, and various institutional protocol changes over this period could have influenced our results. We may not have captured all revisions performed at outside institutions. While this raises the possibility that we underestimated the true revision rate, our findings coincide with those of previous studies, thus missed cases would likely not alter our findings. Lastly, all cases included in this study were from an orthopedic specialty hospital. As a result, our findings may not be generalizable to surgeons who have low DM bearing case volumes or are new to their respective surgical approach of choice.

Conclusion

This study demonstrates that patients undergoing an anterior-based surgical approach compared to a posterior approach did not yield differences in prevalence of iliopsoas injection for groin pain or PROMs when using DM constructs. In current practice, either approach may be used without significant difference in the incidence of postoperative groin pain as well as readmission and revision rates. Future investigation is necessary to determine whether surgical approach influences long-term outcomes in patients receiving DM implants.