INTRODUCTION
Total shoulder arthroplasty (TSA) is a common procedure that produces clinically significant improvements to pain and function for people with glenohumeral arthritis. Multiple glenoid components are available, including all-polyethylene, metal-backed glenoid (MBG), and hybrid components. Failure of the glenoid component remains a significant challenge, contributing to diminished patient outcomes [1] and presenting a major impedance to the long-term survival of TSA in the three types of components [2-4].
MBG components were developed to increase the early stability of components [5]. However, polyethylene components have been shown to have greater long-term durability [6,7]. Their greater durability is hypothesized to be due to their ability to reduce excessive force between the bone and implant interface . One systematic review suggested that modern designs of MBG have shown equal rates of failure compared to polyethylene components [2]. Hybrid glenoid components represent a strategic response to concerns regarding the fixation, complications, and long-term survival rates associated with both polyethylene and MBG designs [8,9]. By integrating features of the two types, hybrid components are intended to mitigate loosening while facilitating supplementary bone growth through central metal pegs for enhanced long-term fixation [10-12]. The optimal selection of the glenoid-type component remains a nuanced decision for the individual patient.
Many studies have compared the efficacy of MBG and polyethylene glenoid components at short- and mid-term follow-up points [2-4,6,7,10,11,13,14]. The present study, on the other hand, was intended to address the lack of direct comparisons among these three glenoid component types by analyzing 5-year outcomes. This study assessed and compared 5-year functional outcomes, patient reported outcomes, and conversion rates to those of reverse TSA (rTSA) between patients who underwent TSA using a modular MBG component, a hybrid glenoid component, and an all-polyethylene component.
METHODS
This study was approved by the Institutional Review Board of Guthrie Clinic (No. Guthrie IRB 2307-43). The requirement for written informed consent was waived due to the retrospective nature of the study.
Patient Population
The patients who were eligible for this study were identified through a retrospective review of electronic health records and had undergone a TSA procedure between November 2017 and December 2018. All procedures were performed by a shoulder and elbow fellowship-trained attending physician at a single institution (JYC). The inclusion criteria were adult patients (≥18 years) who underwent a TSA procedure for glenohumeral arthritis using one of the following glenoid components: a LIMA Hybrid TT Glenoid component, a modular MBG component system, or a LIMA cemented all-polyethylene implant system (LimaCorporate) with 5-year follow up data. The LIMA Hybrid TT glenoid component had a modular press-fit central peg with a polyethylene liner and two pegs (Fig. 1). The SMR metal back was a modular system with a convex base plate of titanium alloy and a press-fit central peg (Fig. 2). The all-polyethylene glenoid component was ultra-high-weight molecular polyethylene with three pegs and was cemented (Fig. 3). The MBG and hybrid glenoid components were indicated for patients who had a higher probability of conversion to rTSA, as these systems were modular and potentially bone-preserving. Cemented all-polyethylene glenoid components were indicated in younger patients who had the option of potential glenoid polyethylene exchange if symptomatic wear occurred and the rotator cuff remained intact. Cemented polyethylene glenoid components were also used in patients who had smaller glenoid vaults identified on preoperative computed tomography imaging. If any concern existed for potential fracture due to the central peg of the MBG, the hybrid glenoid component or the all-polyethylene component with its smaller central and peripheral pegs was chosen. MBG was used in older patients with more robust bone, whereas hybrid components were considered for patients with softer bone. The same diaphyseal fixation stem was utilized in all patients. Patients were excluded from the study if they were deceased or could not be contacted by phone after three separate attempts to complete the 5-year outcome score measurements. Exclusion criteria may have introduced selection bias with data collection, although similar rates of nonresponse were noted in all three groups.
Data Collection
All data were compiled and reviewed by the investigators who did not perform TSA procedures via patient electronic medical records. The data were assessed by two orthopedic surgery residents: the other medical student and one undergraduate student. The data extracted from the patient charts were demographics (sex, age, body mass index [BMI], smoking status, diabetes), date of surgery, total operation time, length of stay, anesthesia type, and physical exam findings, such as range of motion (ROM) and pre- and postoperative American Shoulder and Elbow Surgeons (ASES) scores. The physical exam findings were recorded at 1, 2, and 5 years postoperatively. The clinical data (ASES score and ROM) were collected by the attending orthopedic surgeon (JYC) at each follow-up appointment. If a patient did not have a 5-year follow-up ASES score, they were contacted by telephone for follow-up by the coauthors. The studies on each patient were continued until they were interrupted by patient drop-out or death. The mean ASES scores were compared preoperatively and postoperatively to assess functional improvement. The complications included infection, dislocation, and fracture. We included revisions as a single category because most were performed owing to degenerative rotator cuff tears.
Surgical Technique
The patients were placed in a semi-reclining chair position with the head secured using a McConnel head holder. Following padding of all bony prominences, sterile preparation, and draping of the shoulder, a standard deltopectoral approach was used to gain access to the glenohumeral joint. A lesser tuberosity osteotomy was performed to mobilize the subscapularis, and the humeral head was inspected for arthritic changes. The humeral head was cut in 30° of retroversion using a cutting guide while the supraspinatus and infraspinatus tendons were protected using a Crego retractor. After the cut, a metal cap was placed on the head, and the glenoid was exposed using circumferentially placed retractors. The labrum was removed with cautery and a guide pin was placed in the center of the glenoid. Implants of various sizes were trialed. Once the appropriate size was chosen, the glenoid was concentrically reamed to subchondral bone and a center drill hole was made. A guide was used to drill appropriate holes for the chosen MBG component, hybrid glenoid component, or the all-polyethylene glenoid component (Lima SMR, LimaCorporate). The glenoid implant was secured to the glenoid according to manufacturer technique specifications with screws or cement, respectively. Following the glenoid component placement, the humerus was reamed and broached in 30° of retroversion, and the appropriate size head and components were implanted. The lesser tuberosity osteotomy was re-approximated and fixed with heavy sutures passed through drill holes placed prior to final humeral component implantation. The patient was immobilized in a sling for 4 to 6 weeks, followed by 4 to 6 weeks of physical therapy for ROM.
Statistical Analysis
Statistical analysis was performed using SPSS statistical software version 25 (IBM Corp). The outcomes for this analysis were ASES and visual analog scale (VAS) scores at 5-years of follow-up; other variables collected at 5 years provided insufficient data for analysis. ASES, VAS, forward elevation, internal and external rotation, and rotation at 90° abduction at the 1-year mark were additional outcomes of interest. The primary exposure was glenoid replacement material (MBG, polyethylene, or hybrid).
Data availability was summarized first for the sample characteristics and outcome variables by exposure. This approach was undertaken owing to the varying levels of missing data at each time point, which caused descriptive statistics to be based on differing sample sizes depending on the measure. Next, the summary statistics for demographics, procedure variables, and baseline measures of the outcomes were obtained as means and standard deviations, medians and interquartile ranges, or frequencies and percentages, depending on the variable type. Between-group tests were one-way analysis of variance, Kruskal-Wallis, or Fisher exact tests. No corrections were made for confounding variables.
The outcomes were summarized using the same descriptive statistics and tests, depending on the variable type. Owing to the small sample size at follow-up, no multivariate analysis was performed. Missing values were treated with list-wise deletion, and significance was defined as two-tailed alpha less than 0.05.
RESULTS
This study included a total of 77 patients: 44 patients with all-polyethylene glenoid components, 17 with hybrid components, and 16 with MBG components. The overall average age was 66.9 years (standard deviation, 9.74), with the mean age being significantly younger in the polyethylene group (62.6 [8.6] vs. 72.6 [9.2] MBG and 72.8 [7.2] hybrid, P<0.001) (Table 1). This significant difference in age should be acknowledged as a potential source of confounding. Of the sample, 46.8% were male, 25% were diabetic, and 3.9% (n=3) were tobacco smokers. The average BMI was 32.07 kg/m2, with similar BMIs reported among the three groups. The operative time was significantly longer in the all-polyethylene group, with a median time of 120.0 minutes (interquartile range [IQR], 111.5–132.0) compared to 104.0 minutes (99.0–114.0)) in the MBG group and 109.0 minutes (104.0–130.5) in the hybrid group (P=0.005). The results of reoperative forward elevation, internal rotation, external rotation, and rotation of 90° abduction were similar among the three groups. At 5 years, no difference existed in ASES and VAS scores, although a trend of lower ASES scores was evident for the patients with all-polyethylene components (polyethylene, 87.0 vs. hybrid, 100.0 vs. MBG, 97.0; P=0.086) (Table 2). At 1 year, no difference existed in ASES or VAS scores of forward elevation, internal rotation, and rotation of 90° abduction among the component types. A significant difference existed in external rotation results between the all-polyethylene and hybrid components compared to MBG (polyethylene 30.0 vs. hybrid 30.0 vs. MBG 20.0, P=0.028). Of note, the minimal clinically important difference in external rotation was considered to be between three and four degrees. Hybrid components tended to require more frequent revision at 12% vs. 6.8% of all-polyethylene glenoid components vs. 0% MBGs (P=0.310), although this difference was not significant (Table 3). In three cases, complications occurred in the all-polyethylene group (P=0.310): polyethylene wear, improper union at the implant site following fracture, and methicillin-resistant Staphylococcus aureus infection. No dislocations were reported. No differences in length of hospitalization were evident.
DISCUSSION
The findings of this study contribute to the growing body of evidence showing patient improvement in postoperative ROM, ASES score, and VAS score in all-polyethylene, MBG, and hybrid components. Our primary endpoint showed similar ASES and VAS outcome measurements among component types at the 5-year follow up. At the 1-year follow-up point, no significant difference in ASES or VAS score or in forward elevation, internal rotation, or rotation 90° abduction was evident among the glenoid component types. A significant difference was apparent in external rotation between the all-polyethylene and hybrid components compared to the MBG components. No significant difference was evident in revision rates or complications among the groups.
Many previous studies have shown that polyethylene, MBG, and hybrid components improve patient-reported ASES and VAS scores [10,12,15-18]. A meta-analysis by Kim et al. [2] comparing polyethylene and MBG components showed a mean postoperative improvement in ASES score of 44.5 and 56.5, respectively. While our study did not produce significant improvements in outcome scores among the groups, previous studies found a significant increase in postoperative ASES scores at a mean follow-up time of 50.3 months in a comparison of hybrid and all-polyethylene components (53.3 vs. 46.3) [19]. Gulotta et al. [11] found no difference in ASES scores between hybrid and all-polyethylene components at 24 months. Our data support those of previous studies; i.e., significant improvement occurs in ASES and VAS scores with TSA. Meanwhile, our research contributes to the literature by comparing the effect of component type on final VAS and ASES scores.
At 1-year follow-up, the secondary endpoints were similar among the groups, with external rotation showing the only significant difference. At 50.3 months, Marigi et al. [19] found significant difference in postoperative abduction, forward flexion, and external rotation in postoperative ROM between all-polyethylene and hybrid components. When they compared preoperative to postoperative improvement, only forward flexion remained statistically significant. In a study by Merolla et al. [16], MBG showed ROM improvement between preoperative and postoperative active anterior elevation, active lateral elevation, active external rotation, and active internal rotation at an average follow-up of 38 months. In our study, the MBG group was too small to conduct a statistical comparison. It is unclear why patients with MBG in our study demonstrated less external rotation; nonetheless, we theorized that the greater thickness of the MBG component may have limited the ROM.
In our study, five patients (6.5%) required revision surgery to rTSA. Of these, three were in the all-polyethylene group and two were in the hybrid group. Previous studies demonstrated that the MBG and all-polyethylene revision rates were respectively 14% and 4% after 6 to 7 years of follow-up [6]. The all-polyethylene revision rate was similar to our rate of 6.8%, with our MBG being much lower than those in previous reports. Our results differ from those in the literature in that our cemented polyethylene glenoid component survival rate was three times higher than that of a cementless MBG at 10 years after TSA [1]. Polyethylene components may fail due to loosening, while MBG may fail based on polyethylene wear or instability [3]. We hypothesize that fixation techniques using modern MBG implant designs and technology are superior to those in previous generations and have lower rates of component loosening and revision. One possible explanation for this decreased rate of revision in the MBG group is the success of modern MBG designs in fixation achieved through bony ingrowth [2]. We suspect this may be due to modern designs of MBGs with decreased loosening and failure compared to the all-polyethylene components [2], an effect found at midterm follow-up. The Lima SMR TT MBG system used in this study had a curved-backed and less conforming shape of the glenoid, a stiff and thick metal back to reduce wear on the PE, hydroxyapatite pegs, and a peg with two screws that optimizes stability and longevity. Contemporary MBGs show significantly lower loosening and revision rates compared to conventional MBG implants [3]. In studies comparing modern vs. conventional MBGs, heterogeneity exists in designs of modern implants that may obscure previous findings regarding MBGs [3]. Murphy et al. [12] found 100% implant survival of MBG at 7 years, similar to our study.
The hybrid glenoid had relatively higher revision rates than rTSA in our study (12%). Hybrid glenoid design combines the initial fixation of polyethylene glenoids with the long-term stability of boney ingrowth in MBGs [15]. Low rates of mechanical failure of hybrid glenoids have been described in the literature. At 5-year follow-up, Nelson et al. reported failure rates at 2%, while Hakeem et al. described revision rates of hybrid glenoids to be as low as 2.5% at an average of 4.2 years of follow-up [19,20]. Friedman et al. [10] noted favorable results of hybrid cage glenoids with lower rates of radiolucent glenoid lines and rates of revision compared to all-polyethylene glenoids. The higher revision rate observed in this study was possibly due to the relatively small sample size of 17, as each additional patient requiring revision could have impacted the overall revision rate and may not have been clinically significant.
We demonstrated no significant differences between ASES and VAS score outcomes at 5-year follow-up between glenoid systems, and one could argue for increased modular glenoid component usage to facilitate less complicated revisions. Hybrid and MBG components provide modularity compared to all-polyethylene designs, which facilitates potentially less complex future revisions. Traditional rTSA can be difficult, with high complication rates associated with removal of the humeral stem. For this reason, studies suggest using modular implants that allow conversion to reverse shoulder arthroplasty while maintaining the humeral stem and the glenoid metal back, decreasing the risk of significant bone loss [20,21]. In a study of 26 patients, the authors noted improved ROM, a simplified surgical procedure, no major complications, and reduced recovery time using modular components at 28 months of follow-up [21]. Bitzer et al. [20] noted a significant reduction in LOS in modular revision groups compared to non-modular groups (26.3 to 36.6 hours, P<0.05, respectively) and a decreased operating time (109 to 159 minutes, P=0.02, respectively). They also found significant improvement in ASES and VAS scores and forward flexion in the modular group. Modular glenoid implants may provide the further advantage when revising to rTSA with an easier operative course with comparable patient outcome measures.
Our study had a notable contribution and limitations. The contribution was its inclusion of 5 years of follow-up, which was longer than previously reported when comparing all three glenoid components. One limitation was that the indications varied by impact type, complicating patient selection and corresponding implant selection across these groups. All surgeries were performed by one surgeon. While this may have limited heterogeneity, it may have induced some degree of bias. In addition, our study was retrospective in nature and had relatively low rates of follow-up at the 5-year point. The number of hybrid component patients in our study was limited, as this was a newer system and had fewer patients with long-term follow-up. With the rural nature of our institution and increased cost, radiographic follow-up was not feasible. This would have better characterized the reasons for revision while identifying any loosening or polyethylene wear. Owing to the limited number of patients, we were not able to conduct meaningful multivariate analysis, which would have eliminated bias and strengthened our conclusions. Last, there was a significant difference in age between the component groups, which may have influenced our results.
CONCLUSIONS
At 5-year follow-up, patient-reported outcomes were similar between MBG, all-polyethylene, and hybrid glenoid components. Given the similarity of outcome measures, surgeons performing TSA could consider MBG or hybrid glenoid systems for primary TSA in patients at risk for future conversion to rTSA for the potential bone sparing and less complicated revision surgery.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
