Translate this page into:
Biceps tenodesis in the setting of total shoulder arthroplasty: A matched cohort analysis
, Akiro H. Duey, Troy Li, Akshar V. Patel, William A. Ranson, Evan M. Michaelson, Bradford O. Parsons, Evan L. Flatow, Paul J. Cagle
⁎Corresponding author: Paul J. Cagle. Paul.Cagle@mountsinai.org
-
Received: ,
Accepted: ,
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.
Abstract
Abstract
Studies have theorized that pain refractory to anatomic total shoulder arthroplasty (aTSA) in patients who did not undergo a concomitant biceps tenodesis or tenotomy may be as a result of the pathologic interaction between the long head of the biceps tendon and the humeral prosthesis. This study evaluates the effect of concomitant biceps tenodesis during aTSA on long-term clinical and radiographic patient outcomes.
This retrospective study analyzed patients who underwent aTSA with or without biceps tenodesis from 2000 to 2017. Preoperative and postoperative range of motion (ROM), American Shoulder and Elbow Surgeons (ASES), Visual Analog Scale (VAS), and Simple Shoulder Test (SST) scores were collected. Patients in the biceps tenodesis group were matched with those who underwent aTSA without tenodesis based on age, follow-up interval, sex, and preoperative ASES score.
The study included 88 shoulders (44 tenodesis, 44 no tenodesis), with median follow-up of 9.9 years (range 3.3–19.0). No significant differences were observed between cohorts in postoperative forward elevation, external/internal rotation, VAS, ASES, or SST scores at most recent follow-up. There were no significant differences in pre-to postoperative improvement in ROM and patient-reported outcome scores. Both groups demonstrated excellent implant survival, with similar revision rates at nearly 10 years postoperatively.
Patients who underwent aTSA with concomitant biceps tenodesis did not demonstrate inferior long-term clinical outcomes or pain scores in comparison to patients who underwent aTSA with preservation of the biceps tendon. Appropriately selected patients may experience equivalent, satisfactory outcomes after aTSA, independent of concomitant biceps tenodesis.
Keywords
Total shoulder arthroplasty
Biceps tenodesis
Patient reported outcomes
Range of motion
Implant survival
Clinical outcomes
1 Introduction
Anatomic total shoulder arthroplasty (aTSA) has long been the treatment of choice for patients with end-stage glenohumeral arthritis and absent rotator cuff pathology.1 This procedure has proven to be an effective, reliable solution for pain and restricted motion in many patients. A small percentage, however, have endorsed persistent postoperative shoulder pain without identifiable intra- or postoperative complication.2–7 It has been theorized that this pain may be linked to biceps tendon pathology – either pre-existing or developing over time after implant placement – but few studies have investigated this relationship.8–118–11
Long before the popularization of shoulder arthroplasty, proximal biceps tendon pathology was an identified and accepted cause of anterior pain in native shoulders. In a 2002 landmark study by Godenèche et al. including senior author Dr. Gilles Walch, this group first hypothesized that biceps tendon pathology was a potential source of residual postoperative pain, and ultimately treatment failure, in patients undergoing arthroplasty for glenohumeral osteoarthritis.11 Dr. Walch postulated that failing to perform a biceps tenotomy or tenodesis at the time of shoulder arthroplasty may lead to suboptimal patient outcomes. He therefore began systematically performing a biceps tenodesis during all shoulder arthroplasty cases.9 This change was largely based on anecdotal evidence, as there was minimal literature to support or refute this supposition at the time.
In 2004, Tuckman et al. further investigated this theory of a painful preserved biceps tendon. This study retrospectively assessed the clinical course of 7 patients (8 shoulders) with persistent anterior shoulder pain following uncomplicated shoulder arthroplasty. Ultimately, all 7 patients were diagnosed with biceps tendon pathology - to which the pain was attributed - and all 7 patients underwent reoperation with biceps tenodesis. All 7 patients were reported to experience subsequent relief of their anterior shoulder pain and improvement in function. These results led this group to theorize that inflammation and bowstringing of the intra-articular biceps tendon was the residual pain generator.8
Despite the drastic shift in favor of biceps tenodesis during aTSA observed clinically over the last two decades, the impact on patient outcomes is relatively understudied.9,10 In 2007, Simmen et al. published the first and only prospective study investigating this topic to date. Their analysis found that concomitant biceps tenodesis had a statistically significant favorable effect on treatment success at one year following shoulder arthroplasty as compared to retention of the native intra-articular biceps origin. They reported an odds ratio of 2.29 in favor of performance of a biceps tenodesis for achieving a Constant Score they deemed to be a clinical success.10 Despite this study's results, no long-term outcomes have been explored and no further comparative investigation performed. The aim of this study was to complete the first matched cohort study evaluating the effect of performing biceps tenodesis at the time of aTSA on long-term patient reported outcomes (PROs), range of motion (ROM), and radiographic measures. We hypothesized that patients who receive concomitant biceps tenodesis fare better with regards to PROs and shoulder function in comparison to patients who did not receive a concomitant biceps tenodesis at the time of aTSA.
2 Material and methods
2.1 Study population
This study was a retrospective, matched analysis utilizing data from a single institution and approved by the Institutional Review Board. All procedures were performed in compliance with relevant laws and institutional guidelines. Patients who underwent aTSA with or without concomitant biceps tenodesis by a single fellowship-trained orthopedic surgeon between September 2000 and June 2017 were considered for inclusion. A total of 161 patients were identified from the records of the investigating surgeon. Of these, 47 had less than 2 years of postoperative follow up, were missing postoperative ROM or PRO measurements, or were missing radiographic imaging and were thus excluded from the study. Of the remaining 114 patients, 44 patients underwent aTSA with concomitant biceps tenodesis while 70 patients underwent aTSA with preservation of the biceps tendon. All 44 patients who underwent tenodesis were included in the biceps tenodesis cohort. To minimize the effects of confounding, matching was utilized to identify a similar group of patients in whom the native biceps was preserved. The 44 biceps-preservation patients most like the biceps tenodesis patient cohort based on age, follow-up duration, sex, and preoperative American Shoulder and Elbow Surgeons (ASES) score were included in the no biceps tenodesis cohort, for a total of 88 patients included in the study. Matching was performed using the MatchIt package in R Version 4.1.0.
After participation consent was obtained, retrospective data was collected from the electronic medical record in the form of pre-operative progress notes, patient questionnaires, and plain radiographs for all 88 patients included in the study. Recorded characteristics included patient age, sex, follow-up duration, Body Mass Index (BMI; kg/m2), American Society of Anesthesiologists (ASA) physical status, glenoid implant type (Keeled, Pegged, or Metal-Backed), and preoperative ASES score. BMI data was unable to be obtained for 3 patients in the biceps tenodesis cohort and 7 patients in the no biceps tenodesis cohort while ASA status could not be obtained in 5 patients in the biceps tenodesis cohort and 10 patients in the no biceps tenodesis cohort. These baseline demographic and operative characteristics, separated by study cohort, are presented in Table 1.
| Patient Characteristic | Tenodesis (n = 44) | No Tenodesis (n = 44) | Intergroup P-Value |
| Age (Years) | 65.4 (9.3) | 65.1 (8.9) | 0.866 |
| Follow-Up Interval (Years) | 10.0 (3.1) | 10.2 (4.8) | 0.873 |
| Female Sex | 20 (45.5%) | 20 (45.5%) | 1.000 |
| BMI (kg/m2) | 28.2 (5.5) | 26.5 (5.7) | 0.183 |
| ASA Status | |||
| I | 2 (4.5%) | 3 (6.8%) | 1.000 |
| II | 25 (56.8%) | 20 (45.5%) | 0.394 |
| III | 12 (27.2%) | 11 (25.0%) | 1.000 |
| IV | 0 (0.0%) | 0 (0.0%) | N/A |
| Glenoid Implant Type | |||
| Keeled | 10 (22.7%) | 11 (25.0%) | 1.000 |
| Pegged | 34 (77.3%) | 31 (70.5%) | 0.628 |
| Metal-Backed | 0 (0.0%) | 2 (4.5%) | N/A |
| Preoperative ASES Score | 33.4 (21.9) | 34.0 (17.0) | 0.874 |
2.2 Clinical evaluation
The primary outcome of interest was implant survival, defined as no subsequent occurrence of revision surgery for replacement or removal of either the glenoid or humeral components. Secondary outcome measures included shoulder ROM and PROs. Specifically, ROM measurements consisted of active forward elevation, internal rotation, and external rotation. Internal rotation was measured as the highest vertebral level to which the patient could actively reach dorsally, as described by Amroodi et al.12 The recorded PROs included the American Shoulder and Elbow Surgeons (ASES) score, Visual Analogue Scale (VAS) pain score, and the Simple Shoulder Test (SST) score. Preoperative measures of ROM and PROs were collected from the medical record retrospectively when possible. Postoperative clinical outcome data was collected during each patient's most recent follow-up visit. Both preoperative and postoperative clinical outcome data was separately tabulated for each cohort and compared between cohorts (Table 2).
| Clinical Outcome | Preoperative | Postoperative | P-Value |
| Tenodesis Cohort (n = 44) | |||
| Forward Elevation (degrees) | 117.5 (25.6) | 148.2 (21.3) | <0.001 |
| External Rotation (degrees) | 24.8 (22.1) | 52.2 (14.0) | <0.001 |
| Internal Rotation | 6.4 (4.0) | 11.4 (3.4) | <0.001 |
| VAS | 6.4 (3.0) | 2.2 (2.8) | <0.001 |
| ASES | 33.4 (21.9) | 74.9 (23.7) | <0.001 |
| SST | 3.2 (2.4) | 7.8 (3.9) | <0.001 |
| No Tenodesis Cohort (n = 44) | |||
| Forward Elevation (degrees) | 117.3 (29.1) | 148.3 (24.2) | <0.001 |
| External Rotation (degrees) | 18.0 (22.0) | 53.1 (13.5) | <0.001 |
| Internal Rotation | 6.9 (4.0) | 10.7 (3.6) | <0.001 |
| VAS | 6.7 (2.1) | 2.3 (2.6) | <0.001 |
| ASES | 34.0 (17.0) | 75.9 (20.1) | <0.001 |
| SST | 3.6 (2.5) | 8.8 (2.7) | <0.001 |
2.3 Radiological assessment
Radiographic analysis was independently performed pre- and postoperatively by two fellowship-trained orthopedic surgeons. Plain radiographs of standard AP, Grashey, and Axillary projections were utilized. Radiographic measurements included the acromiohumeral interval (AHI) and a categorical measure of peri-glenoid component lucency. The AHI was measured as the vertical distance (mm) between the superior articular cortex of the humeral head and the inferior cortical surface of the acromion, as described by Lehtinen et al.13 Radiolucency about the glenoid component was graded in accordance with the classification system proposed by Lazarus et al. with each series of radiographs receiving a score from 0 to 5, in which 0 indicates no radiolucency and 5 indicates maximal gross glenoid component loosening.14,15
2.4 Statistical analysis
The cardinal matching algorithm was performed with the MatchIt package in R Core Team 2021, Version 4.1.0 (Vienna, Austria).16,17 A t-test was used to compare continuous variables between groups, and chi-square test was used to compare categorical variables between groups. Implant survivorship and estimates of 5-, 10-, and 15-year survival were assessed for both the biceps tenodesis and no biceps tenodesis cohorts. A Kaplan-Meier survival curve was generated utilizing Python (Fig. 1). An alpha of less than 0.05 was considered significant.
2.5 Operative technique
A deltopectoral approach was used for all aTSA procedures. Superficial dissection was performed in-line with the incision via the deltopectoral intermuscular interval to the level of the conjoint tendon. The circumflex vessels were identified and ligated, and the biceps tendon was palpated within the bicipital groove just proximal to the pectoralis major tendon. The transverse humeral ligament was opened along the medial border of the groove just superior to the pectoralis major tendon insertion and a Mixter clamp was used to bluntly expose the tendon. The biceps tendon was mobilized free from any underlying adhesions throughout the length of the groove. The decision to perform tenodesis was made if the tendon appeared injured, torn, inflamed, or associated with severe synovitis. A #2 Ethibond suture was passed from deep to superficial through the biceps tendon and overlying pectoralis major tendon in a figure-of-8 fashion and the suture was tested to ensure adequate capture of the biceps and pectoralis tendons. The suture was tied, and the biceps tendon was transected approximately 1 cm proximal to the tenodesis site. The transverse ligament was incised vertically along the medial border of the bicipital groove from the level of the tenodesis to the articular surface proximally, with the tendon cut at the level of the rotator interval inside the shoulder joint to excise the proximal tendon fully. Following subscapularis take down using either a lesser tuberosity osteotomy or subscapularis peel technique, the origin of the long head of the biceps was clearly visible at the supraglenoid tubercle. The remaining bicep origin stump was removed later during exposure of the glenoid surface. The remainder of the arthroplasty procedure was performed in standard fashion. The subscapularis, having been released earlier, via either osteotomy or peel, was subsequently repaired at the end of each case prior to closure of the overlying soft tissues. There were a few significant differences in technique for the group not undergoing biceps tenodesis. In this group the biceps tendon was not sutured to the pectoralis major tendon, the proximal length of the tendon was not excised, and the origin of the tendon at the supraglenoid tubercle was not removed during glenoid exposure. The initial exposure and implantation techniques were otherwise unchanged.
3 Results
3.1 Study population
There were 44 shoulders in the biceps tenodesis group and 44 shoulders in the no biceps tenodesis group for a total of 88 shoulders. The main indication for aTSA for the tenodesis patients was osteoarthritis (42, 95.5%), with others including rheumatoid arthritis (1, 2.3%) and post-traumatic arthritis (1, 2.3%). Indications for the no tenodesis group included osteoarthritis (41, 93.2%), rheumatoid arthritis (2, 4.5%) and avascular necrosis (1, 2.3%). The overall median follow-up was 9.9 years (range 3.3–19). The mean BMI at time of surgery was 28.2 ± 5.5 and 26.5 ± 5.7 kg/m2 for the tenodesis and no tenodesis groups, respectively (P = 0.183). There were no significant differences between cohorts in any of the variables used for matching including age at surgery (65.4 ± 9.3 tenodesis, 65.1 ± 8.9 no tenodesis in years), follow-up duration (10.0 ± 3.1 tenodesis, 10.2 ± 4.8 no tenodesis in years), sex, and preoperative ASES score (Table 1). 56.8% of the tenodesis group and 45.5% of the no tenodesis group had an ASA classification of II. 10 (22.7%) and 34 (77.3%) patients of the tenodesis group had keeled and pegged glenoid implants, respectively, with the no tenodesis group including 11 (25.0%), 31 (70.5%), and 2 (4.5%) keeled, pegged, and metal-backed glenoid implants, respectively. The preoperative ASES scores were similar between patients in the tenodesis group (33.4 ± 21.9) and no tenodesis group (34.0 ± 17.0) (P = 0.874).
3.2 Clinical outcomes
At final follow-up, both groups experienced statistically significant improvements in all clinical outcome measures (p < 0.001) (Table 2), with no significant differences in ROM or PROs between the tenodesis and no tenodesis cohorts. There were also no significant differences between the two cohorts in the pre-to postoperative improvement for all ROM and PRO measures: forward elevation (30.7 vs. 30.0, P = 0.932), external rotation (27.4 vs. 34.3, P = 0.139), internal rotation (4.8 vs. 3.9, P = 0.386), VAS score (−4.1 vs. −4.4, P = 0.623), ASES score (40.9 vs. 42.0, P = 0.836), and SST score (4.5 vs. 5.2, P = 0.425) (Table 3).
| Clinical Outcome | Tenodesis (n = 44) | No Tenodesis (n = 44) | Intergroup P-Value |
| Postoperative Only | |||
| Forward Elevation (degrees) | 148.2 (21.3) | 148.3 (24.2) | 0.988 |
| External Rotation (degrees) | 52.2 (14.0) | 53.1 (13.5) | 0.767 |
| Internal Rotation | 11.4 (3.4) | 10.7 (3.6) | 0.415 |
| VAS | 2.2 (2.8) | 2.3 (2.6) | 0.945 |
| ASES | 74.9 (23.7) | 75.9 (20.1) | 0.846 |
| SST | 7.8 (3.9) | 8.8 (2.7) | 0.201 |
| Preoperative to Postoperative Increase | |||
| Forward Elevation (degrees) | 30.7 (34.5) | 30.0 (39.6) | 0.932 |
| External Rotation (degrees) | 27.4 (22.7) | 34.3 (19.5) | 0.139 |
| Internal Rotation | 4.8 (5.1) | 3.8 (4.8) | 0.386 |
| VAS | −4.1 (3.3) | −4.4 (3.1) | 0.623 |
| ASES | 40.9 (25.7) | 42.0 (24.0) | 0.836 |
| SST | 4.5 (4.0) | 5.2 (3.5) | 0.425 |
3.3 Radiographic outcomes
There was no significant difference in mean AHI values between the two cohorts at both the immediate postoperative time point and at final follow-up (Final mean AHI: tenodesis 8.8 ± 2.8 mm, no tenodesis 8.5 ± 4.7 mm, P = 0.704). There was also no difference in change in pre-to postoperative AHI at final follow-up between the two cohorts (tenodesis -2.3 ± 2.3 mm, no tenodesis -3.1 ± 4.8 mm, P = 0.348). In the tenodesis and no tenodesis cohorts, 32 patients (72.7%) and 38 patients (86.4%), respectively, demonstrated no evidence of radiographic loosening of the glenoid component at most recent follow up and the difference in glenoid radiolucencies between the two groups did not reach statistical significance (P = 0.186). (Table 4).
| Radiographic Outcome | Tenodesis (n = 44) | No Tenodesis (n = 44) | P-Value |
| Acromiohumeral Interval (mm) | |||
| Immediate Postoperative | 11.07 (3.58) | 12.19 (5.96) | 0.295 |
| Final Postoperative | 8.83 (2.76) | 8.51 (4.69) | 0.704 |
| Postoperative Change | −2.31 (2.32) | −3.09 (4.80) | 0.348 |
| Glenoid Loosening Score | |||
| 0 | 32 (72.7%) | 38 (86.4%) | 0.186 |
| 1 | 1 (2.3%) | 0 (0.0%) | N/A |
| 2 | 1 (2.3%) | 2 (4.5%) | 1.000 |
| 3 | 1 (2.3%) | 0 (0.0%) | N/A |
| 4 | 1 (2.3%) | 0 (0.0%) | N/A |
| 5 | 6 (13.6%) | 4 (9.1%) | 0.737 |
3.4 Implant survival and revision
5-, 10-, and 15-year survival estimates for the tenodesis group based on the Kaplan-Meier analysis were 100% (95% CI: 100-100%), 93.4% (95% CI: 75.3–98.4%), and 66.6% (95% CI: 18.2–90.7%), respectively, while the 5-, 10-, and 15-year survival estimates for the no tenodesis group were 97.7% (95% CI: 84.9–99.7%), 97.7% (95% CI: 84.9–99.7%), 92.3% (95% CI: 69.7–98.2%), respectively (Fig. 1). There was no significant difference in rates of revision surgery between the two cohorts. Amongst the tenodesis cohort, 4 (9.1%) patients required a revision procedure while in the no tenodesis cohort, 3 (6.8%) patients underwent a revision procedure (P = 1.000).
4 Discussion
Despite the success of aTSA in treating the pain and shoulder motion restriction associated with severe glenohumeral osteoarthritis, some patients without diagnosed complications still experience refractory shoulder pain postoperatively. It has been theorized that these postoperative pain symptoms may arise from friction between the biceps tendon and humeral prosthesis.8–11 Additionally, authors have suggested that postoperative bicipital groove adhesions may restrict bicep tendon excursion, resulting in decreased ROM after shoulder arthroplasty without biceps tenodesis.9,18 In our matched cohort, there were no observed differences in pain, ROM, or PROs, implying that with appropriate patient selection it is possible to retain the native biceps tendon and still avoid postoperative pain and motion restriction.
In 2006, Tuckman et al. described a series of patients with anterior shoulder pain after shoulder arthroplasty. All 8 patients in the series underwent subsequent surgery resulting in eventual pain resolution and ROM improvements. Seven shoulders underwent arthroscopic treatment including biceps tendon resection or tenodesis, subacromial decompression, and associated procedures. The biceps tendon was seen scarred into the tuberosity repair in 3 patients who had undergone hemiarthroplasty for fracture, whereas the other patients were found to have scarring and fibrosis of the tendon along the bicipital groove. The authors proposed that the development of scar around the proximal biceps tendon resulted in decreased tendon excursion. They theorized that the proximal biceps tendon was thus effectively anchored at both its anatomic origin proximally and pathologically at the biceps groove distally. They went on to hypothesize that with certain motions and arm positions, contact with and displacement by the humeral prosthesis caused the tendon to experience supraphysiological tension resulting in anterior shoulder pain. The authors labeled this pathological mechanism the ‘bowstring effect’ and advocated for tenodesis at the time of primary arthroplasty.8 Although some instances of postoperative pain following aTSA may be related to the biceps tendon, there are a host of alternative explanations that should not be discounted. Other reports have, for example, described subacromial impingement or capsulitis as causing ROM restrictions and pain postoperatively, and both have been successfully treated with arthroscopic subacromial decompression or shoulder debridement.19,20
A randomized trial from Soliman et al. investigated the idea that the biceps tendon may cause pain in 37 patients after hemiarthroplasty for fracture. The cohort undergoing biceps tenodesis reported significantly higher Constant scores with fewer patients in this group reporting postoperative pain.21 There was no significant difference in postoperative forward elevation, the only reported ROM measure. In the study by Soliman, two patients in the no tenodesis group ultimately underwent subsequent biceps tenodesis for persistent pain, with resolution of symptoms in both patients. In contrast to the study by Soilman et al. the study presented here demonstrates no difference in revision rates between the tenodesis and no tenodesis cohorts. The presence or absence of trauma is the most likely explanation for the reported differences between the two studies. No patients underwent shoulder arthroplasty for a traumatic indication in the study presented here. In contrast, all 37 patients evaluated by Soliman et al. underwent surgery for a traumatic indication and it is certainly plausible that the initial trauma injured the biceps directly, created a pro-inflammatory environment leading to increased scarring and fibrosis, or that the biceps tendon ultimately scarred into the tuberosity repair as noted by Tuckman et al.8
Two retrospective studies reported outcomes of patients undergoing elective shoulder arthroplasty with or without biceps tenodesis. Fama et al. compared a group of 108 patients with biceps tenodesis and 481 patients without biceps tenodesis in the setting of concomitant aTSA and found tenodesis significantly better than no tenodesis in most outcomes including Constant score, forward elevation, and external rotation. A closer inspection of the results shows only the mobility subsection of the Constant score to be different by more than 1 point. Additionally, the average follow-up was only 43 months, with a range of 1–110 months.9 This short and wide-ranging follow-up could easily influence ROM values, as patients often take up to a year for maximal ROM gains after shoulder arthroplasty.22,23 Godeneche et al. reported on 268 patients undergoing aTSA for glenohumeral arthritis, with 75 of these patients undergoing concomitant biceps tenodesis. This group found a significant, but small decrease of 13.7 compared to 12.4 in the Constant pain subscore for the tenodesis group. Both of these scores would be in the range of mild to no pain.11 Both of these retrospective studies advocated for routine biceps tenodesis during shoulder arthroplasty.
Simmen et al. reported the only previous prospective study examining biceps tenodesis after primary total shoulder arthroplasty. Biceps tenodesis was performed in 54 out of 136 (39.7%) shoulders with at least 1 year follow-up in their prospective cohort. The results of this study were organized by treatment success or failure, which was reported as final Constant score greater than 80 or less than 80, respectively. Final analysis found an odds ratio of 2.97 for performance of concomitant biceps tenodesis and ultimate treatment success, suggesting that concomitant biceps tenodesis should be performed in all patients. This is not in agreement with the current study that found no significant differences in PROs or ROM with or without biceps tenodesis. Recent studies have examined the concept of treatment success in terms of Constant score after aTSA, finding a minimal clinically important difference of 5.7 points, a substantial clinical benefit of 19.1 points, and a substantial clinically important percentage of maximal possible improvement of 39%.24–26 These values are a much more accurate way to determine success compared with the arbitrary Constant score of 80, especially considering that Simmen et al. did not include preoperative Constant values to determine postoperative improvement. Additionally, only 33.8% of patients attained “treatment success”, calling into question the validity of this judgment. The authors were aware of these concerns, stating in the discussion that “clearly, our study needs confirmation from other prospective cohorts and perhaps randomized studies”.10
This is the first study to our knowledge correlating concomitant biceps tenodesis with radiographic findings including glenoid loosening and AHI, and differences were identified with regard to these radiographic outcomes. The change in AHI was not different between the groups, indicating that an intact biceps tendon did not influence postoperative position of the humeral head in the axial plane. Additionally, concomitant biceps tenodesis did not significantly influence glenoid radiolucent lines at final follow-up or implant survivorship, with strong survivorship numbers found in both groups. No patients in our cohort required a subsequent biceps tenodesis, and there was no significant difference in revision rates at final follow-up. These results indicate that the presence or absence of the native biceps tendon in the shoulder joint does not significantly influence the glenohumeral interaction or survivorship of aTSA.
There are limitations present in the current study despite the matched groups and long follow-up period. As with all retrospective studies, there is innate bias with patient selection and data reporting. Additionally, in the earlier portion of the study period, it was standard practice for the investigating surgeon to preserve the biceps unless pathological changes were grossly evident intraoperatively and nearly all patients in the no tenodesis cohort come from this period. Later in the study period, it became standard practice for the investigating surgeon to perform a biceps tenodesis at the time of total shoulder arthoplasty in all patients and nearly all patients in the tenodesis cohort come from this period. This difference in surgical date is a potential confounding variable as changes to implant design over time may have altered interactions between the humeral head and biceps tendon. Further, the current study did not include any postoperative MRI or ultrasound imaging to evaluate the status of the biceps tendon at most recent follow-up. It is possible that one or more patients did have symptoms arising from the biceps postoperatively, they sustained a biceps rupture with relief of pain, and this rupture went undetected. Finally, this is a single-surgeon study from a single academic institution and may not be generalizable to other surgeons or surgical techniques.
5 Conclusions
This matched, long-term comparison of patients undergoing aTSA with or without concomitant biceps tenodesis found no significant differences in shoulder pain, functional outcome, or ROM scores. The presented data indicates that it is possible for patients with a retained native biceps tendon to have equivalent outcomes after aTSA. Simultaneously, the presence of the native biceps tendon postoperatively does not appear to have an impact on long-term implant related outcomes. Further study would be needed to determine if retaining a scarred, tenosynovitic, injured, or inflamed biceps tendon would allow for similar positive outcomes.
CRediT authorship contribution statement
Priya Singh: Conceptualization, Data curation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing. Akiro H. Duey: Conceptualization, Formal analysis, Methodology, Project administration, Visualization, Writing – review & editing. Troy Li: Conceptualization, Formal analysis, Methodology, Writing – review & editing. Akshar V. Patel: Conceptualization, Data curation, Methodology. William A. Ranson: Supervision, Visualization, Writing – review & editing. Evan M. Michaelson: Visualization, Writing – review & editing. Bradford O. Parsons: Conceptualization, Writing – review & editing. Evan L. Flatow: Conceptualization, Methodology, Writing – review & editing. Paul J. Cagle: Conceptualization, Methodology, Writing – review & editing.
Guardian/patient consent
Not applicable to original research article.
IRB
Icahn School of Medicine at Mount Sinai STUDY-17-00684-CR005.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References
- A systematic review of proposed rehabilitation guidelines following anatomic and reverse shoulder arthroplasty. J Orthop Sports Phys Ther. 2019;49(5):337-346.
- [Google Scholar]
- Survivorship of anatomic total shoulder arthroplasty. J Am Acad Orthop Surg. 2022;30(10):457-465.
- [Google Scholar]
- Anatomic total shoulder arthroplasty for primary glenohumeral osteoarthritis is associated with excellent outcomes and low revision rates in the elderly. J Shoulder Elbow Surg. 2021;30(7S):S131-S139.
- [Google Scholar]
- Total shoulder arthroplasty: long-term survivorship, functional outcome, and quality of life. J Shoulder Elbow Surg. 2005;14(5):471-479.
- [Google Scholar]
- Clinical and radiological follow-up of the Aequalis third-generation cemented total shoulder replacement: a minimum ten-year study. J Bone Joint Surg Br. 2009;91(12):1594-1600.
- [Google Scholar]
- Total shoulder arthroplasty with the Neer prosthesis: long-term results. J Shoulder Elbow Surg. 1997;6(6):495-505.
- [Google Scholar]
- Long head of the biceps pathology as a cause of anterior shoulder pain after shoulder arthroplasty. J Shoulder Elbow Surg. 2006;15(4):415-418.
- [Google Scholar]
- The role of concomitant biceps tenodesis in shoulder arthroplasty for primary osteoarthritis: results of a multicentric study. Orthopedics. 2004;27(4):401-405.
- [Google Scholar]
- Usefulness of concomitant biceps tenodesis in total shoulder arthroplasty: a prospective cohort study. J Shoulder Elbow Surg. 2008;17(6):921-924.
- [Google Scholar]
- Prosthetic replacement in the treatment of osteoarthritis of the shoulder: early results of 268 cases. J Shoulder Elbow Surg. 2002;11(1):11-18.
- [Google Scholar]
- Long-term results, functional outcomes and complications after open reduction and internal fixation of neglected and displaced greater tuberosity of humerus fractures. Archives of Bone and Joint Surgery. 2016;4(4):330.
- [Google Scholar]
- Subacromial space in the rheumatoid shoulder: a radiographic 15-year follow-up study of 148 shoulders. J Shoulder Elbow Surg. 2000;9(3)
- [Google Scholar]
- A comparison of pegged and keeled glenoid clinical outcomes at long-term follow-up after total shoulder arthroplasty. J Orthop. 2023;36:120-124.
- [Google Scholar]
- The radiographic evaluation of keeled and pegged glenoid component insertion. J Bone Joint Surg Am. 2002;84(7)
- [Google Scholar]
- MatchIt: nonparametric preprocessing for parametric causal inference. J Stat Software. 2011;42:1-28.
- [Google Scholar]
- The effect of biceps adhesions on glenohumeral range of motion: a cadaveric study. J Shoulder Elbow Surg. 2013;22(5)
- [Google Scholar]
- The role of arthroscopy for the problem shoulder arthroplasty. J Shoulder Elbow Surg. 2002;11(3)
- [Google Scholar]
- The role of arthroscopy in painful shoulder arthroplasty: is revision always necessary? Arthroscopy. 2020;36(6):1508-1514.
- [Google Scholar]
- Proximal humeral fractures treated with hemiarthroplasty: does tenodesis of the long head of the biceps improve results? Injury. 2013;44(4)
- [Google Scholar]
- Postoperative stiffness after reverse total shoulder arthroplasty. Shoulder Elbow. 2022;14(2)
- [Google Scholar]
- Speed of recovery after shoulder arthroplasty: a comparison of reverse and anatomic total shoulder arthroplasty. J Shoulder Elbow Surg. 2014;23(12)
- [Google Scholar]
- Quantifying success after anatomic total shoulder arthroplasty: the substantial clinically important percentage of maximal possible improvement. J Shoulder Elbow Surg. 2023;32(11)
- [Google Scholar]
- Quantifying success after total shoulder arthroplasty: the substantial clinical benefit. J Shoulder Elbow Surg. 2018;27(5)
- [Google Scholar]
- Quantifying success after total shoulder arthroplasty: the minimal clinically important difference. J Shoulder Elbow Surg. 2018;27(2):298-305.
- [Google Scholar]