As the volume of shoulder arthroplasty procedures performed in the United States continues to increase, the predicted number of revision shoulder arthroplasties grows even higher.1,2 Conversion of failed shoulder arthroplasty to reverse total shoulder arthroplasty has become common. Because reverse total shoulder arthroplasty only received US Food and Drug Administration approval in 2003,3 there has been significant investment and advancement in commercially available reverse shoulder arthroplasty systems.
One of the major marketed advantages of many popular shoulder arthroplasty systems is a convertible platform system in which the same universal humeral stem is used for both total shoulder arthoplasty and reverse total shoulder arthroplasty.4,5 Most reports on the treatment of failed shoulder arthroplasty with reverse total shoulder arthroplasty describe removal of the humeral stem in nearly every patient.6–11 If total shoulder arthroplasty or hemiarthroplasty fails and requires revision to reverse shoulder arthroplasty, then the humeral stem can be retained and only the glenoid must be converted to a glenosphere. Removal of the humeral stem can result in significant morbidity because removing a well-fixed humeral component can lead to gross bone loss in an often osteopenic proximal humerus.7,12 However, limited data support the utility of avoiding humeral stem revision by using a shoulder arthroplasty system with a convertible humeral stem. Few studies have compared humeral stem retention vs revision in the conversion of failed shoulder arthroplasty to reverse total shoulder arthroplasty.
This study investigated whether retaining the humeral stem in the conversion of failed shoulder arthroplasty to reverse total shoulder arthroplasty is preferable to revision of the humeral stem. The null hypothesis was that there is no difference between the groups in patient-reported outcomes, intraoperative variables, or complications.
Materials and Methods
Institutional review board approval was obtained before data collection. A retrospective review was performed that identified 32 shoulders in 32 patients with failed shoulder arthroplasty that were revised to reverse total shoulder arthroplasty in the senior author’s practice (L.D.H.) from June 2006 to May 2011. Inclusion criteria were patient age older than 18 years at the time of revision surgery and revision of stemmed hemiarthroplasty or total shoulder arthroplasty to reverse total shoulder arthroplasty. Exclusion criteria were patient age younger than 18 years at the time of revision surgery, obligate humeral stem removal because of prosthetic infection, use of an articulating antibiotic cement spacer, and use of stemless resurfacing shoulder arthroplasty. Patients were excluded from analysis of patient-reported outcomes or postoperative complications if they had follow-up of less than 2 years, if they had inadequate follow-up data, or if they died within 2 years of follow-up. The study group included 12 patients who had the humeral stem retained. The control group included 14 patients who underwent humeral stem revision.
Patient data were reviewed, including age at revision surgery, sex, operative side, type of primary prosthesis, and length of clinical follow-up. Intraoperative data were reviewed, including total operative time, intraoperative blood loss, method of humeral stem removal if performed, and perioperative complications. Postoperative clinical outcome scores of both the operative side and the nonoperative side were reviewed, including Subjective Shoulder Value,13,14 American Shoulder and Elbow Surgeons,15 and Simple Shoulder Test scores.16 Only patients with follow-up of more than 2 years were included in the analysis of patient-reported outcomes. Length of inpatient hospital stay and postoperative complications were also reviewed.
All procedures were performed in the beach chair position by the senior author (L.D.H.). A deltopectoral approach was used in all patients. The prosthesis was exposed by peeling the subscapularis directly off the lesser tuberosity, if present. All patients underwent revision to a Zimmer Anatomical Shoulder Inverse/Reverse System and the Zimmer Trabecular Metal Reverse Shoulder System (Zimmer, Warsaw, Indiana). The humeral stem was retained if it was placed at the appropriate height and version for reverse total shoulder arthroplasty (Figure 1). If removal was required, initial extraction was attempted after the head was removed, and complete debridement of all bone, soft tissue, or cement incarcerating the stem at the top of the metaphysis was performed. The stem was then removed with a mallet with a stem extractor or a carbide-tipped punch in an axial direction. If the stem could not be removed by this method, longitudinal or window osteotomy was performed. An osteotome was used to create a single, controlled fracture line along the anterior humerus from the metaphysis to the stem tip. Osteotomes were further used to open the osteotomy and free the prosthesis. The stem extractor or punch was then used to remove the humeral component. The osteotomy was repaired with a single 18-gauge cerclage wire around the metaphysis.
Preoperative anteroposterior radiograph of the left shoulder showing total shoulder arthroplasty with superior migration of the humeral component indicative of rotator cuff failure after arthroplasty had been performed for glenohumeral arthritis (A). Cement was used for fixation of the humeral stem from a system that allows conversion to reverse shoulder arthroplasty with retention of the humeral stem. Revision to reverse shoulder arthroplasty was subsequently performed because of pain and disability associated with rotator cuff failure and resultant glenoid component loosening. Postoperative anteroposterior radiograph obtained 18 months after revision showing a well-fixed reverse prosthesis with retention of the original humeral stem (B).
The glenoid was circumferentially exposed, and the prosthetic glenoid component was carefully removed with an osteotome through the pegs of the glenoid. The baseplate and glenosphere were placed after glenoid preparation. A trial of the humeral component was then performed for appropriate tension and impingement-free functional range of motion. The final components were placed, and the subscapularis was repaired when possible before closure of the deltopectoral interval and skin. Cement fixation was used for all revised humeral stems (Figure 2).
Preoperative anteroposterior radiograph of the left shoulder in a patient who had a cemented hemiarthroplasty placed for a comminuted proximal humerus fracture with a humeral stem that was not convertible to reverse shoulder arthroplasty (A). High stem placement resulted in non-anatomic position of the tuberosities around the neck of the humeral stem that led to significant pain and disability. When the patient underwent revision to reverse shoulder arthroplasty, removal of the cemented stem was needed to place a humeral component that would mate with a glenosphere as demonstrated on the postoperative radiograph. An osteotomy was required to remove the stem, resulting in additional surgical dissection, increased blood loss, and increased operative time (B).
All patients were placed in a shoulder immobilizer for 4 weeks, with full elbow, wrist, and hand active range of motion allowed immediately. Full active and passive range of motion with formal physical therapy was begun at 4 weeks.
A Wilcoxon test was performed to compare means of normally distributed continuous variables, and Fisher’s exact test was performed to compare groups of nominal variables. Significant differences between groups were determined with P<.05. Post hoc power analysis was performed. Given the sample size, the study had 52.3% power to detect a 10% difference in complication rates at 90% confidence.
There were 26 patients with failed shoulder arthroplasty who underwent revision to reverse total shoulder arthroplasty, with a mean age of 68.46 years (range, 43.42–83.12 years; SD, 9.07), who met the study inclusion criteria (Table 1). This group included 19 patients with previous hemiarthroplasty and 7 patients with previous total shoulder arthroplasty. In the stem revision group, 7 patients had a cemented index humeral stem and 7 patients had a press-fit humeral stem. In the stem retention group, 7 patients had a cemented stem and 5 patients had a press-fit stem. Of the 26 patients, 24 were women. In addition, 14 (53.85%) patients underwent stem revision and 12 (46.15%) retained the stem at the time of conversion to reverse total shoulder arthroplasty. In the 14 patients who underwent stem revision, osteotomy was required to remove a well-fixed stem in 3 patients (21.43%). Of the 14 patients who had a convertible stem, 12 underwent successful conversion to reverse shoulder arthroplasty without stem revision. Mean operative time was 210.19 minutes (range, 105–335 minutes; SD, 59.79); mean blood loss was 540.00 mL (range, 100–1500 mL; SD, 330.09); and mean length of hospital stay was 3.23 days (range 2–5 days; SD, 0.86). There were 21 patients with follow-up of greater than 2 years who were included in the analysis of patient-reported outcomes (Table 2).
Descriptive Statistics at 2-year Follow-up
Comparison of the cohorts of retained vs revised humeral stems showed no statistically significant differences in age, side, or type of index arthroplasty (Table 3). Patients who required humeral stem revision had greater intraoperative blood loss (596.43 vs 468.18 mL), length of hospital stay (3.50 vs 2.92 days), and operative time (237.00 vs 178.92 minutes). The only recorded variable that showed a statistically significant difference between groups was operative time (P=.0109). There were 7 perioperative complications (Table 4), 5 in the revised stem group (35.71%) and 2 in the retained stem group (19.64%). No cases of postoperative infection or instability occurred. In the 21 patients with more than 2-year clinical follow-up, no significant differences were seen in Subjective Shoulder Value, American Shoulder and Elbow Surgeons, or Simple Shoulder Test scores when the humeral stem retention group and the revision group were compared at a mean follow-up of 34.38 months (Table 3).
Difference in Variables by Stem Type
Of the 26 patients, 14 had a stem that was convertible to reverse total shoulder arthroplasty. The stem was retained in 12 of the 14 patients. In the 2 patients who had a modular or convertible humeral stem, the stem was loose intraoperatively and was easily removed during exposure. Gross stem loosening was the only reason for extraction of a convertible stem. In 12 of the 14 patients who had revision of the humeral stem, the humeral component was removed because of lack of modularity. Soft tissue contracture did not prevent the authors’ ability to retain the humeral stem in any patient after the appropriate releases were performed. In patients whose original humeral version was not optimal with a well-fixed modular stem, the stem was still and impingement-free range of motion was sufficient.
This study was among the first to address the question of whether retaining the humeral stem during conversion of failed shoulder arthroplasty to reverse total shoulder arthroplasty is preferable to revising the humeral stem. The results of this study showed that retaining the humeral stem was associated with a statistically significant decrease in operative time. Decreased perioperative complications, decreased blood loss, and decreased length of hospital stay were also noted in the stem retention group, but the differences were not statistically significant. Many shoulder arthroplasty systems that are available in the United States offer a convertible platform system that allows either anatomic or reverse total shoulder arthroplasty with a universal humeral stem. These newer arthroplasty systems are theoretically useful in revision arthroplasty, but little evidence supports their efficacy over nonplatform systems.
Wieser et al17 recently performed a retrospective case-control study comparing revision shoulder arthroplasty with total shoulder arthroplasty performed with and without humeral stem revision. As in the current study, these authors found that stem retention resulted in less intraoperative blood loss, shorter operative time, fewer intraoperative complications, and fewer revisions at a minimum follow-up of 1 year.17 Ortmaier et al18 reported high patient satisfaction but a 24% complication rate in 50 patients undergoing conversion of failed shoulder arthroplasty to reverse total shoulder arthroplasty. Retention of the humeral stem was not possible or was not used in any of these patients, and osteotomy was required to remove the stem in 40%. Patel et al19 reported 31 patients who were treated with reverse shoulder arthroplasty for failed shoulder arthroplasty. In 6 of these 31 patients (19.4%), the modular components allowed the humeral stem to be retained for reverse total shoulder arthroplasty. Werner et al11 studied the effectiveness of a modular shoulder arthroplasty system in the conversion of 14 patients with a rotator cuff-deficient shoulder who had undergone previous hemiarthroplasty and reported a 14% postoperative complication rate. The humeral stem was retained in all 14 study patients. Castagna et al10 showed satisfactory results in the conversion of 26 cases of failed hemiarthroplasty or unconstrained total shoulder arthroplasty to reverse total shoulder arthroplasty. Unlike the study by Ortmaier et al,18 the modular humeral stem was retained in all patients. Similar to the results in the stem retention group in the current study, Castagna et al10 reported a shorter operative time, few intraoperative complications, and satisfactory clinical outcomes.
Removal of a well-fixed humeral stem has significant potential morbidity, including gross bone loss, additional exposure, need for osteotomy, damage to remaining muscular attachments, and additional surgical time. The current study found increased blood loss and operative time in the stem revision group. In addition, 3 of the 14 (21.4%) patients who required stem removal had a humeral shaft or tuberosity fracture during stem removal. In contrast, no humeral complications occurred in the stem retention group. Several recent reports of revision shoulder arthroplasty showed inferior clinical outcomes in patients who needed humeral stem removal for successful revision. Sassoon et al20 found that stem revision resulted in reduced pain scores in 68 shoulders undergoing revision to total shoulder arthroplasty after hemiarthroplasty for glenoid arthrosis. Postacchini et al21 reported the lowest Constant scores in patients undergoing total shoulder arthroplasty for failed postfracture hemiarthroplasty that necessitated removal of cemented humeral stems. Patel et al19 reported that 18.1% of patients undergoing reverse total shoulder arthroplasty for failed shoulder arthroplasty required osteotomy for removal of the well-fixed stem. The current study found a similar rate of osteotomy in 3 (21.4%) of 14 patients in the stem revision group. Levy et al6,7 and Walker et al8 reported that reverse total shoulder arthroplasty in patients with failed hemiarthroplasty and failed total shoulder arthroplasty provided improvement in function and pain relief, but the procedure was accompanied by a high complication rate. This was especially true in patients with proximal humeral bone loss.7 Unlike these reports, the current study showed no clear differences in patient-reported outcomes with stem retention compared with revision. However, the authors’ experience with patients who required stem revision in conversion to reverse total shoulder arthroplasty was consistent with previous reports indicating that satisfactory results can be achieved in this patient population, but high complication rates can be expected.6,7
Limitations of this study included the retrospective design, limited number of patients, and recording bias. Performing a similar study prospectively with a larger sample size may show greater significance in the differences between the humeral stem revision and retention groups. However, this study showed reduced operative time when the humeral stem can be retained during conversion of failed shoulder arthroplasty to reverse total shoulder arthroplasty. Future research is needed to identify differences in cost between retained and revised humeral stems to determine the value of humeral stem retention.
This study showed that humeral stem retention resulted in decreased operative time compared with humeral stem revision when failed shoulder arthroplasty was converted to reverse total shoulder arthroplasty. The findings also showed decreased complications, blood loss, and length of hospital stay in the stem retention group. However, these differences were not statistically significant. Of patients undergoing humeral stem removal, 21.4% had an intraoperative humeral shaft or tuberosity fracture. No humeral complications occurred in the stem retention group. The results of this study did not support the null hypothesis that there is no difference between patients with retention vs revision of the humeral stem during conversion to reverse total shoulder arthroplasty. By avoiding humeral stem revision, the use of a platform shoulder arthroplasty system may be beneficial in patients with failed shoulder arthroplasty who undergo conversion to reverse total shoulder arthroplasty.
- Day JS, Lau E, Ong KL, Williams GR, Ramsey ML, Kurtz SM. Prevalence and projections of total shoulder and elbow arthroplasty in the United States to 2015. J Shoulder Elbow Surg. 2010; 19(8):1115–1120. doi:10.1016/j.jse.2010.02.009 [CrossRef]
- United States Bone and Joint Initiative. The Burden of Musculoskeletal Diseases in the United States. 2nd ed. Rosemont, IL: American Academy of Orthopaedic Surgeons; 2011.
- Nam D, Kepler CK, Neviaser AS, et al. Reverse total shoulder arthroplasty: current concepts, results, and component wear analysis. J Bone Joint Surg Am. 2010; 92(suppl 2):23–35. doi:10.2106/JBJS.J.00769 [CrossRef]
- Jones RB. Hemiarthroplasty for proximal humeral fractures: indications, pitfalls, and technique. Bull Hosp Jt Dis (2013). 2013; 71(suppl 2):60–63.
- Wright TW. Revision of humeral components in shoulder arthroplasty. Bull Hosp Jt Dis (2013). 2013; 71(suppl 2):77–81.
- Levy J, Frankle M, Mighell M, Pupello D. The use of the reverse shoulder prosthesis for the treatment of failed hemiarthroplasty for proximal humeral fracture. J Bone Joint Surg Am. 2007; 89(2):292–300. doi:10.2106/JBJS.E.01310 [CrossRef]
- Levy JC, Virani N, Pupello D, Frankle M. Use of the reverse shoulder prosthesis for the treatment of failed hemiarthroplasty in patients with glenohumeral arthritis and rotator cuff deficiency. J Bone Joint Surg Br. 2007; 89(2):189–195. doi:10.1302/0301-620X.89B2.18161 [CrossRef]
- Walker M, Willis MP, Brooks JP, Pupello D, Mulieri PJ, Frankle MA. The use of the reverse shoulder arthroplasty for treatment of failed total shoulder arthroplasty. J Shoulder Elbow Surg. 2012; 21(4):514–522. doi:10.1016/j.jse.2011.03.006 [CrossRef]
- Kelly JD II, Zhao JX, Hobgood ER, Norris TR. Clinical results of revision shoulder arthroplasty using the reverse prosthesis. J Shoulder Elbow Surg. 2012; 21(11):1516–1525. doi:10.1016/j.jse.2011.11.021 [CrossRef]
- Castagna A, Delcogliano M, de Caro F, et al. Conversion of shoulder arthroplasty to reverse implants: clinical and radiological results using a modular system. Int Orthop. 2013; 37(7):1297–1305. doi:10.1007/s00264-013-1907-4 [CrossRef]
- Werner BS, Boehm D, Gohlke F. Revision to reverse shoulder arthroplasty with retention of the humeral component. Acta Orthop. 2013; 84(5):473–478. doi:10.3109/17453674.2013.842433 [CrossRef]
- Budge MD, Moravek JE, Zimel MN, Nolan EM, Wiater JM. Reverse total shoulder arthroplasty for the management of failed shoulder arthroplasty with proximal humeral bone loss: is allograft augmentation necessary?J Shoulder Elbow Surg. 2013; 22(6):739–744. doi:10.1016/j.jse.2012.08.008 [CrossRef]
- Gerber C, Fuchs B, Hodler J. The results of repair of massive tears of the rotator cuff. J Bone Joint Surg Am. 2000; 82(4):505–515.
- Gilbart MK, Gerber C. Comparison of the subjective shoulder value and the Constant score. J Shoulder Elbow Surg. 2007; 16(6):717–721. doi:10.1016/j.jse.2007.02.123 [CrossRef]
- Richards RR, An KN, Bigliani LU, et al. A standardized method for the assessment of shoulder function. J Shoulder Elbow Surg. 1994; 3(6):347–352. doi:10.1016/S1058-2746(09)80019-0 [CrossRef]
- Lippitt SB, Harryman DT II, Matsen FA III, . A practical tool for evaluating function: the simple shoulder test. In: Matsen FA III, Fu FH, Hawkins RJ, eds. The Shoulder: A Balance of Mobility and Stability. Rosemont, IL: American Academy of Orthopaedic Surgeons; 1993:501–518.
- Wieser K, Borbas P, Ek ET, Meyer DC, Gerber C. Conversion of stemmed hemi- or total to reverse total shoulder arthroplasty: advantages of a modular stem design. Clin Orthop Relat Res. 2015; 473(2):651–660. doi:10.1007/s11999-014-3985-z [CrossRef]
- Ortmaier R, Resch H, Matis N, et al. Reverse shoulder arthroplasty in revision of failed shoulder arthroplasty: outcome and follow-up. Int Orthop. 2013; 37(1):67–75. doi:10.1007/s00264-012-1742-z [CrossRef]
- Patel DN, Young B, Onyekwelu I, Zuckerman JD, Kwon YW. Reverse total shoulder arthroplasty for failed shoulder arthroplasty. J Shoulder Elbow Surg. 2012; 21(11):1478–1483. doi:10.1016/j.jse.2011.11.004 [CrossRef]
- Sassoon AA, Rhee PC, Schleck CD, Harmsen WS, Sperling JW, Cofield RH. Revision total shoulder arthroplasty for painful glenoid arthrosis after humeral head replacement: the nontraumatic shoulder. J Shoulder Elbow Surg. 2012; 21(11):1484–1491. doi:10.1016/j.jse.2011.11.028 [CrossRef]
- Postacchini R, Castagna A, Borroni M, Cinotti G, Postacchini F, Gumina S. Total shoulder arthroplasty for the treatment of failed hemiarthroplasty in patients with fracture of the proximal humerus. J Shoulder Elbow Surg. 2012; 21(11):1542–1549. doi:10.1016/j.jse.2011.12.007 [CrossRef]
| Male||2 (7.69%)|
| Female||24 (92.31%)|
| Right||14 (53.85%)|
| Left||12 (46.15%)|
|Previous surgery, No.|
| Hemiarthroplasty||19 (73.08%)|
| Total shoulder arthroplasty||7 (26.92%)|
|Method of stem removal (for revised stems), No.|
| No osteotomy||11 (78.57%)|
| Osteotomy||3 (21.43%)|
| Retained||12 (46.15%)|
| Revised||14 (53.85%)|
|Blood loss, mean±SD (range), mL||540.00±330.09 (100–1500)|
|Hospital stay, mean±SD (range), d||3.23±0.86 (2–5)|
|Operative time, mean±SD (range), min||210.19±59.79 (105–335)|
|Age, mean±SD (range), y||68.46±9.07 (43.42–83.12)|
Descriptive Statistics at 2-year Follow-up
| Male||1 (4.76%)|
| Female||20 (95.24%)|
| Right||13 (61.90%)|
| Left||8 (38.10%)|
|Previous surgery, No.|
| Hemiarthroplasty||16 (76.19%)|
| Total shoulder arthroplasty||5 (23.81%)|
|Method of removal (for replaced stems), No.|
| No osteotomy||11 (78.57%)|
| Osteotomy||3 (21.43%)|
| Retained||7 (33.33%)|
| Revised||14 (66.67%)|
|Subjective Shoulder Value score, mean±SD (range)|
| Operative shoulder||68.75±18.91 (20–96)|
| Nonoperative shoulder||81.31±21.72 (10–100)|
|American Shoulder and Elbow Surgeons score, mean±SD (range)|
| Operative shoulder||68.38±16.58 (40–96.7)|
| Nonoperative shoulder||74.94±20.60 (36.7–100)|
|Simple Shoulder Test score, mean±SD (range)|
| Operative shoulder||50.00±22.30 (8.3–75)|
| Nonoperative shoulder||83.34±21.12 (25–100)|
|Age, mean±SD (range), y||67.80±8.72 (43.42–83.12)|
| Follow-up, mean±SD, mo||34.38±11.71|
Difference in Variables by Stem Typea
| Right||4 (33.33%)||10 (71.43%)|
| Left||8 (66.67%)||4 (28.57%)|
|Previous surgery, No.||.6652b|
| Hemiarthroplasty||8 (66.67%)||11 (78.57%)|
| Total shoulder arthroplasty||4 (33.33%)||3 (21.43%)|
|Complications, No.||2 (19.64%)||5 (35.71%)||.3913b|
|Subjective Shoulder Value score, operative shoulder, mean±SD||68.00±29.40||69.15±11.55||.3913c|
|American Shoulder and Elbow Surgeons score, operative shoulder, mean±SD||72.39±22.15||66.23±13.25||.5337c|
|Simple Shoulder Test score, operative shoulder, mean±SD||47.63±27.92||51.28±19.80||.8745c|
|Age, mean±SD, y||69.67±7.75||66.86±9.30||.7406c|
|Blood loss, mean±SD, mL||468.18±257.17||596.43±377.47||.5260c|
|Hospital stay, mean±SD, d||2.92±0.79||3.50±0.85||.0911c|
|Operative time, mean±SD, min||178.92±44.56||237.00±59.32||.0109c|
|Stem Retention||Stem Revision|
|Postoperative glenoid baseplate failure requiring revision||Intraoperative humerus fracture requiring acute conversion to long-stem prosthesis with subsequent nonunion requiring open reduction and internal fixation 10 months later|
|Transient nerve palsy of the posterior cord of the brachial plexus||Intraoperative humerus fracture requiring acute conversion to long-stem prosthesis|
|Cement extravasation requiring acute distal humeral exposure and cement removal|
|Greater tuberosity fracture requiring open reduction and internal fixation|
|Transient radial nerve palsy|