Osteosynthesis of osteoporotic and comminuted proximal humerus fractures (PHFs) can be challenging.1,2 Failures of open reduction and internal fixation have led some surgeons to pursue acute arthroplasty as an alternative treatment option. However, hemiarthroplasty for fracture has traditionally been associated with unpredictable results, causing many surgeons to extend the indications of reverse shoulder arthroplasty (RSA) to displaced PHF.3–6
Cementless fixation of the humeral component is currently preferred by most surgeons when shoulder arthroplasty is performed for other indications. Modern humeral components achieve proximal ingrowth in most patients, and the potential for cement removal is eliminated if revision surgery is required. However, RSA for fracture has traditionally been performed with cemented components7 because the fractured proximal humerus is perceived not to provide adequate support for metaphyseal fixation. Interest has grown regarding the use of uncemented humeral components in RSA for fracture, but reports on this remain limited.8–10 Additionally, no study has directly compared cementless humeral fixation with a control group of cemented humeral fixation for acute PHF treated with RSA.
The purpose of this study was to compare the clinical and radiographic outcomes of RSA using either cemented or uncemented humeral stems in the treatment of acute PHFs. The authors hypothesized that cementless humeral fixation would provide clinical and radiographic outcomes equivalent to those of cemented fixation.
Materials and Methods
Between June 2010 and February 2014, 46 shoulders with an acute PHF were treated with RSA at 3 separate institutions. Inclusion criteria were treatment within 1 month (31 days) of injury and completion of 2 years of clinical follow-up. Shoulders with pathologic fractures and patients with preoperative dementia were excluded. Eight patients were eliminated from clinical analysis because of death prior to 2-year follow-up (n=5) or dementia with the inability to complete clinical follow-up (n=3). The remaining 38 shoulders formed the basis of this study.
Following institutional review board approval, patient records were reviewed for injury mechanism, operative intervention, and clinical and radiographic outcomes. Reverse shoulder arthroplasties were performed an average of 7 days after the initial injury (range, 1–30 days). The most common cause of fracture was a fall from standing height (36 of 38). The mean age at the time of fracture was 74 years (range, 56–93 years). Women were more commonly affected (30 vs 8). Patient operative reports were reviewed to assess the type of humeral fixation and implant. All surgeries were performed by a fellowship-trained shoulder/elbow surgeon. The mode of fixation of the humeral component was selected based solely on surgeon preference. Two surgeons performed all their operations using cementless techniques. Two surgeons used cemented techniques for all cases. The remaining surgeons transitioned from using cemented techniques to cementless techniques over time. Once this change was made, no surgeon returned to using cemented techniques. Nineteen shoulders were treated with cemented humeral stems and 19 were treated with uncemented stems. Owing to disruption of the metaphyseal bone with the fracture, uncemented stems were placed to achieve rotational stability at the meta-diaphyseal junction. The cemented stem group had significantly longer follow-up than the uncemented group (44 vs 28 months, P=.03).
All patients were offered a return visit to assess postoperative range of motion (elevation, external rotation, and internal rotation). In 14 cases, patients refused in-person evaluation at most recent follow-up, and a validated shoulder questionnaire was used to assess range of motion.11 The patients not returning to clinic were also contacted via telephone to obtain visual analog scale pain scores, subjective satisfaction scores, Quick Disabilities of the Arm, Shoulder and Hand scores, and American Shoulder and Elbow Surgeons (ASES) scores. Subjective satisfaction was graded on a 4-point scale, with 1 corresponding to “much better,” 2 “better,” 3 “the same,” and 4 “worse” than before surgery.
Preoperative, immediate postoperative, and the most recent series of shoulder radiographs were reviewed. These routinely included an anteroposterior view with the arm in internal and external rotation as well as an axillary view. Nine patients elected to forgo radiographs at the time of most recent follow-up, and their most recent radiograph was included if performed at least 1 year postoperatively. Three shoulders did not have radiographs available after the 1-year postoperative time point. Radiographs were analyzed at a mean of 29 months (range, 12–69 months), with the 2 groups having similar follow-up (P=.91). All radiographs were reviewed by 2 orthopedic surgeons (B.S., J.S.S.). Preoperative radiographs were classified according to the Neer classification, with the addition of documenting the presence or absence of a fracture-dislocation.12 Immediate postoperative radiographs were used only for comparison with the most recent radiographs. Postoperative radiographs were assessed for tuberosity fixation and graded as anatomically united, malunited, nonunited, or resorbed. Humeral stem loosening was assessed according to the method described by Sanchez-Sotelo et al.13,14 Glenoid loosening was assessed according to the method described by Lazarus et al.15 Notching was assessed according to Sirveaux et al.16
All surgeries were performed through a deltopectoral approach. In most cases with fractures of both tuberosities, exposure was carried through the fracture site. When the lesser tuberosity was not fractured, it was osteotomized from the head. The tuberosities were mobilized, and nonabsorbable sutures were placed through the cuff and around the tuberosities. The fractured head segment was removed and saved for bone grafting.
The glenoid baseplate and glenosphere were placed according to the manufacturer's surgical technique. The glenosphere was sized relative to the patient's stature. Glenosphere sizes included 36 mm (n=22), 38 mm (n=6), 42 mm (n=7), and 44 mm (n=3). The humeral stem was fixed between 0° and 30° of retroversion based on surgeon preference. The decision to use cement for humeral fixation was based on surgeon preference. Uncemented stems included 12 Comprehensive stems (Biomet, Warsaw, Indiana) and 7 Delta Xtend stems (DePuy, Warsaw, Indiana). Cemented stems included 9 Comprehensive fracture stems, 6 Delta Xtend stems, and 4 Aequalis fracture stems (Tornier, St Paul, Minnesota). Antibiotics were added to the cement in 18 of the 19 cemented applications. After implantation of the humeral bearing, the joint was relocated and both tuberosities were repaired through the humeral prosthesis or around the humeral trunnion. Fractured tuberosities were repaired using nonabsorbable sutures passed through the bone–tendon junction and around the humeral stem and tied over the bicipital groove. However, exact techniques varied among surgeons and were modified based on the individual fracture.
Postoperatively, patients were immobilized in a sling with a small abduction pillow for 6 weeks. An abduction pillow was used to decrease tension on the rotator cuff to facilitate tuberosity healing. Active assisted range of motion was started at 6 weeks, with progression to graduated strengthening at 12 weeks.
Descriptive statistics are presented as mean (range) for continuous measures and number (percentage) for discrete variables. All 46 RSAs performed during the study period were included in the Kaplan–Meier survivorship analysis until the time of last follow-up and reported as estimated survival (95% confidence interval). Reoperation for any reason was considered an event for survivorship analysis. The 38 shoulders with minimum 2-year follow-up were included in clinical analyses. A Wilcoxon rank sum test was used to evaluate differences in postoperative pain, range of motion, and validated outcome scores. The alpha level for all tests was set at 0.05 for statistical significance. Statistical analyses were performed using JMP software (SAS Institute Inc, Cary, North Carolina).
Thirty-eight RSAs for acute PHF were analyzed. Fractures were classified as 2-part (n=4), 3-part (n=13), and 4-part (n=21). The four 2-part fractures had documented preoperative arthritic changes, which led the surgeon to directly proceed to RSA rather than consider open reduction and internal fixation. At the most recent follow-up, the mean visual analog scale pain score was 2.0 (range, 0–8). The mean range of motion was 123° of elevation (range, 80°–180°), 39° of external rotation (range, 0°–80°), and internal rotation to L3 (range, T5 to greater tro-chanter). The mean ASES score was 62.2 (range, 0–98), and the mean Quick Disabilities of the Arm, Shoulder and Hand score was 27.3 (range, 2–100). The mean satisfaction score was 1.5, with 33 of 36 patients rating their shoulder as much better or better than before surgery. All 3 unsatisfied patients were in the uncemented humeral fixation group. None of these patients experienced complications, developed radiographic loosening, or underwent reoperation.
Two shoulders (uncemented) developed postoperative complications, with both having reoperation. One shoulder dislocated 1 month after the index arthroplasty, requiring revision to a larger humeral polyethylene liner. It remained stable 2 years following revision surgery. A second patient sustained a periprosthetic fracture around a stable stem after a fall from standing height and was treated with internal fixation. At 5 years, the estimated survivorship for the whole cohort was 92.9% (95% confidence interval, 75.6–98.3).
Cemented Versus Uncemented Groups
The uncemented and cemented humeral stem groups were similar regarding age, sex, body mass index, and Charlson Comorbidity Index (P>.05). The cemented group had significantly longer clinical follow-up (43.9 vs 27.9 months, P=.03). Fracture patterns were also similarly distributed between groups, with each having 2 fracture-dislocations (P=.9).
Pain scores and range of motion were similar in the 2 groups. Quick Disabilities of the Arm, Shoulder and Hand scores were also statistically similar between the groups. However, patients treated with a cemented stem had significantly better ASES scores and subjective satisfaction scores (Table 1). Complications were more common in the uncemented group but not directly related to the humeral fixation technique (P=.08). Differences in ASES scores and subjective satisfaction scores remained significant even after removal of the 2 shoulders with complications (P=.007 and P=.04).
Demographics and Outcomes of Cemented Versus Uncemented Humeral Stems
Radiographic follow-up was similar between cemented and uncemented stems (28.9 vs 30.2 months, P=.9). There was no significant difference in tuberosity union (68% vs 53%, P=.4); however, anatomic tuberosity union was significantly higher with cemented components (n=9) compared with uncemented components (n=3) (P=.02). Radiolucent lines were noted around 1 uncemented humeral stem in 2 zones. Subsidence was not seen in any of the uncemented humeral stems over time. Humeral radiolucent lines occurred around 6 cemented components (P=.05). Lines were present in 1 zone (4 shoulders), 2 zones (1 shoulder), and 4 zones (1 shoulder). Glenoid lucencies were not present around any glenoid baseplate or screws. Notching was present in 4 uncemented stems and 5 cemented stems (P=.8). Notching was assessed as grade 1 (4 uncemented stems and 1 cemented stem) and grade 2 (0 uncemented stems and 4 cemented stems).
When comparing the uncemented Comprehensive stems with the uncemented Delta Xtend stems, there was no significant difference in subjective satisfaction score (1.9 vs 1.4, P=.4). However, ASES scores were significantly better with uncemented Delta Xtend components (37 vs 71, P=.01). Radiographically, the authors observed progressive attenuation of the bone at the junction between the greater tuberosity and the diaphysis with the uncemented Biomet stems, a phenomenon not observed with the uncemented DePuy stems (Figure 1).
Immediate postoperative (A) and 2-year follow-up (B) anteroposterior radiographs of a 71-year-old woman treated with an uncemented Comprehensive stem (Biomet, Warsaw, Indiana) with progressive attenuation of the lateral cortical bone (arrows) and ingrowth surface. Comparative immediate postoperative (C) and 2-year follow-up (D) anteroposterior radiographs of a 74-year-old woman treated with an uncemented Delta Xtend stem (DePuy, Warsaw, Indiana) with preserved lateral bony architecture.
Reverse shoulder arthroplasty has emerged as a commonly considered alternative for displaced PHF in the elderly population. Although cementless fixation of the humeral component has been reported to be successful when RSA is performed for other indications, cemented fixation continues to be recommended when RSA is performed in the setting of an acute PHF.
Cemented humeral component fixation presents several disadvantages, including longer operative time, the possibility of bone necrosis impeding healing, and complications with cement removal if revision surgery becomes necessary. In addition, use of an uncemented stem facilitates the ability to change the intraoperative alignment of the humeral prosthesis if this is initially placed in an undesired version. Recently, many surgeons have been compelled to explore the possibility of cementless fixation of the humeral component when RSA is performed for fractures, but little data exist in the literature on the relative success of cemented vs cementless fixation in the fracture setting.
Overall, there were no significant differences in pain scores and range of motion identified between the cemented and uncemented stem groups. However, patient-reported outcomes were worse with cementless stems compared with cemented stems, indicating that patients receiving a cementless RSA must be less able to perform several activities that require good use of the shoulder. Cementless RSA performed using a classic “Grammont-style” design led to a better outcome when compared with a lateralized humeral design. It is possible that lateralization of the humerus affects the rate of tuberosity healing and malunion, but this study was not powered to detect this difference. In the setting of hemiarthroplasty, tuberosity union has been shown to improve functional outcomes.17 Although tuberosity union is not as critical following RSA, Gallinet et al showed that tuberosity malunion, nonunion, and resorption lead to lower external rotation.7,18,19 It is possible that the higher rates of malunion and resorption seen within the uncemented group led to poorer patient-reported outcomes. Larger studies with a higher power are needed to further assess this theory.
Reports of PHFs treated with cementless RSA remain limited, with only 1 study in the English peer-reviewed literature reporting on these separate from cemented fixation.10 Of note, the humeral component design used in the study by Youn et al10 provided diaphyseal fixation, whereas most of the uncemented stems included in the current study were designed for metaphyseal fixation only. The mean ASES score reported in the study by Youn et al was similar to that of the current cemented group (75.9 vs 76.3), with both being higher than the score reported in the current uncemented group. Regarding mean elevation, Youn et al reported 114° and the current series reported 132°. No external rotation or internal rotation was reported. The study by Youn et al and the current study did not show any humeral stem subsidence or cementless stem failure, suggesting that cementless fixation of RSA for PHFs can achieve reliable humeral fixation.
With the numbers available in the current study, the authors were unable to correlate the differences in patient-reported outcome for the cemented and uncemented groups based on pain scores, range of motion, humeral component subsidence, or tuberosity healing. This may be partly because of lack of statistical power and because of difficulties when judging tuberosity position and healing on radiographs. Stress shielding of uncemented stems is well known in hip and shoulder arthroplasty. Several studies have reported on stress shielding with uncemented humeral components in shoulder arthroplasty without an impact on pain and motion.20,21 The resorption typically occurs in the posterolateral humeral region (Figure 2). Melis et al22 reported that, compared with cemented humeral stems, uncemented stems in RSA have a higher rate of tuberosity resorption and stress shielding without a significant difference in pain and motion. When used for acute PHF, cementless RSA may create an area of stress shielding that interferes with the ability of the tuberosity to heal reliably.
Immediate postoperative anteroposterior radiograph of an uncemented humeral stem with non-fractured lateral metaphysis intact above the distal aspect of the metaphyseal ingrowth surface (A). At 2 years postoperatively, the anteroposterior radiograph shows stress shielding with some resorption of the distal greater tuberosity (B).
The current study had several strengths. It represents the first comparative study between cemented and cementless fixation when RSA is performed in the treatment of acute PHF. Patients were evaluated using validated outcome scores, and the authors had a high rate of clinical and radiographic follow-up in a population commonly affected by dementia and with a high mortality rate.23 This study also had several limitations. Data were collected retrospectively. Stem fixation was based purely on surgeon preferences, with the 38 surgeries using 4 different implant systems and slight variation in tuberosity repair techniques. Radiographic outcomes were also limited because 7 patients elected to forgo radiographs at the 2-year visit, which could have led to undetected radiographic failures. Additionally, patients with preexisting dementia were eliminated, leading to selection bias in this cohort.
Reverse shoulder arthroplasty for acute PHFs has been shown to provide reliable pain relief and improved function in patients who are not candidates for internal fixation. Although cemented and cementless RSAs provided similar pain relief and range of motion, patients receiving a cemented stem had significantly better ASES scores and subjective satisfaction scores in this series.
- Aaron D, Shatsky J, Paredes JC, Jiang C, Parsons BO, Flatow EL. Proximal humeral fractures: internal fixation. Instr Course Lect. 2013;62:143–154.
- Südkamp N, Bayer J, Hepp P, et al. Open reduction and internal fixation of proximal humeral fractures with use of the locking proximal humerus plate: results of a prospective, multicenter, observational study. J Bone Joint Surg Am. 2009;91(6):1320–1328. doi:10.2106/JBJS.H.00006 [CrossRef]
- Boileau P, Krishnan SG, Tinsi L, Walch G, Coste JS, Molé D. Tuberosity malposition and migration: reasons for poor outcomes after hemiarthroplasty for displaced fractures of the proximal humerus. J Shoulder Elbow Surg. 2002;11(5):401–412. doi:10.1067/mse.2002.124527 [CrossRef]
- Khatib O, Onyekwelu I, Yu S, Zuckerman JD. Shoulder arthroplasty in New York State, 1991 to 2010: changing patterns of utilization. J Shoulder Elbow Surg. 2015;24(10):E286–E291. doi:10.1016/j.jse.2015.05.038 [CrossRef]
- Han RJ, Sing DC, Feeley BT, Ma CB, Zhang AL. Proximal humerus fragility fractures: recent trends in nonoperative and operative treatment in the Medicare population. J Shoulder Elbow Surg. 2016;25(2):256–261. doi:10.1016/j.jse.2015.07.015 [CrossRef]
- Schairer WW, Nwachukwu BU, Lyman S, Craig EV, Gulotta LV. National utilization of reverse total shoulder arthroplasty in the United States. J Shoulder Elbow Surg. 2015;24(1):91–97. doi:10.1016/j.jse.2014.08.026 [CrossRef]
- Bufquin T, Hersan A, Hubert L, Massin P. Reverse shoulder arthroplasty for the treatment of three- and four-part fractures of the proximal humerus in the elderly: a prospective review of 43 cases with a short-term follow-up. J Bone Joint Surg Br. 2007;89(4):516–520. doi:10.1302/0301-620X.89B4.18435 [CrossRef]
- Phadnis J, Huang T, Watts A, Krishnan J, Bain GI. Cemented or cementless humeral fixation in reverse total shoulder arthroplasty? A systematic review. Bone Joint J. 2016;98-B(1):65–74. doi:10.1302/0301-620X.98B1.36336 [CrossRef]
- Ross M, Hope B, Stokes A, Peters SE, McLeod I, Duke PF. Reverse shoulder arthroplasty for the treatment of three-part and four-part proximal humeral fractures in the elderly. J Shoulder Elbow Surg. 2015;24(2):215–222. doi:10.1016/j.jse.2014.05.022 [CrossRef]
- Youn SM, Deo S, Poon PC. Functional and radiologic outcomes of uncemented reverse shoulder arthroplasty in proximal humeral fractures: cementing the humeral component is not necessary. J Shoulder Elbow Surg. 2016;25(4):E83–E89. doi:10.1016/j.jse.2015.09.007 [CrossRef]
- Smith AM, Barnes SA, Sperling JW, Farrell CM, Cummings JD, Cofield RH. Patient and physician-assessed shoulder function after arthroplasty. J Bone Joint Surg Am. 2006;88(3):508–513.
- Neer CS II, . Displaced proximal humeral fractures: Part I. Classification and evaluation. Clin Orthop Relat Res. 2006;442:77–82. doi:10.1097/01.blo.0000198718.91223.ca [CrossRef]
- Sanchez-Sotelo J, Wright TW, O'Driscoll SW, Cofield RH, Rowland CM. Radiographic assessment of uncemented humeral components in total shoulder arthroplasty. J Arthroplasty. 2001;16(2):180–187. doi:10.1054/arth.2001.20905 [CrossRef]
- Sanchez-Sotelo J, O'Driscoll SW, Torchia ME, Cofield RH, Rowland CM. Radiographic assessment of cemented humeral components in total shoulder arthroplasty. J Shoulder Elbow Surg. 2001;10(6):526–531. doi:10.1067/mse.2001.118482 [CrossRef]
- Lazarus MD, Jensen KL, Southworth C, Matsen FA III, . The radiographic evaluation of keeled and pegged glenoid component insertion. J Bone Joint Surg Am. 2002;84(7):1174–1182. doi:10.2106/00004623-200207000-00013 [CrossRef]
- Sirveaux F, Favard L, Oudet D, Huquet D, Walch G, Molé D. Grammont inverted total shoulder arthroplasty in the treatment of glenohumeral osteoarthritis with massive rupture of the cuff: results of a multicentre study of 80 shoulders. J Bone Joint Surg Br. 2004;86(3):388–395. doi:10.1302/0301-620X.86B3.14024 [CrossRef]
- Gallinet D, Clappaz P, Garbuio P, Tropet Y, Obert L. Three or four parts complex proximal humerus fractures: hemiarthroplasty versus reverse prosthesis. A comparative study of 40 cases. Orthop Traumatol Surg Res. 2009;95(1):48–55. doi:10.1016/j.otsr.2008.09.002 [CrossRef]
- Klein M, Juschka M, Hinkenjann B, Scherger B, Ostermann PAW. Treatment of comminuted fractures of the proximal humerus in elderly patients with the Delta III reverse shoulder prosthesis. J Orthop Trauma. 2008;22(10):698–704. doi:10.1097/BOT.0b013e31818afe40 [CrossRef]
- Gallinet D, Adam A, Gasse N, Rochet S, Obert L. Improvement in shoulder rotation in complex shoulder fractures treated by reverse shoulder arthroplasty. J Shoulder Elbow Surg. 2013;22(1):38–44. doi:10.1016/j.jse.2012.03.011 [CrossRef]
- Spormann C, Durchholz H, Audigé L, et al. Patterns of proximal humeral bone resorption after total shoulder arthroplasty with an uncemented rectangular stem. J Shoulder Elbow Surg. 2014;23(7):1028–1035. doi:10.1016/j.jse.2014.02.024 [CrossRef]
- Raiss P, Edwards TB, Deutsch A, et al. Radiographic changes around humeral components in shoulder arthroplasty. J Bone Joint Surg Am. 2014;96(7):E54. doi:10.2106/JBJS.M.00378 [CrossRef]
- Melis B, DeFranco M, Lädermann A, et al. An evaluation of the radiological changes around the Grammont reverse geometry shoulder arthroplasty after eight to 12 years. J Bone Joint Surg Br. 2011;93(9):1240–1246. doi:10.1302/0301-620X.93B9.25926 [CrossRef]
- Somersalo A, Paloneva J, Kautiainen H, Lönnroos E, Heinänen M, Kiviranta I. Increased mortality after upper extremity fracture requiring inpatient care. Acta Orthop. 2015;86(5):533–557. doi:10.3109/17453674.2015.1043833 [CrossRef]
Demographics and Outcomes of Cemented Versus Uncemented Humeral Stems
|Characteristic||Cemented Group (n=19)||Uncemented Group (n=19)||P|
|Age, mean, y||74.1||74.8||.69|
|Sex, male/female, No.||5/14||3/16||.69|
|Time from injury to reverse shoulder arthroplasty, mean, d||6.4||8.1||.27|
|Visual analog scale score, preoperative pain, mean||9.3||9.6||.11|
|Follow-up, mean, mo||43.9||27.9||.03|
|Visual analog scale score, postoperative pain, mean||1.4||2.6||.11|
|External rotation, mean||40°||38°||.69|
|Internal rotation, mean||L3||L2||.16|
|Quick Disabilities of the Arm, Shoulder and Hand score, mean||20.4||34.5||.43|
|American Shoulder and Elbow Surgeons score, mean||76.3||48.0||.005|
|Satisfaction score, mean||1.2||1.8||.04|