- Email: [email protected]
Locking plate versus nonsurgical treatment for proximal humeral fractures: better midterm outcome with nonsurgical treatment
J Shoulder Elbow Surg (2011) 20, 1118-1124
www.elsevier.com/locate/ymse
Locking plate versus nonsurgical treatment for proximal humeral fractures: better midterm outcome with nonsurgical treatment Rick J. Sanders, BSa,b, Luc G. Thissen, BSa,b, Jop C. Teepen, BSa, Albert van Kampen, MD, PhDb, Ruurd L. Jaarsma, MD, PhD, FRACSa,* a b
Department of Orthopaedics, Flinders Medical Centre, Adelaide, Australia Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands Background: Since its introduction, there has been controversy about the use of locking plates in the treatment of proximal humeral fractures. Have they really improved the functional outcome after a proximal humeral fracture or should nonsurgical treatment have a more prominent role? In order to evaluate our hypothesis that nonsurgical treatment for proximal humerus fractures should be the first choice of treatment, a matched controlled cohort study was conducted to compare the midterm (>1 year) functional and radiologic outcome of a group of patients treated with a locking plate and a matched group of patients treated nonsurgically. Complications in each group of patients were evaluated. Materials and methods: Through direct matching, 17 patients (1 bilateral fracture) treated with a locking plate were matched to 18 patients treated nonsurgically. Medical records and radiographs were reviewed retrospectively to obtain relevant patient related data and fracture type according to Neer classification (i.e. 2-, 3- and 4-part fractures). At the time of clinical follow-up, EQ-5D, American Shoulder and Elbow Surgeons (ASES) score, visual analog pain (VAS) pain and VAS satisfaction scores were completed. Active range of motion was tested. New radiographs were made to evaluate fracture healing, complications and, in the locking plate group, the position of the plate and screws. Results: No significant differences were found in the characteristics of the patient groups. A significant difference in range of motion was found in favor of the nonsurgically treated patients. Results of ASES and patient satisfaction scores were also tending toward nonsurgical treatment. Furthermore, the complication rate was higher with locking plate treatment. Patients treated with a locking plate needed significantly more additional treatment on their injured shoulder (P ¼ 0.005). Discussion: This study’s main limitation was the fact that the choice of initial fracture management was based on clinical judgement, as well as patient’s fitness for surgery and therefore not randomized. By matching for fracture type this bias was largely overcome. Surgical treatment had a higher complication rate, requiring more additional treatment, which was often related to the initial surgery. Improving surgical technique could possibly lead to better outcomes for the surgically treated patients. In addition to the more favorable outcomes, nonsurgical treatment is also a more cost effective treatment.
This study received approval from the Southern Adelaide Flinders Clinical Human Research Ethics Committee (SAFC HREC).
*Reprint requests: Ruurd L. Jaarsma, MD, PhD, FRACS, Flinders Medical Centre, Department of Orthopaedics and Trauma, Flinders Dr, Bedford Park, SA 5042, Australia. E-mail address: [email protected] (R.L. Jaarsma).
1058-2746/$ - see front matter Ó 2011 Journal of Shoulder and Elbow Surgery Board of Trustees. doi:10.1016/j.jse.2011.01.025
Locking plate vs nonsurgical treatment
1119
Conclusion: Nonsurgical treatment should have a more prominent role in the treatment of proximal humeral fractures. Level of evidence: Level III, Retrospective Case Control Study, Treatment Study. Ó 2011 Journal of Shoulder and Elbow Surgery Board of Trustees. Keywords: Proximal humeral fracture; locking-plate; nonsurgical treatment; active range of motion; EQ-5D score; ASES score; complication rate
Approximately 4% to 5% of all fractures involve the proximal humerus,10 making it a common fracture in the general population. The male/female gender distribution is 3:7, and more than 70% of patients are aged older than 60 years.27,28 Other factors related to proximal humeral fractures are a high risk of falling and a low bone density.19,25 Proximal humeral fractures can be treated nonsurgically in 80% to 90% of patients.15 Approximately 10% to 20% require surgery5 by open reduction and internal fixation (ORIF), according to the indications for operative treatment as outlined by Neer,22 or by hemiarthroplasty. Since the introduction of the locking plate, it has gradually become the most commonly used method for a surgical fixation of a proximal humeral fracture. Several studies have reported favorable short-term clinical results of locked plating for proximal humeral fractures, but these studies also highlight complications related to this technique.1,2,4,5,11,12,16,24,26-28 However, good results for nonsurgical treatment of proximal humeral fractures, including 3- and 4-part fractures, have also been described.7-9,17,29 Despite these results, there still is no consensus on the first choice of treatment for proximal humeral fractures. Studies comparing longer-term results after treatment of patients with a locking plate vs a control group are scarce. The main purpose of this cohort study was to compare midterm outcome (>1 year) of patients with proximal humeral fractures treated nonsurgically and patients treated with a locking plate.
Materials and methods Patients A cohort study was done of patients who were treated nonsurgically or with a locking plate for a proximal humeral fracture. This included a retrospective review and observational follow-up. All patients presented between January 1, 2002, and December 31, 2008, at a level 1 trauma center in Australia. They were identified using the hospital databases. The study excluded patients aged younger than 18. This resulted in a list of 234 patients treated for a proximal humeral fracture, of which 178 patients received nonsurgical treatment, 41 received a locking plate, and 15 received another surgical treatment. Of this group, 18 patients with a locking plate were able to participate in the study.
To get 2 comparable patient groups, patients treated with a locking plate were directly matched with patients from the nonsurgical group. The matching was done primarily based on the fracture type according to Neer classification,22 age, and gender, without having any knowledge about other demographic and clinical data. Fracture type was assessed using the first radiographs made after the incident (Fig. 1). Every patient was evaluated according to the initial treatment received (intention to treat). In all of the patients treated with a locking plate, a deltopectoral approach was used to insert the plate, which were either Synthes (West Chester, PA, USA) or Smith & Nephew (Andover, MA, USA). Postoperative management consisted of treatment with a collar and cuff, early pendular exercises, and physiotherapy. After 2, 6, and 12 weeks, patients were reviewed in the outpatient department and x-rays images were made at each of these occasions. The standard nonsurgical management in this hospital consisted of treatment with a collar and cuff, pendular exercises after 1 week, and referral to physiotherapy after 2 weeks. After 1, 6, and 12 weeks, patients were reviewed in the outpatient department. X-ray images were made at each of these visits.
Data collection Medical records were checked for demographic and clinical information (eg, comorbidity such as osteoporosis). Radiographs were evaluated by the first 2 authors (R.S., L.T.) for fracture type according to the Neer classification. If there was disagreement about the classification, the x-ray images were reevaluated with help of a consultant (R.J.), thereby reducing interobserver variation. Fractures were divided into 2-, 3-, and 4-part fractures.
Follow-up All patients who returned to the hospital for midterm clinical follow-up were evaluated by the first 2 authors. A consent form was signed by each patient. Two standardized questionnaires were completed to review the patients’ current shoulder function and general health state. The current shoulder function was evaluated using the American Shoulder and Elbow Surgeons (ASES) self-assessment score.21 For the general health state of the patient, the EQ-5D score was used.6,23 A visual analog scale (VAS) was used to grade patient satisfaction with the current shoulder function. To measure flexion, extension, abduction, external rotation, and internal rotation, an active range of motion (ROM) test was performed. The ROM of the affected side was compared with the unaffected side. Internal rotation was determined by the highest vertebra level that could be reached by putting the hand behind the
1120
R.J. Sanders et al.
Figure 1 Radiographs show the humerus before treatment. The left fracture was treated with a locking plate, and the right fracture was treated nonsurgically. back. Because patients were not able perform internal rotation in 90 abduction, internal rotation in this position was not evaluated. Finally, new radiographs of the proximal humerus in anteroposterior (AP), lateral, and axial views were obtained. These were evaluated for fracture healing, position of the humeral head, and signs of osteoarthritis (OA). Malunion of the humeral head was defined as an angulation of the fracture fragments of 30 or more, or displacement of 5 mm or more, or both. The degree of OA was graded using the Kellgren-Lawrence score.18 When a locking plate was still in situ, the position of the plate and screws was evaluated. A good position of the plate was assumed when the plate was placed lateral on the proximal humerus and at least 1 cm inferior to the proximal tip of the greater tuberosity. Position of screws was described as normal (located in humeral head), subchondral, or protruding into the joint space. When the locking plate had been removed, the last radiograph in which the plate was still in situ was included for evaluation, allowing observation of possible malposition of the plate or screws, or both, before removal of the metalwork.
Data analysis Statistical analysis of obtained data was performed using IBM PSAW 18 software (SPSS Inc, Chicago, IL, USA). To compare continuous outcomes variables between the locking plate and nonsurgical treatment groups, independent sample t tests were used when data was normally distributed and the Mann-Whitney U test when skewed. Categoric data were compared with c2 tests. A value of P .05 was considered significant. Pearson correlation coefficients were calculated to measure correlations.
Results The study evaluated clinical and radiographic follow-up in 35 patients with 36 proximal humeral fractures (1 patient in the locking plate group had bilateral fractures); of which 18 fractures (50%) had been treated with a proximal humeral locking plate and the other 18 had been treated nonsurgically. No closed reductions were performed in the latter group.
The 2 groups were matched according to fracture type, length of follow-up, gender, and age. No significant differences in demographic and clinical characteristics between the groups were found (Table I).
Clinical follow-up All patients completed the EQ-5D and ASES questionnaires, scored their pain and satisfaction on a VAS scale, and were tested for the ROM in their shoulders. For the EQ-5D questionnaire, no significant differences were found in any of the 5 domains and the EQ5D-VAS score (Table II), indicating that the general health state between the 2 groups was similar. The mean ASES score was 71.6 in the locking plate group and 82.5 in the nonsurgical group, which resulted in a nonsignificant difference (P ¼ .130). Other studies of locking plates have reported a mean ASES score of approximately 70.14,24 Using this ASES score of 70 as a cutoff, scores were higher in 10 patients (56%) of the locking plate group and in 14 patients (78%) of the nonsurgical group. Furthermore, the VAS pain score was 1.5 for locking plate patients vs 1.6 for nonsurgical patients, and the mean patient satisfaction was 5.6 vs 7.2, respectively. Neither result was significantly different (P ¼ .896 and P ¼ .128, respectively). Table III reports the mean degrees of ROM of the injured shoulder for both groups. It was not possible to determine the difference in ROM between the injured and the noninjured shoulder in 1 patient in the locking plate group because the patient had a bilateral fracture. In all other patients, ROM of the noninjured shoulder was not impaired by any other condition. The resulting loss of motion is reported in Table III. These results show a significant difference between the locking plate and nonsurgical group in flexion (P ¼ .003), abduction (P ¼ .000), and external rotation (P ¼ .001). Only 50% of all patients in the locking plate group
Locking plate vs nonsurgical treatment Patient characteristics Characteristic)
1121
Table I
Gender Male Female Age, years Follow-up, mon Neer classification 2 part 3 part 4 part Osteoporosis Arthritis Diabetes mellitus Dominant arm affected ) y
P
Treatment Locking platey
Nonsurgical
(n ¼ 18)
(n ¼ 18)
8 9 58 37
(47.1) (52.9) 14 12
6 12 64 42
(33.3) (66.7) 15 26
3 13 2 8 3 1 8
(16.7) (72.2) (11.1) (44.4) (16.7) (5.6) (44.4)
3 14 1 8 6 4 13
(16.7) (77.8) (5.6) (44.4) (33.3) (22.2) (72.2)
.407
.225 .464 .831
1.000 0.443 0.338 0.091
Data are presented as mean standard deviation, or number (%). One patient in the locking plate group had a bilateral fracture.
Table II
EQ-5D score
Domain and problem Mobility) No problems Some problems Confined to bed Personal care No problems Some problems Unable to do Usual activities No problems Some problems Unable to do Pain or discomfort None Moderate Extreme Anxiety or depression None Moderate Extreme Health VAS scorey
P
Treatment Locking plate
Nonsurgical
13 (72.2) 5 (27.8) 0 (0.0)
14 (77.8) 4 (22.2) 0 (0.0)
14 (77.8) 4 (22.2) 0 (0.0)
17 (94.4) 0 (0.0) 1 (5.6)
9 (50.0) 8 (44.4) 1 (5.60)
11 (61.1) 6 (33.3) 1 (5.6)
9 (50.0) 7 (38.9) 2 (11.1)
6 (33.3) 10 (55.6) 2 (11.1)
.700
.071
.784
.569
.165 10 7 1 78
(55.6) (38.9) (5.6) 13
15 3 0 76
(83.3) (16.7) (0.0) 16
.706
VAS, Visual analog scale. ) Data are number (%) or mean standard deviation. y Scores range from 0-100.
were able to reach above vertebrae L2 compared with 72% in the nonsurgical group. A strong correlation was found between the ASES score and patient satisfaction (R ¼ 0.83), but the correlation between patient satisfaction and the VAS pain score was less strong (R ¼ e0.59).
Radiographic follow-up The position of the humeral head was evaluated for every radiograph. Malunion of the humeral head was present in 8 patients (44%) of the locking plate group. In 5 of these
1122 Table III
R.J. Sanders et al. Range of motion
Movement
Flexion Extension Abduction External rotation Loss of) Flexion Extension Abduction External rotation
P
Treatment Locking plate
Nonsurgical
Mean SD, deg
Mean SD, deg
131 43 106 26
37 11 45 18
157 40 156 36
28 16 30 28
. . . .
48 8 71 27
37 10 43 14
13 4 17 9
26 7 25 13
.003 .141 .000 .001
SD, standard deviation. ) Loss is calculated as noninjured shoulder function minus fractured shoulder function.
Figure 2 Radiographs taken at follow-up in this study show malunion (Left) in a patient treated with a locking plate and (Right) in a patient treated nonsurgically.
patients (63%), displacement of the bony parts was already present on radiographs directly after surgery, indicating nonoptimal reduction of the fracture. In the other 3 patients (37%), displacements occurred during the healing process. In the nonsurgically treated group, 5 patients (28%) showed a malunited humeral head. No significant difference in the incidence of humeral head displacement was found between the locking-plate and nonsurgical group (P ¼ .298; Fig. 2). Results of OA gradation are reported in Table IV. Presence of OA is obvious when the Kellgren-Lawrence score is grade 2 or above. Looking at grade 2 or above, 11 patients (65%) in the locking plate group showed clear signs of OA compared with 6 patients (33%) in the nonsurgical group (P ¼ .06).
Complications In total, 10 patients in the locking plate group (56%) and 2 patients in the non-surgical group (11%) needed
additional treatment on their shoulder, which was a significant difference (P ¼ .005). The metalwork was completely removed in 7 patients (39%) in the locking plate group. The reason for removal was impingement or restricted ROM, or both, in 6 cases, and infection resulting in severe osteonecrosis in 1 patient. An arthrodesis was performed on this last patient. The plates were removed an average of 9 months after placement (range, 4-15 months). In 2 shoulders (11%), screws were replaced because of secondary screw perforation, and in 1 shoulder, the plate was substituted by a new plate due to loss of fixation. In the nonsurgical group, nonunion occurred in 1 patient. Chronic bursitis developed in another patient, which was treated with cortisone injections. Finally, position of the plate and screws was reviewed: 11plates (61%) were still in situ, 4 showed good position (>1 cm under the greater tuberosity and placed lateral to biceps tendon), and 5 plates were placed less than 1 cm under the greater tuberosity, 1 plate was not laterally
Locking plate vs nonsurgical treatment Table IV
1123
Kellgren-Lawrence score for osteoarthritis
Kellgren-Lawrence score None, grade 0 Doubtful, grade 1 Minimal, grade 2 Moderate, grade 3 Severe, grade 4
Treatment, No. (%) Locking plate
Nonsurgical
2 4 7 3 1
5 7 4 2 0
(11.8) (23.5) (41.2) (17.6) (5.9)
(27.8) (38.9) (22.2) (11.1) (0.0)
placed, and 1 plate was placed both less than 1cm under greater tuberosity and not lateral. Positioning of the screws in patients in which the plate was still in situ was normal in 3 cases, subchondral in 7, and protruding into the joint space in 1. In the 7 shoulders where the plate was removed, the position of the plate was suboptimal in all cases at the time of removal. Position of screws was normal in 6 patients and subchondral in 1.
Discussion It is difficult to compare the functional outcome and complication rate between surgical and nonsurgical treatments of proximal humeral fractures. To our knowledge, randomized controlled trials comparing functional outcomes of locking plates with a nonsurgical treatment on a midterm to long-term outcome have never been performed and are difficult to realize. This study tried to overcome this problem by directly matching 2 cohorts of patients, primarily based on fracture type. A significantly better ROM in patients treated nonsurgically was found in the current study. Difference in internal rotation was nonsignificant, but results were tending to a more favorable outcome in nonsurgically treated patients. Results in ASES and patient satisfaction scores were also tending toward nonsurgical treatment. An explanation for the more unfavorable midterm outcomes for patients treated with a locking plate could be due to the high complication rate in these patients. Many of these complications are related to the initial surgery and might therefore be preventable.1,4,5,27 Improving surgical technique could possibly lead to better midterm outcomes in patients with locking plates. Two factors might explain the presence of more profound OA in the locking plate group in this study. First, there is the direct influence of the plate and screws on the bony surface of the proximal humerus. Second, the higher rate of repeat operation in the locking plate group could have caused additional damage to the glenohumeral joint. Although the dominant arm was less often affected in the locking plate group, this difference was not significant. Therefore, it is unlikely that this factor substantially
affected the differences in midterm outcomes between the 2 treatment groups. Furthermore, it would probably underestimate the more favorable results for the nonsurgical group, because a fracture in the dominant arm would give more restrictions in daily functioning. Characteristics and results of the locking plate group in this study were similar to those in recent prospective multicenter studies, indicating that our sample of locking plate patients was representative.4,27 Furthermore, several studies comparing surgical treatment with nonsurgical treatment groups showed similar results as found in this study, concluding that nonsurgical treatment should be considered as a treatment option for proximal humeral fractures.13,17,20,29 This study has some limitations. Choice of fracture treatment was based on clinical judgement, as well as the patient’s fitness at time of treatment decision. Clinical judgement could influence prognosis of midterm outcome, leading to confounding by indication. However, by matching on fracture type, correction for this major clinical prognostic factor was achieved. Fitness at time of treatment could also lead to confounding by indication. Some patients who clinically warranted an operation were not fit for surgery at the time of injury and therefore received nonsurgical treatment. Nevertheless, this would underestimate the functional outcomes of nonsurgical treatment found in this study, because patients with a lower fitness are expected to have worse midterm functional outcomes. This study used the Kellgren-Lawrence score for classification of OA. Although this score has only been validated for use in the knee and hip,3 it was the best available instrument to score OA in the shoulder. Moreover, the number of patients in this study was quite small. Although it was large enough to detect a significant difference in ROM between the patient groups, results in ASES and patient satisfaction scores were tending toward nonsurgical treatment. A study that includes more patients would possibly show significant differences for these outcome measurements also. Finally, when considering the optimal treatment strategy for a proximal humeral fracture, the cost of treatment should always be taken into account. Surgical treatment with a locking plate is far more expensive than nonsurgical treatment. Research on the cost-effectiveness of both treatment strategies is needed to make a deliberate decision.
Conclusion Despite good results for locked plating on short-term follow-up, this study shows that patients treated nonsurgically for a proximal humeral fracture achieve better ROM on midterm follow-up (>1 year) compared with patients treated with a locking plate. Furthermore, OA seems to be more profound when a locking plate is used.
1124 At least equal results were found for other outcome measurements, such as ASES, patient satisfaction, and pain scores. The results of this study indicate that nonsurgical treatment should have a more prominent role in the treatment of proximal humeral fractures, even in cases that present clear indications for locking plate treatment.
Disclaimer This study was fully funded by the hospital budget provided by Government of South Australia to Flinders University Medical Centre. No grants were received and no other source providing funding or equipment were used. The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
References 1. Agudelo J, Sch€ urmann M, Stahel P, Helwig P, Morgan SJ, Zechel W, et al. Analysis of efficacy and failure in proximal humerus fractures treated with locking plates. J Orthop Trauma 2007;21:676-81. doi:10. 1097/BOT.0b013e31815bb09d 2. Bj€ orkenheim JM, Pajarinen J, Savolainen V. Internal fixation of proximal humeral fractures with a locking compression plate: a retrospective evaluation of 72 patients followed for a minimum of 1 year. Acta Orthop Scand 2004;75:741-5. doi:10.1080/00016470410004120 3. Brox JI, Lereim P, Merckoll E, Finnanger AM. Radiographic classification of glenohumeral arthrosis. Acta Orthop Scand 2003;74:186-9. doi:10.1080/00016470310013932 4. Brunner F, Sommer C, Bahrs C, Heuwinkel R, Hafner C, Rillmann P, et al. Open reduction and internal fixation of proximal humerus fractures using a proximal humeral locked plate: a prospective multicenter analysis. J Orthop Trauma 2009;23:163-72. doi:10.1097/ BOT.0b013e3181920e5b 5. Clavert P, Adam P, Bevort A, Bonnomet F, Kempf JF. Pitfalls and complications with locking plate for proximal humerus fracture. J Shoulder Elbow Surg 2010;19:489-94. doi:10.1016/j.jse.2009.09.005 6. Coons SJ, Rao S, Keininger DL, Hays RD. A comparative review of generic quality-of-life instruments. Pharmacoeconomics 2000;17: 13-35. 7. Court-Brown CM, McQueen MM. The impacted varus (A2.2) proximal humeral fracture: prediction of outcome and results of nonoperative treatment in 99 patients. Acta Orthop Scand 2004;75:736-40. doi:10.1080/759369235 8. Court-Brown CM, Cattermole H, McQueen MM. Impacted valgus fractures (B1.1) of the proximal humerus: the results of non-operative treatment. J Bone Joint Surg Br 2002;84:504-8. doi:10.1302/0301620X.84B4.12488 9. Court-Brown CM, Garg A, McQueen MM. The translated two-part fracture of the proximal humerus: epidemiology and outcome in the older patient. J Bone Joint Surg Br 2001;83:799-804. doi:10.1302/ 0301-620X.83B6.11401
R.J. Sanders et al. 10. Court-Brown CM, Garg A, McQueen MM. The epidemiology of proximal humeral fractures. Acta Orthop Scand 2001;72:365-71. 11. Duralde XA, Leddy LR. The results of ORIF of displaced unstable proximal humeral fractures using a locking plate. J Shoulder Elbow Surg 2010;19:480-8. doi:10.1016/j.jse.2009.08.008 12. Egol KA, Ong CC, Walsh M, Jazrawi LM, Tejwani NC, Zuckerman JD. Early complications in proximal humerus fractures (OTA Types 11) treated with locked plates. J Orthop Trauma 2008;22: 159-64. doi:10.1097/BOT.0b013e318169ef2a 13. Fjalestad T, Strømsøe K, Bl€ucher J, Tennøe B. Fractures in the proximal humerus: functional outcome and evaluation of 70 patients treated in hospital. Arch Orthop Trauma Surg 2005;125:310-6. doi:10. 1007/s00402-005-0803-9 14. Friess DM, Attia A, Vallier HA. Locking plate fixation for proximal humerus fractures: a comparison with other fixation techniques. Orthopedics 2008;31:1183. doi:10.3928/01477447-20081201-07 15. Gaebler C, McQueen MM, Court-Brown CM. Minimally displaced proximal humeral fractures: epidemiology and outcome in 507 cases. Acta Orthop Scand 2003;74:580-5. doi:10.1080/000164703100 17992 16. Handschin AE, Cardell M, Contaldo C, Trentz O, Wanner GA. Functional results of angular-stable plate fixation in displaced proximal humeral fractures. Injury 2008;39:306-13. doi:10.1016/j.injury. 2007.10.011 17. Hanson B, Neidenbach P, de Boer P, Stengel D. Functional outcomes after nonoperative management of fractures of the proximal humerus. J Shoulder Elbow Surg 2009;18:612-21. doi:10.1016/ j.jse.2009.03.024 18. Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957;16:494-502. doi:10.1136/ard.16.4.494 19. Lee SH, Dargent-Molina P, Breart G. Risk factors for fractures of the proximal humerus: results from the EPIDOS prospective study. J Bone Miner Res 2002;17:817-25. doi:10.1359/jbmr.2002.17.5.817 20. McLaurin TM. Proximal humerus fractures in the elderly are we operating on too many? Bull Hosp Jt Dis 2004;62:24-32. 21. Michener LA, McClure PW, Sennett BJ. American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form, patient self-report section: reliability, validity, and responsiveness. J Shoulder Elbow Surg 2002;11:587-94. doi:10.1067/mse.2002.127096 22. Neer CS 2nd. Displaced proximal humeral fractures: Part I. Classification and evaluation. J Bone Joint Surg Am 1970;52:1077-89. doi:10. 1055/s-2007-980100 23. Rabin R, de Charro F. EQ-5D: A measure of health status from the EuroQol Group. Ann Med 2001;33:337-43. 24. Ricchetti ET, Warrender WJ, Abboud JA. Use of locking plates in the treatment of proximal humerus fractures. J Shoulder Elbow Surg 2010; 19:66-75. doi:10.1016/j.jse.2010.01.001 25. Ring D. Current concepts in plate and screw fixation of osteoporotic proximal humerus fractures. Injury 2007;38(suppl 3):S59-68. doi:10. 1016/j.injury.2007.08.013 26. Smith AM, Mardones RM, Sperling JW, Cofield RH. Early complications of operatively treated proximal humeral fractures. J Shoulder Elbow Surg 2007;16:14-24. doi:10.1016/j.jse.2007.02.109 27. S€udkamp N, Bayer J, Hepp P, Voigt C, Oestern H, K€a€ab M, 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;91:13208. doi:10.2106/JBJS.H.00006 28. Thanasas C, Kontakis G, Angoules A, Limb D, Giannoudis P. Treatment of proximal humerus fractures with locking plates: a systematic review. J Shoulder Elbow Surg 2009;18:837-44. doi:10.1016/j.jse. 2009.06.004 29. Zyto K. Non-operative treatment of comminuted fractures of the proximal humerus in elderly patients. Injury 1998;29(5):349-52. doi: 10.1016/S0020-1383(97)00211-8
www.elsevier.com/locate/ymse
Locking plate versus nonsurgical treatment for proximal humeral fractures: better midterm outcome with nonsurgical treatment Rick J. Sanders, BSa,b, Luc G. Thissen, BSa,b, Jop C. Teepen, BSa, Albert van Kampen, MD, PhDb, Ruurd L. Jaarsma, MD, PhD, FRACSa,* a b
Department of Orthopaedics, Flinders Medical Centre, Adelaide, Australia Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands Background: Since its introduction, there has been controversy about the use of locking plates in the treatment of proximal humeral fractures. Have they really improved the functional outcome after a proximal humeral fracture or should nonsurgical treatment have a more prominent role? In order to evaluate our hypothesis that nonsurgical treatment for proximal humerus fractures should be the first choice of treatment, a matched controlled cohort study was conducted to compare the midterm (>1 year) functional and radiologic outcome of a group of patients treated with a locking plate and a matched group of patients treated nonsurgically. Complications in each group of patients were evaluated. Materials and methods: Through direct matching, 17 patients (1 bilateral fracture) treated with a locking plate were matched to 18 patients treated nonsurgically. Medical records and radiographs were reviewed retrospectively to obtain relevant patient related data and fracture type according to Neer classification (i.e. 2-, 3- and 4-part fractures). At the time of clinical follow-up, EQ-5D, American Shoulder and Elbow Surgeons (ASES) score, visual analog pain (VAS) pain and VAS satisfaction scores were completed. Active range of motion was tested. New radiographs were made to evaluate fracture healing, complications and, in the locking plate group, the position of the plate and screws. Results: No significant differences were found in the characteristics of the patient groups. A significant difference in range of motion was found in favor of the nonsurgically treated patients. Results of ASES and patient satisfaction scores were also tending toward nonsurgical treatment. Furthermore, the complication rate was higher with locking plate treatment. Patients treated with a locking plate needed significantly more additional treatment on their injured shoulder (P ¼ 0.005). Discussion: This study’s main limitation was the fact that the choice of initial fracture management was based on clinical judgement, as well as patient’s fitness for surgery and therefore not randomized. By matching for fracture type this bias was largely overcome. Surgical treatment had a higher complication rate, requiring more additional treatment, which was often related to the initial surgery. Improving surgical technique could possibly lead to better outcomes for the surgically treated patients. In addition to the more favorable outcomes, nonsurgical treatment is also a more cost effective treatment.
This study received approval from the Southern Adelaide Flinders Clinical Human Research Ethics Committee (SAFC HREC).
*Reprint requests: Ruurd L. Jaarsma, MD, PhD, FRACS, Flinders Medical Centre, Department of Orthopaedics and Trauma, Flinders Dr, Bedford Park, SA 5042, Australia. E-mail address: [email protected] (R.L. Jaarsma).
1058-2746/$ - see front matter Ó 2011 Journal of Shoulder and Elbow Surgery Board of Trustees. doi:10.1016/j.jse.2011.01.025
Locking plate vs nonsurgical treatment
1119
Conclusion: Nonsurgical treatment should have a more prominent role in the treatment of proximal humeral fractures. Level of evidence: Level III, Retrospective Case Control Study, Treatment Study. Ó 2011 Journal of Shoulder and Elbow Surgery Board of Trustees. Keywords: Proximal humeral fracture; locking-plate; nonsurgical treatment; active range of motion; EQ-5D score; ASES score; complication rate
Approximately 4% to 5% of all fractures involve the proximal humerus,10 making it a common fracture in the general population. The male/female gender distribution is 3:7, and more than 70% of patients are aged older than 60 years.27,28 Other factors related to proximal humeral fractures are a high risk of falling and a low bone density.19,25 Proximal humeral fractures can be treated nonsurgically in 80% to 90% of patients.15 Approximately 10% to 20% require surgery5 by open reduction and internal fixation (ORIF), according to the indications for operative treatment as outlined by Neer,22 or by hemiarthroplasty. Since the introduction of the locking plate, it has gradually become the most commonly used method for a surgical fixation of a proximal humeral fracture. Several studies have reported favorable short-term clinical results of locked plating for proximal humeral fractures, but these studies also highlight complications related to this technique.1,2,4,5,11,12,16,24,26-28 However, good results for nonsurgical treatment of proximal humeral fractures, including 3- and 4-part fractures, have also been described.7-9,17,29 Despite these results, there still is no consensus on the first choice of treatment for proximal humeral fractures. Studies comparing longer-term results after treatment of patients with a locking plate vs a control group are scarce. The main purpose of this cohort study was to compare midterm outcome (>1 year) of patients with proximal humeral fractures treated nonsurgically and patients treated with a locking plate.
Materials and methods Patients A cohort study was done of patients who were treated nonsurgically or with a locking plate for a proximal humeral fracture. This included a retrospective review and observational follow-up. All patients presented between January 1, 2002, and December 31, 2008, at a level 1 trauma center in Australia. They were identified using the hospital databases. The study excluded patients aged younger than 18. This resulted in a list of 234 patients treated for a proximal humeral fracture, of which 178 patients received nonsurgical treatment, 41 received a locking plate, and 15 received another surgical treatment. Of this group, 18 patients with a locking plate were able to participate in the study.
To get 2 comparable patient groups, patients treated with a locking plate were directly matched with patients from the nonsurgical group. The matching was done primarily based on the fracture type according to Neer classification,22 age, and gender, without having any knowledge about other demographic and clinical data. Fracture type was assessed using the first radiographs made after the incident (Fig. 1). Every patient was evaluated according to the initial treatment received (intention to treat). In all of the patients treated with a locking plate, a deltopectoral approach was used to insert the plate, which were either Synthes (West Chester, PA, USA) or Smith & Nephew (Andover, MA, USA). Postoperative management consisted of treatment with a collar and cuff, early pendular exercises, and physiotherapy. After 2, 6, and 12 weeks, patients were reviewed in the outpatient department and x-rays images were made at each of these occasions. The standard nonsurgical management in this hospital consisted of treatment with a collar and cuff, pendular exercises after 1 week, and referral to physiotherapy after 2 weeks. After 1, 6, and 12 weeks, patients were reviewed in the outpatient department. X-ray images were made at each of these visits.
Data collection Medical records were checked for demographic and clinical information (eg, comorbidity such as osteoporosis). Radiographs were evaluated by the first 2 authors (R.S., L.T.) for fracture type according to the Neer classification. If there was disagreement about the classification, the x-ray images were reevaluated with help of a consultant (R.J.), thereby reducing interobserver variation. Fractures were divided into 2-, 3-, and 4-part fractures.
Follow-up All patients who returned to the hospital for midterm clinical follow-up were evaluated by the first 2 authors. A consent form was signed by each patient. Two standardized questionnaires were completed to review the patients’ current shoulder function and general health state. The current shoulder function was evaluated using the American Shoulder and Elbow Surgeons (ASES) self-assessment score.21 For the general health state of the patient, the EQ-5D score was used.6,23 A visual analog scale (VAS) was used to grade patient satisfaction with the current shoulder function. To measure flexion, extension, abduction, external rotation, and internal rotation, an active range of motion (ROM) test was performed. The ROM of the affected side was compared with the unaffected side. Internal rotation was determined by the highest vertebra level that could be reached by putting the hand behind the
1120
R.J. Sanders et al.
Figure 1 Radiographs show the humerus before treatment. The left fracture was treated with a locking plate, and the right fracture was treated nonsurgically. back. Because patients were not able perform internal rotation in 90 abduction, internal rotation in this position was not evaluated. Finally, new radiographs of the proximal humerus in anteroposterior (AP), lateral, and axial views were obtained. These were evaluated for fracture healing, position of the humeral head, and signs of osteoarthritis (OA). Malunion of the humeral head was defined as an angulation of the fracture fragments of 30 or more, or displacement of 5 mm or more, or both. The degree of OA was graded using the Kellgren-Lawrence score.18 When a locking plate was still in situ, the position of the plate and screws was evaluated. A good position of the plate was assumed when the plate was placed lateral on the proximal humerus and at least 1 cm inferior to the proximal tip of the greater tuberosity. Position of screws was described as normal (located in humeral head), subchondral, or protruding into the joint space. When the locking plate had been removed, the last radiograph in which the plate was still in situ was included for evaluation, allowing observation of possible malposition of the plate or screws, or both, before removal of the metalwork.
Data analysis Statistical analysis of obtained data was performed using IBM PSAW 18 software (SPSS Inc, Chicago, IL, USA). To compare continuous outcomes variables between the locking plate and nonsurgical treatment groups, independent sample t tests were used when data was normally distributed and the Mann-Whitney U test when skewed. Categoric data were compared with c2 tests. A value of P .05 was considered significant. Pearson correlation coefficients were calculated to measure correlations.
Results The study evaluated clinical and radiographic follow-up in 35 patients with 36 proximal humeral fractures (1 patient in the locking plate group had bilateral fractures); of which 18 fractures (50%) had been treated with a proximal humeral locking plate and the other 18 had been treated nonsurgically. No closed reductions were performed in the latter group.
The 2 groups were matched according to fracture type, length of follow-up, gender, and age. No significant differences in demographic and clinical characteristics between the groups were found (Table I).
Clinical follow-up All patients completed the EQ-5D and ASES questionnaires, scored their pain and satisfaction on a VAS scale, and were tested for the ROM in their shoulders. For the EQ-5D questionnaire, no significant differences were found in any of the 5 domains and the EQ5D-VAS score (Table II), indicating that the general health state between the 2 groups was similar. The mean ASES score was 71.6 in the locking plate group and 82.5 in the nonsurgical group, which resulted in a nonsignificant difference (P ¼ .130). Other studies of locking plates have reported a mean ASES score of approximately 70.14,24 Using this ASES score of 70 as a cutoff, scores were higher in 10 patients (56%) of the locking plate group and in 14 patients (78%) of the nonsurgical group. Furthermore, the VAS pain score was 1.5 for locking plate patients vs 1.6 for nonsurgical patients, and the mean patient satisfaction was 5.6 vs 7.2, respectively. Neither result was significantly different (P ¼ .896 and P ¼ .128, respectively). Table III reports the mean degrees of ROM of the injured shoulder for both groups. It was not possible to determine the difference in ROM between the injured and the noninjured shoulder in 1 patient in the locking plate group because the patient had a bilateral fracture. In all other patients, ROM of the noninjured shoulder was not impaired by any other condition. The resulting loss of motion is reported in Table III. These results show a significant difference between the locking plate and nonsurgical group in flexion (P ¼ .003), abduction (P ¼ .000), and external rotation (P ¼ .001). Only 50% of all patients in the locking plate group
Locking plate vs nonsurgical treatment Patient characteristics Characteristic)
1121
Table I
Gender Male Female Age, years Follow-up, mon Neer classification 2 part 3 part 4 part Osteoporosis Arthritis Diabetes mellitus Dominant arm affected ) y
P
Treatment Locking platey
Nonsurgical
(n ¼ 18)
(n ¼ 18)
8 9 58 37
(47.1) (52.9) 14 12
6 12 64 42
(33.3) (66.7) 15 26
3 13 2 8 3 1 8
(16.7) (72.2) (11.1) (44.4) (16.7) (5.6) (44.4)
3 14 1 8 6 4 13
(16.7) (77.8) (5.6) (44.4) (33.3) (22.2) (72.2)
.407
.225 .464 .831
1.000 0.443 0.338 0.091
Data are presented as mean standard deviation, or number (%). One patient in the locking plate group had a bilateral fracture.
Table II
EQ-5D score
Domain and problem Mobility) No problems Some problems Confined to bed Personal care No problems Some problems Unable to do Usual activities No problems Some problems Unable to do Pain or discomfort None Moderate Extreme Anxiety or depression None Moderate Extreme Health VAS scorey
P
Treatment Locking plate
Nonsurgical
13 (72.2) 5 (27.8) 0 (0.0)
14 (77.8) 4 (22.2) 0 (0.0)
14 (77.8) 4 (22.2) 0 (0.0)
17 (94.4) 0 (0.0) 1 (5.6)
9 (50.0) 8 (44.4) 1 (5.60)
11 (61.1) 6 (33.3) 1 (5.6)
9 (50.0) 7 (38.9) 2 (11.1)
6 (33.3) 10 (55.6) 2 (11.1)
.700
.071
.784
.569
.165 10 7 1 78
(55.6) (38.9) (5.6) 13
15 3 0 76
(83.3) (16.7) (0.0) 16
.706
VAS, Visual analog scale. ) Data are number (%) or mean standard deviation. y Scores range from 0-100.
were able to reach above vertebrae L2 compared with 72% in the nonsurgical group. A strong correlation was found between the ASES score and patient satisfaction (R ¼ 0.83), but the correlation between patient satisfaction and the VAS pain score was less strong (R ¼ e0.59).
Radiographic follow-up The position of the humeral head was evaluated for every radiograph. Malunion of the humeral head was present in 8 patients (44%) of the locking plate group. In 5 of these
1122 Table III
R.J. Sanders et al. Range of motion
Movement
Flexion Extension Abduction External rotation Loss of) Flexion Extension Abduction External rotation
P
Treatment Locking plate
Nonsurgical
Mean SD, deg
Mean SD, deg
131 43 106 26
37 11 45 18
157 40 156 36
28 16 30 28
. . . .
48 8 71 27
37 10 43 14
13 4 17 9
26 7 25 13
.003 .141 .000 .001
SD, standard deviation. ) Loss is calculated as noninjured shoulder function minus fractured shoulder function.
Figure 2 Radiographs taken at follow-up in this study show malunion (Left) in a patient treated with a locking plate and (Right) in a patient treated nonsurgically.
patients (63%), displacement of the bony parts was already present on radiographs directly after surgery, indicating nonoptimal reduction of the fracture. In the other 3 patients (37%), displacements occurred during the healing process. In the nonsurgically treated group, 5 patients (28%) showed a malunited humeral head. No significant difference in the incidence of humeral head displacement was found between the locking-plate and nonsurgical group (P ¼ .298; Fig. 2). Results of OA gradation are reported in Table IV. Presence of OA is obvious when the Kellgren-Lawrence score is grade 2 or above. Looking at grade 2 or above, 11 patients (65%) in the locking plate group showed clear signs of OA compared with 6 patients (33%) in the nonsurgical group (P ¼ .06).
Complications In total, 10 patients in the locking plate group (56%) and 2 patients in the non-surgical group (11%) needed
additional treatment on their shoulder, which was a significant difference (P ¼ .005). The metalwork was completely removed in 7 patients (39%) in the locking plate group. The reason for removal was impingement or restricted ROM, or both, in 6 cases, and infection resulting in severe osteonecrosis in 1 patient. An arthrodesis was performed on this last patient. The plates were removed an average of 9 months after placement (range, 4-15 months). In 2 shoulders (11%), screws were replaced because of secondary screw perforation, and in 1 shoulder, the plate was substituted by a new plate due to loss of fixation. In the nonsurgical group, nonunion occurred in 1 patient. Chronic bursitis developed in another patient, which was treated with cortisone injections. Finally, position of the plate and screws was reviewed: 11plates (61%) were still in situ, 4 showed good position (>1 cm under the greater tuberosity and placed lateral to biceps tendon), and 5 plates were placed less than 1 cm under the greater tuberosity, 1 plate was not laterally
Locking plate vs nonsurgical treatment Table IV
1123
Kellgren-Lawrence score for osteoarthritis
Kellgren-Lawrence score None, grade 0 Doubtful, grade 1 Minimal, grade 2 Moderate, grade 3 Severe, grade 4
Treatment, No. (%) Locking plate
Nonsurgical
2 4 7 3 1
5 7 4 2 0
(11.8) (23.5) (41.2) (17.6) (5.9)
(27.8) (38.9) (22.2) (11.1) (0.0)
placed, and 1 plate was placed both less than 1cm under greater tuberosity and not lateral. Positioning of the screws in patients in which the plate was still in situ was normal in 3 cases, subchondral in 7, and protruding into the joint space in 1. In the 7 shoulders where the plate was removed, the position of the plate was suboptimal in all cases at the time of removal. Position of screws was normal in 6 patients and subchondral in 1.
Discussion It is difficult to compare the functional outcome and complication rate between surgical and nonsurgical treatments of proximal humeral fractures. To our knowledge, randomized controlled trials comparing functional outcomes of locking plates with a nonsurgical treatment on a midterm to long-term outcome have never been performed and are difficult to realize. This study tried to overcome this problem by directly matching 2 cohorts of patients, primarily based on fracture type. A significantly better ROM in patients treated nonsurgically was found in the current study. Difference in internal rotation was nonsignificant, but results were tending to a more favorable outcome in nonsurgically treated patients. Results in ASES and patient satisfaction scores were also tending toward nonsurgical treatment. An explanation for the more unfavorable midterm outcomes for patients treated with a locking plate could be due to the high complication rate in these patients. Many of these complications are related to the initial surgery and might therefore be preventable.1,4,5,27 Improving surgical technique could possibly lead to better midterm outcomes in patients with locking plates. Two factors might explain the presence of more profound OA in the locking plate group in this study. First, there is the direct influence of the plate and screws on the bony surface of the proximal humerus. Second, the higher rate of repeat operation in the locking plate group could have caused additional damage to the glenohumeral joint. Although the dominant arm was less often affected in the locking plate group, this difference was not significant. Therefore, it is unlikely that this factor substantially
affected the differences in midterm outcomes between the 2 treatment groups. Furthermore, it would probably underestimate the more favorable results for the nonsurgical group, because a fracture in the dominant arm would give more restrictions in daily functioning. Characteristics and results of the locking plate group in this study were similar to those in recent prospective multicenter studies, indicating that our sample of locking plate patients was representative.4,27 Furthermore, several studies comparing surgical treatment with nonsurgical treatment groups showed similar results as found in this study, concluding that nonsurgical treatment should be considered as a treatment option for proximal humeral fractures.13,17,20,29 This study has some limitations. Choice of fracture treatment was based on clinical judgement, as well as the patient’s fitness at time of treatment decision. Clinical judgement could influence prognosis of midterm outcome, leading to confounding by indication. However, by matching on fracture type, correction for this major clinical prognostic factor was achieved. Fitness at time of treatment could also lead to confounding by indication. Some patients who clinically warranted an operation were not fit for surgery at the time of injury and therefore received nonsurgical treatment. Nevertheless, this would underestimate the functional outcomes of nonsurgical treatment found in this study, because patients with a lower fitness are expected to have worse midterm functional outcomes. This study used the Kellgren-Lawrence score for classification of OA. Although this score has only been validated for use in the knee and hip,3 it was the best available instrument to score OA in the shoulder. Moreover, the number of patients in this study was quite small. Although it was large enough to detect a significant difference in ROM between the patient groups, results in ASES and patient satisfaction scores were tending toward nonsurgical treatment. A study that includes more patients would possibly show significant differences for these outcome measurements also. Finally, when considering the optimal treatment strategy for a proximal humeral fracture, the cost of treatment should always be taken into account. Surgical treatment with a locking plate is far more expensive than nonsurgical treatment. Research on the cost-effectiveness of both treatment strategies is needed to make a deliberate decision.
Conclusion Despite good results for locked plating on short-term follow-up, this study shows that patients treated nonsurgically for a proximal humeral fracture achieve better ROM on midterm follow-up (>1 year) compared with patients treated with a locking plate. Furthermore, OA seems to be more profound when a locking plate is used.
1124 At least equal results were found for other outcome measurements, such as ASES, patient satisfaction, and pain scores. The results of this study indicate that nonsurgical treatment should have a more prominent role in the treatment of proximal humeral fractures, even in cases that present clear indications for locking plate treatment.
Disclaimer This study was fully funded by the hospital budget provided by Government of South Australia to Flinders University Medical Centre. No grants were received and no other source providing funding or equipment were used. The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
References 1. Agudelo J, Sch€ urmann M, Stahel P, Helwig P, Morgan SJ, Zechel W, et al. Analysis of efficacy and failure in proximal humerus fractures treated with locking plates. J Orthop Trauma 2007;21:676-81. doi:10. 1097/BOT.0b013e31815bb09d 2. Bj€ orkenheim JM, Pajarinen J, Savolainen V. Internal fixation of proximal humeral fractures with a locking compression plate: a retrospective evaluation of 72 patients followed for a minimum of 1 year. Acta Orthop Scand 2004;75:741-5. doi:10.1080/00016470410004120 3. Brox JI, Lereim P, Merckoll E, Finnanger AM. Radiographic classification of glenohumeral arthrosis. Acta Orthop Scand 2003;74:186-9. doi:10.1080/00016470310013932 4. Brunner F, Sommer C, Bahrs C, Heuwinkel R, Hafner C, Rillmann P, et al. Open reduction and internal fixation of proximal humerus fractures using a proximal humeral locked plate: a prospective multicenter analysis. J Orthop Trauma 2009;23:163-72. doi:10.1097/ BOT.0b013e3181920e5b 5. Clavert P, Adam P, Bevort A, Bonnomet F, Kempf JF. Pitfalls and complications with locking plate for proximal humerus fracture. J Shoulder Elbow Surg 2010;19:489-94. doi:10.1016/j.jse.2009.09.005 6. Coons SJ, Rao S, Keininger DL, Hays RD. A comparative review of generic quality-of-life instruments. Pharmacoeconomics 2000;17: 13-35. 7. Court-Brown CM, McQueen MM. The impacted varus (A2.2) proximal humeral fracture: prediction of outcome and results of nonoperative treatment in 99 patients. Acta Orthop Scand 2004;75:736-40. doi:10.1080/759369235 8. Court-Brown CM, Cattermole H, McQueen MM. Impacted valgus fractures (B1.1) of the proximal humerus: the results of non-operative treatment. J Bone Joint Surg Br 2002;84:504-8. doi:10.1302/0301620X.84B4.12488 9. Court-Brown CM, Garg A, McQueen MM. The translated two-part fracture of the proximal humerus: epidemiology and outcome in the older patient. J Bone Joint Surg Br 2001;83:799-804. doi:10.1302/ 0301-620X.83B6.11401
R.J. Sanders et al. 10. Court-Brown CM, Garg A, McQueen MM. The epidemiology of proximal humeral fractures. Acta Orthop Scand 2001;72:365-71. 11. Duralde XA, Leddy LR. The results of ORIF of displaced unstable proximal humeral fractures using a locking plate. J Shoulder Elbow Surg 2010;19:480-8. doi:10.1016/j.jse.2009.08.008 12. Egol KA, Ong CC, Walsh M, Jazrawi LM, Tejwani NC, Zuckerman JD. Early complications in proximal humerus fractures (OTA Types 11) treated with locked plates. J Orthop Trauma 2008;22: 159-64. doi:10.1097/BOT.0b013e318169ef2a 13. Fjalestad T, Strømsøe K, Bl€ucher J, Tennøe B. Fractures in the proximal humerus: functional outcome and evaluation of 70 patients treated in hospital. Arch Orthop Trauma Surg 2005;125:310-6. doi:10. 1007/s00402-005-0803-9 14. Friess DM, Attia A, Vallier HA. Locking plate fixation for proximal humerus fractures: a comparison with other fixation techniques. Orthopedics 2008;31:1183. doi:10.3928/01477447-20081201-07 15. Gaebler C, McQueen MM, Court-Brown CM. Minimally displaced proximal humeral fractures: epidemiology and outcome in 507 cases. Acta Orthop Scand 2003;74:580-5. doi:10.1080/000164703100 17992 16. Handschin AE, Cardell M, Contaldo C, Trentz O, Wanner GA. Functional results of angular-stable plate fixation in displaced proximal humeral fractures. Injury 2008;39:306-13. doi:10.1016/j.injury. 2007.10.011 17. Hanson B, Neidenbach P, de Boer P, Stengel D. Functional outcomes after nonoperative management of fractures of the proximal humerus. J Shoulder Elbow Surg 2009;18:612-21. doi:10.1016/ j.jse.2009.03.024 18. Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957;16:494-502. doi:10.1136/ard.16.4.494 19. Lee SH, Dargent-Molina P, Breart G. Risk factors for fractures of the proximal humerus: results from the EPIDOS prospective study. J Bone Miner Res 2002;17:817-25. doi:10.1359/jbmr.2002.17.5.817 20. McLaurin TM. Proximal humerus fractures in the elderly are we operating on too many? Bull Hosp Jt Dis 2004;62:24-32. 21. Michener LA, McClure PW, Sennett BJ. American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form, patient self-report section: reliability, validity, and responsiveness. J Shoulder Elbow Surg 2002;11:587-94. doi:10.1067/mse.2002.127096 22. Neer CS 2nd. Displaced proximal humeral fractures: Part I. Classification and evaluation. J Bone Joint Surg Am 1970;52:1077-89. doi:10. 1055/s-2007-980100 23. Rabin R, de Charro F. EQ-5D: A measure of health status from the EuroQol Group. Ann Med 2001;33:337-43. 24. Ricchetti ET, Warrender WJ, Abboud JA. Use of locking plates in the treatment of proximal humerus fractures. J Shoulder Elbow Surg 2010; 19:66-75. doi:10.1016/j.jse.2010.01.001 25. Ring D. Current concepts in plate and screw fixation of osteoporotic proximal humerus fractures. Injury 2007;38(suppl 3):S59-68. doi:10. 1016/j.injury.2007.08.013 26. Smith AM, Mardones RM, Sperling JW, Cofield RH. Early complications of operatively treated proximal humeral fractures. J Shoulder Elbow Surg 2007;16:14-24. doi:10.1016/j.jse.2007.02.109 27. S€udkamp N, Bayer J, Hepp P, Voigt C, Oestern H, K€a€ab M, 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;91:13208. doi:10.2106/JBJS.H.00006 28. Thanasas C, Kontakis G, Angoules A, Limb D, Giannoudis P. Treatment of proximal humerus fractures with locking plates: a systematic review. J Shoulder Elbow Surg 2009;18:837-44. doi:10.1016/j.jse. 2009.06.004 29. Zyto K. Non-operative treatment of comminuted fractures of the proximal humerus in elderly patients. Injury 1998;29(5):349-52. doi: 10.1016/S0020-1383(97)00211-8