Musculoskelet Surg DOI 10.1007/s12306-015-0371-2
ORIGINAL ARTICLE
Proximal humeral fracture fixation: multicenter study with carbon fiber peek plate R. Rotini1 • M. Cavaciocchi1 • D. Fabbri1 • G. Bettelli1 • F. Catani2 • G. Campochiaro2 • M. Fontana3 • A. Colozza3 • C. F. De Biase4 • G. Ziveri4 C. Zapparoli5 • F. Stacca5 • R. Lupo6 • S. Rapisarda6 • E. Guerra1
•
Received: 4 February 2015 / Accepted: 30 March 2015 Istituto Ortopedico Rizzoli 2015
Abstract Background Locking plate fixation is a reliable treatment for many displaced proximal humeral fractures. Carbon fiber-reinforced–poly-ether-ether-ketone (CFR-PEEK) plates have recently been introduced as an alternative to traditional metallic plates. Methods In a multicenter study involving the Orthopedic Services of 6 Italian hospitals, 182 patients with a proximal humeral fracture were treated with a Diphos H (Lima Corporate, San Daniele del Friuli, Italy) CFR-PEEK plate, 160 of whom were followed clinically and radiographically for 2 years or more. Fractures were classified by Neer’s system. The functional results were assessed by Constant and DASH scores. Results The average time to radiographic healing was 5.6 months in 158 of 160 cases. Mean Constant score was 76, and mean DASH score was 28 at 2 years. There were two nonunions (one septic and one aseptic) and 13 cases of partial (9) or massive (4) humeral head necrosis. In three of the 78 patients treated with the first-generation plates, hardware breakage happened during the operation and the
& R. Rotini [email protected] 1
Shoulder and Elbow Unit, Rizzoli Orthopedic Institute, Bologna, Italy
2
Orthopedic Clinic, Modena University, Modena, Italy
3
Orthopedic and Trauma Unit, Faenza Hospital, Faenza, Italy
4
Orthopedic Clinic, Parma University, Parma, Italy
5
Orthopedic and Trauma Unit, New Civil Hospital, Modena, Italy
6
Orthopedic and Trauma Unit, San Giovanni di Dio Hospital, Agrigento, Italy
plate was replaced. There was no failure among the cases treated with the thicker second-generation plate. In eight cases, there was a perforation of the humeral head by the cephalic screws. Conclusions CFR-PEEK plates proved as reliable as metallic plates in the treatment of proximal humeral fractures. The advantages of these new devices include a better visualization of fracture reduction during intraoperative fluoroscopic assessment and easy hardware removal due to the absence of screw-plate cold fusion. Keywords Proximal humeral fracture Locking plate CFR-PEEK plate
Introduction Proximal humerus fractures represent about 5 % of all fractures [1, 2], and about 70 % involve patients over 60 years in which the effects of trauma can be complicated by preexisting osteoporosis. Recent studies show that their incidence in the population group over 75 years is raising exponentially, with a probable increase of 250 % in the next 30 years [3, 4]. Displaced fractures which can represent an indication for surgical treatment are between 13 and 16 % of all cases [5]. For these fractures, it is difficult to define a universally accepted treatment, even because retrospective studies and meta-analyses have not yet clarified whether one protocol of treatment is superior to others [6– 10]. Specific indications such as conservative treatment and prosthesis implant are strictly related to clinical and radiographic parameters. Internal fixation can be performed by several techniques (pins, endomedullary nailing, plates). Although all the fixation systems if correctly applied can produce good results, locking plates are nowadays the most
123
Musculoskelet Surg
widespread device for proximal humerus fractures [11–14]. Thanks to a dedicated precontoured design reflecting the anatomy of the proximal humerus, these plates can act as a mold for fracture reduction and also as a support for the greater tuberosity. Compared to the other osteosynthesis devices, locking plates guarantee a more stable fracture fixation and do not damage the rotator cuff tendons. In the recent years, carbon fiber-reinforced–poly-etherether-ketone (CFR-PEEK) has been tested and studied for orthopedic applications, in spinal and then in trauma surgery, as an alternative to metallic devices [15, 16]. CFRPEEK has a number of positive features. (1) Its elastic modulus is similar to the one of the cortical bones; however, its resistance to torsion is close to steel. (2) Its radiolucency allows an easier fluoroscopic and radiographic visualization of the fracture, reducing the ‘starburst effect’ and the artifacts and determining a better reading of CT scans and MRIs. (3) The risk of cold fusion between the plate and the locking screws is deleted, making hardware removal easier and safer. (4) The risk of allergic reaction to metals in sensible patients is avoided. CFR-PEEK locking plates (Diphos H, Lima Corporate, San Daniele del Friuli, Italy) have been available in market in the last 4 years. Aim of this work is to report the results of a multicenter retrospective study in which Diphos H CFR-PEEK plates were used in proximal humerus fractures in six Italian trauma centers. Clinical and functional results as well as implant-related and non-implant-related complications were studied.
Main aim of the study has been the clinical and radiographic evaluation (with plain radiographs and in selected cases CT scan) at a minimum 2-year follow-up. Twentytwo patients were lost to follow-up, of which 10 at the firstyear follow-up and 12 at the 2-year follow-up. Among the remaining 160 patients, there were 119 women (74.3 %) and 41 men, and their average age was 64 years (range from 23 to 84). The mechanism of trauma was low-energy or mid-energy in the vast majority of cases. Preoperative radiograms and CT scans were examined by two independent expert orthopedic surgeons for each hospital subgroup, and the fractures were classified according to Neer’s system. All the operations were performed by surgeons trained in the treatment of proximal humerus fractures (nine surgeons). Clinical and radiographic follow-up was obtained at 1, 3, 6 months, 1 and 2 years from surgery. Functional results were evaluated by Constant-Murley score and DASH score at 1 and 2 years from surgery. At surgery, the following data were recorded: presence of rotator cuff lesions, surgical time, use of a three- or five-hole plate. The first 78 plates were 3.3 mm thick in the diaphyseal body and 4.7 mm thick in the edges, while in the remaining 82 plates the thickness was 4.4 and 5.2 mm, respectively (Fig. 1). Hardware removal was done in 31 cases where humeral head necrosis, acromial impingement, shoulder arthrosis or fracture nonunion were diagnosed. In the cases of plate removal performed at the Rizzoli Orthopaedic Institute, a laboratory study with optical and electronic microscopy examination of the tissues surrounding the plate was done.
Materials and methods Surgical technique Patient recruitment and clinical assessment From April 2010 to May 2012, 182 patients with a proximal humeral fracture were enrolled in the orthopedic departments of six Italian hospitals (Istituto Ortopedico Rizzoli, Bologna; Orthopedic Clinic of the University, Modena; Orthopedic and Trauma Unit, Faenza; Orthopedic Clinic of the University, Parma; Orthopedic and Trauma Unit, New Civil Hospital, Modena; Orthopedic and Trauma Unit, San Giovanni di Dio Hospital, Agrigento). Inclusion criteria were closed fractures with indication for surgical open reduction and internal fixation, in skeletally mature patients at least 18 years old. Exclusion criteria were nonunions, pathologic fractures, concomitant fractures of the same upper limb, brachial plexus lesions, patients refusing to enter the study, patients unable to express a valid consensus. The patients were treated with a Diphos H plate (Lima Corporate, San Daniele del Friuli, Italy), and they were prospectively followed according to a definite protocol.
123
The patient is placed in beach-chair position, arm draped free, with a stable head support, on a radiolucent surgical table with side placement of an image intensifier to allow viewing of the humeral head in two planes. In all cases, a deltopectoral approach is followed with identification and lateral displacement of the cephalic vein, removal of the subscapularis bursa for a correct visualization of the underlying muscle and of the fracture, and exposure of the metaphyseal bone. The fracture is then reduced with indirect maneuvers, placing if necessary bone sutures for a better control of the tuberosities, leaving the soft tissue connections of the bony fragments and mainly of the greater tuberosity. After obtaining a satisfactory alignment of the fracture, a three- or five-hole plate, depending on the distal extension of the fracture, is placed 5–8 mm distal to the top of the greater tuberosity and 3–4 mm lateral to the bicipital groove, in order to avoid the impingement with the acromion proximally and with the long head of the biceps medially. Temporary reduction is
Musculoskelet Surg Fig. 1 Comparison between the thickness of the first Diphos Humeral plate model (4.2 mm) and the current one (5.2 mm). The numerical data, expressed in millimeters, include the plate edges
Fig. 2 a, b Valgus impacted three-part fracture of the right proximal humerus; c temporary reduction with k-wire and Diphos H plate: insertion of a non-locking screw in the oval hole, the aiming device is positioned on the plate. d Diphos H plate in situ with the sutures for
the tuberosities, passing through the dedicated plate holes; blue screws, locking, for epiphysis; red screws, locking, for diaphysis. e, f Final X-rays at 36-month follow-up (color figure online)
fixed by 1.8 mm k-wires passing through the dedicated plate holes and by a screw in the oblong hole of the plate and then checked by the image intensifier. Once the reduction is deemed adequate, the k-wires are replaced by locking head screws and finally locking or standard screws are placed in the remaining diaphyseal holes (Fig. 2).
were done without arriving overhead. After 3–4 weeks, active mobilization was added, while continuing passive mobilization by the therapist. Joint loading was allowed at 60 days, after checking possible complications and fracture healing. Statistical analysis
Postoperative treatment All the patients wore a sling for 3–4 weeks after surgery. In all cases, an early (24 or 48 h based on some factors such as age, bleeding, general conditions) postoperative rehabilitation was performed, with passive self-assisted exercises or, in older patients, with the assistance of a physical therapist. Forward elevation and abduction movements
Study monitoring, database management and statistics were performed by a central monitoring organization. Data were entered in an Access database (Microsoft Corporation, Washington, DC) and exported into Excel (Microsoft Corporation, Washington, DC). Statistical analyses were conducted with Intercooled Stata/IC (Stata Corporation LP, College Station, TX).
123
Musculoskelet Surg
Results Among the 160 cases with a minimum 2-year follow-up, there were 55 two-part (34.4 %), 76 three-part (47.5 %) and 29 four-part (18.1 %) fractures. The average surgical time for internal fixation was 84 min (range 35–180 min). Three-hole plates were implanted in most of cases (90 %). Radiographic healing was evident on average after 5.6 months in 158 of 160 cases. At the 12 month clinical examination, the average DASH score was 30 points and the average Constant score was 71. At the 2-year evaluation, average DASH score was 28 points and average Constant score was 76 points. The clinical scores obtained at 2-year follow-up are summarized in Table 1. Complications in our series (Table 2) can be separated as implant-related and non-implant-related. The first group includes three first-generation (4.2 mm thickness) plate breakages (3/78, 3.8 %), happened during insertion of a round head screw inside the oblong hole, trying to force fracture reduction by tightening the plate to bone. In all cases, failure was intraoperative and the plate was replaced by another identical one after improving bone alignment. There were eight cases (5 %) of head perforation by the head screws; in all cases, these screws were removed or changed. In two of these patients, an early glenohumeral degeneration was seen, treated by a shoulder prosthesis. In two cases (1.2 %), we observed asymptomatic backing out of screws from the plate. Among non-implant-related complications, there were two cases of nonunion. In one case, the nonunion was aseptic and was treated with plate removal and new fixation by metallic plate and bone graft, a cortical strut opposite to the plate. The second nonunion was septic (S. Epidermidis) and needed plate removal, temporary antibiotic-loaded cement spacer and general antibiotic therapy, and then secondary fixation by metallic plate and homologous endomedullary fibula bone graft (Fig. 3). Both nonunions healed uneventfully in about 9 months. Table 1 Functional results at 2-year follow-up 1 year
2 years
R.O.M. Forward elevation
119 ± 29
137 ± 28
Abduction
110 ± 29
129 ± 25
Ext. rotation in adduction
36 ± 13
40 ± 13
Int. rotation in adduction Ext. rotation in abduction
40 ± 21 42 ± 12
48 ± 19 46 ± 10
52 ± 23
56 ± 26
Int. rotation in abduction Constant score
71
76
DASH
30
28
123
Thirteen cases of avascular necrosis (AVN) (8.1 %) were found in our series, of which nine partial (superior pole) necroses, all treated with simple plate and screws removal, and four massive necroses (all in four-part fractures), treated by reverse shoulder prosthesis in two cases and by simple plate and screws removal in two cases (Fig. 4). There were 16 patients (10.0 %) complaining about rotator cuff symptoms, of which 14 related to the supraspinatus tendon and two to supraspinatus and infraspinatus. Adhesive capsulitis was diagnosed in five cases (3.1 %), of which three were treated by arthroscopic release and two conservatively, in one case the recovery was incomplete, with residual symptoms at the latest follow-up. In 18 patients (11.2 %), a symptomatic impingement between plate and acromion was diagnosed and 15 of these were treated with plate removal after definitive fracture healing. Two patients (1.2 %) had a displacement of the greater tuberosity. In five cases (3.1 %), there was a postoperative loss of reduction; however, these patients did not need surgical revision, and in all of them, the screws supporting the medial buttress were absent or insufficient. There were no cases of perioperative or postoperative superficial infection and no cases of neurological or vascular complications. The optical and electronic microscopy study of the tissue surrounding the plate did not reveal any foreign body reaction, thus confirming the material biocompatibility.
Discussion Fractures of the proximal humerus, especially three- and four-part fractures, are still nowadays a challenge for orthopedic surgeons aiming to obtain a stable anatomic reduction that can allow an early functional recovery. Several complications can happen in this type of fractures, especially in the most vulnerable age group, over 65 years, due to the poor bone quality. A different setting is represented by young or middle-aged patients in whom proximal humeral fractures are caused by high-energy trauma and often show complex patterns. In this subgroup, it is more frequent to have to face the problem of unsatisfactory functional results after healing, which can be a reason, together with the cases of humeral head necrosis, for considering a second operation for hardware removal. This procedure can sometimes cause complications mainly related to screw blockage (so-called cold fusion) with consequent lengthening of surgical time, metal cutting instruments are not always available inside a surgical room, and, even worse, possible retention of metallic parts (e.g., broken screws inside the medullary canal) can represent an obstacle for a future operation like
Musculoskelet Surg Table 2 Complications
Complications
Implant-related
Non-implant-related
All
Reoperation
Nonunion Aseptic
0
1
1
1
Septic
0
1
1
1
Impingement
0
18
18
15
Frozen shoulder
0
5
5
3
Rotator cuff lesion
0
16
16
0
Neurologic complication
0
0
0
0
Secondary loss of reduction
5
0
5
0
Secondary dislocation of tuberosity
0
2
2
0
Soft tissue
Reduction
Plates and screws Secondary screw perforation
8
0
8
8
Loosening
0
0
0
0
Screw backing out Breakage
2 3
0 0
3
0 0
Proximal screw pull out
2
1
Head necrosis Partial
0
9
9
Total
0
4
4
4
18
56
74
42
Total number of complications
a shoulder prosthesis. This has caused an increasing interest for a biomaterial like CFR-PEEK that shares the biomechanical advantages of metallic locking plates and at the same time can solve the problems related to surgical removal. Our study, in our knowledge, is the first one analyzing clinical and radiographic results of patients affected with a proximal humeral fracture treated with CFR-PEEK locking plates, with a 2-year follow-up, relative to a multicentric study from first- and second-level trauma centers. Experimental studies [17] have compared CFR-PEEK and titanium rods in spine surgery and have concluded that the higher flexibility of CFR-PEEK reduces stresses at the bone–screw interface. Other studies comparing titanium and CFR-PEEK plates of similar shapes lead to the hypothesis that the lower rigidity of CFR-PEEK fixation may reduce the risk of screw pullout/pushout [16]. Unfortunately, a CFR-PEEK plate, having a form memory, cannot be modified and adapted on the surgical field and cannot be used in long constructs. Clinical results and complications as showed by our 2-year study with Diphos H plate (available in market since 2010) are comparable with those reported in literature both on the point of view of functional recovery and on clinical tests results [18, 19]. Ricchetti [20] maintains that the frequency of complications has varied in the literature, with reported overall complication rates ranging from 10 to 81 %, depending on the definitions and inclusion criteria
9
utilized. The great majority of fractures (158/160, 98.7 %) healed before 6 months from surgery. The analysis of complications shows that they can be separated in two groups, one relative to the implant and the other unrelated to the implant [19]. In the first group, breakage of the plate has a major role, related to the nonmetallic type of material. Plate failure happened in three cases in two different institutions, and all these plates were lower thickness type used in the first part of this experience (3/78, 3.8 %). Breakage happened at the height of the oval hole, which was the weak point of the first-generation plates, at the moment of tightening the spherical head screw while exploiting the plate to force the reduction in the fracture. With second-generation plates, in which the thickness of the plate body has been increased to 4.4 mm and the edges to 5.2 mm, and that are the only Diphos H plates available in market since October 2011, no case of breakage has been reported. Fatigue failure has never happened with both generations of plates. About screws problems, there were eight cases (5 %) of secondary perforation of the humeral head, two cases (1.2 %) of backing out and two cases (1.2 %) of pull out of the cephalic screws. These complications, leading to a loss of fracture reduction, are related in part to an excessive implant stiffness and in part to other factors such as severe osteoporosis, uncorrect screw placement, or excessive screw length, or insufficient medial support of the lower cephalic screws [21].
123
Musculoskelet Surg Fig. 3 a, b X-rays of a left proximal humeral fracture after 6-month follow-up, with a septic nonunion. c X-rays after Diphos H plate removal, cement spacer and temporary fixation with metal wires. d X-rays at 9 months from fixation with metal plate and intramedullary homologous bone graft (fibula)
Fig. 4 a, b Valgus impacted four-part fracture of the right proximal humerus. c, d Postoperative X-rays, with reduction and fixation with Diphos H plate. e, f At 6-month follow-up, signs of head necrosis. g, h X-rays of the head necrosis at 4-year follow-up, after plate and screws removal
123
Musculoskelet Surg
Among non-implant-related complications, which proved to be similar to those reported in literature with other plates, the incidence of head necrosis (Fig. 4) in our study was 8.1 % (13 cases, of which nine partial and four massive). It has to be mentioned that 18.1 % of the fractures were four-part, notoriously having a higher risk of AVN. Recent literature data indicate an overall risk of AVN of 10 [22], 16.4 [23], 9 [14] and 1.85 % [20]. The incidence of fracture nonunion, as reported by Thanasas et al. [18] in his review, is about 1.6 %; however, other authors [23] found higher figures (5.5 %). Similar to the data by Thanasas, we found in our study two cases of nonunion (of which one septic), representing 1.2 %. We think that this is normal in relation to the pattern of fractures included in the study, rarely involving the upper diaphyseal part of the humerus. In fact, three-hole plates were used in most of our cases, adequate for obtaining a stable reduction in fractures. The Diphos H plate has been proved to provide a stable construct, similar to metallic plates. Another point that was carefully analyzed is the impingement between plate and acromial bone. The incidence of this complication, secondary to a high plate positioning, is quite variable in literature from 2 [19] to 22 % [18]. Basing on our experience with several different types of plate, we believe that this problem is strictly related to the system of plate fixation. In fact we regularly check intraoperatively the presence of acromial impingement in 90 elevation and external rotation, and a contact between plate and acromion is almost constant in our experience. We therefore think that in active patients, a future surgery for plate removal has often to be foreseen in order to improve the range of motion once the fracture is healed. This point stresses the importance to use a plate guaranteeing low chances of complications like screw-plate cold fusion during removal. In our study, we have recorded an 11.2 % incidence of symptomatic impingement between plate and acromion. In all cases where the plate was removed, there was a symptom benefit with increase in the range of motion. One advantage of the Diphos H plate as evidenced by this multicenter study is its radiolucency, allowing an optimal visualization of the fracture during the surgical procedure and then during follow-up. Another point is represented by the multidirectional self-taping locking screws, allowing the surgeon to choose for each screw the optimal axis of introduction. A third main advantage of this plate is the easiness of removal. In the two cases in which during removal one blocked screw was found, the problem was solved removing all the other screws and cutting the plate around the blocked screw by a simple bone saw, and then, the screw was grasped and unscrewed.
Some limits of this study deserve a mention. •
• • • •
The series of the single institutions are not completely homogeneous relative to age, fracture types, antibiotic prophylaxis and rehabilitation protocols. The fractures were classified only according to the Neer’s system. The structure of the plate was modified as mentioned during the study. A mechanical evaluation of the holding strength between plate and locking screws is missing. The functional evaluation through the Constant score, considering the operated limb only, conditions the results as the study population is not homogeneous in relation to age, and therefore, data are influenced by the lower basic scores of the elderly patients.
Conclusions The experience with the Diphos H plate compared with other locking plate types led to some interesting conclusions. •
•
•
•
The functional results as measured through Constant and DASH scores are comparable, and the incidence of complications is similar (plate failure never happened after modifying the plate thickness). The advantages include plate radiolucency, mainly during reduction maneuvers of complex fractures, especially in the axial radiographic view, multidirectional self-taping locking screws and resolution of the problem of screw-plate cold fusion. The limits include the impossibility to model the plate to adapt it to the bony contour that makes it unfit for fractures extending to the mid-third of the diaphysis and for malunions and nonunions. The debris generation at the insertion of the screws does not represent a biocompatibility risk.
Conflict of interest of interest.
The authors declare that they have no conflict
Human and animal rights This article does not contain any studies with human or animal subjects performed by any of the authors.
References 1. Palvanen M, Kannus P, Niemi S, Parkkari J (2006) Update in the epidemiology of proximal humeral fractures. Clin Orthop Relat Res 442:87–92. doi:10.1097/01.blo.0000194672.79634.78 2. Bell J-E, Leung BC, Spratt KF et al (2011) Trends and variation in incidence, surgical treatment, and repeat surgery of proximal humeral fractures in the elderly. J Bone Joint Surg Am 93:121–131. doi:10.2106/JBJS.I.01505
123
Musculoskelet Surg 3. Obrant KJ, Bengne´r U, Johnell O, Nilsson BE, Sernbo I (1989) Increasing age-adjusted risk of fragility fractures: a sign of increasing osteoporosis in successive generations? Calcif Tissue Int 44:157–167. doi:10.1007/BF02556558 4. Hertel R (2005) Fractures of the proximal humerus in osteoporotic bone. Osteoporos Int. doi:10.1007/s00198-004-1714-2 5. Rose SH, Melton LJ, Morrey BF, Ilstrup DM, Riggs BL (1982) Epidemiologic features of humeral fractures. Clin Orthop Relat Res 168:24–30 6. Resch H (2011) Proximal humeral fractures: current controversies. J Shoulder Elbow Surg 20:827–832. doi:10.1016/j.jse.2011. 01.009 7. Boileau P, Pennington SD, Alami G (2011) Proximal humeral fractures in younger patients: fixation techniques and arthroplasty. J Shoulder Elbow Surg 20:S47–S60. doi:10.1016/j.jse. 2010.12.006 8. Murray IR, Amin AK, White TO, Robinson CM (2011) Proximal humeral fractures: current concepts in classification, treatment and outcomes. J Bone Joint Surg Br 93:1–11. doi:10.1302/0301620X.93B1.25702 9. Iacobellis C, Berizzi A, Biz C, Camporese A (2014) Treatment of proximal humeral fractures with reverse shoulder arthroplasty in elderly patients. Musculoskelet Surg. doi:10.1007/s12306-0140331-2 10. Verdano MA, Lunini E, Pellegrini A, Corsini T, Marenghi P, Ceccarelli F (2014) Can the osteosynthesis with locking plates be a better treatment for unstable fractures of the proximal humerus? Musculoskelet Surg 98:27–33. doi:10.1007/s12306-013-0267-y 11. Edwards SL, Wilson NA, Zhang L-Q, Flores S, Merk BR (2006) Two-part surgical neck fractures of the proximal part of the humerus. A biomechanical evaluation of two fixation techniques. J Bone Joint Surg Am 88:2258–2264. doi:10.2106/JBJS.E.00757 12. Seide K, Triebe J, Faschingbauer M et al (2007) Locked vs. unlocked plate osteosynthesis of the proximal humerus—a biomechanical study. Clin Biomech 22:176–182. doi:10.1016/j. clinbiomech.2006.08.009 13. Siffri PC, Peindl RD, Coley ER, Norton J, Connor PM, Kellam JF (2006) Biomechanical analysis of blade plate versus locking plate fixation for a proximal humerus fracture: comparison using cadaveric and synthetic humeri. J Orthop Trauma 20:547–554. doi:10.1097/01.bot.0000244997.52751.58
123
14. Duralde XA, Leddy LR (2010) The results of ORIF of displaced unstable proximal humeral fractures using a locking plate. J Shoulder Elbow Surg 19:480–488. doi:10.1016/j.jse.2009.08.008 15. Steinberg EL, Rath E, Shlaifer A, Chechik O, Maman E, Salai M (2012) Carbon fiber reinforced PEEK Optima–a composite material biomechanical properties and wear/debris characteristics of CF-PEEK composites for orthopedic trauma implants. J Mech Behav Biomed Mater 17:221–228. doi:10.1016/j.jmbbm.2012.09. 013 16. Feerick EM, Kennedy J, Mullett H, FitzPatrick D, McGarry P (2012) Investigation of metallic and carbon fibre PEEK fracture fixation devices for three-part proximal humeral fractures. Med Eng Phys. doi:10.1016/j.medengphy.2012.07.016 17. Ponnappan RK, Serhan H, Zarda B, Patel R, Albert T, Vaccaro AR (2009) Biomechanical evaluation and comparison of polyetheretherketone rod system to traditional titanium rod fixation. Spine J 9:263–267. doi:10.1016/j.spinee.2008.08.002 18. Thanasas C, Kontakis G, Angoules A, Limb D, Giannoudis P (2009) Treatment of proximal humerus fractures with locking plates: a systematic review. J Shoulder Elbow Surg 18:837–844. doi:10.1016/j.jse.2009.06.004 19. Brunner F, Sommer C, Bahrs C et al (2009) Open reduction and internal fixation of proximal humerus fractures using a proximal humeral locked plate: a prospective multicenter analysis. J Orthop Trauma 23:163–172. doi:10.1097/BOT.0b013e318 1920e5b 20. Ricchetti ET, Warrender WJ, Abboud JA (2010) Use of locking plates in the treatment of proximal humerus fractures. J Shoulder Elbow Surg 19:66–75. doi:10.1016/j.jse.2010.01.001 21. Charalambous CP, Siddique I, Valluripalli K et al (2007) Proximal humeral internal locking system (PHILOS) for the treatment of proximal humeral fractures. Arch Orthop Trauma Surg 127:205–210. doi:10.1007/s00402-006-0256-9 22. Sproul RC, Iyengar JJ, Devcic Z, Feeley BT (2011) A systematic review of locking plate fixation of proximal humerus fractures. Injury 42:408–413. doi:10.1016/j.injury.2010.11.058 23. Clavert P, Adam P, Bevort A, Bonnomet F, Kempf JF (2010) Pitfalls and complications with locking plate for proximal humerus fracture. J Shoulder Elbow Surg 19:489–494. doi:10. 1016/j.jse.2009.09.005
ORIGINAL ARTICLE
Proximal humeral fracture fixation: multicenter study with carbon fiber peek plate R. Rotini1 • M. Cavaciocchi1 • D. Fabbri1 • G. Bettelli1 • F. Catani2 • G. Campochiaro2 • M. Fontana3 • A. Colozza3 • C. F. De Biase4 • G. Ziveri4 C. Zapparoli5 • F. Stacca5 • R. Lupo6 • S. Rapisarda6 • E. Guerra1
•
Received: 4 February 2015 / Accepted: 30 March 2015 Istituto Ortopedico Rizzoli 2015
Abstract Background Locking plate fixation is a reliable treatment for many displaced proximal humeral fractures. Carbon fiber-reinforced–poly-ether-ether-ketone (CFR-PEEK) plates have recently been introduced as an alternative to traditional metallic plates. Methods In a multicenter study involving the Orthopedic Services of 6 Italian hospitals, 182 patients with a proximal humeral fracture were treated with a Diphos H (Lima Corporate, San Daniele del Friuli, Italy) CFR-PEEK plate, 160 of whom were followed clinically and radiographically for 2 years or more. Fractures were classified by Neer’s system. The functional results were assessed by Constant and DASH scores. Results The average time to radiographic healing was 5.6 months in 158 of 160 cases. Mean Constant score was 76, and mean DASH score was 28 at 2 years. There were two nonunions (one septic and one aseptic) and 13 cases of partial (9) or massive (4) humeral head necrosis. In three of the 78 patients treated with the first-generation plates, hardware breakage happened during the operation and the
& R. Rotini [email protected] 1
Shoulder and Elbow Unit, Rizzoli Orthopedic Institute, Bologna, Italy
2
Orthopedic Clinic, Modena University, Modena, Italy
3
Orthopedic and Trauma Unit, Faenza Hospital, Faenza, Italy
4
Orthopedic Clinic, Parma University, Parma, Italy
5
Orthopedic and Trauma Unit, New Civil Hospital, Modena, Italy
6
Orthopedic and Trauma Unit, San Giovanni di Dio Hospital, Agrigento, Italy
plate was replaced. There was no failure among the cases treated with the thicker second-generation plate. In eight cases, there was a perforation of the humeral head by the cephalic screws. Conclusions CFR-PEEK plates proved as reliable as metallic plates in the treatment of proximal humeral fractures. The advantages of these new devices include a better visualization of fracture reduction during intraoperative fluoroscopic assessment and easy hardware removal due to the absence of screw-plate cold fusion. Keywords Proximal humeral fracture Locking plate CFR-PEEK plate
Introduction Proximal humerus fractures represent about 5 % of all fractures [1, 2], and about 70 % involve patients over 60 years in which the effects of trauma can be complicated by preexisting osteoporosis. Recent studies show that their incidence in the population group over 75 years is raising exponentially, with a probable increase of 250 % in the next 30 years [3, 4]. Displaced fractures which can represent an indication for surgical treatment are between 13 and 16 % of all cases [5]. For these fractures, it is difficult to define a universally accepted treatment, even because retrospective studies and meta-analyses have not yet clarified whether one protocol of treatment is superior to others [6– 10]. Specific indications such as conservative treatment and prosthesis implant are strictly related to clinical and radiographic parameters. Internal fixation can be performed by several techniques (pins, endomedullary nailing, plates). Although all the fixation systems if correctly applied can produce good results, locking plates are nowadays the most
123
Musculoskelet Surg
widespread device for proximal humerus fractures [11–14]. Thanks to a dedicated precontoured design reflecting the anatomy of the proximal humerus, these plates can act as a mold for fracture reduction and also as a support for the greater tuberosity. Compared to the other osteosynthesis devices, locking plates guarantee a more stable fracture fixation and do not damage the rotator cuff tendons. In the recent years, carbon fiber-reinforced–poly-etherether-ketone (CFR-PEEK) has been tested and studied for orthopedic applications, in spinal and then in trauma surgery, as an alternative to metallic devices [15, 16]. CFRPEEK has a number of positive features. (1) Its elastic modulus is similar to the one of the cortical bones; however, its resistance to torsion is close to steel. (2) Its radiolucency allows an easier fluoroscopic and radiographic visualization of the fracture, reducing the ‘starburst effect’ and the artifacts and determining a better reading of CT scans and MRIs. (3) The risk of cold fusion between the plate and the locking screws is deleted, making hardware removal easier and safer. (4) The risk of allergic reaction to metals in sensible patients is avoided. CFR-PEEK locking plates (Diphos H, Lima Corporate, San Daniele del Friuli, Italy) have been available in market in the last 4 years. Aim of this work is to report the results of a multicenter retrospective study in which Diphos H CFR-PEEK plates were used in proximal humerus fractures in six Italian trauma centers. Clinical and functional results as well as implant-related and non-implant-related complications were studied.
Main aim of the study has been the clinical and radiographic evaluation (with plain radiographs and in selected cases CT scan) at a minimum 2-year follow-up. Twentytwo patients were lost to follow-up, of which 10 at the firstyear follow-up and 12 at the 2-year follow-up. Among the remaining 160 patients, there were 119 women (74.3 %) and 41 men, and their average age was 64 years (range from 23 to 84). The mechanism of trauma was low-energy or mid-energy in the vast majority of cases. Preoperative radiograms and CT scans were examined by two independent expert orthopedic surgeons for each hospital subgroup, and the fractures were classified according to Neer’s system. All the operations were performed by surgeons trained in the treatment of proximal humerus fractures (nine surgeons). Clinical and radiographic follow-up was obtained at 1, 3, 6 months, 1 and 2 years from surgery. Functional results were evaluated by Constant-Murley score and DASH score at 1 and 2 years from surgery. At surgery, the following data were recorded: presence of rotator cuff lesions, surgical time, use of a three- or five-hole plate. The first 78 plates were 3.3 mm thick in the diaphyseal body and 4.7 mm thick in the edges, while in the remaining 82 plates the thickness was 4.4 and 5.2 mm, respectively (Fig. 1). Hardware removal was done in 31 cases where humeral head necrosis, acromial impingement, shoulder arthrosis or fracture nonunion were diagnosed. In the cases of plate removal performed at the Rizzoli Orthopaedic Institute, a laboratory study with optical and electronic microscopy examination of the tissues surrounding the plate was done.
Materials and methods Surgical technique Patient recruitment and clinical assessment From April 2010 to May 2012, 182 patients with a proximal humeral fracture were enrolled in the orthopedic departments of six Italian hospitals (Istituto Ortopedico Rizzoli, Bologna; Orthopedic Clinic of the University, Modena; Orthopedic and Trauma Unit, Faenza; Orthopedic Clinic of the University, Parma; Orthopedic and Trauma Unit, New Civil Hospital, Modena; Orthopedic and Trauma Unit, San Giovanni di Dio Hospital, Agrigento). Inclusion criteria were closed fractures with indication for surgical open reduction and internal fixation, in skeletally mature patients at least 18 years old. Exclusion criteria were nonunions, pathologic fractures, concomitant fractures of the same upper limb, brachial plexus lesions, patients refusing to enter the study, patients unable to express a valid consensus. The patients were treated with a Diphos H plate (Lima Corporate, San Daniele del Friuli, Italy), and they were prospectively followed according to a definite protocol.
123
The patient is placed in beach-chair position, arm draped free, with a stable head support, on a radiolucent surgical table with side placement of an image intensifier to allow viewing of the humeral head in two planes. In all cases, a deltopectoral approach is followed with identification and lateral displacement of the cephalic vein, removal of the subscapularis bursa for a correct visualization of the underlying muscle and of the fracture, and exposure of the metaphyseal bone. The fracture is then reduced with indirect maneuvers, placing if necessary bone sutures for a better control of the tuberosities, leaving the soft tissue connections of the bony fragments and mainly of the greater tuberosity. After obtaining a satisfactory alignment of the fracture, a three- or five-hole plate, depending on the distal extension of the fracture, is placed 5–8 mm distal to the top of the greater tuberosity and 3–4 mm lateral to the bicipital groove, in order to avoid the impingement with the acromion proximally and with the long head of the biceps medially. Temporary reduction is
Musculoskelet Surg Fig. 1 Comparison between the thickness of the first Diphos Humeral plate model (4.2 mm) and the current one (5.2 mm). The numerical data, expressed in millimeters, include the plate edges
Fig. 2 a, b Valgus impacted three-part fracture of the right proximal humerus; c temporary reduction with k-wire and Diphos H plate: insertion of a non-locking screw in the oval hole, the aiming device is positioned on the plate. d Diphos H plate in situ with the sutures for
the tuberosities, passing through the dedicated plate holes; blue screws, locking, for epiphysis; red screws, locking, for diaphysis. e, f Final X-rays at 36-month follow-up (color figure online)
fixed by 1.8 mm k-wires passing through the dedicated plate holes and by a screw in the oblong hole of the plate and then checked by the image intensifier. Once the reduction is deemed adequate, the k-wires are replaced by locking head screws and finally locking or standard screws are placed in the remaining diaphyseal holes (Fig. 2).
were done without arriving overhead. After 3–4 weeks, active mobilization was added, while continuing passive mobilization by the therapist. Joint loading was allowed at 60 days, after checking possible complications and fracture healing. Statistical analysis
Postoperative treatment All the patients wore a sling for 3–4 weeks after surgery. In all cases, an early (24 or 48 h based on some factors such as age, bleeding, general conditions) postoperative rehabilitation was performed, with passive self-assisted exercises or, in older patients, with the assistance of a physical therapist. Forward elevation and abduction movements
Study monitoring, database management and statistics were performed by a central monitoring organization. Data were entered in an Access database (Microsoft Corporation, Washington, DC) and exported into Excel (Microsoft Corporation, Washington, DC). Statistical analyses were conducted with Intercooled Stata/IC (Stata Corporation LP, College Station, TX).
123
Musculoskelet Surg
Results Among the 160 cases with a minimum 2-year follow-up, there were 55 two-part (34.4 %), 76 three-part (47.5 %) and 29 four-part (18.1 %) fractures. The average surgical time for internal fixation was 84 min (range 35–180 min). Three-hole plates were implanted in most of cases (90 %). Radiographic healing was evident on average after 5.6 months in 158 of 160 cases. At the 12 month clinical examination, the average DASH score was 30 points and the average Constant score was 71. At the 2-year evaluation, average DASH score was 28 points and average Constant score was 76 points. The clinical scores obtained at 2-year follow-up are summarized in Table 1. Complications in our series (Table 2) can be separated as implant-related and non-implant-related. The first group includes three first-generation (4.2 mm thickness) plate breakages (3/78, 3.8 %), happened during insertion of a round head screw inside the oblong hole, trying to force fracture reduction by tightening the plate to bone. In all cases, failure was intraoperative and the plate was replaced by another identical one after improving bone alignment. There were eight cases (5 %) of head perforation by the head screws; in all cases, these screws were removed or changed. In two of these patients, an early glenohumeral degeneration was seen, treated by a shoulder prosthesis. In two cases (1.2 %), we observed asymptomatic backing out of screws from the plate. Among non-implant-related complications, there were two cases of nonunion. In one case, the nonunion was aseptic and was treated with plate removal and new fixation by metallic plate and bone graft, a cortical strut opposite to the plate. The second nonunion was septic (S. Epidermidis) and needed plate removal, temporary antibiotic-loaded cement spacer and general antibiotic therapy, and then secondary fixation by metallic plate and homologous endomedullary fibula bone graft (Fig. 3). Both nonunions healed uneventfully in about 9 months. Table 1 Functional results at 2-year follow-up 1 year
2 years
R.O.M. Forward elevation
119 ± 29
137 ± 28
Abduction
110 ± 29
129 ± 25
Ext. rotation in adduction
36 ± 13
40 ± 13
Int. rotation in adduction Ext. rotation in abduction
40 ± 21 42 ± 12
48 ± 19 46 ± 10
52 ± 23
56 ± 26
Int. rotation in abduction Constant score
71
76
DASH
30
28
123
Thirteen cases of avascular necrosis (AVN) (8.1 %) were found in our series, of which nine partial (superior pole) necroses, all treated with simple plate and screws removal, and four massive necroses (all in four-part fractures), treated by reverse shoulder prosthesis in two cases and by simple plate and screws removal in two cases (Fig. 4). There were 16 patients (10.0 %) complaining about rotator cuff symptoms, of which 14 related to the supraspinatus tendon and two to supraspinatus and infraspinatus. Adhesive capsulitis was diagnosed in five cases (3.1 %), of which three were treated by arthroscopic release and two conservatively, in one case the recovery was incomplete, with residual symptoms at the latest follow-up. In 18 patients (11.2 %), a symptomatic impingement between plate and acromion was diagnosed and 15 of these were treated with plate removal after definitive fracture healing. Two patients (1.2 %) had a displacement of the greater tuberosity. In five cases (3.1 %), there was a postoperative loss of reduction; however, these patients did not need surgical revision, and in all of them, the screws supporting the medial buttress were absent or insufficient. There were no cases of perioperative or postoperative superficial infection and no cases of neurological or vascular complications. The optical and electronic microscopy study of the tissue surrounding the plate did not reveal any foreign body reaction, thus confirming the material biocompatibility.
Discussion Fractures of the proximal humerus, especially three- and four-part fractures, are still nowadays a challenge for orthopedic surgeons aiming to obtain a stable anatomic reduction that can allow an early functional recovery. Several complications can happen in this type of fractures, especially in the most vulnerable age group, over 65 years, due to the poor bone quality. A different setting is represented by young or middle-aged patients in whom proximal humeral fractures are caused by high-energy trauma and often show complex patterns. In this subgroup, it is more frequent to have to face the problem of unsatisfactory functional results after healing, which can be a reason, together with the cases of humeral head necrosis, for considering a second operation for hardware removal. This procedure can sometimes cause complications mainly related to screw blockage (so-called cold fusion) with consequent lengthening of surgical time, metal cutting instruments are not always available inside a surgical room, and, even worse, possible retention of metallic parts (e.g., broken screws inside the medullary canal) can represent an obstacle for a future operation like
Musculoskelet Surg Table 2 Complications
Complications
Implant-related
Non-implant-related
All
Reoperation
Nonunion Aseptic
0
1
1
1
Septic
0
1
1
1
Impingement
0
18
18
15
Frozen shoulder
0
5
5
3
Rotator cuff lesion
0
16
16
0
Neurologic complication
0
0
0
0
Secondary loss of reduction
5
0
5
0
Secondary dislocation of tuberosity
0
2
2
0
Soft tissue
Reduction
Plates and screws Secondary screw perforation
8
0
8
8
Loosening
0
0
0
0
Screw backing out Breakage
2 3
0 0
3
0 0
Proximal screw pull out
2
1
Head necrosis Partial
0
9
9
Total
0
4
4
4
18
56
74
42
Total number of complications
a shoulder prosthesis. This has caused an increasing interest for a biomaterial like CFR-PEEK that shares the biomechanical advantages of metallic locking plates and at the same time can solve the problems related to surgical removal. Our study, in our knowledge, is the first one analyzing clinical and radiographic results of patients affected with a proximal humeral fracture treated with CFR-PEEK locking plates, with a 2-year follow-up, relative to a multicentric study from first- and second-level trauma centers. Experimental studies [17] have compared CFR-PEEK and titanium rods in spine surgery and have concluded that the higher flexibility of CFR-PEEK reduces stresses at the bone–screw interface. Other studies comparing titanium and CFR-PEEK plates of similar shapes lead to the hypothesis that the lower rigidity of CFR-PEEK fixation may reduce the risk of screw pullout/pushout [16]. Unfortunately, a CFR-PEEK plate, having a form memory, cannot be modified and adapted on the surgical field and cannot be used in long constructs. Clinical results and complications as showed by our 2-year study with Diphos H plate (available in market since 2010) are comparable with those reported in literature both on the point of view of functional recovery and on clinical tests results [18, 19]. Ricchetti [20] maintains that the frequency of complications has varied in the literature, with reported overall complication rates ranging from 10 to 81 %, depending on the definitions and inclusion criteria
9
utilized. The great majority of fractures (158/160, 98.7 %) healed before 6 months from surgery. The analysis of complications shows that they can be separated in two groups, one relative to the implant and the other unrelated to the implant [19]. In the first group, breakage of the plate has a major role, related to the nonmetallic type of material. Plate failure happened in three cases in two different institutions, and all these plates were lower thickness type used in the first part of this experience (3/78, 3.8 %). Breakage happened at the height of the oval hole, which was the weak point of the first-generation plates, at the moment of tightening the spherical head screw while exploiting the plate to force the reduction in the fracture. With second-generation plates, in which the thickness of the plate body has been increased to 4.4 mm and the edges to 5.2 mm, and that are the only Diphos H plates available in market since October 2011, no case of breakage has been reported. Fatigue failure has never happened with both generations of plates. About screws problems, there were eight cases (5 %) of secondary perforation of the humeral head, two cases (1.2 %) of backing out and two cases (1.2 %) of pull out of the cephalic screws. These complications, leading to a loss of fracture reduction, are related in part to an excessive implant stiffness and in part to other factors such as severe osteoporosis, uncorrect screw placement, or excessive screw length, or insufficient medial support of the lower cephalic screws [21].
123
Musculoskelet Surg Fig. 3 a, b X-rays of a left proximal humeral fracture after 6-month follow-up, with a septic nonunion. c X-rays after Diphos H plate removal, cement spacer and temporary fixation with metal wires. d X-rays at 9 months from fixation with metal plate and intramedullary homologous bone graft (fibula)
Fig. 4 a, b Valgus impacted four-part fracture of the right proximal humerus. c, d Postoperative X-rays, with reduction and fixation with Diphos H plate. e, f At 6-month follow-up, signs of head necrosis. g, h X-rays of the head necrosis at 4-year follow-up, after plate and screws removal
123
Musculoskelet Surg
Among non-implant-related complications, which proved to be similar to those reported in literature with other plates, the incidence of head necrosis (Fig. 4) in our study was 8.1 % (13 cases, of which nine partial and four massive). It has to be mentioned that 18.1 % of the fractures were four-part, notoriously having a higher risk of AVN. Recent literature data indicate an overall risk of AVN of 10 [22], 16.4 [23], 9 [14] and 1.85 % [20]. The incidence of fracture nonunion, as reported by Thanasas et al. [18] in his review, is about 1.6 %; however, other authors [23] found higher figures (5.5 %). Similar to the data by Thanasas, we found in our study two cases of nonunion (of which one septic), representing 1.2 %. We think that this is normal in relation to the pattern of fractures included in the study, rarely involving the upper diaphyseal part of the humerus. In fact, three-hole plates were used in most of our cases, adequate for obtaining a stable reduction in fractures. The Diphos H plate has been proved to provide a stable construct, similar to metallic plates. Another point that was carefully analyzed is the impingement between plate and acromial bone. The incidence of this complication, secondary to a high plate positioning, is quite variable in literature from 2 [19] to 22 % [18]. Basing on our experience with several different types of plate, we believe that this problem is strictly related to the system of plate fixation. In fact we regularly check intraoperatively the presence of acromial impingement in 90 elevation and external rotation, and a contact between plate and acromion is almost constant in our experience. We therefore think that in active patients, a future surgery for plate removal has often to be foreseen in order to improve the range of motion once the fracture is healed. This point stresses the importance to use a plate guaranteeing low chances of complications like screw-plate cold fusion during removal. In our study, we have recorded an 11.2 % incidence of symptomatic impingement between plate and acromion. In all cases where the plate was removed, there was a symptom benefit with increase in the range of motion. One advantage of the Diphos H plate as evidenced by this multicenter study is its radiolucency, allowing an optimal visualization of the fracture during the surgical procedure and then during follow-up. Another point is represented by the multidirectional self-taping locking screws, allowing the surgeon to choose for each screw the optimal axis of introduction. A third main advantage of this plate is the easiness of removal. In the two cases in which during removal one blocked screw was found, the problem was solved removing all the other screws and cutting the plate around the blocked screw by a simple bone saw, and then, the screw was grasped and unscrewed.
Some limits of this study deserve a mention. •
• • • •
The series of the single institutions are not completely homogeneous relative to age, fracture types, antibiotic prophylaxis and rehabilitation protocols. The fractures were classified only according to the Neer’s system. The structure of the plate was modified as mentioned during the study. A mechanical evaluation of the holding strength between plate and locking screws is missing. The functional evaluation through the Constant score, considering the operated limb only, conditions the results as the study population is not homogeneous in relation to age, and therefore, data are influenced by the lower basic scores of the elderly patients.
Conclusions The experience with the Diphos H plate compared with other locking plate types led to some interesting conclusions. •
•
•
•
The functional results as measured through Constant and DASH scores are comparable, and the incidence of complications is similar (plate failure never happened after modifying the plate thickness). The advantages include plate radiolucency, mainly during reduction maneuvers of complex fractures, especially in the axial radiographic view, multidirectional self-taping locking screws and resolution of the problem of screw-plate cold fusion. The limits include the impossibility to model the plate to adapt it to the bony contour that makes it unfit for fractures extending to the mid-third of the diaphysis and for malunions and nonunions. The debris generation at the insertion of the screws does not represent a biocompatibility risk.
Conflict of interest of interest.
The authors declare that they have no conflict
Human and animal rights This article does not contain any studies with human or animal subjects performed by any of the authors.
References 1. Palvanen M, Kannus P, Niemi S, Parkkari J (2006) Update in the epidemiology of proximal humeral fractures. Clin Orthop Relat Res 442:87–92. doi:10.1097/01.blo.0000194672.79634.78 2. Bell J-E, Leung BC, Spratt KF et al (2011) Trends and variation in incidence, surgical treatment, and repeat surgery of proximal humeral fractures in the elderly. J Bone Joint Surg Am 93:121–131. doi:10.2106/JBJS.I.01505
123
Musculoskelet Surg 3. Obrant KJ, Bengne´r U, Johnell O, Nilsson BE, Sernbo I (1989) Increasing age-adjusted risk of fragility fractures: a sign of increasing osteoporosis in successive generations? Calcif Tissue Int 44:157–167. doi:10.1007/BF02556558 4. Hertel R (2005) Fractures of the proximal humerus in osteoporotic bone. Osteoporos Int. doi:10.1007/s00198-004-1714-2 5. Rose SH, Melton LJ, Morrey BF, Ilstrup DM, Riggs BL (1982) Epidemiologic features of humeral fractures. Clin Orthop Relat Res 168:24–30 6. Resch H (2011) Proximal humeral fractures: current controversies. J Shoulder Elbow Surg 20:827–832. doi:10.1016/j.jse.2011. 01.009 7. Boileau P, Pennington SD, Alami G (2011) Proximal humeral fractures in younger patients: fixation techniques and arthroplasty. J Shoulder Elbow Surg 20:S47–S60. doi:10.1016/j.jse. 2010.12.006 8. Murray IR, Amin AK, White TO, Robinson CM (2011) Proximal humeral fractures: current concepts in classification, treatment and outcomes. J Bone Joint Surg Br 93:1–11. doi:10.1302/0301620X.93B1.25702 9. Iacobellis C, Berizzi A, Biz C, Camporese A (2014) Treatment of proximal humeral fractures with reverse shoulder arthroplasty in elderly patients. Musculoskelet Surg. doi:10.1007/s12306-0140331-2 10. Verdano MA, Lunini E, Pellegrini A, Corsini T, Marenghi P, Ceccarelli F (2014) Can the osteosynthesis with locking plates be a better treatment for unstable fractures of the proximal humerus? Musculoskelet Surg 98:27–33. doi:10.1007/s12306-013-0267-y 11. Edwards SL, Wilson NA, Zhang L-Q, Flores S, Merk BR (2006) Two-part surgical neck fractures of the proximal part of the humerus. A biomechanical evaluation of two fixation techniques. J Bone Joint Surg Am 88:2258–2264. doi:10.2106/JBJS.E.00757 12. Seide K, Triebe J, Faschingbauer M et al (2007) Locked vs. unlocked plate osteosynthesis of the proximal humerus—a biomechanical study. Clin Biomech 22:176–182. doi:10.1016/j. clinbiomech.2006.08.009 13. Siffri PC, Peindl RD, Coley ER, Norton J, Connor PM, Kellam JF (2006) Biomechanical analysis of blade plate versus locking plate fixation for a proximal humerus fracture: comparison using cadaveric and synthetic humeri. J Orthop Trauma 20:547–554. doi:10.1097/01.bot.0000244997.52751.58
123
14. Duralde XA, Leddy LR (2010) The results of ORIF of displaced unstable proximal humeral fractures using a locking plate. J Shoulder Elbow Surg 19:480–488. doi:10.1016/j.jse.2009.08.008 15. Steinberg EL, Rath E, Shlaifer A, Chechik O, Maman E, Salai M (2012) Carbon fiber reinforced PEEK Optima–a composite material biomechanical properties and wear/debris characteristics of CF-PEEK composites for orthopedic trauma implants. J Mech Behav Biomed Mater 17:221–228. doi:10.1016/j.jmbbm.2012.09. 013 16. Feerick EM, Kennedy J, Mullett H, FitzPatrick D, McGarry P (2012) Investigation of metallic and carbon fibre PEEK fracture fixation devices for three-part proximal humeral fractures. Med Eng Phys. doi:10.1016/j.medengphy.2012.07.016 17. Ponnappan RK, Serhan H, Zarda B, Patel R, Albert T, Vaccaro AR (2009) Biomechanical evaluation and comparison of polyetheretherketone rod system to traditional titanium rod fixation. Spine J 9:263–267. doi:10.1016/j.spinee.2008.08.002 18. Thanasas C, Kontakis G, Angoules A, Limb D, Giannoudis P (2009) Treatment of proximal humerus fractures with locking plates: a systematic review. J Shoulder Elbow Surg 18:837–844. doi:10.1016/j.jse.2009.06.004 19. Brunner F, Sommer C, Bahrs C et al (2009) Open reduction and internal fixation of proximal humerus fractures using a proximal humeral locked plate: a prospective multicenter analysis. J Orthop Trauma 23:163–172. doi:10.1097/BOT.0b013e318 1920e5b 20. Ricchetti ET, Warrender WJ, Abboud JA (2010) Use of locking plates in the treatment of proximal humerus fractures. J Shoulder Elbow Surg 19:66–75. doi:10.1016/j.jse.2010.01.001 21. Charalambous CP, Siddique I, Valluripalli K et al (2007) Proximal humeral internal locking system (PHILOS) for the treatment of proximal humeral fractures. Arch Orthop Trauma Surg 127:205–210. doi:10.1007/s00402-006-0256-9 22. Sproul RC, Iyengar JJ, Devcic Z, Feeley BT (2011) A systematic review of locking plate fixation of proximal humerus fractures. Injury 42:408–413. doi:10.1016/j.injury.2010.11.058 23. Clavert P, Adam P, Bevort A, Bonnomet F, Kempf JF (2010) Pitfalls and complications with locking plate for proximal humerus fracture. J Shoulder Elbow Surg 19:489–494. doi:10. 1016/j.jse.2009.09.005
