Analysis of failure with the use of locked plates for stabilization of proximal humerus fractures.
Patients and Methods
After IRB approval, a retrospective review was performed on patients who sustained a proximal humerus fracture for which 78 patients were stabilized using a locked plate between June 2007 and December 2011 at a single level one trauma center. Inclusion criteria for this study included any patient older than 18 years of age with a displaced two-part, three-part, and four-part fracture without associated dislocation. Exclusion criteria included patients younger than 18 years of age and less than 6 months of clinical follow-up. Five fellowship-trained orthopaedic trauma surgeons performed the surgeries. Medical records were reviewed for mechanism of injury, patient demographics, past medical history, smoking and alcohol use, date of surgery, perioperative complications, preoperative and postoperative radiographs, and follow-up evaluation. All surgeries were performed through a deltopectoral approach with a Synthes 3.5 mm proximal humerus locking plate (West Chester, Pennsylvania, USA). All patients followed a similar post surgical rehabilitation protocol that emphasized early range of shoulder motion. Bone graft was not used in the patient portion of the study group.
As part of our standard protocol, pre- and postoperative radiographs after proximal humerus fracture fixation included an anteroposterior view with the arm in neutral rotation (AP), transscapular lateral, and axillary. These radiographs were evaluated to determine fracture type, fracture alignment, union time, and implant complications. Fracture type was based on that described by Neer. (8,9) Intraoperative fluoroscopy was used to assess intraoperative fracture reduction, medial cortical support, humeral head-shaft angle, and plate height. Medial cortical support was defined as either anatomic reduction of the medial cortex or an oblique inferomedial locking screw in the proximal humeral head as described by Gardner and colleagues. (10)
The humeral head-shaft angle was determined on the AP radiograph by drawing a line from the superior to inferior border of the articular surface and followed by a line perpendicular to it. The neck-shaft angle is formed from the intersection of a line bisecting the humeral shaft and this perpendicular line (Fig. 1). (11) A head shaft angle less than 130[degrees] on initial postoperative x-rays was considered a varus malreduction and on final radiographs as a varus malunion. (12,13)
The head height relative to plate placement was assessed for loss of reduction as described by Gardner and coworkers. (10) This measurement was made on the AP x-ray by drawing two lines perpendicular to a line parallel to the humeral shaft--one line at the superior end of the plate and another line at the most proximal portion of the humeral head. The distance between these two lines determined the humeral head height (Fig. 2). Radiographic measurements made were standardized for magnification based on locking screw diameter. Humeral head migration was assessed from initial postoperative humeral head height and that at fracture union. A humeral head height change of 5 mm was considered a loss of reduction.
Fracture union was defined as radiographic findings of bridging bone across the fracture. A healing complication was defined as either: loss of reduction, varus malunion, avascular necrosis, screw penetration into the glenohumeral joint, or plate impingement based on symptoms and radiographic findings. Statistical analysis was performed using SPSS software (Chicago, Illinois) to assess the association between patient injury and surgical factors and healing complications. Statistical significance was set at p < 0.05.
Between June 2007 and December 2011, 78 proximal humerus fractures in 78 patients were treated with open reduction and internal fixation using a locked plate. Twenty-four patients were lost to follow-up (mean age: 54 years old), while 54 patients were available for 6-month minimum follow-up (range: 6 to 60 months) and comprised the study group. There were no differences in demographics or fracture patterns between the two groups of patients. In the study group, there were 29 men and 25 women with an average age of 52 years (range: 18 to 82 years). Mechanism of injury included motor vehicle accident (15 patients), slip and fall (11 patients), pedestrian struck (8 patients), motorcycle accident (7 patients), fall from a standing height (7 patients), gunshot wound (3 patients), seizure (2 patients) and bicycle accident (1 patient). Using the Neer Classification System, there were 17 two-part, 19 three-part, and 18 four-part fractures. The most common comorbidities in the patients were hypertension (25 patients), diabetes (11 patients), dyslipidemia (10 patients), and coronary artery disease (4 patients).
Uneventful fracture union occurred in 34 patients (63%) at an average time of 4.3 months (range: 2 to 8 months). A healing complication occurred in 20 patients (37%) and consisted of varus malunion (16 patients), loss of humeral height greater than 5 mm (16 patients), avascular necrosis (6 patients), and implant penetration (1 patient). Some of the 20 patients had more than one healing complication (Table 1). Of the 16 patients with varus malunion, 10 patients had initial varus malreduction on postoperative x-rays (p < 0.001). Eleven of 54 patients required secondary surgery: revision ORIF (5), implant removal (5), and hemiarthroplasty (1). Revision surgery was a result of impingement pain and functional impairment.
On analysis, the factors associated with a healing complication were fracture type, number of comorbidities, and initial varus malreduction. As the number of fracture parts increased, a higher rate of healing complications occurred, p < 0.029. Of the study group patients, 10% of two-part fractures, 50% of three-part, and 40% of four-part fractures had a healing complication. When combined together, three-part and four-part fractures had a significantly higher rate of healing complications compared to two-part fractures, (90% vs. 10%, p < 0.016). Patients who had one or more and three or more comorbidities had a higher rate of healing complications compared to those without medical conditions, p < 0.016 and p < 0.038, respectively. Patients who had a varus malreduction had a higher rate of healing complications than those not reduced in varus (p < 0.001). The odds ratio for a healing complication based on a Neer Type three- or four-part fracture versus a two-part fracture was 12.7 (1.72 to 94.0), p < 0.013, based on the presence of associated comorbidities versus none was 38.9 (3.63 to 417.9), p < 0.003, and for a varus malreduction was >10,000 (0 to 10,000), p = 0.99. No significant association was found between patient age, gender, tobacco use, alcohol use, or the presence or lack of medial cortex support and healing complication.
In this study, we found an overall complication rate of 37% in patients stabilized using a locked plate after sustaining a proximal humerus fracture. Revision surgery was required in 20% of patients. On initial postoperative radiographs, 10 patients had a varus malreduction, which all went on to varus malunion. Varus malunion is a representation of loss of initial fixation. Factors associated with a healing complication included Neer three- and four-part fracture types, increasing number of comorbidities, and initial varus malreduction. No association was found between patient age, gender, tobacco use, alcohol use or lack of medial cortex support and healing complication.
Complication rates after locked plating of the proximal humerus range in the literature from 9.7% to 39% and revision surgery rates have been reported as high as 29%. (14-19) Clavert described complications based on two factors: 1. technical errors due to plate positioning, screw length, and screw cut-out; and 2. the locked plate technology itself. (20) Sudkamp and associates (4) reported a complication rate of 34% (52 patients) in 155 patients who sustained a proximal humerus fracture stabilized using a locked plate and were available for 1-year follow-up. Similar to our results, these investigators reported an association between poor surgical technique and a healing complication; however, the most common complication in their series was intraoperative screw joint penetration, while in our study it was loss of reduction and varus malunion. Their reported revision surgery rate of 19% is similar to the 20% reported in our series.
Jost and colleagues (21) reported a series of 121 patients who sustained a proximal humerus fracture stabilized using a locked plate; these investigators reported that increased number of fracture parts was associated with higher complication rates. This was particularly evident in the patients who had Neer type three- and four-part fractures. A study by Ye and coworkers (22) reported a complication rate of 20.2% in a series of 89 patients over the age of 50 years who sustained a Neer type three- or four-part proximal humerus fracture stabilized using a locked plate. Subacrominal impingement was the most prevalent complication; only two patients healed with a malunion. Similar to our results, these investigators found an association between increased number of fracture parts and higher complication rates.
We found that a technical error, such as malreduction following locked plating, was associated with a subsequent higher rate of varus malunion and an increased overall complication rate (p < 0.001). Agudelo also reported a correlation between malreduction and loss of fixation. (23) These results highlight the importance of sound surgical technique even when using a locked plate to stabilize a proximal humerus fracture and concur with those results reported by Bell and associates. (24) In a Medicare data analysis, they reported that the significant increase in the rate of surgical treatment after proximal humerus since the introduction of locked plates was associated with significantly higher rates of revision surgery.
The current literature has a vast range of percentages for the common complications associated with locked plating after proximal humerus fracture. Cornell and colleagues (25) in a meta-analysis reported fracture displacement due to loss of fixation ranges between 0% to 14.8% while intra-articular screw penetration ranges between 11% to 15% of cases. We had one intra-articular screw, which was removed, and the patient went on to heal.
The current literature does not describe an association with medical comorbidities and increased rate of complications. This is a unique finding to our study, which showed a direct relationship with the number of comorbidities. This association may reflect an increased amount of osteoporosis in this sicker population as well as more patients with malunions, avascular necrosis, and other negative events. Therefore, patients with multiple comorbidities may be treated better conservatively, as they are at higher risks for complications.
Postoperative x-rays of the 54 patients were reviewed for medial cortical support. Forty-eight fractures (89%) had medial cortical support on initial postoperative radiographs. No association was found between lack of medial support and healing complication. These results are different from those reported by Gardner and colleagues. The investigators found a significant relationship between medial cortical support and loss of reduction. (10) Patients who had medial cortical support on initial postoperative radiographs had an average loss of humeral height of 1.2 mm verse 5.8 mm in the group without medial cortical support (p < 0.001).
Limitations of our study include the retrospective chart review design and a high number of patients lost to follow-up. Twenty-four of our original 78 patients (31%) were lost to follow-up. Furthermore, if the sample size was larger, there may have been more findings with statistical significance. The mean age of patients in this study is younger compared to the more senior age in most literature, which could possibly make population comparison difficult.
We found an overall complication rate of 37% in patients stabilized using a locked plate after sustaining a proximal humerus fracture. Revision surgery was required in 20% of the total patients. Factors associated with a healing complication included Neer three- and four-part fracture type, increasing number of comorbidities, and initial varus malreduction. Based on these results, patient selection for locked plating after proximal humerus fracture should incorporate many factors with meticulous attention to surgical technique.
None of the authors have a financial or proprietary interest in the subject matter or materials discussed, including, but not limited to, employment, consultancies, stock ownership, honoraria, and paid expert testimony.
Michael P Silverstein, M.D., Resident, Department of Orthopaedic Surgery, The Cleveland Clinic, Cleveland, Ohio. Kwadwo Yirenkyi, M.D., Spine Fellow, SUNY Upstate, Syracuse, New York. George Haidukewych, M.D., Chairman and Program Director, Department of Orthopaedic Surgery, Orlando Health, Orlando, Florida. Kenneth J. Koval, M.D., Director of Orthopaedic Research, Department of Orthopaedic Surgery, Orlando Health, Orlando, Florida.
Correspondence: Michael P Silverstein, M.D., 12911 Cedar Road, Cleveland Heights, Ohio 44118; email@example.com.
(1.) Neuhaus V, Swellengrebel CH, Bossen JK, et al. What are the factors influencing outcome among patients admitted to a hospital with a proximal humeral fracture? Clin Orthop Relat Res. 2013 May; 471(5):1698-706.
(2.) Rothberg D, Higgins T. Fractures of the proximal humerus. Orthop Clin North Am. 2013 Jan; 44(1):9-19.
(3.) Nho SJ, Brophy RH, Barker JU, et al. Management of proximal humeral fractures based on current literature. J Bone Joint Surg Am. 2007 Oct; 89 Suppl 3:44-58.
(4.) Sudkamp N, Bayer J, Hepp P, et al. Open reduction and internal fixation of proximal humeral fractures with use of the locking proximal humerus plate. Results of a prospective, multicenter, observational study. J Bone Joint Surg Am. 2009 Jun; 91(6):1320-8.
(5.) Khmelnitskaya E, Lamont LE, Taylor SA, et al. Evaluation and management of proximal humerus fractures. Adv Orthop. 2012 Dec; 2012:861598.
(6.) Hirzinger C, Tauber M, Resch H. [Proximal humerus fracture: new aspects in epidemiology, fracture morphology, and diagnostics]. Unfallchirurg. 2011 Dec; 114(12):1051-8.
(7.) Solberg BD, Moon CN, Franco DP, et al. Locked plating of 3- and 4-part proximal humerus fractures in older patients: the effect of initial fracture pattern on outcome. J Orthop Trauma. 2009 Feb; 23(2):113-9.
(8.) Neer CS 2nd. Displaced proximal humeral fractures. I. Classification and evaluation. J Bone Joint Surg Am. 1970 Sep; 52(6):1077-89.
(9.) Marsh JL, Slongo TF, Agel J, et al. Fracture and dislocation classification compendium--2007: Orthopaedic Trauma Association classification, database and outcomes committee. J Orthop Trauma. 2007 Nov-Dec; 21(10 Suppl):S1-133.
(10.) Gardner MJ, Weil Y, Barker JU, et al. The importance of medial support in locked plating of proximal humerus fractures. J Orthop Trauma. 2007 Mar; 21(3):185-91.
(11.) Paavolainen P, Bjorkenheim JM, Slatis P, et al. Operative treatment of severe proximal humeral fractures. Acta Orthop Scand, 1983 Jun; 54(3):374-9.
(12.) Robinson CM, Wylie JR, Ray AG, et al. Proximal humeral fractures with a severe varus deformity treated by fixation with a locking plate, J Bone Joint Surg Br, 2010 May; 92(5):672-8.
(13.) Ockert B, Braunstein V, Kirchhoff C, et al. Monoaxial versus polyaxial screw insertion in angular stable plate fixation of proximal humeral fractures: radiographic analysis of a prospective randomized study, J Trauma, 2010 Dec; 69(6):154-551.
(14.) Aksu N, Gogus A, Kara AN, et al. Complications encountered in proximal humerus fractures treated with locking plate fixation. Acta Orthop Traumatol Turc. 2010 Mar; 44(2):89-96.
(15.) Burke NG, Kennedy J, Green C, et al. Locking plate fixation for proximal humerus fractures. Orthopedics. 2012 Feb; 35(2):e250-4.
(16.) Lu Y, Wang MY, Zhu YM, et al, Complications of the locking plate for displaced proximal humeral fractures, Chin Med J (Engl). 2010 Oct; 123(19):2671-5.
(17.) Kumar C, Gupta AK, Nath R, Ahmad J. Open reduction and locking plate fixation of displaced proximal humerus fractures. Indian J Orthop. 2013 Mar; 47(2):156-60.
(18.) Schliemann B, Siemoneit J, Theisen C, et al, Complex fractures of the proximal humerus in the elderly--outcome and complications after locking plate fixation. Musculoskelet Surg, 2012 May; 96 Suppl 1:S3-11.
(19.) Konrad G, Bayer J, Hepp P, et al. Open reduction and internal fixation of proximal humeral fractures with use of the locking proximal humerus plate, Surgical technique, J Bone Joint Surg Am, 2010 Mar; 92 Suppl 1 Pt 1:85-95.
(20.) Clavert P, Adam P, Bevort A, et al, Pitfalls and complications with locking plate for proximal humerus fracture, J Shoulder Elbow Surg, 2010 Jun; 19(4):489-94.
(21.) Jost B, Spross C, Grehn H, Gerber C. Locking plate fixation of fractures of the proximal humerus: analysis of complications, revision strategies and outcome, J Shoulder Elbow Surg, 2013 Apr; 22(4):542-9.
(22.) Ye T, Wang L, Zhuang C, et al. Functional outcomes following locking plate fixation of complex proximal humeral fractures. Orthopedics. 2013 Jun; 36(6):e715-22.
(23.) Agudelo J, Schurmann M, Stahel P, et al. Analysis of efficacy and failure in proximal humerus fractures treated with locking plates. J Orthop Trauma. 2007 Nov-Dec; 21(10):676-81.
(24.) Bell JE, Leung BC, Spratt KF, et al. Trends and variation in incidence, surgical treatment, and repeat surgery of proximal humeral fractures in the elderly, J Bone Joint Surg Am, 2011; 93(2):121-31.
(25.) Cornell CN, Ayalon O. Evidence for success with locking plates for fragility fractures, HSS J, 2011 Jul; 7(2):164-9.
Caption: Figure 1 Humeral neck-shaft angle. The humeral neck-shaft angle was determined on AP radiograph by drawing a line from superior to inferior border of the articular surface, followed by a line perpendicular to it. The neck-shaft angle is formed from the intersection of a line bisecting the humeral shaft and the perpendicular line. To view this image in color see www.nyuhjdbulletin.org.
Caption: Figure 2 Humeral head height. A horizontal line was drawn across the superior portion of the humeral head and across the superior portion of the locking plate. Between the horizontal lines, the head height was determined. An acceptable length of head height was determined as a standardized baseline value less than or equal to 10 mm. To view this image in color see www.nyuhjdbulletin.org.
Table 1 Complications following Locked Plating Complication Amount Varus Malunion 16 Loss of Humeral Height 16 Avascular Necrosis 6 Implant Penetration 1
Please note: Illustration(s) are not available due to copyright restrictions.
|Printer friendly Cite/link Email Feedback|
|Author:||Silverstein, Michael P.; Yirenkyi, Kwadwo; Haidukewych, George; Koval, Kenneth J.|
|Publication:||Bulletin of the NYU Hospital for Joint Diseases|
|Date:||Jul 1, 2015|
|Previous Article:||A physician checklist to record patient levels of inflammation, damage and distress as quantitative data rather than as narrative impressions.|
|Next Article:||Limb sparing resection for tumors involving the distal humerus and reconstruction with a modular endoprosthesis.|