Pediatric Medial Epicondyle Fractures: Are We There Yet?
The medial humeral epicondyle apophysis begins to ossify between 4 and 7 years of age, ultimately fusing at 15 to 20 years--the elbow is the final ossification center to do so. (1,2) The medial epicondyle serves as the attachment point for the flexor-pronator mass of the forearm and the medial collateral ligament (MCL) of the elbow. The posterior oblique ligament of the MCL forms the floor of the cubital tunnel, while the anterior oblique ligament of the MCL serves as the primary static restraint to valgus stress. A cadaver study also demonstrated that, in the absence of MCL contribution, the flexor carpi ulnaris and flexor digitorum superficialis are additional dynamic contributors to valgus stability of the elbow. (3) In combination, the MCL and flexor-pronator muscle mass provide 290 N of tensile static and dynamic stabilizing force. (4) The ulnar nerve courses directly posterior to the medial epicondyle. Therefore, trauma to the epicondyle has the potential to affect the ulnar nerve.
Medial epicondyle fractures most commonly result from a valgus moment to the elbow leading to an apophyseal avulsion. It can be seen in overhead, upper-extremity athletes or from a fall onto an outstretched hand with the elbow extended. (5) This injury represents up to 12% of all pediatric elbow fractures. (6) They are most common in patients 7 to 15 years of age due to the incomplete ossification of the apophysis in that age range. (2,7) Fractures of the medial epicondyle occur approximately twice as frequently in males than in females. (8) Presenting complaints typically include medial pain and tenderness of the elbow, and patients may also report feeling or hearing a "pop" at the moment of injury. Medial epicondyle fractures are also associated with valgus elbow instability, ulnar nerve symptoms (10% of patients), and elbow dislocation (50% of patients). (5,8,9)
Initial Radiographic Evaluation
The typical workup consists of anteroposterior (AP) and lateral x-rays of the injured joint. Internal oblique views may also be helpful. Stress radiographs can be used to evaluate elbow stability but can be limited in practicality in an acutely injured child. (10) Traditionally, these x-rays are used to assess fracture displacement, check for an incarcerated intra-articular fracture fragment, and rule out an elbow dislocation or other associated fractures. An anatomic descriptive study of 171 plain films of normal distal humeri of children aged 4 to 15 years demonstrated that the average location of the center of the intact medial epicondyle on AP radiographs is 0.5 mm below the olecranon fossa line and 1.2 mm anterior to the posterior humeral line in lateral radiographs. There was minimal variation with skeletal maturation. (11) However, when dealing with fractured and displaced medial epicondyles, the reliability of the two-dimensional AP and lateral views in the assessment of this three-dimensional phenomenon has been shown to be of limited use. (12,13) Advanced imaging can be considered in such cases, but may risk radiation exposure or require anesthesia in the youngest patients, in addition to cost concerns.
Measuring Fracture Displacement
Although treatment decisions are often based on the magnitude of fracture displacement, multiple studies have demonstrated the inaccuracy and inconsistency of measuring displacement on traditional AP, lateral, and oblique radiographs. (12,13) In a study by Pappas et al., (12) five reviewers were asked to measure fracture displacement on 38 AP, lateral, and oblique radiographs of medial epicondyle fractures. Their intra- and inter-observer agreement was then measured. The reviewers disagreed with their own measurements 26% of the time and disagreed with each other on AP measurements 54% of the time, lateral measurements 87% of the time, and oblique measurements 64% of the time. (12) One potential explanation for this high rate of both intra- and inter-observer disagreement is that without a standardized system, one can theoretically measure the distance between any two points on the fragment and its donor site, thus producing a variety of values. (10) As previously noted, internal oblique radiographs may provide additional information in the evaluation of medial epicondyle fractures. Analysis of 45[degrees] and 60[degrees] internal oblique views in cadavers with medial epicondyle fractures established that, while intra- and inter-observer reliability in these views was excellent, the accuracies of 60% and 35%, respectively, were not much better than those of AP and lateral films. Additionally, accuracy and reliability changed according to the magnitude of fracture displacement and obliquity of the x-ray projection, and none of the views were reliable for purely anterior displacement. (9,13)
Edmonds published a series of 11 fractures originally deemed to be minimally to non-displaced on x-ray that then underwent computed tomography (CT) scanning with three-dimensional reconstructions. He found that anterior displacement was underestimated by an average of 7.9 mm on lateral x-rays (6 out of 11 patients had more than 1 cm anterior displacement) and medial displacement was overestimated on the AP view by an average of 3.6 mm. (13) Considering that a recent discrete choice experiment indicated that fracture displacement is one of the only attributes significantly influencing surgeons to perform operative interventions, the demonstrated lack of accuracy and consistency is even more disconcerting. (14)
In an attempt to accurately measure fracture displacement without relying on advanced imaging, Souder et al. (15) developed the distal humerus axial x-ray. For this novel view, the child rests the elbow on the table at nearly 90[degrees] of flexion with the x-ray tube positioned 25[degrees] from the long axis of the humerus, centered on the distal humerus. A cadaveric analysis of this technique demonstrated only 0.8 mm error in displacements that were greater than 10 mm and 1.5 mm error in displacements less than 10 mm, with excellent intra-class correlation coefficient (0.974) for the axial view compared to 0.257 for the AP and oblique views. (15) A more recent cadaveric study comparing the standard AP view to the distal humerus axial view showed that the AP radiograph significantly underestimated the displacement of the 5 mm (-1.6 mm), 10 mm (-4.5 mm), and 15 mm (-7.1 mm) fractures at 90[degrees] elbow flexion relative to the more accurate axial view. (16)
To Operate or Not to Operate?
Although novel radiographic views or advanced imaging may improve the ability to accurately measure displacement, how to best use this information is still unclear. Few, if any, absolute operative indications exist for medial epicondyle fractures. Relative indications for non-operative treatment include minimal displacement with a stable elbow. Surgery may be considered for an incarcerated fracture fragment, associated elbow instability, ulnar nerve symptoms, and significant fracture displacement. As one might presume, it is this last ambiguous qualifier that begs the question of what the threshold for operative treatment should be. Thresholds reported in the literature include 2 mm, 5 mm, 9 mm, or an undefined "significant amount" of displacement. (17-19) Furthermore, most of these studies measured displacement on traditional radiographic views, which have been proven to be inaccurate. Thus, the role of fracture displacement in clinical decision making must be scrutinized.
A number of available level III and IV studies suggest similar clinical outcomes between surgical and non-surgical patients. A commonly cited retrospective study of 42 patients with displacement greater than 5 mm and mean follow-up of 34 years found no difference in long-term outcomes between those treated with a cast and those undergoing open reduction and internal fixation (ORIF). Poor outcomes were found after fragment excision. (20) Of the 19 patients treated with a long arm cast without reduction, 16 reported good and three fair outcomes. Of the 17 who underwent ORIF, 15 reported good and two reported fair outcomes. Six patients underwent fragment excision, and four reported poor and two fair outcomes. A more critical analysis of this study suggests that while the magnitude of follow-up was impressive, the cohort was likely quite underpowered to detect meaningful differences in outcomes. Additionally, the classification of outcomes into good, fair, and poor were somewhat arbitrary--one patient reported full range of motion and a good outcome despite an abnormal carrying angle. Furthermore, the surgical patients in this study underwent fixation with T-nails or pins since they were treated before widespread availability of more rigid implants like screws. Therefore, it is difficult to arrive at a firm conclusion from these results. Another long-term retrospective study (21) examined 56 children treated nonoperatively at a mean of 35 years later and found that 31 had pseudarthroses seemingly independent of the initially measured displacement. While the clinical relevance of this result may be questionable, 12 of the 51 subjects (21%) had residual symptoms independent of bony or fibrous healing. These symptoms included activity related discomfort, subjective weakness, tenderness, hand numbness, and subjective instability. Furthermore, 11 patients reportedly had a "small loss" of elbow extension. All three patients reporting ulnar symptoms were in the pseudarthroses group. (21) While the authors conclude that outcomes were good in all cases, one can argue that one in every five patients had a suboptimal outcome due to residual symptoms decades later.
A systematic review of 14 studies (498 patients) sought to compare the outcomes of operative versus nonoperative management of medial epicondyle fractures and discovered that surgery leads to 9.33 times higher odds of radiographic union than non-surgical treatment, but that this did not translate into any difference in pain or ulnar nerve symptoms at final follow-up. (8) However, the analysis was limited by the level III and IV studies it reviewed. Furthermore, functional outcomes, range of motion, and deformity were difficult to evaluate due to differences in reporting throughout the literature. Most recently, a retrospective review of 22 non-operative subjects matched to 22 operative subjects at moderate displacements (mean 9.7 mm in both groups) showed no statistically significant differences in average length of immobilization, median time to full range of motion, proportion of patients needing physical therapy, or complications. However, the study was likely underpowered to detect relevant clinical differences. (22)
While the available literature suggests that outcomes appear to be similar after operative or non-operative treatment in the general population, one may hypothesize that there may be differences in young athletes, especially overhead athletes that place high demands on the elbow. A study of 20 adolescent athletes with at least 2 years of follow-up found no difference in outcomes, with patients reporting similarly excellent Disabilities of the Arm, Shoulder, and Hand (DASH) scores, satisfaction rate, and rate of return to sport regardless of the type of intervention. These results also held true for overhead athletes specifically. Because treatment decisions were based on the energy level of the mechanism of injury and resulted in excellent outcomes, the study suggested consideration of a surgical approach for high energy injuries and non-operative care for lower energy injuries. (23) A more recent retrospective review matched 28 non-operative to 14 operative pediatric upper-extremity athletes being treated for medial epicondyle fractures and found no statistically significant difference in proportions of subjects returning to sport, duration of time required to return to sport, pain, range of motion, need for physical therapy, or complications. (7) However, like several other retrospective studies on the subject, these analyses were likely underpowered to detect significant differences.
A common theme in most of the aforementioned studies is an emphasis on fracture displacement in clinical decision making and statistical analysis. However, since radiographic measurement is fraught with inaccuracy and may not have a dominant effect on outcomes, other factors must be considered. One such consideration is elbow stability. A systematic review of seven studies (81 patients), sought to compare the outcomes of non-operative treatment between 42 subjects with a fracture-dislocation and 39 with an isolated fracture. The authors found a significantly higher rate of stiffness among the fracture-dislocations (43%) than fractures alone (15%) despite varying follow-up times and measures of stiffness. (5) While gross instability at initial presentation may be a useful guideline for treatment, prevention of long-term laxity is another consideration. Though the literature is unclear about who is truly at risk for this, it may be more common with non-operative patients, as seen in several reports and small series. (24-27) The most striking might be a retrospective review of 20 patients with associated elbow dislocations who report "definite" laxity in flexion in 45% and in extension in 35% of their patients. (27) Further research may help identify which patients are at risk for these long-term complications and surgical intervention could be beneficial for them.
While one can contend that elbow laxity may be a more important consideration than fracture displacement, the anatomy of the medial epicondyle suggests that specific patterns of displacement may in fact result in laxity. If the epicondyle is significantly displaced anteriorly, the anterior oblique ligament of the MCL may lose its tension and the posterior oblique ligament of the MCL may tighten. This can potentially lead to valgus instability in flexion and a block to full extension. In fact, a computer simulation model demonstrated that for every 1 mm of anterior displacement of a medial epicondyle fracture, approximately 2% of wrist flexion strength is lost. In other words, 1 cm of purely anterior displacement would result in greater than 20% loss of wrist flexion strength. (28) In vivo, one might expect additional loss of strength as the shortened flexor-pronator muscle mass begins to atrophy due to immobilization and disuse. Therefore, while elbow stability and energy of injury are likely important factors to consider, the magnitude of anterior displacement may also play a role in determining outcomes.
Regardless of the many controversies surrounding medial epicondyle fractures, a decision must ultimately be made with the patient and family regarding the treatment plan. First, the goals of treatment must be established, understanding that these may vary depending on the specific patient. Generally, these include achievement of a stable, painless elbow with acceptable motion and strength. As noted, "acceptable" motion and strength may vary from patient to patient. Osseous union intuitively seems preferable, but fibrous union may be adequate for some patients as demonstrated in the available literature.
Next, a review of advantages and disadvantages of surgical versus non-surgical treatment should be discussed. Based on the literature, the potential benefits of operative treatment are bony fracture union, likely greater valgus stability, and good clinical outcomes even for high-energy injuries, athletes, and those with ulnar nerve symptoms at presentation. A shorter period of postoperative immobilization may be feasible, resulting in earlier elbow motion. On the other hand, the child must be subjected to general anesthesia and, though relatively low, there is a risk of infection and iatrogenic neurologic injury. A recent article reported two cases of iatrogenic radial nerve injury due to guide wire placement during cannulated screw insertion. (29) Surgical treatment also may incur higher cost.
The benefits of non-operative treatment are that it avoids anesthesia and surgical complications while potentially yielding good clinical outcomes. However, these outcomes have yet to be evaluated in a rigorous way, especially in comparison with modern surgical implants and in patients that place high demands on their elbow. Cast treatment may also cost less than surgery. Potential disadvantages of non-operative treatment include somewhat longer immobilization, higher rates of stiffness for some patients, and the possibility of laxity or weakness. Bony union is less likely than with surgical management, but this may not be clinically significant for many patients.
While much of the previous literature focuses on the similarly positive outcomes of operative and non-operative treatment, Mehlman and Howard (30) conducted a systematic review to evaluate the risk of harm with either intervention. The authors found that complications were generally higher in non-operative patients and that the "number needed to harm" was quite low for some of these issues. Specifically, 80 patients would need to be treated non-surgically for one to develop valgus instability, 4 for one to develop stiffness, and 1.6 for one to result in non-union. These complications were exceedingly rare in operative patients. (30) The authors note that any systematic review is only as strong as the studies included and that all of the included literature was retrospective. A discussion of these advantages and disadvantages with the patient and family is of paramount importance. A shared decision can then be made regarding surgical or non-surgical treatment.
As demonstrated in this review, numerous controversies persist in the evaluation and treatment of medial epicondyle fractures in children. Traditional radiographic views are inaccurate in measuring fracture displacement. Furthermore, the role of displacement in guiding treatment is unclear. Though the available evidence suggests similar outcomes for operative and non-operative management, all the literature is retrospective in nature and thereby carries significant potential for bias and power issues. In the face of uncertainty, clinicians must review the goals of treatment and discuss the advantages and disadvantages of various options with the patient and family. Prospective studies on this topic are underway and will hopefully provide stronger evidence for clinical decision making in the future.
None of the authors have a financial or proprietary interest in the subject matter or materials discussed in the manuscript, including, but not limited to, employment, consultancies, stock ownership, honoraria, and paid expert testimony.
(1.) Stevenson RA, Perry DC. Paediatric medial epicondyle fractures of the distal humerus. Orthopedics and Trauma. 2018;32(5):360.
(2.) Beck JJ, Bowen RE, Silva M. What's New in Pediatric Medial Epicondyle Fractures? J Pediatr Orthop. 2018;38(4):e202-e206.
(3.) Park MC, Ahmad CS. Dynamic contributions of the flexorpronator mass to elbow valgus stability. J Bone Joint Surg Am. 2004;86(10):2268-74.
(4.) Werner SL, Fleisig GS, Dillman CJ, Andrews JR. Biomechanics of the elbow during baseball pitching. J Orthop Sports Phys Ther. 1993;17(6):274-8.
(5.) Knapik DM, Fausett CL, Gilmore A, Liu RW. Outcomes of Nonoperative Pediatric Medial Humeral Epicondyle Fractures With and Without Associated Elbow Dislocation. J Pediatr Orthop. 2017;37(4):e224-e228.
(6.) Wilkins KE. Fractures Involving the Medial Epicondylar Apophysis (3rd ed). Philadelphia: J.B. Lippincott, 1991.
(7.) Axibal DP, Carry P, Skelton A, Mayer SW. No Difference in Return to Sport and Other Outcomes Between Operative and Nonoperative Treatment of Medial Epicondyle Fractures in Pediatric Upper-Extremity Athletes. Clin J Sport Med. 2018 Oct 1.
(8.) Kamath A, Baldwin K, Horneff J, Hosalkar H. Operative versus non-operative management of pediatric medial epicondyle fractures: a systematic review. J Child Orthop. 2009;3(5):345-57.
(9.) Gottschalk HP, Bastrom TP, Edmonds EW. Reliability of internal oblique elbow radiographs for measuring displacement of medial epicondyle humerus fractures: a cadaveric study. J Pediatr Orthop. 2013;33(1):26-31.
(10.) Patel NM, Ganley TJ. Medial epicondyle fractures of the humerus: how to evaluate and when to operate. J Pediatr Orthop. 2012;32 Suppl 1:S10-3.
(11.) Klatt JB, Aoki SK. The location of the medial humeral epicondyle in children: position based on common radiographic landmarks. J Pediatr Orthop. 2012;32(5):477-82.
(12.) Pappas N, Lawrence JT, Donegan D, et al. Intraobserver and interobserver agreement in the measurement of displaced humeral medial epicondyle fractures in children. J Bone Joint Surg Am. 2010;92(2):322-7.
(13.) Edmonds EW. How displaced are "nondisplaced" fractures of the medial humeral epicondyle in children? Results of a three-dimensional computed tomography analysis. J Bone Joint Surgery Am. 2010;92(17):2785-91.
(14.) Hughes M, Dua K, O'Hara NN, et al. Variation Among Pediatric Orthopedic Surgeons When Treating Medial Epicondyle Fractures. J Pediatr Orthop. 2017 Oct 18.
(15.) Souder CD, Farnsworth CL, McNeil NP, et al. The Distal Humerus Axial View: Assessment of Displacement in Medial Epicondyle Fractures. J Pediatr Orthop. 2015;35(5):449-54.
(16.) Cao J, Smetana BS, Carry P, et al. A Pediatric Medial Epicondyle Fracture Cadaveric Study Comparing Standard AP Radiographic View With the Distal Humerus Axial View. J Pediatr Orthop. 2019;39(3):e205-e209.
(17.) Hines RF, Herndon WA, Evans JP. Operative treatment of medial epicondyle fractures in children. Clin Orthop Relat Res. 1987;(223):170-4.
(18.) Lee HH, Shen HC, Chang JH, et al. Operative treatment of displaced medial epicondyle fractures in children and adolescents. J Shoulder Elbow Surg. 2005;14(2):178-85.
(19.) Kobayashi Y, Oka Y, Ikeda M, Munesada S. Avulsion fracture of the medial and lateral epicondyles of the humerus. J Shoulder Elbow Surg. 2000;9(1):59-64.
(20.) Farsetti P, Potenza V, Caterini R, Ippolito E. Long-term results of treatment of fractures of the medial humeral epicondyle in children. J Bone Joint Surg Am. 2001;83A(9):1299-305.
(21.) Josefsson PO, Danielsson LG. Epicondylar elbow fracture in children: 35-year follow-up of 56 unreduced cases. Acta Orthop Scand. 1986;57(4):313-5.
(22.) Axibal DP, Ketterman B, Skelton A, et al. No difference in outcomes in a matched cohort of operative versus nonoperatively treated displaced medial epicondyle fractures. J Pediatr Orthop B. 2018 Dec 11.
(23.) Lawrence JTR, Patel NM, Macknin J, et al. Return to Competitive Sports After Medial Epicondyle Fractures in Adolescent Athletes: Results of Operative and Nonoperative Treatment. Am J Sports Med. 2013;41(5):1152-7.
(24.) Gilchrist AD, McKee MD. Valgus instability of the elbow due to medial epicondyle nonunion: Treatment by fragment excision and ligament repair--a report of 5 cases. J Shoulder Elbow Surg. 2002;11(5):493-7.
(25.) Woods GW, Tullos HS. Elbow instability and medial epicondyle fractures. Am J Sports Med. 1977;5(1):23-30.
(26.) Smith JT, McFeely ED, Bae DS, et al. Operative fixation of medial humeral epicondyle fracture nonunion in children. J Pediatr Orthop. 2010;30(7):644-8.
(27.) Dias JJ, Johnson GV, Hoskinson J, Sulaiman K. Management of severely displaced medial epicondyle fractures. J Orthop Trauma. 1987;1(1):59-62.
(28.) Edmonds EW, Santago AC, Saul KR. Functional Loss With Displacement of Medial Epicondyle Humerus Fractures: A Computer Simulation Study. J Pediatr Orthop. 2015;35(7):666-71.
(29.) Marcu DM, Balts J, McCarthy JJ, et al. Iatrogenic radial nerve injury with cannulated fixation of medial epicondyle fractures in the pediatric humerus: a report of 2 cases. J Pediatr Orthop. 2011;31(2):e13-6.
(30.) Mehlman CT, Howard AW. Medial epicondyle fractures in children: clinical decision making in the face of uncertainty. J Pediatr Orthop. 2012;32(2):S135-42.
Konstantin Brnjos, BS, and Neeraj M. Patel, MD, MPH, MBS
Konstantin Brnjos, BS, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA. Neeraj M. Patel, MD, MPH, MBS, Northwestern University Feinberg School of Medicine, and Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, Illinois, USA.
Correspondence: Neeraj M. Patel, MD, MPH, MBS, Ann & Robert H. Lurie Children's Hospital of Chicago, 225 East Chicago Avenue, Box 69, Chicago, Illinois 60611, USA; firstname.lastname@example.org.
Please Note: Illustration(s) are not available due to copyright restrictions.
|Printer friendly Cite/link Email Feedback|
|Author:||Brnjos, Konstantin; Patel, Neeraj M.|
|Publication:||Bulletin of the NYU Hospital for Joint Diseases|
|Date:||Jan 1, 2019|
|Previous Article:||The Anterolateral Ligament (ALL): The New Ligament?|
|Next Article:||Endoscopic Spine Surgery: Past, Present, and Future.|