Radial neck fractures

5% to 10% of all elbow fractures in children

8 to 11 years old (range 2-16yrs)

Controversy remains regarding:

acceptability of displacement
method of treatment
classification of initial displacement
acceptable reduction
radiologic and functional outcome.

Mechanism of injury

Fall on an outstretched hand with the elbow extended and the forearm supinated,. The resulting valgus force compresses the capitellum against the radial head. Radial neck fracture may occur in association with a posterior dislocation of the elbow.

The majority of these fractures involve the proximal radial physis, usually producing a Salter-Harris Type II injury. Salter-Harris Type I fractures are also common, as are fractures a few millimetres distal to the physis, which are entirely metaphyseal.

Associated injuries

Olecranon fracture
Proximal ulna fracture
Medial epicondyle fracture
Lateral epicondyle fracture
Rupture of the medial collateral ligament and concomitant dislocation of the elbow.

Diagnosis

History of fall on outstretched hand. On examination, there is usually tenderness to palpation over the radial head, exacerbated by supination and pronation. Crepitus, swelling, and bruising may or may not be present, depending on the severity of the fracture.

Radiographs

Anteroposterior, lateral, and oblique views. To avoid the off profile physis, which may mimic a fracture, radiographs should be obtained with the elbow in full extension.

There is a normal angulation to the radial neck, which also can be mistaken for a fracture, and comparison radiographs of the uninvolved extremity may be useful. The degree of forearm rotation can influence the measurement of angulation and translation, thus rotating the forearm under fluoroscopy to determine displacement accurately is recommended.

Determining displacement

The amount of displacement needs to be assessed to determine treatment.

Several methods have been described:

Angulation

The angle between "a perpendicular line drawn through the epiphyseal plate with another line drawn along the midline of the radial metaphysis"
The angle between the superior articular surface of the radial head and radial shaft
The angle between a line perpendicular to the articular surface of the radial head with a line drawn through the centre of the radial shaft.

Suggest measure the angle between a line perpendicular to the articular surface of the radial head with a line down the shaft of the proximal radius. The line perpendicular to the articular surface is chosen rather than to the physis because these fractures often involve the physis and may render its use as an anatomic landmark more difficult.

Translation

percentage, where the width of the uncovered radial metaphysis is divided by the total width of the metaphysis (multiplied by 100).
absolute measurement, as the distance from the center of the radial head to the center of the long axis of the proximal radial shaft in millimeters.

Suggest measuring translation as a percentage because these fractures occur in a wide age range in which absolute measurements may not be comparable in radii of significantly different widths.

Angulation and translation seem to be correlated; most treatment algorithms are based on angulation alone.

Elbow seems to tolerate translation less than angulation.

Classification

Most authors classify based on the amount of initial angulation.

mild (0°-30°)
moderate(30°-60°)
severe (>60°)

Some authors include the combination of angulation and/or translation in their classification system.

Steinberg et al and Rodríguez-Merchán

mild (10°-29°, <30% translation)
moderate (30°-59°,<50% translation)
severe (60°-90°, >50% translation)

Steele and Graham

Grade 1 (0°-30°, 0%-10% translation)
Grade 2 (31°-60°, 11%-50% translation)
Grade 3(61°-90°, 51%-90% translation)
Grade 4 (>90°, >90% translation)

Metaizeau et al and Judet et al

Grade 1 (0°, with translation)
Grade 2 (<30°)
Grade 3 (30°-60°)
Grade 4a (60°-80°)
Grade 4b (80°-90°).

Controversy exists as to how much initial displacement is acceptable when deciding operative and non operative management. As such no consensus exists on the ideal classification system.

Because most authors classify initial displacement in 30° increments of angulation, perhaps that system should be adopted as the standard.

Treatment

Their is a wide variation in what is considered acceptable initial displacement, requiring no treatment before immobilization.

Acceptable degrees of angulation range from 0° to 60°, with most authors opting for less than 30°.

Age at presentation and remodeling potential are important considerations.

Most authors recommend an attempt at closed reduction for fractures that are angulated more than 30(range 20° -45°)

Fractures that are angulated less than 20° to 45° seem to do well and that the trauma from surgery impacts negatively on the outcome.

Some authors place a limit on the degree of angulation above which closed reduction should not be attempted. range (45°-70°)

Closed reduction techniques:

The general reduction technique begins with traction and counter traction applied to the extended forearm. The arm is rotated until the maximum amount of lateral displacement is appreciated. A varus force then is applied to the elbow and the proximal fragment is reduced by digital pressure.
Kaufman et al reported success using a closed reduction maneuver in which the elbow is flexed to 90° in maximum supination. Digital pressure then is applied to the lateral aspect of the proximal fragment while the forearm is rotated gradually into full pronation.

Percutaneous reduction techniques

Recommended after failed closed reduction in an attempt to avoid the high incidence of complications after open reduction.

Three methods described none shown to be superior to another in the literature.

Direct pressure on the proximal fragment with a Steinmann pin or similar instrument that is introduced percutaneously or though a small stab wound. Care must be taken to avoid the deep branch of the radial nerve, which traverses the arcade of Frohse. Fluoroscopy is used to assess the position of maximum displacement, guide reduction, and determine stability through a range of motion (ROM). If the reduction is stable, internal fixation is unnecessary. The arm then is immobilized for 3 to 6 weeks in 90° flexion.
A thinner Kirschner (K) wire is used which is introduced into and across the fracture site. Rather than pressing directly on the proximal fragment, the reduction is achieved by a cephalad sweep of the surgeon's hand which levers the radial head back into place. The wire is removed gently and the stability of the reduction is assessed. Again, internal fixation may or may not be necessary and the arm is immobilized.
Intramedullary manipulation of the radial head. A small incision is made 1 to 2 cm proximal to the distal radial physis, taking care to avoid injuring the sensory branch of the radial nerve. The lateral cortex is penetrated with a drill and a specially bent, long K wire is introduced into the medullary canal. Under fluoroscopy, the wire is advanced cephalad to engage the proximal fragment. By rotating the K wire, the radial head is reduced. Occasionally, direct pressure on the radial head with a percutaneous wire is necessary to aid the reduction. The intramedullary wire is left in place for fixation. The elbow is mobilized at 2 to 3 weeks and the wire is removed at 8 weeks.

Open Reduction techniques

Recommended for severe initial displacement but also for failure of closed or percutaneous attempts.

Precisely how much residual angulation constitutes a failed attempt at reduction remains controversial. Reports in the literature vary from 15° to 45° residual angulation.
Most authors agree that the posterolateral (Kocher) approach to the radial head, being careful to avoid injuring the posterior interosseous nerve, is the approach of choice. After reduction of the proximal fragment, internal fixation should be used if there is evidence of instability.

Various methods of internal fixation have been described

Suture fixation of the fragment has been used but is not a popular method.
Transcapitellar pin fixation was used but no longer is recommended because these pins have a tendency to break in the joint.
The current recommendation is to place one or two K wires obliquely from the lateral edge of the radial head, across the fracture site, into the distal fragment. The wires are left protruding from the wound for simple removal. Before closing the incision, the capsule and orbicular ligament should be repaired. The arm then is immobilized in 90° flexion and neutral rotation, because these fractures have a tendency to lose some supination and pronation. Remove wires at 3 to 4 weeks.

Outcome measures

D'Souza et al based on ROM

Excellent: full motion
Good: loss of <20° in any plane
Fair: loss of 20°-60° in any plane
Poor: loss of >60° in any plane

Metaizeau et al similar but considered loss in all planes, rather than any plane, and chose 40° as the distinction between fair and poor.

Steinberg et al and Rodríguez-Merchán classified outcome into three categories

Good: full motion, no pain;
Fair: loss of <20° in any plane or slight pain
Poor: loss of >20° in any plane or definite pain

Vahvanen and Gripenberg classified outcome into three categories

Good: loss of <10° in any plane and a carrying angle <10° different from the contralateral side
Fair: loss of 10°-30° in any plane
Poor: loss >30° in any plane

Steele and Graham used a much more comprehensive classification system, which categorizes loss of pronation (in increments of 15°), supination (in increments of 10°-15°), flexion and extension (in increments of 5°), and carrying angle (in increments of 5°). The lowest grade in any category determined the outcome as excellent, good, fair, or poor. The presence of pain automatically downgraded the outcome to poor.

There is currently no scientific basis for preferring one classification system to another. Radomisli recommended that the relatively simple classification system used by Steinberg et al and Rodríguez-Merchán be adopted as the standard to make comparisons between studies possible.

OUTCOME

Given the controversies in the literature concerning the measurement of initial displacement, determination of severity, choice of treatment, and classification of outcome, it is no wonder that the assessment and comparison of result is problematic. A review of the recent literature shows that most authors agree that worse results follow more aggressive treatment, but clearly it is the more severe fracture that warrants the more aggressive approach. Whether poor results are a consequence of the treatment or the severity of the fracture is not entirely clear. Most probably both factors play a role in outcome, but there are no prospective outcome studies that compare different treatment protocols for specific fracture patterns.

Closed treatment of the paediatric radial neck fracture yields better results than does operative treatment, most likely because it is indicated for fractures of less severity. In a retrospective analysis of 100 patients, D'-Souza et al reported excellent or good results in 99% of the patients treated by closed methods in contrast to only 55 of those treated with open reduction. Fifteen percent of the patients who underwent closed treatment reported pain at follow-up, in contrast to 66% of those who underwent open treatment. Similarly, 15% of the patients treated by closed methods lost some amount of forearm rotation, and 10% lost some flexion or extension at the elbow. In contrast, 79% of the patients treated with open reduction lost some rotation, whereas 34% lost some flexion or extension. Vahvanen and Gripenberg reported good results in 83% of patients treated nonoperatively, 61% of who had an initial angulation less than 30°. Of 14 patients who underwent operative treatment, the authors reported good results in five patients (one patient, <30°; one patient, 30°-60°; three patients, >60°), fair results in six patients (four patients, 30°-60°; two patients, >60°), and poor results in three patients (one patient, <30°; two patients,>60°). In 10 children with radial neck fractures angulated less than 60°, Kaufman et al reported an anatomic closed reduction in eight patients, accepting 10° displacement in two patients. At follow-up, 2 to 5 years later, all 10 patients had full ROM. In a review of five series, including their own, Steinberg et al reported similar results. Of 50 patients treated nonoperatively, the results were good in 34 (68%), fair in seven (14%), and poor in nine (18%). By comparison, of the 104 patients treated with an open reduction, the results were good in 60 (58%), fair in 17 (16%), and poor in 27 (26%). Steinberg et al further analyzed the results in terms of the patient's age at the time of the fracture and the angulation at the time of presentation. Overall, patients with fractures displaced 30° to 59° had better results than patients with fractures displaced more than 60°. Operative treatment yielded better results than did closed treatment in fractures angulated more than 60°, whereas closed treatment resulted in a better outcome than operative treatment in fractures angulated 30° to 59°. Steinberg et al found no difference in outcome comparing patients who were 5 to 9 years old with patients older than 10 years. Reidy and Van Gorder and Rodríguez-Merchán, in contrast, reported better overall results in children younger than 10 years old.

The percutaneous methods of reduction, developed in an effort to avoid the trauma inherent in open treatment of these fractures, generally have met with favourable results when compared with open reduction. Rodríguez-Merchán, using a percutaneous Steinmann pin to apply direct pressure on the proximal fragment, reported a successful reduction in 20 of 23 patients, with 75% good results. Using the same technique, Bernstein et al reported a successful reduction in 15 of 18 patients, resulting in a residual angulation of 8° to 20°. Using a percutaneous K wire to lever the proximal fragment into place, Steele and Graham reported a successful reduction in 33 of 36 patients presenting with an initial angulation more than 30°, 94% with an excellent or good final result. Metaizeau et al, using a specially bent intramedullary wire, reported an excellent or good functional result in 30 of 31 patients who had an initial angulation of 30° to 80° and in 11 of 16 patients with an initial angulation greater than 80°. Most authors agree that percutaneous methods are a simple, safe, and effective alternative to open reduction in the majority of fractures, provided there is some contact between the proximal fragment and radial metaphysis.

The poorer outcome of operative treatment has been alluded to above. In addition to the studies already mentioned by D'Souza et al, Steinberg et al, and Vahvanen and Gripenberg, Fowles and Kassab reported a loss of motion in 50% of children treated with open reduction and internal fixation. A limitation of motion was more common after radial neck fractures associated with other injuries. Rodríguez-Merchán reported 50% fair and poor results in children treated operatively. The loss of motion, which accounted for the majority of fair and poor results, was thought to be related more to the injury itself and not to the method of reduction.

Complications

Overall 20%-60% complication rate. Complications are thought to be a result of aggressive treatment and a result of the initial severity of the fracture, particularly vascular and physeal injury.

Common complications

Loss of motion (0%-79%)
Pain (9%-66%)
Avascular necrosis (7%-44%)
Premature physeal closure (0%-31%)
Periarticular ossification (2%-13%)
Nonunion
Malunion
Radioulnar synostosis
Residual angulation
Cubitus valgus
Nerve palsy
Pin tract infection
Hardware breakage (after transarticular fixation)
Compartment syndrome

Summary

Radial neck fractures may account for as much as 20% of elbow fractures in children.

Associated with a relatively high complication rate.

It is recommended that displacement be measured as the angle between a line perpendicular to the articular surface with another down the shaft of the proximal radius.

30° is a reasonable cut off above which some form of reduction should be attempted. The classification used by Steinberg et al and Rodríguez-Merchá, described above should perhaps be used as the standard incorporating angulation and percentage translation.

References

Radomisli, Timothy E. MD; Rosen, Andrew L. MD Controversies Regarding Radial Neck Fractures in Children. Clinical Orthopaedics & Related Research. 1(353):30-39, August 1998.