Introduction

A vast majority of undisplaced or minimally displaced radial neck fractures can be treated non-operatively with good outcome [1]. Angulations of up to even 30° are expected to remodel, more so in younger children. Severely displaced or angulated fractures, fractures in children with age more than 10 years, increase the risk of unsatisfactory outcomes after non-operative management [2–4]. Even after open reduction in displaced fragment, the results are cited as fair or poor [5]. Transcapitellar pinning across the elbow has been associated with high incidence of pin breakage and nerve injury [6]. Closed extracapsular reduction by Metaizeau technique followed by intramedullary pinning avoids both these adverse conditions, becoming the logical choice in this subset of radial neck fractures [7, 8]. We did a retrospective study to analyze the outcome of intramedullary pinning by modified Metaizeau technique in severely displaced radial neck fractures.

Materials and methods

Fourteen children presenting with Judet type IV radial neck fracture to our institute between 2004 and 2012 were included in the study. Cases with previous elbow injury, closed physis were excluded.

The operative technique used was intramedullary pinning according to Metaizeau [7] with modifications. A 1.2–2.0-mm K-wire was bent at one end with its sharp end cut and was introduced into the medullary canal through a 3.2-mm drill hole made 4 cm proximal to the distal physis and then hammered upward until its tip reached the epiphyseal plate, to create a tract (Fig. 1a, b). This is as compared to original Metaizeau technique where the point of entry is 2 cm proximal to the distal radial physis [9]. The wire was removed and a sharp K-wire with slightly bent anterior tip was introduced in the same track till it reached the fracture site. We primarily used percutaneous K-wire assistance to achieve reduction, and then intramedullary K-wire was pushed across the physis. Fine tuning of the reduction in the radial head to anatomic position by rotating the wire by appropriate degrees around its axis was not required in any of our patients. Cases where closed reduction assisted with K-wire joystick method was not achieved in two attempts, attempt to reposition the head fragment that was done by levering the head onto the shaft fragment like a ‘tyre maneuver’, which if unsuccessful in two attempts was subjected to open reduction; the method of fixation remained same. One severely displaced type IV b (125°) fracture neck radius and one type IV a fracture required open reduction, as closed radial head reposition could not be achieved. Postoperative immobilization in an above elbow back slab was given for 3 weeks. K-wires were removed after the fracture had consolidated.

Fig. 1
figure 1

a Intraoperative photograph showing the technique of introducing the K-wire and making of tract. b Intraoperative photograph showing the technique of introducing the K-wire and method of reduction used

Full size image

The evaluation of clinical results was performed using Mayo elbow performance score [10]. Initial radiographs and radiographs after K-wire removal were examined for residual angulation and translation, peri-articular ossification, avascular necrosis of radial head, enlargement of radial head and premature physeal closure. Radiographic results were graded as excellent, if the reduction was anatomic, good, if a simple shift or inclination not exceeding 20° persisted, fair, if the tilt was between 20° and 40° and poor, if it was beyond 40°.

Results

There were nine girls and five boys, age range 5–13 years, with average age of 9 years. Left-sided involvement was seen in 9 (64.3 %) and right sided in 5 (35.7 %) patients. All children had type IV Judet fracture with mean angulation of 72.8° (range 60–125), 3 type IV B and 11 type IV A (Fig. 2a, b). There was associated ipsilateral elbow dislocation in one child which resulted in Salter–Harris epiphyseal separation type I of the proximal radius. Follow-up period ranged from 6–80 months with a mean follow-up of 39 months. In all patients, postoperative course was uneventful. The K-wire was removed after radiological consolidation (10–16 weeks, average 12 weeks).

Fig. 2
figure 2

a AP and lateral radiographs of patient with type IV A fracture neck radius. b AP and lateral radiographs of patient with type IV B fracture neck radius, resulting from ipsilateral elbow dislocation (reduced in this radiograph)

Full size image

Based on Mayo elbow performance score, all elbows had excellent outcome with mean score of 98.2 (95–100). Final radiographic analysis showed 11 (78.6 %) excellent and 03 (21.4 %) good results. Two patients, one type IV B and other type IV A had translation had to undergo open reduction. At final follow-up of both these patients, clinical outcome was excellent (MES-95) with good radiological outcome, inspite of evidence of heterotopic ossification in one patient and transient radial nerve palsy in the other. There was no patient with avascular necrosis or enlargement of radial head at final follow-up. No patient at final follow-up had a growth arrest and the carrying angle was equal to the uninjured side (Fig. 3a–c).

Fig. 3
figure 3

a AP and lateral radiographs of patient with type IV A fracture neck radius at final follow-up. b AP and lateral radiographs of patient with type IV B fracture neck radius, requiring open reduction at final follow-up. Note the heterotrophic ossification. c Clinical photograph of the same patient showing range of motion at elbow

Full size image

Discussion

Management of displaced radial neck fractures in pediatric patients remains controversial. A plethora of surgeries have been described ranging from percutaneous pinning, transcapitellar pinning, intramedullary pinning and the modes (open or closed) and methods (joystick/leverage/rotational force of intramedullary bent wires) of reduction. Open reduction has been recommended conventionally for failed closed manipulation in displaced radial neck fractures. However, since worse outcomes following open reduction have been encountered, it is desirable to reserve it for fractures where possible functional problems secondary to deformity may outweigh the risk of disturbing the blood supply [11]. In consensus, close reduction yield better functional results than open reduction; in severely angulated fractures, closed reduction alone has unpredictable results ranging from inability to achieve complete reduction or even loss of reduction inside the plaster [12, 13]. Percutaneous Kirschner-wire manipulation and fixation although enables satisfactory reduction but does not guarantee adequate stability when used alone for fixation [3, 14]. Moreover, repeated manipulations and leverage may cause further injury to the fractured fragment.

Although several authors have reported excellent results with pinning of grade-III and grade-IV fractures, but indications for reduction (angulation >30° and >35°) were different from our study group involving only type IV [13, 15]. Percutaneous reduction and pinning yielded excellent results in series by Biyani et al. [16]; however, they still concluded that this technique may, however, not be suitable for radial neck fractures with major translocation. Although the technique of Metaizeau is well described and published by various authors, we differ from them by primarily using a joystick manipulation in all the cases, evaluating only type IV fractures and with a different point of entry from earlier described technique [17]. Open reduction was required in two patients; both the patients had complications, 1 had heterotrophic ossification and the other had a transient nerve palsy of the posterior interosseous nerve. We believe that inability to achieve close reduction in these two patients was due to loss of intact periosteal hinge; even leverage maneuver could not achieve reduction.

We believe the above modifications to the well-described Metaizeau technique have some significant advantages. Firstly, shifting the entry portal proximally prevents iatrogenic injury to the superficial radial nerve which has been reported earlier in series of cases treated by Metaizeau technique, [9] but was not encountered in any of our cases. Secondly, the use of a smaller diameter Kirschner-wire in our technique prevents the crushing of the soft cancellous bone of the head of radius that might be associated with the Metaizeau technique with a intramedullary elastic nail [11]. Thirdly, the use of joystick manipulation allows more precise reduction. Reduction in Metaizeau technique relies only on the arc of rotation of the bent wire which might be sufficient to reduce Judet type III fractures but may be unable to fully reduce type IV fractures.

We recommend these modifications in the classical Metaizeau technique in Judet’s type IV fractures of neck radius. Failure to achieve closed reduction may occur in cases where lateral periosteal hinge is broken and these cases usually require open reduction. The presence of a periosteal hinge as a marker of success of closed reduction needs to be further evaluated in a larger series.