Lateral condylar humerus fractures are the second most common fractures in children, accounting for approximately 17% of pediatric elbow fractures [1]. Treatment of the fractures depends on the extent of displacement of the distal fragment and stability of the fracture. Usually, for nondisplaced and minimally (less than 2 mm) displaced stable fractures, posterior plaster splint or long-arm cast followed by close observation usually produces good results; while for those displaced unstable fractures, open reduction and internal fixation is usually recommended to obtain a congruent articular surface and sufficient fracture reduction. This treatment strategy has been widely accepted by majority of pediatric surgeons [1–8].
Few studies reported on CRPP for treatment of pediatric lateral condylar humerus fractures. Foster et al reported that percutaneous pin fixation of nondisplaced and minimally displaced (༜2 mm) fractures is an acceptable alternative in any situation in which close scrutiny cannot be ensured [9]. Mintzer et al treated 12 children with lateral condyle fractures with moderate displacement (2–4 mm) using CRPP. The intraoperative arthrograms showed no incongruity of the articular surface and the results were good without complications. They advocated CRPP for unstable, moderately displaced lateral condylar fractures (Jakob type II) [10]. Weiss et al reported 65 patients with lateral condyle fractures displaced > 2 mm but with an intact articular surface as assessed by an arthrogram, treated using CRPP. They claimed that fractures with displacement more than 4 mm were more likely to have incongruent articular surface, which should be treated with open reduction [11]. However, all these cases treated with CRPP were fractures with minimal or moderate displacement (༜4 mm) [9–11]. It appears that it is impossible to treat such fractures as Jakob type III using CRPP.
Song et al. showed that significant displaced unstable lateral condylar humerus fractures, even those with totally rotation, could be managed without open reduction [12–14]; at this point, CRPP truly become a viable option. First, Song et al. introduced a new classification of LCHF. Stage I indicated a fracture through the lateral humeral condyle with a minimal lateral gap and ≤ 2 mm of displacement. Stage II indicated a fracture through the lateral humeral condyle to the epiphyseal articular cartilage with a lateral gap and ≤ 2 mm of displacement. Stage III indicated a fracture through the lateral humeral condyle into the joint, a fracture gap that was as wide laterally as it was medially, ≤ 2 mm of displacement, and a high risk of further displacement. Stage IV indicated a fracture with > 2 mm of displacement without rotation of the distal fragment. Stage V indicated a fracture with > 2 mm of displacement with rotation of the distal fragment. Minimal displacements (≤ 2 mm) with fracture lines extending into the joint and a fracture gap as wide laterally as medially on any radiographic views were categorized as unstable fractures. They attempted CRPP in all cases; 13 (76%) of 17 stage III fractures, 30 (75%) of the 40 stage IV fractures, 3 (50%) of the 6 stage IV fractures were successfully treated by CRPP with residual displacement less than 2 mm. The follow-up showed excellent or good results in all cases with minor complications like bone spurs at the lateral side of the distal humerus without symptoms. Their later papers, including more fractures with completely and rotational displacement, showed higher success rates. Eighteen of 24 (75%) completely displaced and rotated fractures were reduced within 2 mm of residual displacement using closed reduction and internal fixation (CRIF). Therefore, they claimed that CRIF is an effective treatment for completely displaced and rotated LCHFs in many children [12–14].
Our study showed similar results in treating LCHFs with displacement more than 4 mm by CRPP. Approximately 78% (36/46) of our series were managed with CRPP, with success rate of 76% (19/25) for Song stage IV and 78% (14/18) for Song stage V fractures, and 78% (22/28) in Jakob type II and 78% (14/18) in Jakob type III fractures. The final follow-up showed no functional disfunction or complications except lateral bone spur formation without symptoms in most cases.
Now this CRPP technique has been introduced in the latest edition of a classic book, Rockwood and Wilkins Fractures in Children [1]. A few later studies reported their results using this technique [15–18]. Silva et al. claimed that CRPP is an alternative treatment of pediatric LCFs with displacement between 2 and 4 mm. The advantages of this technique include avoiding an unaesthetic scar, decreased surgical times, and not significantly increasing the incidence of complications [15]. The study by Pennock et al. reported that only one patient with displacement between 4 and 5 mm in their series was treated with CRPP; the authors concluded that CRPP may be suitable for lateral condyle fractures with displacement between 2–4 mm, because of high incidence of joint congruity in these cases, thereby avoiding an incision and decreasing surgical time [16].
However, almost all cases, in these studies were LCHFs with displacement less than 4 mm. The reason may be the difficult learning curve, as pointed out by Song et al. [12]. In the present study, the average preoperative displacement was 11.8 mm in Song stage V and 5.0 mm in Song stage IV LCHFs, with an average postoperative displacement was 1.2 mm in Song stage V and 1.1 mm in Song stage IV LCHFs. All cases were treated by a single surgeon (Li-Wei Xie) in our department, who had little experience in treating LCHFs with CRPP, but abundant experience with ORIF and in treating supracondylar fractures with CRPP. The authors’ experiences in this study showed that CRPP in treating LCHFs is technically feasible, and that more than 3/4 of cases can avoid an incision. The surgeon should prepare himself to make sure he is familiar with the technique of manual reduction before starting his first case.
The following are several useful tips to achieve a satisfied reduction of the fractures. First, never do an arthrogram before reduction, as this will interfere with your judgement regarding the relative position of the fragments. Because the origin of the extensor of the forearm is always disrupted, contrast medium will flow into the lateral soft tissue and this will blur the pictures of lateral side of the distal fragment. Second, a lateral puncture point for arthrogram is optimal. The position is just at the lateral side of the humeroradial joint. In this way, contrast medium will directly flow into the articular surface and the fracture gap with little disturbance of judgement of the reduction. Third, the distal fragment will displace anteriorly or posteriorly on the lateral fluoroscopy after reduction, which could be reduced by extension or flexion of the elbow joint. Fourth, the pinning site is a little more distal and anterior than the regular position in pinning a flexed elbow, because this manipulation usually needs to maintain the reduction in an extensional elbow with a valgus force. Fifth, for Jakob type III and Song stage V, the difficulty lies in the correction of rotation of the distal fragment. In most cases, an anterior backward and inward compression of the distal fragment with the elbow joint flexion could correct the rotational displacement in most cases. Occasionally, a 2.0-mm Kirschner wire could be used as a joystick to help reduce the rotation. Sixth, a thinner 1.0-mm Kirschner wire could be used first to avoid too much pinning holes in the small nuclear ossification of distal physeal, which could be changed to 1.5-mm Kirschner wires for final fixation. Last but not least, if the reduction of a LCHF cannot be obtained by several attempts, convert to ORIF directly.
The present study has several limitations. The number of cases was small and the time of follow-up was short. Because the study was not controlled, we could not tell the difference between cases treated with CRPP and the cases treated with ORIF directly. Moreover, the cases in this study were performed by one surgeon, and we are not sure about the repeatability among surgeons.