We present our experience of forearm lengthening in children with various conditions performed by a single surgeon between 1995 and 2009. A total of 19 children with a mean age of 9.8 years (2.1 to 15.9) at the time of surgery had 22 forearm lengthenings using either an Ilizarov/spatial and Ilizarov circular frame or a monolateral external fixator. The patients were divided into two groups: group A, in whom the purpose of treatment was to restore the relationship between the radius and the ulna, and group B, in whom the objective was to gain forearm length. The mean follow-up after removal of the frame was 26 months (13 to 53).
There were ten patients (11 forearms) in group A with a mean radioulnar discrepancy of 2.4 cm (1.5 to 3.3) and nine patients (11 forearms) in group B. In group A, the mean lengthening achieved was 2.7 cm (1.0 to 5.5), with a lengthening index of 11.1 weeks/cm. Equalisation or overcorrection of the discrepancy was achieved in seven of 11 forearms, but lengthening was only partially successful at preventing subluxation or dislocation of the radial head. In group B, the mean lengthening achieved was 3.8 cm (1.9 to 6.8), with a lengthening index of 7.25 weeks/cm. Common complications in both groups were pin-site infection and poor regenerate formation.
Forearm lengthening by distraction osteogenesis is a worthwhile procedure in children that can improve cosmesis and function, particularly in patients with shortening of both radius and ulna.
Limb lengthening has been used in the forearm in the management of a variety of conditions, such as physeal arrest,1 multiple exostoses,2 Madelung deformity3 and radial hemimelia.4 The upper limb has different alignment requirements from those in the lower limb, and moderate discrepancies in forearm length do not usually produce a significant functional deficit.5 However, the combination of shortening and angular deformities can reduce the ability to carry out simple activities of daily living. A deficiency in forearm length reduces the volume of space available for the hand6 and may cause functional and cosmetic problems. A relative discrepancy between the length of the radius and ulna not only causes shortening of the forearm but may result in subluxation of proximal and/or distal joints, and limitation of supination and/or pronation.7-9 Deformities are worsened by soft-tissue contractures that can further reduce the range of movement and dexterity of the limb.
Treatment recommendations for relative discrepancies in the length of the forearm bones have included excision of exostoses, shortening of the longer bone and epiphysiodesis. However, these procedures do not correct the underlying abnormality and may result in further shortening of the forearm. There are several reports of the use of distraction osteogenesis in the forearm, with complication rates ranging from 0% to 100%.1,2,6,9-13
The aim of our study was to report the experience of a single surgeon with distraction osteogenesis of the forearm in two different paediatric populations. In group A the aim was to restore the normal relationship in length between the radius and the ulna, and in group B the aim was to gain forearm length.
Patients and Methods
From 1995 to 2010, 22 forearms in 19 patients (11 girls and eight boys) underwent lengthening by distraction osteogenesis. There were 18 ulnar lengthenings, two radial lengthenings, and in two forearms both bones were lengthened. In two patients bilateral lengthening was undertaken and in one a second lengthening was undertaken after 2.75 years, giving a total of 22 forearms. The mean age at the time of initial lengthening was 9.8 years (2.1 to 15.9). Patients were divided into two groups: group A (ten patients, lengthening for relative discrepancy) (Fig. 1) and group B (nine patients, lengthening for a short forearm) (Fig. 2). Details of the patients are given in Table I. The medical records and radiographs of each patient were reviewed retrospectively.
All patients were assessed for function by examining limitation of activities of daily living in relation to their deformity. No validated function outcome was used. All patients in group A had some degree of ulnar deviation of the wrist (mean 18° (5° to 25°)). In group B, four patients had radial deviation of the wrist (mean 36° (33° to 41°)). One patient had 15° ulnar deviation and anterior bowing, and two patients had instability of the wrist, tested by stress in the ulnar direction. All patients had some degree of bowing of the forearm. The two patients with cutis aplasia had soft-tissue tightness and forearm shortening of 9 cm and 10 cm compared with the contralateral side, respectively. Among the two groups, 14 patients had normal or near normal function in terms of activities of daily living, despite the forearm being markedly deformed or short.
The indication for surgical treatment in group A was deformity in the wrist, forearm or elbow in association with shortening of the radius or ulna by > 2 cm, particularly when an increasing discrepancy with growth was anticipated. Subluxation or dislocation of the radial head was also an indication for treatment. The goal of intervention was to correct the discrepancy between the radius and the ulna and to prevent subluxation or dislocation of the radial head. If subluxation was present the aim was to prevent progression to dislocation, and if dislocation was present the aim was to achieve reduction of the radial head. In group B, the indication for surgical treatment was to correct deformities, which in most patients affected the wrist, and lengthen the forearm sufficiently to improve function and appearance. Full restoration of forearm length was not considered a requirement for a successful outcome, and each patient was assessed individually to achieve a balance of treatment duration against the risk of increasing complications.
The patients were assessed pre-operatively by a multidisciplinary team including occupational therapists and physiotherapists. The degree of deformity and function of the limb was documented prior to treatment and followed for the duration of the lengthening procedure. All procedures were performed by the senior author (RAH).
Either a monolateral external fixator (Orthofix Ltd, Maidenhead, United Kingdom) or a circular Ilizarov or spatial/Ilizarov hybrid fixator (Smith & Nephew Orthopaedics Ltd, Warwick, United Kingdom) was used (Table I). In 15 lengthenings a two-ring construct was used, with a 5/8 ring at the elbow and a full ring at the wrist, constructed with a combination of wires and half-pins (Fig. 3). In three lengthenings two full rings were used. The osteotomy was performed percutaneously by pre-drilling the bone prior to attaching the frame, and then completing the osteotomy with an osteotome after the frame was attached. When lengthening the radius, the preferred level for osteotomy was the distal radial metaphysis. In ulnar lengthening, the preferred level for osteotomy was the proximal metaphysis just distal to the coronoid process. The distraction followed a prearranged protocol, starting five to seven days post-operatively with initial distraction of 0.75 mm/day to 1.0 mm/day. Radiographs were taken post-operatively every two weeks during the lengthening period to monitor new bone formation, and thereafter every three to four weeks until consolidation was achieved. The fixators were removed when corticalisation of three out of four cortices was visible on radiographs. After removal of the frame, cast immobilisation was used selectively for patients who had developed flexion contractures during lengthening, or in whom there was concern about the strength of the regenerate.
All patients had supervised post-operative hand and elbow therapy to maintain function. Patients initially carried out passive range of movement exercises, and as pain decreased, active exercises were encouraged. Splints and occupational therapy were used to prevent flexion contractures in the wrist and fingers.
The mean follow-up for all patients was 26 months (13 to 53) after removal of the frame. The mean time in the fixator was 172 days (105 to 325). There was considerable variation in the pre-operative discrepancy in forearm length among patients with a mean of 5.7 cm (1.5 to 12.2). The mean length gained was 3.2 cm (1.0 to 6.8). The mean lengthening index, defined as the time from frame fixation to frame removal, divided by gain in length, was 9.2 weeks/cm (3.4 to 34.9). The results for both groups are shown in Tables II and III.
Complications were similar in both groups. Pin-site infections were treated with local care and oral antibiotics (three lengthenings in group A, six in group B). In no patient was there progression to deep infection.
In group A, four patients (four forearms) had poor regenerate formation and delayed consolidation. In these patients the rate of distraction was either reduced to 0.5 mm/day or paused and restarted after a mean of six days. After cessation of lengthening, a low-intensity pulsed ultrasound device (Exogen, Smith & Nephew Inc, Memphis, Tennessee) was used in three patients to improve consolidation of the regenerate, at 20 minutes per day until the regenerate consolidated or failed to improve after six weeks of treatment. This proved to be successful in two patients, but the third required bone grafting from the iliac crest. In the fourth patient, who had a discrepancy of 3.3 cm, only 1.5 cm of lengthening was possible due to pin-site infection resulting in loosening of the frame and poor regenerate formation; an open reduction and internal fixation with iliac crest bone graft was required. In group B there was one case of poor regenerate formation resulting in cessation of lengthening.
The status of the radiocapitellar joint before and after lengthening is shown in Table IV. In seven lengthenings there was no change in the position of the radial head after lengthening. In one patient with multiple hereditary exostoses (MHE) the radial head was normally located prior to lengthening but was subluxed after lengthening, and in a further patient with MHE the radial head was normally located prior to lengthening but was dislocated at the last follow-up. In one patient with a short ulna, the radial head was subluxed prior to lengthening and fully located at the last follow-up. Lengthening in a patient with osteochondroma of the distal ulna resulted in subluxation of a previously located radial head. One patient with MHE with a dislocated radial head prior to lengthening underwent open reduction combined with a radial neck osteotomy after lengthening that successfully reduced the radial head, but it re-dislocated during follow-up.
With the description of the principles of distraction osteogenesis using the Ilizarov fixator, lengthening of the radius and ulna with simultaneous correction of deformity became a surgical option.14 Most of the published literature on distraction osteogenesis has concentrated on the lower limb. Little has been published on upper limb lengthening, particularly in the forearm. Long-term outcomes have not been widely reported and there is a lack of consensus regarding the indications for surgery. Careful patient selection is necessary for a successful outcome in terms of functional gain, and those with contractures restricting the use of the arm (particularly a stiff elbow) or with a non-functional hand may not be appropriate for lengthening.15 In addition, forearm lengthening predisposes to flexion contractures of adjacent joints, and hand function may decrease because of digital stiffness.6 Nevertheless, several studies of forearm lengthening using the Ilizarov technique in children have reported functional and cosmetic benefits.6,11,15
In our study, rather than group cases by specific pathology, we chose to take an anatomical perspective and divided the patients into those with a relative discrepancy in the length of the radius and ulna, and those with overall shortening of the forearm. Patients who have a discrepancy in the length of the radius and ulna are at risk of joint subluxation or dislocation, most commonly of the radial head, when the ulna is relatively short. MHE is one of the commonest diagnoses presenting in this way (four out of 11 cases in this series), but it may be seen in other causes of ulnar shortening. We wanted to see whether ulnar lengthening would either prevent this happening or enable reduction of a previously subluxed or dislocated radial head. The natural history of radial head dislocation in patients with MHE is unknown. The condition is of variable severity and there may be deformities in both bones, as well as shortening, making it difficult to draw conclusions on the prophylactic value of lengthening. The disease may also progress in the period of remaining growth. However, it seems likely that a radial head that is subluxed at presentation is at risk of dislocation. Overall, in group A (all pathologies) there were two lengthenings with a subluxed radial head prior to lengthening: one remained subluxed and the other reduced. Two lengthenings had dislocated radial heads prior to lengthening: both remained dislocated at follow-up, despite an open reduction and a radial neck osteotomy after ulnar lengthening in one case. In seven lengthenings the radial heads were located prior to lengthening; of these, five remained located, one subluxed (MHE) and one dislocated (MHE). The numbers are too small to draw conclusions, but our experience was that once the radial head was dislocated we were unable to obtain a stable reduction. Our single success in converting a subluxation to a reduction was in a patient with a short ulna, which did not have the concomitant deformities associated with MHE. MHE seems to be particularly difficult because of the complexity of the deformity and progression of the disease during growth; even if the radial head is located after lengthening it may sublux or dislocate with further growth. Akita et al16 showed that excision of the dislocated radial head was associated with a better outcome than open reduction in patients with multiple cartilaginous exostoses.
There were two lengthenings for growth arrest due to localised physeal damage and in both the radial head was located prior to lengthening, and remained so.
A total of seven patients in group B had radial hemimelia. None of these had a radial head and most had complete absence of the radius, and so radial head subluxation or dislocation was not an issue. The other two patients in this group had cutis aplasia: one had a dislocated radial head prior to lengthening which remained dislocated, and the other had a located radial head throughout the lengthening procedure. In this group of patients the aim was to achieve forearm length and, where appropriate, to correct deformity and place the wrist in a more functional position. A worthwhile gain in length was achieved in most patients in this group. One of the forearms with the least gain in length had thrombocytopenia absent radius syndrome, where problems with bleeding precluded further lengthening. In one other case with poor gain in length, distraction had to be stopped because of pain associated with abnormal soft tissues, including scarring from previous surgery. There was no loss in function after lengthening, possibly because considerable emphasis was placed on post-operative hand and elbow therapy and dedicated splintage. This group in particular described improvement in cosmesis and function.
Full restoration of forearm length is not a requirement for a successful outcome, as the highly mobile joints of the upper limb, particularly the shoulder, can compensate for residual deformity, and these young patients are able to adapt well to most situations. The goal is to lengthen the forearm and correct deformity sufficiently to improve function while minimising complications, including those that might follow overenthusiastic lengthening.
Our mean lengthening index of 9.2 weeks/cm is similar to that of other studies,17 although Pickford and Scheker18 reported a much shorter lengthening index of 3.8 weeks/cm. We consider that the lengthening index is of limited value in a group of such disparate pathologies. It is also somewhat arbitrary, in that it depends on the surgeon’s estimation as to when the regenerate has consolidated sufficiently to allow safe removal of the frame.
The timing of upper limb lengthening in children remains controversial. The underlying pathology, the functional deficit, the severity of the deformity and the age and likely co-operation of the patient all need to be taken into account. In some cases the deformities may progress until skeletal maturity. Abe et al1 recommended postponing lengthening until approximately ten years of age, and suggested overcorrection of 0.5 to 1.0 cm to allow for progression of the deformity. Pritchett12 observed a recurrence of deformity of 0.25 cm/year at the age of 12 years for patients with MHE, necessitating a second procedure. However, some authors believe that the forearm can tolerate a greater amount of lengthening at one time than the lower limb.19 We do not have a fixed age to start lengthening, but recommend that the timing of treatment should be individualised and should start when functional problems develop. With respect to forearm bone discrepancy, it remains our policy to commence lengthening as soon as there is evidence of radial head subluxation, either clinically or radiologically. However, at present we cannot draw firm conclusions of the value of ulnar lengthening in preventing dislocation of the radial head, particularly in MHE.
In forearm lengthening, complications are usually transient and do not affect outcome. Injury to the neurovascular structures could be a potential problem, as there are few completely safe anatomical planes for wire insertion and many patients have abnormal anatomy. The risk can be minimised by semi-open insertion techniques, particularly for the radius. Delayed consolidation is a feature in most forearm lengthening series. Neocorticalisation is relatively slow, and premature removal of the apparatus can lead to deformity or post-removal fracture of the regenerate. Even though the arm is not a weight-bearing limb, a period of protection is usually necessary after removal, especially in cases with a high percentage of lengthening or significantly tight soft tissues. In our series, no patient with radial hemimelia developed nonunion. This is in contrast to other studies that report a rate of nonunion of 23% to 67%.11,15,17,18 Forearm lengthening is usually complicated by frequent pin-site infections. Catagni, Szabo and Cattaneo6 reported a 100% complication rate for congenital cases using the Ilizarov technique.
In summary, we support forearm lengthening by distraction osteogenesis in children, as worthwhile improvements in cosmesis and function can be achieved, particularly in overall shortening (group B patients). However, the role of forearm lengthening in patients with radioulnar discrepancy (group A) is less clear, particularly in attempted prophylaxis against radial head dislocation in MHE. In this group, lengthening often has to be combined with correction of deformity.
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
- Received May 9, 2011.
- Accepted July 20, 2011.
- ©2011 British Editorial Society of Bone and Joint Surgery