Congenital clavicular pseudarthrosis: comparison of two treatment methods

Abstract

Background

Congenital pseudarthrosis of the clavicle is rare and popular surgical options include excision of the non-union, iliac crest bone grafting and stabilisation with either a fully threaded pin or stabilisation with a reconstruction plate.

Methods

Between 1995 and 2009, ten patients with congenital pseudarthrosis of the clavicle were retrospectively reviewed for outcome of two different forms of surgical management. Five patients had excision of the non-union, iliac crest bone grafting and stabilisation with a fully threaded pin (group A) and the other five patients had excision of the non-union, iliac crest bone grafting and stabilisation with a reconstruction plate (group B). One patient in the second group had bilateral pseudarthrosis.

Results

In group A, three patients achieved radiological union at a median of 6 months. Two patients failed to unite at the pseudarthrosis and one of them had further surgery with a reconstruction plate in order to achieve union. In group B, all five patients (six clavicles) achieved radiological union at a median of 3 months. All patients had painless full range of shoulder movement and were engaging in unrestricted activities.

Conclusion

Excision of the pseudarthrosis, autogenous iliac bone grafting and stabilisation with plate achieves union quicker and with lower incidence of complications compared to excision of the pseudarthrosis, autogenous iliac bone grafting and stabilisation with fully threaded pins.

Introduction

Congenital pseudarthrosis of the clavicle is a rare condition; following its first report in 1910 [1], many individual cases and case series have been published. Congenital pseudarthrosis of the clavicle usually presents as a swelling in the clavicular region at or soon after birth. The swelling is usually painless and occasionally present as weakness and inability to push the arm while crawling due to excessive movement at the pseudarthrosis site. It is believed to be caused by the failure of the two primary ossification centres to fuse in the 6th week of intrauterine life [2]. However, this view has been disputed by subsequent anatomical studies. Its occurrence has also been associated with the close proximity of the subclavian artery to the clavicle, particularly when the first rib is prominent [3–7].

O’Leary et al. [8] recommended non-operative management of the condition. However, surgical intervention is now widely accepted. Surgery includes excision of the non-union with or without bone grafting and stabilisation [9, 10]. The stabilisation methods used are either K-wires [11], Steinmann pin [12], threaded pins, compression plating [13] or external fixation [14].

We present a case series of consecutive patients diagnosed with pseudarthrosis of the clavicle. We have reviewed the outcome of two different forms of surgical management: excision of the non-union, iliac crest bone grafting and stabilisation with a fully threaded pin, and excision of the non-union, iliac crest bone grafting and stabilisation with a reconstruction plate (Figs. 1, 2, 3 and 4).

Fig. 1.

Clavicular pseudarthrosis pre-operatively

Fig. 2.

Healing of pseudarthrosis following surgical stabilisation with a reconstruction plate

Fig. 3.

Healed clavicle, following the removal of metal work

Fig. 4.

Clinical picture of clavicular pseudarthrosis following surgery

Materials and methods

We reviewed all patients presenting to our institution with a diagnosis of congenital pseudarthrosis of the clavicle between 1995 and 2009. Twelve patients fulfilled the inclusion criteria. The policy of our institution at the time under review was to offer surgical management to all patients.

Two patients did not undergo surgery. The first patient deferred surgery due to associated complex cardiac malformation and the second patient presented to us at the age of 14 years with a painless lump in the middle of the clavicle with no functional limitations and did not prefer surgery.

The remaining ten patients had operative intervention and were studied. The first group of five patients (group A) had excision of the non-union, structural iliac crest bone grafting and stabilisation with a single fully threaded pin. The second group of five patients (group B) had excision of the non-union, structural iliac crest bone grafting and stabilisation with a reconstruction plate. Following excision of the non-union, all patients had a gap; primary union could have been achieved by shortening the clavicle in some cases. Bone grafting helped to achieve adequate length; hence, all patients received bone grafting. One patient in group B had bilateral pseudarthrosis. No patient had cleidocranial dysplasia.

All surgery was performed through a single transverse incision over the superior surface of the clavicle. The pseudarthrosis was excised and all cases had fresh autogenous cortico-cancellous structural iliac crest bone grafting. Stabilisation was achieved either by a single fully threaded pin or by a reconstruction plate with a minimum purchase of six cortices. Post-operatively, all patients were advised sling use for 3 weeks followed by gentle mobilisation of the shoulder. Metal work was removed following radiological union or due to complications.

Data were collected retrospectively from case notes, operative records, clinic letters and radiographs. The following demographic details were recorded and analysed: age at presentation, associated conditions, age at surgery, details of surgery, post-operative management, time to radiological union, associated complications, further surgery, time to the removal of metal work and functional outcome.

The two groups were compared to search for any differences in the age at operation, time to radiological union, time to metal work removal and any associated complications.

Results

Comparative demographics and outcome of the patients in the two groups are illustrated in Table 1. In group A, three patients achieved clinical and radiological union at a median of 6 months (range 4–6). Two patients failed to unite at the pseudarthrosis; the first patient had a protruding pin which was removed at 2 months post-operation and the pseudarthrosis failed to progress to union. Radiographs showed no progress in healing at 8 months following surgery. The patient refused further surgery as she had a well-healed scar with minimal cosmetic deformity and good function in her upper limb. The second patient in this group had wound breakdown with infection at 4 weeks post-operation and was managed with rest and antibiotics. At 8 weeks, the wound was still unsettled and the radiographs showed no evidence of bony healing. Removal of metal work, debridement and stabilisation with a reconstruction plate achieved union with good functional outcome. In the three patients who had satisfactory union, one patient had superficial infection post-operatively that settled with a course of antibiotics and no further intervention was required. Pins were removed following radiological evidence of consolidation at an average of 3 months. All patients had painless full range of shoulder movement and good functional outcome.

Table 1.

Summary of results

	Group A	Group B
Age at presentation, median [range]	4 months [2–36]	2 months [1–48]
Age at operation, median	5 years	5 years
Time to radiological union, median [range]	6 months [4–6]	3 months [3–22]
Non-union	2	0
Infection/prominent metal work	1	1
Metal work removal	3 months	7 months

In group B, one patient had bilateral pseudarthrosis; all five patients (six clavicles) achieved radiological union at a median of 3 months (range 3–22). One patient had a prominent plate and had premature removal of metal work at 5 months. This patient also had neurofibroma and bony healing was prolonged. Complete radiological union was achieved at 22 months from the initial surgery. Four patients have had their metal work removed at an average of 6 months following the initial surgery and one patient is awaiting metal work removal. All patients had painless full range of shoulder movement and were engaging in unrestricted activities.

Analysis showed both groups had surgery at the same median age (5 years). The time to radiological union was twice as long in the pin fixation group compared to the plate fixation group. The pin fixation group was associated with higher complication rates.

Discussion

Congenital pseudarthrosis is uncommon. We acknowledge that our study represents a small number of patients and believe that a multi-centre study over a long period would give more significant numbers and statistically significant data.

Lorente Molto et al. [11] achieved healing in 6–8 weeks and recommended surgical excision and stabilisation with pins. Persiani et al. [15] made similar recommendations following a comparative study between pin and plate fixation. They found lower re-operation rates when treated with Kirschner wires and bone grafting. Contrary to these studies, we observed higher complication rates in cases stabilised with pins; they consisted of non-union and infection.

Plate internal fixation achieved bony union at an average of 3 months. All patients in our study with plate fixation had full range of pain-free movement at the shoulder and were engaging in unrestricted activities. Similar findings following internal fixation with plate stabilisation was observed at an average of 3 months by other authors [13, 16, 17]. A minimal rate of complication has been reported when treated with this method.

Comparing the pin fixation group with the plate fixation group, we observed that, in group A, three of the five patients had complications; two cases with pin fixation failed to unite and one patient had superficial infection related to pin protrusion. One of these patients required revision to plate osteosynthesis to achieve union. Although studies have reported satisfactory healing with pin fixation, we had 40% [2/5] non-union in this group and in those who had satisfactory healing, the mean time for radiological union was 5 months. In group B, all patients achieved union at a mean of 3 months. One patient with prominent metal work required removal.

Plate fixation achieved more reliable union quicker and with fewer complications compared to pin fixation. The dangers of operation are related to the proximity of the subclavian artery, brachial plexus [12], splintering of the bone if the operation is done at a very early age and infection [18]. It is probably wise to defer operation until pre-school age, as it is then technically easy. The mean age in our study group at operation was 5 years and we did not have any of the above-mentioned complications.

Conclusions

1. Excision of the pseudarthrosis, autogenous iliac bone grafting and stabilisation with plates achieves union more quickly compared to excision of the pseudarthrosis, autogenous iliac bone grafting and stabilisation with fully threaded pins.

• 2.Stabilisation with plates achieved union in all patients compared to the pin fixation group, which had a 60% union rate.

• 3.Pin fixation had a higher rate of complications in the form of infection and non-union compared to the plate fixation group.