Introduction

The aetiology of congenital pseudarthrosis of the tibia (CPT) has not been clearly understood yet but its association with neurofibromatosis type 1 (NF1), fibrous dysplasia or osseofibrous dysplasia is known [1–3]. Numerous studies reported on the outcomes of its management with the use of available operative techniques that include autologous free bone grafting, vascularized fibular grafting, and the Ilizarov method that are supported by internal fixation and adjuvant chemical, physical and biological means [2–9].

The Ilizarov method and vascularized fibular grafting are the two alternative methods that are able to provide better bone union rates in the treatment of CPT. They are based on the use of a tissue complex that is capable of preserving blood supply [2, 4, 5]. However, the Ilizarov method was recognized as the optimal one, having the highest rate of fusion (75.5%) as well as success in bone lengthening, correction of the associated deformities and foot malposition [10]. Intramedullary nailing was added to the arsenal of the Ilizarov method for stability and retaining the union achieved [6–8].

Repetitive operations are quite frequent in the management of CPT due to the severity of its pathological complex. Dysplastic changes in the bone tissue of the tibia and fibula, limb-length discrepancy (LLD), residual tibial deformities and foot malposition are their main causes [2–5]. Therefore, management of this multi-component orthopaedic pathology in a paediatric population remains challenging and needs further study and discussion [8].

Our work was aimed at a comparative analysis of the outcomes (refracture-free rate, number of reoperations per patient, union rate by the completion of the study) achieved with the Ilizarov techniques in the management of CPT that was associated with NF1 versus CPT and featured tibial dysplasia, but its aetiology was idiopathic or unconfirmed.

Materials and methods

We retrospectively studied the outcomes of CPT management with the Ilizarov method in 28 patients who had been treated between 2005 and 2013 by our team. Inclusion criteria were paediatric age and CPT that was associated either with NF1 (group 1, n = 14, mean age = 9.7) established according to the guidelines of the National Institutes of Health Consensus Development Conference Statement on neurofibromatosis 1 [11], or tibial dysplasia due to idiopathic or unconfirmed aetiology (group 2, n = 14, mean age = 8.6). Group 2 patients had radiographic confirmation of dysplastic bone lesions in the tibia and fibula but lacked specific clinical NF1 manifestations. The demographic data and pre-operative findings are given in Tables 1 and 2. There was no statistical difference in the mean age and number of operations per patient before admission (2.4 operations in group 1 versus 2.0 in group 2, p > 0.05).

Table 1 Group 1 patients’ data
Full size table
Table 2 Group 2 patients’ data
Full size table

Medical records and radiographs were examined. Radiographic findings revealed that CPT was located in the middle or distal third of the tibia (the middle third was involved in one case of group 1 and in six cases of group 2). Three cases were bowed dysplastic tibias that were not fractured (a latent type). Twenty-four cases were atrophic CPT (Crawford Type IV) associated with fibular pseudarthrosis. They featured tibial bone osteoporosis, local areas of sclerosis, bone marrow canal eburnation of different extension, cortical resoption, and sclerotic end plates on the non-united fragments that had the ends which were incongruent in shape. Clinical examination revealed that 75% of patients had rough soft tissue scars due to previous operations at other hospitals. According to Dopplerography, blood flow in the anterior and posterior tibial arteries was compromised in atrophic CPT type.

Clinical goals were bone union, correction of tibial deformities and of foot malposition, as well as complete or partial LLD compensation in order to achieve weight bearing.

Ilizarov method technical solutions for implementing these goals are presented in Table 1 and Table 2. Individual technical solutions were planned for each patient. Coaptation of tibial fragments aimed to thicken the tibia and increase the contact area of the fragments at the nonunion level in atrophic CPT as well as deformity correction were the means to achieve union and weight-bearing. Open coaptation was a preferable method. It included refreshment of sclerotic end plates and two variants of their plunging into each other: (1) the sharper bone fragment end into the wider one (Fig. 1) or (2) splitting of each bone fragment end into two parts (“swallow’s tail”) for mutual introduction. Extensive resections were not used as most of the patients had had them at previous operations. A local split from the proximal tibial area was gradually transported using wire traction to the pseudarthrosis site to serve as a local autologous graft in one case of group 1. An autologous fibular graft to cover the pseudarthrosis and intramedullary wires were applied in one case of group 2. Khakhutov’s bone plasty with a local rotational tibial graft to bridge the pseudarthrosis was performed in one case of group 2. Acute compression was produced intraoperatively. Supportive compression was maintained every week or two weeks (Fig. 2). LLD was compensated by lengthening in the tibial metaphysis of a longer bone fragment after its osteotomy (Fig. 3) but bone union and correction of deformities was a priority to lengthening in eight patients who suffered severe conditions (Fig. 4). Twenty-two patients needed correction of associated deformities.

Fig. 1
figure 1

Coaptation by plunging a sharper bone fragment end into a wider one in a 12-year-old male patient (Table 2, case 10) with CPT in the distal right tibia due to nonspecific tibial dysplasia (a-b) treated in 2010. X-rays show the outcomes following the removal of the apparatus (c) and at a one year follow-up (d)

Full size image
Fig. 2
figure 2

A 12-year-old female patient (Table 1, case 10) with CPT Crowford type IV associated with NF1 (a) who had three failed operations at another hospital was treated in 2008 by closed osteosynthesis with the Ilizarov apparatus (b) and gradual correction of the deformity for 30 days (c). X-rays show the outcome after 100 days of fixation following deformity correction (d). Re-fracture occurred seven months after the removal of the apparatus (e). Re-operation comprised spared resection and coaptation of fragments (f). Supportive compression of 1 mm every two weeks was maintained at the fragments contact. Consolidation was obtained after 130 days (g); the leg was immobilized with a plaster cast. X-rays of the outcome in 2011 (h) demonstrate the healing

Full size image
Fig. 3
figure 3

A 15-year-old male patient (Table 2, case 14) with CPT due to nonspecific tibial dysplasia (a) treated in 2012 by coaptation and lengthening in the tibial metaphysis of a longer bone fragment and simultaneous deformity correction (b). X-rays after the apparatus removal (c) and at one-year follow-up show good healing (d)

Full size image
Fig. 4
figure 4

A 14-year-old patient (Table 1, case 14) with CPT associated with NF1 (a) operated in 2013. Osteosynthesis with the Ilizarov apparatus continued for 245 days (b). X-rays of a one-year outcome show good consolidation (c)

Full size image

The Ilizarov apparatus was removed based on a clinical test for absence of pathological movements at the pseudarthrosis level and pain during axial loading. Radiographic union was judged by an interrupted continuity of the cortex at the fragments junction and along the regenerated bone and by similar bone density in the regenerated area to the adjacent tibial areas. Orthotic bracing was advised for a minimum of four months following treatment.

Re-fracture-free rate, number of re-operations per patient in the groups, CPT union in the age subgroups under ten and ≤ten years of age after the index operation, as well as union rate by the completion of this study were compared between the groups. Pearson’s chi-squared and Mann–Whitney tests were used for revealing the significance of difference between the groups.

Informed consent statements were given by children’s parents. The study was approved by the ethic committee of our institution and was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki.

Results

Outcomes after the index operation

Bone union was obtained in all the cases after the first operation. However, eight patients of group 1 and nine patients in group 2 had re-fractures within a year post-operatively (Table 1 and Table 2). The refracture-free rate was higher in group 1 versus group 2 after the first operation (42.86 % and 35.71% respectively) but it was not statistically significant (p > 0.05).

Within the subgroups, two children aged ≥three years with a latent CPT type who underwent their first treatment recovered completely. Only one case healed in the age from three to ten years that had bilateral involvement. In children aged ≤ ten years, re-fractures did not happen in 50% (4 out of 8 cases) and 57% (4 out of 7 cases) in group 1 and group 2, respectively.

The union achieved in 11 patients after the first operation was preserved at follow-ups (range, 2–9 years).

Axial deformities were corrected in all the cases. LLD was compensated completely or partially in six patients of group 1 and in seven patients of group 2.

Outcomes of subsequent operations

In group 1, refractures resulted in a greater number of reoperations as compared to group 2 (14 versus 11) over the period studied but the difference between the groups was not statistically significant (1.07 and 0.78 re-operations per patient respectively, p > 0.05).

Thirteen patients (92.86%) out of 14 in each group retained bone union by the completion of this study (follow-up range, 2–9 years).

Three patients in group 1 that had achieved complete union after the first operation underwent only lengthening procedures at subsequent stages.

One patient discontinued treatment. None of the patients had amputation.

Complications

Complications other than refracture were residual deformities and LLD. Wire-tract infection was observed but it was resolved using local anti-infection injections, wire removal or its reinsertion.

Discussion

Histological examinations in 192 patients from several countries showed a nonspecific CPT appearance in 45.3% of cases, resemblance to fibrous dysplasia in 15.6%, and in 39% there was a histologic evidence of neurofibromatosis [3]. Tibial dysplasia in children with NF1, osseous fibrous or fibrous dysplasia frequently results in tibial bowing and a pathological fracture as these children grow [12, 13]. Treatment of pseudarthrosis starts before skeletal maturity but often fails and results in repetitive operations in the paediatric population. As the research showed, the problems of bone reconstruction in CPT are related to a low bone tissue potential for regeneration though the aetiological causes of this affected capacity to regenerate differ [14–17]. Recent innovations in molecular and histological studies have helped in understanding the aetiological mechanisms of the CPT pathogenesis but an appropriate treatment method has not been found. Therefore, surgical interventions with the objective to achieve union, salvage the limb, and improve patients’ quality of life remain the main means.

The two alternative ways of CPT management that yield better outcomes are vascularized fibular autologous grafting and the Ilizarov method [4]. This fact can be explained by the main feature of the techniques: they both use a complex of tissues that preserves blood supply to the nonunion area.

Ilizarov proposed his method for CPT treatment in the early 1970s and described residual angulations and refractures in cases of thinned tibial fragments that happened within five to six months after frame removal [18]. There is no doubt that the Ilizarov method is capable of producing rigid and controlled fixation for compression, distraction, bone transport to the resection defect, lengthening, deformity correction, and union. However, considerable changes in the bone tissue of the tibia make proper consolidation difficult. Therefore, refractures are frequently inevitable [15, 19]. The largest multicentre study of CPT treatment found that the Ilizarov method was the most optimal method for CPT management as far as it provides a comprehensive approach to the pathological complex [3, 4]. It was concluded that segmental bone transport alone resulted in a lower fusion rate when compared with either: (1) simple resection, acute shortening, compression; or (2) resection, acute shortening, metaphyseal lengthening [10].

Our groups matched in CPT types, mean age and number of previous operations per patient

Several technical approaches with the Ilizarov method were used in our series after having considered the individual condition in each of the patients. Unfortunately, the orthopaedic pathology had been worsened by previous operations in the majority of the cases. We did not do extensive resections in the pseudarthrosis area but largely used coaptation of bone fragments and various types of bridging the nonunion with local tissues to thicken the bone and increase the fragments contact area in atrophic cases, corrected deformities and lengthened in the metaphyseal area. In our opinion, bone union and axial alignment are of primary goals. Lengthening can be postponed to the next stage in severe atrophic CPT cases, in particular after multiple previous surgery [19–21]. However, our approaches did not gain the outcomes that would satisfy us as the refracture rate was rather high. Altogether, CPT consolidated in 11 children after the first intervention and in 26 cases out of the total of 28 in the period studied.

It was advised that the surgery for CPT should be avoided before the third year of life [10, 21]. We cannot draw conclusions for outcomes at this early age as far since there were few cases of this age. The risk of re-fracture was significantly higher in children under the age of ten years in cases when the tibial cross-section at the pseudarthrosis was narrow and CPT was associated with persistent fibular pseudarthrosis. We should acknowledge that the results were poor in this age subgroup. We should also admit that patients should be braced for a longer period upon initial union if complete bone remodeling does not happen in the pseudarthrosis area, or even through skeletal maturity [7].

Most of the latest studies agree that the Ilizarov method was more efficient when it was supported by intramedullary nailing [6, 9]. The advantages of both methods such as high union rate with alignment control achieved with the Ilizarov method and protection against refracture provided by intramedullary nailing are combined together. Nevertheless, the reported rates of union and re-fracture differ in the available literature that studied combined fixation [2, 9, 21, 22]. Infected cases have also been reported [2, 9].

A combination of biological and mechanical solutions was suggested for CPT healing which included a free periosteal autologous graft along with bone graft, internal nailing and Ilizarov fixation [7]. The periosteal graft was used as a biological envelope to promote osteogenesis. However, the rate of refractures in that study was also high. It is remarkable that all the patients in our study and in the study mentioned [7] obtained union after the primary operation. It proves the fact that a greater volume of bone tissue that is achieved either by autologous free grafting or local tissues around the nonunion plays a positive role [9]. Unfortunately, it is not sufficient for the resolution of the disease as far as the tibial bone tissue remains pathological.

The search for new optimal methods of improving bone tissue quality in CPT continues [20]. It seems mandatory to study the effects of the Masquelet two-stage membrane technique using an autologous graft and internal fixation as a procedure to be recommended for consideration in achieving and maintaining union in the early age [8, 23], an “unexpected effect” of CPT healing by lengthening over a rod without compression of the nonunion [24], and to test the benefits from regenerative strategies based on mesenchymal stromal cells, platelet-rich fibrin [25], rh-BMP-7 [26, 27], or calcium and vitamin D supplementation [28].

Nevertheless, currently in the hands of clinicians the Ilizarov method alone or supported by IMN and autologous grafting remains a method of choice for CPT treatment in children.

To our knowledge, our study is one of the latest studies of a large group of children with CPT and the first one that compared the results of CPT management with the Ilizarov techniques according to the etiological factor. We have found that the treatment does not result in a significant difference between the group of CPT with NF1 and the group without it.

Conclusion

The unsupported Ilizarov method for management of CPT that is associated with NF1 and CPT of nonspecific or unconfirmed origin yields comparable outcomes: high rate of bone union but high re-fracture rate. More research is necessary to seek optimal solutions to support the Ilizarov method in order to reduce the number of repetitive surgery or eliminate them.