Clinical and radiological follow-up after free vascularised fibula transplantation for bone defects of the upper extremity with MRI and MR-angiography


 Background Vascular patency of free vascularised fibula grafts can be postoperatively assessed by island flap which is not always possible or angiography which is invasive. Bone healing is examined based on radiographs with scoring systems. We present data on MRI with contrast agent and a comparison of the Giessler bone healing score by X-ray and MRI and clinical scores in patients with operation on the upper extremity. Methods We reviewed the clinical and radiological outcome of 13 patients with a follow-up of at least 1 year after free vascularised fibula graft of the upper extremity. The examination included the DASH, MSTS, SF-36, Rosén-Lundberg-scores and MRI with contrast agent. We determined the Giessler bone healing score by most recent x-ray and MRI and the Weiland score and its modification for the upper extremity. For statistics, we used the Wilcoxon test and Pearson correlation. Results Vascular patency was directly or indirectly detectable all cases in the MRI with contrast agent. 5 patients did not return to work and 5 patients had ongoing pain medication. Median DASH was 21.7, MSTS of the upper extremity 76.7, of the lower extremity 93.3, the Rosén-Lundberg score for median nerve 2.43, for ulnar nerve 2.32. Giessler score for x-ray and MRI showed no significant difference for proximal or distal junction. Conclusions We could prove with MRI that it was indeed a vascularised graft and its vitality at time of follow-up by assessment of vascular patency, and that MRI can be used to determine the Giessler score similarly to X-ray.


Abstract
Background Vascular patency of free vascularised fibula grafts can be postoperatively assessed by island flap which is not always possible or angiography which is invasive. Bone healing is examined based on radiographs with scoring systems. We present data on MRI with contrast agent and a comparison of the Giessler bone healing score by X-ray and MRI and clinical scores in patients with operation on the upper extremity.
Methods We reviewed the clinical and radiological outcome of 13 patients with a follow-up of at least 1 year after free vascularised fibula graft of the upper extremity. The examination included the DASH, MSTS, SF-36, Rosén-Lundberg-scores and MRI with contrast agent. We determined the Giessler bone healing score by most recent x-ray and MRI and the Weiland score and its modification for the upper extremity. For statistics, we used the Wilcoxon test and Pearson correlation.
Results Vascular patency was directly or indirectly detectable all cases in the MRI with contrast agent. 5 patients did not return to work and 5 patients had ongoing pain medication. Median DASH was 21.7, MSTS of the upper extremity 76.7, of the lower extremity 93.3, the Rosén-Lundberg score for median nerve 2.43, for ulnar nerve 2.32. Giessler score for x-ray and MRI showed no significant difference for proximal or distal junction.
Conclusions We could prove with MRI that it was indeed a vascularised graft and its vitality at time of follow-up by assessment of vascular patency, and that MRI can be used to determine the Giessler score similarly to X-ray.

Background
After surgical treatment of osteomyelitis, bone tumours or pseudarthrosis, the resulting bone defect may be filled with a vascularised fibula graft (1,2). This has been shown to be effective for defects of the femoral shaft (3) and after resection of giant cell tumour of the distal radius (4). But compared to the lower limb or shoulder, there less information on free vascularised fibula grafts of the forearm (5,6).
The value of preoperative MRI imaging for identification of the peroneal artery has been shown (7).
The postoperative controls usually include plain X-ray for determination of bony consolidation and hypertrophy (7). Postoperative control of vascular patency can be performed by an indicator flap (8) or by angiography.

Methods
We searched our electronic database for patients who received a free vascularised fibula bone graft for the upper extremity without shoulder between January 2002 and September 2018 in the Universitätsmedizin Greifswald (Germany) and Unfallkrankenhaus Berlin (Germany) and invited them for follow-up with MR-angiography.
We documented sex, age at follow-up, date of first operation, fibula graft operation and follow-up, indication for fibular graft transplantation, duration of hospitalization, side of operation, hand dominance, complications, occupational status, current pain medication, graft receiving area, donor side, length of graft, type of vascular connection, and type of additional operative procedures.
Clinical examination included range of motion and items for the Rosén-Lundborg-Score (table 1) (9).
We averaged force grip (Jamar dynamometer), two-point pinch force between tip of thumb and index (Pinchmeter) from three trials alternating between operated and contralateral hand as part of the Rosén-Lundborg-score.
Most recent X-ray images and MRI with and without contrast agent were assessed for bony consolidation, calculation of the bone hypertrophy index after de Boer and Wood, and bone healing using the Giessler modification of the scale published by Taira, the original Weiland score and its modification by Höpfner (14)(15)(16)(17)(18). Clinical function comparison was performed using the contralateral hand as control. Here, we present data only for patients with a follow-up of at least 1 year but show imaging of a patient with more recent MRI. Values were rounded according to the relevance of decimals. Due to the small sample size, we used the Wilcoxon-test for comparison of paired quantitative variables. The Pearson correlation coefficient was used to show the degree of linear correlation. Significant differences were assumed for p < = 0.05.

Results
We identified 24 patients who received a vascularised fibula graft for the upper extremity. Eleven patients were lost to follow-up. Of the presenting 13 patients, two patients had a contraindication for MRI diagnostics. Two patients had claustrophobia, resulting in 9 MRI examinations. Table 2 and 3 show information on the nine male and four female patients. Table 4

Discussion
We could show direct or indirect patency of the vessels supplying the fibula grafts in MRI. Therefore, we can conclude that it was indeed a vascularised graft in all cases. The advantages over nonvascularised grafts are assumed to include more rapid healing, reduction of non-union, fatigue fracture and infection (17,19).
For postoperative monitoring of the vascularity, scintigraphy has been used but may give false-positive results (20)(21)(22). Angiography or indicator flaps are other possible methods (8,23). In cases of arterial interposition, doppler analysis of blood flow distally would prove patency of the anastomosis (24). Biopsy of live osteocytes is an invasive method that shows viability of the graft (22,25).
In a postoperative angiographic control of vascular patency at 13.9 weeks, 3 out of 74 patients showed a occlusion of the anastomosis (26). Angiography up to six weeks postoperatively were performed by Moore and Weiland but not continued due to potential morbidity (19,25). Additionally, the radiation dose would not permit frequent repeats of the angiography.
Viability of the graft is assumed in cases with evident union (27). It is assumed that bony union will occur within 12 months when vascularity is retained (28). But this was not yet proven. In children, failure of revascularization would lead to premature fusion of the growth plate (29). Complications can range up to 52 % in fibula grafting for tumour resection (30,31).
The low number of hypertrophies in the upper extremity is consistent with previous findings. In a study with 21 humeral and 12 forearm reconstructions, no bone hypertrophy was seen after one year follow up (26). Solid consolidation may not always appear at one junction which was explained by occlusion of blood outflow, which convert the fibula into an end-organ (24,26,27). Sparks et al.
discussed that a fibula, that was osteotomized on several sides and lost its periosteum, may show decreased or arrested perfusion, resulting in resorption and patchy necrosis (32).
Weakness of extensor hallucis longus was reported as intermittent donor site complication (4, 33), but did not appear in our series.
In a study with 32 patients who had an average defect of 12 cm with a follow-up of 5 months to 14.6 years, six received a fibular graft of the radius and three of the humerus (33). They found a hypertrophy in three out of six fibulas in the upper limb and a not significant lower degree of 6 hypertrophy compared to the lower limb. That study did not report functional outcomes, and complications were given in total. But results can vary a lot regarding clinical outcome and bony union (34).

Consent for publication
We obtained written consent from all patients who presented for follow-up for publication of research data, including radiographic data.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no conflict of interest.    Patient 2 and 3 received the fibula graft for the construction of a one bone forearm. For motor function, patient 6 and 12 were excluded as one has a wrist arthrodesis and the other a fibula graft operation for defect of the humerus. Table 5 Radiographic scores