MANAGEMENT OF TRAUMATIC TIBIAL DEFECTS USING FREE VASCULARIZED FIBULA OR ILIZAROV BONE TRANSPORT: A COMPARATIVE STUDY TAREK A. EL-GAMMAL, M.D.,1* ANIS E. SHIHA, M.D.,2 MOHAMED ALAM EL-DEEN, M.D.,2 AMR EL-SAYED, M.D.,1 MOHAMED M. KOTB, M.D.,1 AHMAD I. ADDOSOOKI, M.D.,2 YASSER FAROUK RAGHEB, M.D.,1 and WALEED RIAD SALEH, M.D.1
There are several options for the treatment of traumatic tibial defects. Among these options, free vascularized fibula and Ilizarov bone transport are well-known and effective techniques. The differences between both techniques and the indications for each of them are not well studied in the literature. Between September 1995 and December 2004, Ilizarov bone transport and free vascularized fibula were used to treat 25 traumatic bone. Patients were divided into two groups, Ilizarov group (12) and free fibula group (13). Preoperative data, operative data, duration of treatment, functional outcome, range of motion loss, number of secondary procedures, and type and rate of complications were compared in both groups. All the cases were ultimately united in both groups. Operative time and blood loss were significantly higher in the free fibula group. Although external fixation time was longer in the Ilizarov group, the overall duration of treatment was similar. Range of motion loss was less in the Ilizarov group, but the overall functional outcome was similar. There was no difference in complication rate, or number secondary procedures. Defect size was found to have the most significant effect on the results. Results were much better in the free fibula group when the defect length was 12 cm or more while the results were better in Ilizarov group when the defect length was less than 12 cm. We recommend using free vascularized fibula for traumatic tibial defects of 12 cm or more, whenever C 2008 Wiley-Liss, Inc. Microsurgery 28:339–346, 2008. experience is available. V
Both Ilizarov bone transport and free vascularized fibula have proved successful for the treatment of traumatic tibial bone defects. The advantages of free vascularized fibula include maintained graft vascularity and so ability to hypertrophy in response to load,1 ability to treat concomitant soft tissue defects by osteocutaneous flaps2,3, relatively short external fixation time (EFT) and resistance to infection.2 Its disadvantages are the need for microsurgical skills,2–4 possibility of total necrosis due to anastomotic complications,5 donor site morbidity,2,3,6 and occasional stress fracture.7 On the other hand, Ilizarov bone transport can correct associated deformity8 and shortening, address small areas of soft tissue defects,9 and allow immediate mobilization. Its disadvantages are long duration of treatment especially in long defects10,11 pain accompanying the transport,11 frequency of pin tract infection,10–12 and occasional nonunion at the docking site.10 The only comparison between free fibula and Ilizarov bone transport in the treatment of traumatic tibial bone defect was carried out by Yokoyama et al. 2001.13 Four free fibulas and four callus distractions were performed for traumatic tibial defects at the authors’ institute. They
attempted to discern any differences of results between both techniques. They selected as contributing factors: EFT, complication rate, hospital charge, union rate, and functional score. They concluded that no clear differences between the two treatment groups could be determined and that more cases are needed to establish statistically significant differences between both methods. We think that the small number of patients and the statistically significant difference between the defect lengths in both groups make the results inconclusive. Comparison between the published results of using the free fibula or Ilizarov bone transport in traumatic tibial bone defects is not possible because of the different preoperative data (patients’ age, defect length, and duration of trauma), lack of enough follow-up data [functional outcome, range of motion loss (ROML), FWBT, etc], and using different evaluation systems. In this study, both techniques were compared as regards preoperative data, operative data, duration of treatment, functional outcome, ROML, number of secondary procedures and type and rate of complications. The indications for the use of either technique were outlined. MATERIALS AND METHODS
1
Department of Orthopedics and Traumatology, Reconstructive Microsurgery Unit, Assiut University School of Medicine, Assiut, Egypt 2 Department of Orthopedics, Sohag University, Sohag, Egypt *Correspondence to: Tarek Abdalla El-Gammal, M.D., Reconstructive Microsurgery Unit, Assiut University Hospitals and School of Medicine, Assiut 71526, Egypt. E-mail:
[email protected] Received 2 January 2008; Accepted 12 February 2008 Published online 6 June 2008 in Wiley InterScience (www.interscience.wiley. com). DOI 10.1002/micr.20501 C V
2008 Wiley-Liss, Inc.
Between September 1995 and December 2004, Ilizarov bone transport or free fibula were used to treat 25 traumatic bone defects in the Department of Orthopedics and Traumatology, Sohag University Hospital and Reconstructive Microsurgery Unit, Assiut University Hospitals. The inclusion criterion is traumatic tibial bone defect. The exclusion criteria are: (1) Defects less than 6 cm or more than 25 cm. (2) Defects involving the knee or ankle
340
El-Gammal et al. Table 1. Preoperative Data
Age (years) Sex (male/female) Duration of trauma (months) Previous operationsa Defect length (cm)
Free fibula
Ilizarov
31.5 11/2 4.7 2.2 12.6
32.7 11/1 12 2.3 9.7
a Previous operations include debridement, removal of hardware, and external fixation.
joint. (3) Partial circumference defects. (4) Patient’s age below 10 years or above 50 years. The patients were divided into two groups, the Ilizarov group and the free fibula group. The choice of either the free fibula or Ilizarov bone transport was made according to available experience and not according to specific criteria; i.e., all cases operated at the first center were treated with bone transport, while all cases operated at the second center were treated with vascularized fibula. Table 1 summarizes the preoperative data. In the Ilizarov group, follow-up ranged from 1.7 to 8.1 years (mean 4.3 6 2.2 years). X-rays were done every 2–3 weeks during the transport phase and every 3–4 weeks during the consolidation phase. At every visit patients were evaluated for knee and ankle range of motion, and pin tract infection. In the free fibula group, follow-up ranged from 1.7 to 7.3 years (mean 3.2 6 1.8 years). Patients had clinical follow-up with or without radiological review at monthly intervals for the first 6 months, 2-monthly intervals for the rest of the first year, and then 3-monthly intervals for the second year. Patients who had a follow-up more than 2 years were reviewed every 6 months. Nonweight bearing crutch walking was allowed 2–3 weeks after the operation. Partial weight bearing began after radiological union of the graft. The allowance of full weight bearing was based on individual criteria according to age and weight of the patient, method of fixation, and signs of graft hypertrophy. Complications in each group were divided into minor, moderate, and major complications, Table 2. Minor complications are the complications that needed no operative treatment (e.g., pin tract infection). Moderate complications are the complications that needed operative treatment (e.g., nonunion). Major complications are the residual complications that could not be corrected (e.g., residual shortening and joint contracture). After the end of follow-up, the patients of both groups were evaluated clinically and radiologically using a simple and complete seven-scale scoring system proposed by Puno et al., designed for assessing functional outcome of severe open fractures of the tibia.14 The results of both techniques were compared as regards operative time, blood loss, EFT, external fixation index (EFI)10 which is EFT divided by defect length, full Microsurgery DOI 10.1002/micr
weight bearing time (FWBT) which is time from index operation until full weight bearing, full weight bearing index (FWBI) which is FWBT divided by the defect length, knee, ankle, and ROML, functional outcome using Puno’s criteria, and complications. EFT and FWBT, rather than union time, were used for comparison as they are more representative of complete healing especially in the free fibula group where union usually occurs much earlier before full weight bearing and removal of external fixation are possible. EFI and FWBI were used to neutralize the defect length factor as they measure the external fixation and FWBTs for every 1 cm of defect length. The effect of preoperative data (age, duration of trauma, number of previous operations, and defect length) on the results was studied in each group. Mann-Whitney and Spearman correlation tests were used for data analysis. The significance level was set at P 0.05. RESULTS
Table 3 shows the comparative results. Operative time and blood loss were significantly higher in the free fibula group than the Ilizarov group. EFT and EFI was significantly lower in the free fibula group compared to the Ilizarov group. FWBT was not significantly different in both groups, while FWBI was significantly lower in the free fibula group. There was no significant difference between both groups as regards number of secondary procedures, functional score, functional grade, ROML or number and quality of complications. The most common complication in the Ilizarov group was superficial pin tract infection (10 cases) representing nearly one third (28.6%) of a total 35 complications in this group. The most common complication in the free fibula group was stress fracture (six cases) representing about one forth (24%) of a total 24 complications in this group. Effect of the Preoperative Data on the Results
Table 4 shows the correlation between the preoperative data and the results in both groups. Age did not correlate with any of the results. The defect length had the most significant effect on the results especially in the Ilizarov group where it correlated positively with blood loss, ROML, external fixation and FWBT, complication rate and number of secondary procedures. In the free fibula group, the defect length correlated negatively with external fixation and full weight bearing indices. When the defect length is plotted against the complication rate and total ROML, there was a gradual linear increase of both complication rate and total ROM loss as the defect length increased up to 12 cm, after which there was a sharp rise (Fig. 1) therefore, patients were subdivided, according to whether their defect length is < or 12 cm
Management of Traumatic Tibial Defects
341
Table 2. Complications Complications
Free Fibula
Ilizarov
Minor
Superficial infection Exposure of peroneal tendons
Moderate
Pin tract infection G 1,2a Stress fracture Non-union Venous congestion
Major
Knee flexion contracture >10 Ankle fusion Varus deformity Shortening
Superficial infection Transient knee flexion contracture Mal-alignment of transported segment Delayed maturation of regenerate Pin tract infection G 1,2a Stress fracture Soft tissue interposition Malreduction at docking site Partial frame failure Knee flexion contracture 10 Loss of ankle motion 20 Mal-alignment 5
a
Pin tract infection G1 5 superficial infection, G2 5 deep infection not reaching bone, G3 5 bone infection.
Table 3. The Overall Comparison Results Item Operative Data Operative time (hours) Blood loss (ml) Hospital stay (days) Healing Data EFT (months) EFI (month/cm) FWB time (months) FWB index (months/cm) Secondary procedures Functional Outcome Puno functional score Freedom from pain Restoration of function Range of motion loss (knee and ankle) Residual deformity X-ray changes (AP and lateral) Sensation Muscle strength Total Puno functional grade Excellent Good Fair Poor ROM loss Knee ROM loss Ankle ROM loss Total range of motion loss Complications Minor Moderate Major Total
Ilizarov 1.23 6 0.21 169.17 6 30.29 28.08 6 23.48 10.58 1.14 11.88 1.28 1.08
Significance
10.54 6 1.14 846.15 6 199.44 17.92 6 5.01
P < 0.001 P < 0.001 NS
6 6 6 6 6
P 5 0.018 P 5 0.003 NS P 5 0.014 NS
4.81 0.41 5.34 0.47 1
6.92 0.65 9.04 0.86 0.85
1.81 0.33 2.83 0.42 0.8
12.25 6 2.38 12.25 6 4.33
12.69 6 2.18 12.69 6 2.18
NS NS
12.50 6 2.81 14.00 6 1.95
8.31 6 5.22 12.69 6 4.09
P 5 0.029 NS
6 6 6 6
NS NS NS NS
12.58 8.17 12.25 84.00 6 2 2 2
18 11 6 35
6 6 6 6 6
Free fibula
6 6 6 6
3.09 2.95 4.33 15.86
11.08 9.31 14.08 80.85
5.11 1.32 1.44 16.32
(50%) (16.7%) (16.7%) (16.7%)
5 (38%) 4 (31%) 1 (8%) 3 (23%)
8.75 6 15.97 6.67 6 14.67 15.42 6 28.08
11.54 6 14.05 20 6 18.82 31.54 6 30.58
NS NS NS
6 6 6 6
NS NS NS NS
(1.50 (0.92 (0.50 (2.92
6 6 6 6
0.67) 0.90) 0.80) 2.11)
into group A or B, respectively. The results of Ilizarov bone transport and free fibula were compared in each group. Table 5 shows the results of comparison.
13 5 7 25
(1.00 (0.38 (0.54 (1.92
1.00) 0.51) 0.66) 1.19)
Operative time and blood loss were significantly higher using free fibula in both groups A and B. External fixation and FWBT, and indices were not significantly Microsurgery DOI 10.1002/micr
342
El-Gammal et al. Table 4. Effect of the Preoperative Patients’ Data
Group Ilizarov group
Number of previous operations
Age
Operative time Blood loss Treatment duration
NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
NS NS NS 0.85 NS 0.87 NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
NS NS NS 0.84 NS 0.76 NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
NS 0.69 0.83 NS 0.78 NS NS NS 20.75 20.62 NS NS 20.58 NS 0.85 0.65 0.68 0.87 0.83 0.84 0.94 0.81
NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
NS NS NS 20.75 NS 20.57 NS NS 0.70 NS NS NS NS NS NS 0.61 0.60 NS NS NS NS NS
NS NS NS NS 0.64 NS NS 20.69 NS 20.68 NS NS 20.73 20.64 NS NS NS NS NS NS NS NS
NS NS NS 20.88 NS 20.79 NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
Functional score
ROML
Complications
Free fibula group
Duration of trauma
Results’ items
EFT EFI FWBT FWBI Pain Function ROM Deformity X-ray Sensation Muscle strength Total score Knee Ankle Total Minor Moderate Major Total
Number of secondary operations Operative time Blood loss Treatment duration
EFT EFI FWBT FWBI Functional score Pain Function ROM Deformity X-ray Sensation Muscle strength Total score ROML Knee Ankle Total Complications Minor Moderate Major Total Number of secondary operations
Defect length
NS, not significant, the numbers are the values of correlation coefficient of significant correlation; EFT, external fixation time; EFI, external fixation index; FWBT, full weight bearing time; FWBI, full weight bearing index; ROM, range of motion; ROML, range of motion loss.
different between Ilizarov bone transport and free fibula in group A. In group B, EFT and full weigh bearing and external fixation indices were significantly higher using Ilizarov bone transport. Results did not differ, qualitatively, between Ilizarov bone transport and free fibula in group A. However, in group B, the results of the free fibula were better than that of Ilizarov bone transport. The percents of excellent and good results of Ilizarov bone transport and free fibula Microsurgery DOI 10.1002/micr
were 25 and 50%, respectively, and the percents of fair and poor results were 75 and 50%, respectively. Total ROM loss was significantly lower using Ilizarov bone transport (mean 21.88 6 8.84) versus free fibula (mean, 23.57 6 38.70) in group A, while in group B there was no significant difference. Complication rate was not significantly different, in group A, using either technique. However, qualitatively, major complications were significantly higher using free
Management of Traumatic Tibial Defects
Figure 1. defect length plotted against the total range of motion loss (TROML) (A), and complication rate (B) showing that complication rate and range of motion loss increase acutely when the defect length reached 12 cm.
fibula (mean, 0.57 6 0.53) versus Ilizarov bone transport (mean, 0.00). In group B, complication rate was significantly higher using Ilizarov bone transport (5.5 6 1.29) versus free fibula (1.67 6 1.21). Qualitative comparison also revealed that minor and moderate complications were significantly higher using Ilizarov bone transport. Number of secondary procedures was not significantly different between Ilizarov bone transport and free fibula in both groups A and B. Representative cases of both techniques are shown in Figures 2 and 3. DISCUSSION
In this study, there were no statistically significant differences between Ilizarov bone transport and free fibula groups as regards patient age, sex, defect length, duration of trauma, and number of previous operations (Table 1). The overall treatment duration indicated by FWBT was not significantly different between both techniques. Although the EFT was significantly longer in the Ilizarov group than in the free fibula group, the EFT is not an actual indicator of healing in the free fibula group because of long duration required between removal of external fixation and the start of full weight bearing. While the healing indices (treatment duration for every cm of defect length) indicated by EFI and FWBI were significantly shorter in the free fibula group than in the
343
Ilizarov group. This can be explained by the fact that, in the Ilizarov cases, the transport/consolidation time, and hence the EFT and FWBT, were increased by increasing defect length, while this was not the case in free fibula cases, where union and subsequent hypertrophy are not affected by the defect length. In a trial to shorten the duration of transport/consolidation and EFT in long defects, Paley and Maar10 used double level transport (trifocal) in 6 cases of 19 cases with a mean defect length of 15 cm and one level transport (bifocal) in 13 cases with a mean defect length of 7.6 cm. The EFI were significantly lower in double-level transport (mean, 1.2 months) compared with single-level transport (mean, 2.1 months). Using Puno et al. criteria,14 the mean overall functional score in the Ilizarov group and the free fibula group was not statistically significant different. There was no statistically significant difference between individual items of the scoring system in both groups except for the range of motion item, where the score was significantly higher in the Ilizarov group. This can be explained by the fact that three cases in the free fibula group underwent ankle arthrodesis (Table 3). The increased ROM loss in free fibula group might result from the use of cast as a main method of fixation in two cases. The knee ROM loss, ankle ROM loss, and total ROM loss were higher in the free fibula group than in the Ilizarov group, though not statistically significant. The overall higher ROM loss in the free fibula group can be explained by the fact that patients in the Ilizarov group were allowed immediate mobilization and early partial weight bearing while in the free fibula group, mobilization was delayed until complete union of the graft and full weight bearing was delayed until adequate hypertrophy of the transferred fibula. To solve the problem of long standing immobilization and non-weight bearing, Banic and Hertel15 used double vascularized fibular grafts as a double strut to increase its mechanical strength. Complications did not differ either quantitatively or qualitatively between both groups. Stress fracture was the most frequent complication in free fibula group. El-Gammal et al.3 reported the use or rigid internal fixation (bridging plate or intramedullary nail) in 9 out of 11 cases of free fibula for lower limb defects. Only, one case was complicated by stress fracture following plate removal. They used rigid internal fixation for defects resulting from tumor resection, but in traumatic tibial defects the presence of infection may limit the use of such rigid internal fixation. Baumgart et al. in 1997 and 199916,17 reported implantable intramedullary nail for bone transport to treat bone defects. This method of bone transport was not associated with the problems of troublesome pain accompanying bone transport, and pin tract infection which is the most common complication in the Ilizarov group. Microsurgery DOI 10.1002/micr
344
El-Gammal et al. Table 5. Comparison of the Results Depending on the Defect Length
Group Defect length < 12 cm
Defect length 12 cm
Item Defect length (cm) Operative time (hours) Blood loss (ml) EFT (moths) EFI (months/cm) FWBT (months) FWBI (months/cm) Total ROM Loss Secondary procedures Functional score (points) Functional results Excellent Good Fair Poor Complications Minor Moderate Major Total complications Secondary procedures Defect length (cm) Operative time (hrs) Blood loss (ml) EFT (moths) EFI (months/cm) FWB time (months) FWB index (months/cm) Total ROM Loss Secondary procedures Functional score (points) Functional results Excellent Good Fair Poor Complications Minor Moderate Major Total Secondary procedures
Ilizarov 6.94 1.18 158.75 8.75 1.24 9.94 1.40 21.88 0.63 87.50
6 6 6 6 6 6 6 6 6 6
0.68 0.22 27.48 3.96 0.44 4.52 0.51 8.84 0.52 15.75
5 (63.5%) 2 (25%) 0 1 (12.5%) 1.13 6 0.50 6 0.00 1.63 6 0.63 6 15.25 6 1.3325 6 190 6 14.25 6 0.94 6 15.75 6 1.04 6 50 6 26 77 6
0.35 0.53 0.74 0.52 3.40 0.16 23.45 4.65 0.27 5.19 0.30 17.8 1.15 15.68
1 (25%) 0 2 (50%) 1 (25%) 2.25 1.75 1.50 5.50 2.00
6 6 6 6 6
0.50 0.96 0.58 1.0 1.15
Free fibula
Significance
6 6 6 6 6 6 6 6 6 6
NS P 5 0.001 P 5 0.001 NS NS NS NS P 5 0.019 NS NS NS NS NS NS NS
8.71 10.43 914.29 7.14 0.87 8.93 1.08 23.57 0.71 85.00
1.98 1.27 226.78 0.75 0.28 1.24 0.32 38.70 0.49 14.07
3 (43%) 3 (43%) 0 1 (14%) 1.14 0.43 0.75 2.14 0.71 17.17 10.67 766.67 6.67 0.39 9.17 0.55 40.83 1.00 76.0 2 1 1 2
6 6 6 6 6 6 6 6 6 6 6 6 6 6 6
1.21 0.53 0.53 1.21 0.49 3.66 1.08 140.24 2.66 0.11 4.17 0.25 15.94 1.10 18.69
P P P P P
NS NS 0.016 NS NS NS 5 0.010 5 0.010 5 0.030 5 0.011 NS 5 0.042 NS NS NS
(33%) (17%) (17%) (33%)
0.83 0.33 0.50 1.67 1.10
6 6 6 6 6
0.75 0.52 0.84 1.21 0.45
0.019 0.023 NS 0.011 NS
EFT, external fixation time; EFI, external fixation index; FWBI, full weight bearing index; ROM, range of motion; NS, not significant.
The defect length was found to be the most important factor that affected the results. There was no difference in the EFT, EFI, FWBT, and FWBI when the defect length was
