A biomechanical study of proximal tibia bone grafting through the lateral approach

A biomechanical study of proximal tibia bone grafting through the lateral approach

G Model JINJ 6903 No. of Pages 8 Injury, Int. J. Care Injured xxx (2016) xxx–xxx Contents lists available at ScienceDirect Injury journal homepage:...

1MB Sizes 21 Downloads 79 Views

G Model JINJ 6903 No. of Pages 8

Injury, Int. J. Care Injured xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Injury journal homepage: www.elsevier.com/locate/injury

A biomechanical study of proximal tibia bone grafting through the lateral approach Chin Tat Lim, MBBS, MMed (Orth,) MRCS, FRCSa,c,* ,1, David Q.K. Ng, M. Enga,b,1, Ken Jin Tan, MBBS, MMed (Orth), MRCS, FRCSc,1, Amit K. Ramruttun, M.Scc,1, Wilson Wang, MBBS, MRCS, FRCS, PhDa,c,1, Desmond Y.R. Chong, PhDb,d,2 a

Department of Orthopaedic Surgery, University Orthopaedics, Hand and Reconstructive Microsurgery Cluster, National University Hospital, Singapore Department of Biomedical Engineering, National University of Singapore, Singapore c Department of Orthopaedics, Yong Loo Lin School of Medicine, National University Health System, Singapore d Engineering Design and Innovation Centre, National University of Singapore, Singapore b

A R T I C L E I N F O

Keywords: Tibia fracture Bone graft Post operative care Injury Prevention Biomechanics

A B S T R A C T

Background: Autologous bone graft remains the gold standard source of bone graft. Iliac crest has traditionally been the most popular source for autologous bone graft. However, iliac crest bone graft harvesting is associated with high donor site morbidity. Bone graft harvesting from the proximal tibia has shown great potential with reported low complication rates. However, there is a paucity of biomechanical studies concerning the safety as well as yield of proximal bone graft harvesting. Purpose: This biomechanical study was designed to investigate (1) the stability of the harvested proximal tibial during physiological loading, and (2) the maximum size of the cortical window that can be safely created and (3) volume of accessible bone graft. Methods: Bone grafts were harvested from eleven cadaveric tibiae using a circular cortical window along the lateral proximal tibia. These harvested proximal tibiae were then loaded under physiological conditions (mean 2320N, range 1650–3120N) using a customized test fixture. Strain rosettes were mounted at 7 locations in the harvested proximal tibia to record the changes in strain at the harvested proximal tibia. The change in strain with increasing cortical window size (10–25 mm diameter) was also studied. Bone principal strains as well as volume of bone harvested were recorded. Results: A repeated measures ANOVA was used to analyze the change in bone strains with the cortical window size. Statistically significant (p < 0.05) increases in bone strains at the anterior and medial aspects of the tibia were observed with increasing size of osteotomies ( 328.85 me, SD = 232.21 to 964.78 me, SD = 535.89 and 361.64 me, SD = 229.90 to 486.08 me, SD = 270.40 respectively), and marginally significant changes in strain at the lateral and posterior aspects. None of the tibiae failed under normal walking loads even with increasing osteotomies size of 10–25 mm diameter. A smaller osteotomy of 10 mm diameter yielded an average volume of 7.15 ml of compressed bone graft, while a larger osteotomy of 25 mm diameter yielded on average an additional 3.64 ml of bone graft. Bone grafting of the proximal tibia through the lateral approach with a circular osteotomy is a feasible option even with osteotomies of 25 mm diameter. Even though increased bone strains were observed, the strains did not exceed the yield strain of cortical bone when loaded under normal walking conditions. The quantity of bone harvested from the proximal tibia is comparable to that harvested from the iliac crest. Conclusions: This biomechanical study demonstrated the stability of the harvested proximal tibia under conditions of full weight bearing ambulation. It has also refined the technique of proximal bone graft harvesting by determining the maximum size of the cortical window. The findings of this study add to the overall understanding of proximal tibial bone graft harvesting, providing objective data regarding stability as well as yield. This information would be useful during selection of source of autologous bone graft. ã 2016 Elsevier Ltd. All rights reserved.

* Corresponding author. E-mail addresses: [email protected] (C.T. Lim), [email protected] (D.Q.K. Ng), [email protected] (K.J. Tan), [email protected] (A.K. Ramruttun), [email protected] (W. Wang), [email protected] (D.Y.R. Chong). 1 NUHS Tower Block, Department of Orthopaedic Surgery, Level 11, 1E Kent Ridge Road, 119228, Singapore. 2 Block E4, #04-08, 4 Engineering Drive 3, 117583, Singapore. http://dx.doi.org/10.1016/j.injury.2016.09.017 0020-1383/ ã 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: C.T. Lim, et al., A biomechanical study of proximal tibia bone grafting through the lateral approach, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.09.017

G Model JINJ 6903 No. of Pages 8

2

C.T. Lim et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx

Introduction 2.2 million bone grafts are performed each year by different specialties for a wide range of indications ranging from oral and maxillofacial reconstruction to revision arthroplasty surgeries [1]. While several synthetic bone grafts have become available over recent years, autogenous bone grafts remain gold standard as it exhibits the desirable qualities of osteoinduction, osteoconduction, and osteogenesis with minimal adverse reactions [2]. The illiac crest is the most popular site for autogenous bone graft harvesting, but reports show that complication rates and donor site morbidity remain high [3]. The proximal tibia is an alternative site for autogenous corticocancellous bone graft harvesting. Compared to iliac crest bone graft harvesting, it has a low complication rate with a comparable quantity of corticocancellous bone available [1–3]. Numerous surgical techniques exist for this procedure. They differ mainly in the approach (lateral or medial), size and shape of cortical window osteotomy created to harvest the cancellous bone [2–9]. The more widely accepted method is the lateral approach with a circular osteotomy to reduce the formation of stress risers in the tibia [2–4]. No significant differences in the amount of bone graft harvested were reported between the medial and lateral approach [7]. The amount of bone graft that can be harvested is largely related to the size of osteotomy introduced. Most authors suggest an osteotomy of approximately 1 cm2 or a 10 mm diameter circular osteotomy for harvesting [4,8]. Zouhary et al. recommended a bigger osteotomy window of between 1 cm–2 cm2. The ideal technique of proximal tibia bone graft harvesting which minimize instability and maximizes the amount of bone graft harvested remains inconclusive. Long term clinical studies on proximal tibia bone graft harvesting have reported a relatively low rate (0.5–2%) of serious complications [2,3]. The largest reported clinical study involving 230 proximal tibiae showed that the procedure yielded good results for patients that are limited to non-weight bearing for a period of six weeks, with only 1 patient out of 230 cases developing a fracture and 2 others developing other serious complications [10]. Kim et al. found 4 cases of fracture out of 105 cases for patients who were not restricted in weight bearing [11]. Weight bearing appears to increase the complication rate of this procedure. However, these reports include study populations that are heterogeous and make it difficult to apply to clinical practice. Well conducted biomechanical studies could demonstrate the safety of proximal tibial bone graft harvesting. In vitro biomechanical studies allow us to standardize the procedure and evaluate the effect of loading without concern of patient’s harm. However, there is a paucity of biomechanical studies on the harvested tibia. Alt

et al. [12] and Vittayakittipong et al. [13] both independently found that the mean maximal compressive axial strength of the tibia was not significantly reduced between intact and harvested conditions using the lateral and medial approaches respectively. However, Gerressen et al. had contradictory findings when a different surgical technique with a larger osteotomy was used [14]. These results suggested that the outcome of the procedure is affected by surgical technique and the size of osteotomy introduced. While a greater quantity of bone graft may be harvested with a larger cortical bone osteotomy, the maximal size of the cortical window that can be safely created has not been determined. Bottlang found that variations in the location of the osteotomy and the size of the osteotomy affected tibial plateau deformation, but did not make conclusion about tibia stability [15]. The safety of proximal tibia bone graft harvesting needs to be further defined. The primary aim of this study was to determine if bone graft can be harvested from the proximal tibia without compromising the mechanical stability of the tibia and to determine the maximum feasible size of the cortical window that can be created while maintaining the stability of the tibia. The secondary aim was to correlate the amount of bone graft that can be harvested with the size of osteotomy introduced. This study adopted the more widely used lateral approach with a circular osteotomy as described by Myeroff et al. [3]. Materials and methods Specimen preparation 12 fresh cadaveric tibiae with a mean age of 61 years old (range 36–87) were used for this study and selected in accordance with national and institutional ethical standards. There were 8 male cadavers and 4 female cadavers (Table 1). Records were checked to ensure that there was no history of bone disease. DEXA scans were performed on the proximal tibiae to determine their bone quality. The bone quality was evaluated according to the reference tibia Bone Mineral Density (BMD) values as established by Popp et al. [16]. The proximal 20 cm of each tibia was dissected, keeping the menisci intact. The tibiae were then cleaned with ethanol and acetone. A set of reference axes as described by Ruff and Hayes [17] was marked in indelible ink on the surface of the tibia. The anterior-posterior and medio-lateral dimensions of the tibial plateau are measured with a set of vernier calipers. 5 tri-axial stacked strain rosettes and 2 uniaxial train gauges (models FRA-5-11-3L and FLA-5-11-1L, TML, Tokyo, Japan) were mounted at 7 locations on the tibia with cyanoacrylate cement as indicated in Fig. 1. These locations were chosen to document strain changes at various anatomical locations including the osteotomy

Table 1 Volumes of Bone Graft Harvested. Sex

Age BMD (g/ cm2)

A: Size of Tibial Plateau

Uncompressed Bone Volume at 10 mm window (ml)

B:Compressed Bone Volume at 10 mm window (ml)

B/A (ml/ cm2)

Uncompressed Bone Volume at 25 mm window(ml)

Compressed Bone Volume at 25 mm window (ml)

C: Total Compressed Volume (ml)

C/A (ml/ cm2)

M F M F M M F M M M M

77 69 59 48 61 49 59 54 36 75 87

37.63 26.81 33.09 30.28 33.97 28.64 23.97 32.45 35.28 34.09 28.67

25 8.5 12 11 17.9 16.4 17.9 16.7 18.4 12.9 17.3

10.4 3 7.4 5.4 10.1 6.6 6 7.2 7.6 7 8

0.276 0.112 0.224 0.178 0.297 0.23 0.25 0.222 0.215 0.205 0.279

9.2 5.2 10.3 5 2 6.8 8.9 5.3 19.1 4.4 11

4.2 2.6 5.8 1.7 1.8 2.2 2.9 2.2 7.6 3.2 5.8

14.6 5.6 13.2 7.1 11.9 8.8 8.9 9.4 15.2 10.2 13.8

0.388 0.209 0.399 0.234 0.35 0.307 0.371 0.290 0.431 0.299 0.481

0.812 0.521 0.618 0.581 0.776 0.727 0.593 0.729 0.795 0.792 0.698

Please cite this article in press as: C.T. Lim, et al., A biomechanical study of proximal tibia bone grafting through the lateral approach, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.09.017

G Model JINJ 6903 No. of Pages 8

C.T. Lim et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx

3

site, and the tibia plateau. Each tibia was affixed to a stainless steel pot with dental cement (Meliodent, Heraeus Kulzer, Germany) and load tested by a servohydraulic materials test system (MTS 858 Mini Bionix, MTS systems, Eden Prarie,MN). Load testing The loading system consisted of a custom designed fixture incorporating a femoral knee arthroplasty component (Gender Solutions Natural Knee Flex System, Zimmer, Warsaw, USA) (Fig. 2). The loading was designed to replicate physiological loading conditions of 60–40% medio-lateral tibia plateau load distribution. The loading jig also allows controlled tri-axial rotation to account for anatomical differences in the cadavers. Several sizes of the femoral arthroplasty component were available for use, allowing matching of the of the tibia specimen size. The femoral component was aligned to the meniscus of the tibia to ensure an accurate loading area. The meniscii were then removed. A pressure sensor (Model K-Scan 4000CR, Tekscan Inc, Boston, USA) was used to confirm the physiological load distribution across the tibial plateau (Fig. 2). A maximum load condition of 3 times body weight was applied in the axial direction to replicate normal loading conditions at the knee joint during normal walking. A preload of 25N was applied and increased to 3 times body weight or failure of the specimen was reached. This maximum force was held on the specimen for a period of 10 s before unloading. Strain values were recorded throughout the load cycle at a rate of 1 Hz by a data logger (TMS-530, TML, Tokyo, Japan), while load-displacement readings were recorded by the MTS Bionix system at 2 Hz. Failure of the specimen was defined by a sudden, marked decrease in the force-displacement curve or any visible fracture of the specimen. Each test was conducted three times with an interval of five minutes allowing for strain relaxation at room temperature (25  C). The specimen was hydrated regularly with saline solution.

Fig. 2. Tibia specimen loaded in a custom fixture with a Tekscan 4000CR (red arrow) and strain rosettes attached. The femoral arthroplasty component and stainless steel pot are marked with a yellow and blue arrow respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Bone graft harvesting and subsequent load testing After each intact tibia was load tested, a 10 mm circular cortical window was created using a bone graft harvesting trephine (Bone Graft Harvesting Set, Depuy Synthes, Zuchwil, Switzerland) over the proximal lateral tibia 1.5 cm posterior to the tibial tuberosity and 2.5 cm below the tibial plateau. Bone curettes were used to harvest the cancellous bone until the curette could no longer pass through. Cancellous bone 2 cm below the tibial plateau was not harvested to preserve the stability of the tibial plateau. This was similar to techniques used by Alt et al. [12], Kushner [4], and Myeroff [3]. The uncompressed volume of bone harvested is collected and measured by the amount of displaced water in a

graduated cylinder with 10 ml of distilled water (Fig. 3). The harvested bone was then placed into a 10 ml syringe and compressed with maximal thumb pressure, and this volume measured was recorded as compressed volume. During harvesting, care was taken not to damage the strain gauges and the tibia specimen (Fig. 3). After harvesting, the specimen was loaded again as described in the previous section. After the loading, the diameter of the cortical window was widened in increments of 2.5 mm up to a maximum of 25 mm with a surgical burr (Colibri II, DepuySynthes, Zuchwil,

Fig. 1. Strain rosette and gauge assignment locations. Levels 1–3 indicate horizontal axes 10, 25, and 100 mm rom the tibial plateau respectively. Strain rosettes were placed at locations 1AL, 1L, 2L, 2P, and 1M. Uniaxial strain gauges were placed at locations 3L and 3M, oriented parallel to the vertical axis. Strain rosette 1AL is 15 mm laterally from the anterior vertical axis.

Please cite this article in press as: C.T. Lim, et al., A biomechanical study of proximal tibia bone grafting through the lateral approach, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.09.017

G Model JINJ 6903 No. of Pages 8

4

C.T. Lim et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx

Fig. 3. Harvested cancellous bone graft as measured in a graduated cylinder. The left most figure shows uncompressed volume, while the middle figure shows compressed volume. The osteotomy location is kept consistent and care is taken not to damage the strain gauges as show in the right most figure.

Switzerland), with the most proximal edge of the window kept at 2 cm from the tibial plateau. The process of widening the cortical window and re-loading the specimen was continued up to a maximum window diameter size of 25 mm. If the specimen failed before reaching 25 mm window diameter, the last window size before failing was recorded. If the specimen was able to withstand the load with a maximum cortical window size of 25 mm, more bone graft is harvested again with bone curettes and the additional volumes harvested are recorded. A final load was then prescribed to the re-harvested specimen (Fig. 4). Results Of the 12 tibiae, one was found to be severely osteoporotic (Tscore of 5.5) and excluded from the study. Of the remaining 11 tibiae, the average BMD at the tibia epiphysis was 0.693 g/cm2 with a standard deviation of 0.103 g/cm2. This corresponds to a T-score of 0.74 (Table 1). Load testing and strain measurements The mean maximum load that each tibia was subjected to was 2320N (range 1650–3120N). None of the 11 tibiae tested experienced failure during loading. Principal strain values were analyzed to observe any changes in the loading response of the tibia after bone harvesting. Maximum and minimum principal strains were calculated from the strain values recorded from the strain rosettes. A repeated measures ANOVA with the Greenhouse-Geisser correction was conducted to analyze the significance of the change in strain values across cortical window sizes (p < 0.05) (SPSS statistics software version 22, IBM systems, USA). Statistically significant changes in strain were found in minimum principal strain values at 1AL (p = 0.002) and the maximum principal strain values at M1 with increasing cortical window size (p = 0.004) (Fig. 5). Post-hoc tests using the Bonferroni correction showed that at location 1AL the minimum principal strain values dropped significantly from the intact condition to the harvested condition at 10 mm cortical hole window size. ( 328.85 me, SD = 232.21 to 964.78 me, SD = 535.89). For location 1 M, the maximum principal

Fig. 4. A tibia with an enlarged osteotomy undergoing load testing.

Please cite this article in press as: C.T. Lim, et al., A biomechanical study of proximal tibia bone grafting through the lateral approach, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.09.017

G Model JINJ 6903 No. of Pages 8

C.T. Lim et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx

5

Fig. 5. Changes in principal strains with window size locations 1AL and 1 M. A cortical window size of 0 indicates the intact condition, while 25* indicates a 25 mm cortical window with a second cancellous harvest.

strain values increased significantly from 10 mm window size to a 22.5 mm window size (361.64 me, SD = 229.90 to 486.08 me, SD = 270.40). Changes in maximum principal strain values at 1L and 2P and minimum principal strain values at 1 M were marginally significant (p = 0.074, p = 0.079 and p = 0.055 respectively). Changes in maximum principal strain at 1AL and 2L, as well as minimum principal strain values at 1L, 2P, 2L, were not found to be statistically significant and are not reported (Fig. 6). Uniaxial strain values at 3 M increased slightly, then decreased as the cortical bone window size was increased while uniaxial strain values at 3L increased (Fig. 7). These changes were not, however, statistically significant (p = 0.101 p = 0.149 respectively).

average compressed volume was 7.15 ml (range of 3–10.4 ml). The average ratio of the compressed volume to area of the tibial plateau was 0.23 ml/cm2 (range 0.112–0.297 ml/cm2) (Table 1). For cortical window size of 25 mm, an additional uncompressed volume of 7.9 ml (range 2–19.1 ml) and compressed volume of 3.64 ml (range 1.7–7.6 ml) were harvested. The average total compressed volume was 10.79 ml (range 5.6–15.2 ml) and the average ratio of compressed volume to tibial plateau area was 0.34 ml/cm2 (range 0.209–0.481 ml/cm2) (Table 1). Student’s t-test for equal variances (p < 0.05) showed that the average amount of compressed bone graft harvested from a 25 mm cortical window was significantly more than that harvested at 10 mm (p = 0.004). Discussion

Relation of bone graft volume to cortical window size The average uncompressed volume of bone graft harvested with a 10 mm cortical window was 15.8 ml (ranging from 12 to 25 ml). The

The proximal tibia has shown great promise as an alternative bone graft harvest site with low complications and comparable bone graft harvest volume [2–4,18,19]. The primary aim of this

Fig. 6. Changes with principal strains with window size at locations 1L, 2P and 1 M. A cortical window size of 0 indicates the intact condition, while 25* indicates a 25 mm cortical window with a second cancellous harvest.

Please cite this article in press as: C.T. Lim, et al., A biomechanical study of proximal tibia bone grafting through the lateral approach, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.09.017

G Model JINJ 6903 No. of Pages 8

6

C.T. Lim et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx

Fig. 7. Changes in medial and lateral uniaxial strains with a cortical window size. A cortical window size of 0 indicates the intact condition, while 25* indicates a 25 mm cortical window with a second cancellous harvest.

study was to determine if bone graft can be safely harvested from the proximal tibia using circular cortical windows, without compromising the mechanical stability of the tibia. This study employed a lateral approach with a circular osteotomy introduced at the proximal tibia to harvest the cancellous bone. A circular osteotomy window is ideal as it will avoid formation of stress riser in the proximal tibia. The circular osteotomy can be performed with a commercially available bone graft harvesting trephine. The osteotomy is performed on the lateral aspect of the proximal tibia because a smaller proportion of the body weight passes through the lateral tibial plateau. The loading conditions were chosen in order to simulate loads that a human tibia would experience during normal gait. Under these loading conditions, although there were significant changes in strains at various locations in the tibia, none of the tibiae failed. The significant decrease in minimum principal strain at the 1 M location, in conjunction with the increase in strain over the lateral strain gauge suggested that some of the load was transferred from the lateral compartment to the medial compartment of the knee. It was suspected that the unequal loads on the medial and lateral compartments, coupled with the osteotomy on the lateral side introduce bending moments onto the tibia. This finding was not reported in previous biomechanical studies [12– 15]. These studies focused mainly on the compressive strength of the tibia post-harvesting. This is a crucial finding as post-operative fracture was always assumed to be due to compression. Instead, the findings suggest that fracture could occur due to secondary tensile stress introduced from bending moments, since bone is weaker in tension than compression. There is also a possibility that patients may complain of pain on other aspects of the tibia rather than the operated side. Clinicians should have a high level index of suspicious for stress fracture if patient complains pain around medial aspect of proximal tibia even if pain is not around the operated side. However this could only be confirmed if the full strain state of the tibia is known. Changes in principal strains with cortical window size at other sites were largely inconclusive. Some increases in strain were observed but these were mostly deemed not to be statistically

significant. The magnitudes of principal bone strains were similar to those measured in vivo by Burr and colleagues [20], who measured in vivo bone strains during activities such as uphill, level, and downhill walking, running, and loaded walking. Even though bone is weaker in tension, our reported maximum principal strains observed (3063 me) were well below the tensile yield strains of cortical bone as reported by various researchers [21–24]. These results, in conjunction with the fact that none of the tibiae failed under the loading conditions prescribed indicate that the tibia is able to sustain normal walking loads even after large osteotomies have been introduced for bone grafting. Catastrophic failure of the tibia in such an instance is unlikely. However, we did note that the strains encountered are within the range of strain values reported by Taylor where fatigue failure is possible in the absence of bone remodeling [25]. The amount of bone graft harvested by this technique is largely dependent upon the cortical window size. Studies report a range of 5–11.3 ml of compressed volume bone grafts that can be harvested from the proximal tibia from the lateral approach based on differing cortical window size [1,18,19]. Many studies have deemed the volumes of bone graft harvested from the proximal tibia to be comparable to that of other commonly used sites [1–3,18]. The reported average compressed volumes of bone graft harvested at 10 mm diameter cortical window (7.15 ml) were higher than those reported by Alt et al. [12] and Vittayakittipong et al. [13] (5.39 and 6.62 ml respectively), but were similar to those reported by Mauffrey [1] (7.3 ml). On widening the osteotomy to a 25 mm diameter, we were on average able to harvest an additional 50.8% of compressed bone graft volume (3.64 ml). There is a positive correlation between cortical window size and amount of bone graft that can be harvested, indicating that more bone graft can be harvested with a larger cortical window. The amount of compressed bone graft harvested through a 10 mm (7.15 ml) and 25 mm osteotomy (10.79 ml) were also comparable to the amount of bone harvested at the anterior illiac crest as reported by Mauffrey (6.2 ml) [1] and Englested (7.0 ml) [19]. Average volumes of compressed bone graft harvested from the posterior illiac crest as reported by Engelsted (10.1 ml) were quite similar to those

Please cite this article in press as: C.T. Lim, et al., A biomechanical study of proximal tibia bone grafting through the lateral approach, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.09.017

G Model JINJ 6903 No. of Pages 8

C.T. Lim et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx

recorded in our procedure performed with a large osteotomy. Given that a similar or larger amount of bone graft from the proximal tibia was harvested when compared to the iliac crest for a lower complication rate and less invasive surgery, proximal tibia offers a superior alternative source of bone graft. It has been established that proximal tibia remains stable after harvesting even with a 25 mm circular cortical window under normal walking loads. It has also been determined the volume of bone graft harvested is comparable to reported volume of bone graft from iliac crest. However, there are clinical considerations with this method of harvesting. Clinically, it may not be feasible to introduce such a large osteotomy to the region due to the exposure limitations. Hence, it is advisable for the surgeon to determine the volume of cancellous bone graft that is needed before making the appropriate cortical window size. This study has correlated the volume of bone graft that can be harvested to the area of tibia plateau. The size of the tibia could be used as an estimate for the amount of accessible bone. The bone quality of the patient should also be considered if patient is selected for proximal tibia bone graft harvesting. The cadavers available for this experiment showed overall good bone quality. Caution should be taken in older or osteoporotic individuals where bone quality is poor. As the specimens were only tested to 3 times of body weight, these results are not able to apply to more vigorous activities such as running, jumping, or playing contact sports. Patients are advised to refrain such activities during the healing period. Based on reported biomechanical testing, Gerressen and colleagues [12] advised only partial weight bearing of one half of body weight on the operated limb in the week immediately postsurgery. The current results suggest that normal level walking is safe for patients immediately post operation. Some limitations of this study were observed. Firstly, soft tissue effects were not taken into account. This was due to the difficulty in recording accurate strain readings with the viscoelastic properties of soft tissues such as the meniscus. It is known that the meniscus acts as an important shock absorber at the knee, and hence the forces transmitted in this experiment may be slightly higher than expected from in vivo bone strain in normal walking. This would lead to actual in-vitro loads on the tibia being lower than what that experienced in the experiment. The current study showed that there were no failures in the harvested tibiae even without the meniscii. The presence of meniscii in vitro should make bone graft harvesting in this site safer. The effects of the fibula on the stability of the harvested tibia are also unknown. The fibula has been known to provide additional weight bearing stability to the tibia [26–28]. This study showed that the harvested tibia remains stable even without fibula. Different modes of loading, such as bending and torsion of the tibia after bone grafting, were not studied. Large standard deviations are observed in the recorded strain values. This may be due to the anatomical variances between specimens as well as differing bone mineral densities, which in turn affect bone strength. It is acknowledged that fatigue fracture is a possible complication that may occur over repeated loading of the tibia, such as continued walking over extended periods of time. In conclusion, bone grafting of the proximal tibia through the lateral approach with a circular osteotomy is a feasible option. This study showed that a tibia can withstand a load of up to three times of body weight even when large osteotomies of up to 25 mm diameter were introduced. Adequate volumes of bone graft were able to be harvested from the proximal tibia. These volumes are comparable to amounts harvested from the illiac crest. Osteotomy can be widened safely to a maximum of 25 mm if larger volumes of bone grafts are needed. However, the patient should adhere strictly to the physician’s restriction of activity to normal weight-bearing ambulation.

7

Conflict of interests There are no known conflicts of interest in the current manuscript in accordance with the International Committee of Medical Journal Editors Guidelines. Ethical review committee statement This manuscript used human material but was exempted from Ethical Review in accordance with the Institution’s Ethical Review Board. The work in this manuscript was done at the National University of Singapore, Yong Loo Lin School of Medicine, Orthopaedic Biomechanics Research Laboratory. Acknowledgements This study was funded by the National Medical Research Council, Singapore, grant number NMRC/CNIG/1097/2012. The authors would also like to thank Nicole Soh, Alex Choh, Hazlan bin Sanusi, Adib Rizhuan, Saw Jhen Jin, Gordon Ko, and Yang Hua for their valuable contributions to this project. References [1] Mauffrey C, Madsenp M, Bowles RJ, Seligson D. Bone graft harvest site options in orthopaedic trauma: a prospective in vivo quantification study. Injury 2012;43(3):323–6. [2] Zouhary KJ. Bone graft harvesting from distant sites: concepts and techniques. Oral Maxillofac Surg Clin North Am 2010;22(3):301–16. [3] Myeroff C, Archdeacon M. Autogenous bone graft: donor sites and techniques. J Bone Joint Surg Am 2011;93(23):2227–36. [4] Kushner GM. Tibia bone graft harvest technique. Atlas Oral Maxillofac Surg Clin North Am 2005;13(2):119–26. [5] Besly W, Ward Booth P. Technique for harvesting tibial cancellous bone modified for use in children. Br J Oral Maxillofac Surg 1999;37(2):129–33. [6] Catone GA, Reimer BL, McNeir D, Ray R. Tibial autogenous cancellous bone as an alternative donor site in maxillofacial surgery. J Oral Maxillofac Surg 1992;50(12):1258–63. [7] Herford AS, King BJ, Audia F, Becktor J. Medial approach for tibial bone graft: anatomic study and clinical technique. J Oral Maxillofac Surg 2003;61(3):358– 63. [8] Mauffrey CPC, Seligson D, Krikler S. Surgical techniques: how I do it? Bone graft harvest from the proximal lateral tibia. Injury 2010;41(2):242–4. [9] Jakse N, Seibert FJ, Lorenzoni M, Eskici A, Pertl C. A modified technique of harvesting tibial cancellous bone and its use for sinus grafting. Clin Oral Implants Res 2001;12(5):488–94. [10] O’Keefe Jr RM, Riemer BL, Butterfield SL. Harvesting of autogenous cancellous bone graft from the proximal tibial metaphysis: a review of 230 cases. J Orthop Trauma 1991;5(4):469–74. [11] Kim IK, Cho HY, Pae SP, Jung BS, Cho HW, Seo JH. Tibial bone fractures occurring after medioproximal tibial bone grafts for oral and maxillofacial reconstruction. J Korean Assoc Oral Maxillofac Surg 2013;39(6):257–62. [12] Alt V, Meeder PJ, Seligson D, Schad A, Atienza CJ. The proximal tibia metaphysis: a reliable donor site for bone grafting? Clin Orthop Relat Res 2003;414:315–21. [13] Vittayakittipong P, Nurit W, Kirirat P. Proximal tibial bone graft: the volume of cancellous bone, and strength of decancellated tibias by the medial approach. Int J Oral Maxillofac Surg 2012;41(4):531–6. [14] Gerressen M, Riediger D, Saxe J. Stability behavior of human tibias after bone removal—comparative examination in 15 cadaver tibia pairs. J Oral Maxillofac Surg 2010;68(1):60–7. [15] Bottlang M. Effect of tibial graft harvest on plateau compliance: a biomechanical study. J Oral Maxillofac Surg 2008;66(8):52. [16] Popp AW, Senn C, Franta O, Krieg MA, Perrelet R, Lippuner K. Tibial or hip BMD predict clinical fracture risk equally well: results from a prospective study in 700 elderly Swiss women. Osteoporos Int 2009;20(8):1393–9. [17] Ruff CB, Hayes WC. Cross sectional geometry of Pecos Pueblo femora and tibiae: a biomechanical investigation: method and general patterns of variation. Am J Phys Anthropol 1983;60(3):359–81. [18] Alt V, Nawab A, Seligson D. Bone grafting from the proximal tibia. J Trauma 1999;47(3):555–7. [19] Englested ME, Morse T. Anterior iliac crest, posterior iliac crest, and proximal tibia donorsites: a comparison of cancellous bone volumes in fresh cadavers. J Oral Maxillofac Surg 2010;68(12):3015–21. [20] Burr DB, Milgrom C, Fyhrie D, Forwood M, Nyska M, Finestone A, et al. In vivo measurement of human tibial strains during vigorous activity. Bone 1996;18 (5):405–10.

Please cite this article in press as: C.T. Lim, et al., A biomechanical study of proximal tibia bone grafting through the lateral approach, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.09.017

G Model JINJ 6903 No. of Pages 8

8

C.T. Lim et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx

[21] Evans FG. Mechanical properties and histology of cortical bone from younger and older men. Anat Rec 1976;185(1):1–11. [22] Carter D, Spengler D. Mechanical properties and composition of cortical bone. Clin Orthop Relat Res 1978;135:192–217. [23] Morgan EF, Keaveny TM. Dependence of yield strain of human trabecular bone on anatomic site. J Biomech 2001;34(5):569–77. [24] H.Bayraktar H, F.Morgan E, L.Niebur G, E.Morris G, K.Wong E, M.Keaveny T. Comparison of the elastic and yield properties of human femoral trabecular and cortical bone tissue. J Biomech 2004;37(1):27–35.

[25] Taylor D. Fatigue of bone and bones: an analysis based on stressed volume. J Orthop Res 1998;16(2):163–9. [26] Lambert KL. The weight bearing function of fibula. A strain gauge study. J Bone Joint Surg Am 1971;53(3):507–13. [27] Takebe K, Nakagawa A, Minami H, Kanazawa H, Hirohata K. Role of the Fibula in weight bearing. Clin Orthop Relat Res 1984;184:289–92. [28] Weinert Jr CR, Mcmaster JH, Ferguson RJ. Dynamic function of the human fibula. Am J Anat 1973;138(2):145–9.

Please cite this article in press as: C.T. Lim, et al., A biomechanical study of proximal tibia bone grafting through the lateral approach, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.09.017